lion-ai/arctic-l-medembed

SentenceTransformer based on Snowflake/snowflake-arctic-embed-l

This is a sentence-transformers model finetuned from Snowflake/snowflake-arctic-embed-l on the med_embed-training-triplets-v1 dataset. It maps sentences & paragraphs to a 1024-dimensional dense vector space and can be used for semantic textual similarity, semantic search, paraphrase mining, text classification, clustering, and more.

Model Details

Model Description

• Model Type: Sentence Transformer
• Base model: Snowflake/snowflake-arctic-embed-l
• Maximum Sequence Length: 512 tokens
• Output Dimensionality: 1024 dimensions
• Similarity Function: Cosine Similarity
• Training Dataset:
  • med_embed-training-triplets-v1

Model Sources

• Documentation: Sentence Transformers Documentation
• Repository: Sentence Transformers on GitHub
• Hugging Face: Sentence Transformers on Hugging Face

Full Model Architecture

SentenceTransformer(
  (0): Transformer({'max_seq_length': 512, 'do_lower_case': False, 'architecture': 'BertModel'})
  (1): Pooling({'word_embedding_dimension': 1024, 'pooling_mode_cls_token': True, 'pooling_mode_mean_tokens': False, 'pooling_mode_max_tokens': False, 'pooling_mode_mean_sqrt_len_tokens': False, 'pooling_mode_weightedmean_tokens': False, 'pooling_mode_lasttoken': False, 'include_prompt': True})
  (2): Normalize()
)

Usage

Direct Usage (Sentence Transformers)

First install the Sentence Transformers library:

pip install -U sentence-transformers

Then you can load this model and run inference.

from sentence_transformers import SentenceTransformer

# Download from the 🤗 Hub
model = SentenceTransformer("lion-ai/arctic-l-medembed")
# Run inference
queries = [
    "moderate level of depression diagnosis",
]
documents = [
    "The patient's symptoms are suggestive of depression. The Beck Depression Inventory score was 24, indicating a moderate level of depression.",
    'The patient was admitted due to symptoms of Major Depressive Disorder (MDD) which included restlessness, sleep problems, depressed mood, decreased social function, and suicidal thoughts as reported by the patient.',
    'The right upper limb was examined and revealed a morphological variant of the coracobrachialis muscle (CBM) that corresponded to the classical description, originating from the apex of the coracoid process together with the short head of the biceps brachii.',
]
query_embeddings = model.encode_query(queries)
document_embeddings = model.encode_document(documents)
print(query_embeddings.shape, document_embeddings.shape)
# [1, 1024] [3, 1024]

# Get the similarity scores for the embeddings
similarities = model.similarity(query_embeddings, document_embeddings)
print(similarities)
# tensor([[0.5307, 0.3000, 0.0426]])

Evaluation

Metrics

Triplet

• Dataset: medembed-triplets-dev-300
• Evaluated with TripletEvaluator

Metric	Value
cosine_accuracy	0.96

Training Details

Training Dataset

med_embed-training-triplets-v1

• Dataset: med_embed-training-triplets-v1 at 0b344f0
• Size: 230,357 training samples
• Columns: anchor, positive, and negative
• Approximate statistics based on the first 1000 samples:

  	anchor	positive	negative
  type	string	string	string
  details	min: 5 tokens mean: 11.84 tokens max: 28 tokens	min: 5 tokens mean: 39.33 tokens max: 216 tokens	min: 7 tokens mean: 40.03 tokens max: 171 tokens

• Samples:

  anchor	positive	negative
  MTHFR homozygous A223V mutation symptoms	The patient was admitted to our institution where MTHFR homozygous A223V mutation was identified. Folic acid intake was increased to 800 mcg/d, and no other coagulation tests were abnormal.	The patient had several symptoms of MM, including hypercalcemia, bone fractures, anemia, and renal insufficiency. A biopsy showed atypical clonal plasma cells with Cluster of Differentiation (CD)138 positive infiltration.
  Causes of spindle cell malignancy in the duodenal wall	Histological analysis revealed a spindle cell malignancy that was positive for CD21, CD23, and vimentin, but negative for CD20, CD34, CD35, CD117, DOG 1, and smooth muscle actin.	Based on immunohistochemical analysis of the tumor cells, the primary buttock tumor was diagnosed as a skeletal muscle metastasis of the primary small intestine gastrointestinal stromal tumor (GIST).
  What was the patient's main complaint during hospital admission?	This 27-year-old pregnant woman was admitted to the hospital at 36 weeks gestation with acute vision loss in her left eye and severe onset headache.	The patient was discharged the next day

• Loss: MultipleNegativesRankingLoss with these parameters:

  {
      "scale": 20.0,
      "similarity_fct": "cos_sim",
      "gather_across_devices": false
  }
  

Evaluation Dataset

med_embed-training-triplets-v1

• Dataset: med_embed-training-triplets-v1 at 0b344f0
• Size: 300 evaluation samples
• Columns: anchor, positive, and negative
• Approximate statistics based on the first 300 samples:

  	anchor	positive	negative
  type	string	string	string
  details	min: 4 tokens mean: 11.88 tokens max: 32 tokens	min: 7 tokens mean: 38.67 tokens max: 127 tokens	min: 8 tokens mean: 40.55 tokens max: 108 tokens

• Samples:

  anchor	positive	negative
  What was the initial presentation of the patient?	The 45-year-old female patient presented to the department with an enlarging lesion in her upper abdomen.	The patient was transferred to this hospital for further evaluation.
  giant omphalocele symptoms	The patient, a 9-year-old female, presented to the hospital with a large lump in the anterior abdominal wall extending from the xiphisternum to the level of iliac crest.	The patient presented with bilateral nasovestibular lumps which grew in size over several months, occluding nasal entrance and protruding outside the nose.
  granulomatous lymphocytic interstitial lung disease treatment	The patient had clubbing and chronic lung findings, and thorax CT revealed extended and severe bronchiectasis with thickened bronchial walls, some granulomatous nodules and mosaic appearance, compatible with granulomatous lymphocytic interstitial lung disease (GLILD). Regular intravenous immunoglobulin (IVIG) replacement was started.	The patient was treated with methylprednisolone pulse therapy followed by oral prednisolone (PSL) and cyclophosphamide intravenously. After treatment, arthralgia, renal function, proteinuria, and skin manifestations improved.

• Loss: MultipleNegativesRankingLoss with these parameters:

  {
      "scale": 20.0,
      "similarity_fct": "cos_sim",
      "gather_across_devices": false
  }
  

Training Hyperparameters

Non-Default Hyperparameters

• eval_strategy: steps
• per_device_train_batch_size: 16
• per_device_eval_batch_size: 16
• gradient_accumulation_steps: 8
• learning_rate: 1e-05
• weight_decay: 0.01
• num_train_epochs: 2
• warmup_ratio: 0.1
• dataloader_num_workers: 4
• load_best_model_at_end: True
• ddp_find_unused_parameters: False
• batch_sampler: no_duplicates

All Hyperparameters

Click to expand

• overwrite_output_dir: False
• do_predict: False
• eval_strategy: steps
• prediction_loss_only: True
• per_device_train_batch_size: 16
• per_device_eval_batch_size: 16
• per_gpu_train_batch_size: None
• per_gpu_eval_batch_size: None
• gradient_accumulation_steps: 8
• eval_accumulation_steps: None
• torch_empty_cache_steps: None
• learning_rate: 1e-05
• weight_decay: 0.01
• adam_beta1: 0.9
• adam_beta2: 0.999
• adam_epsilon: 1e-08
• max_grad_norm: 1.0
• num_train_epochs: 2
• max_steps: -1
• lr_scheduler_type: linear
• lr_scheduler_kwargs: {}
• warmup_ratio: 0.1
• warmup_steps: 0
• log_level: passive
• log_level_replica: warning
• log_on_each_node: True
• logging_nan_inf_filter: True
• save_safetensors: True
• save_on_each_node: False
• save_only_model: False
• restore_callback_states_from_checkpoint: False
• no_cuda: False
• use_cpu: False
• use_mps_device: False
• seed: 42
• data_seed: None
• jit_mode_eval: False
• bf16: False
• fp16: False
• fp16_opt_level: O1
• half_precision_backend: auto
• bf16_full_eval: False
• fp16_full_eval: False
• tf32: None
• local_rank: 0
• ddp_backend: None
• tpu_num_cores: None
• tpu_metrics_debug: False
• debug: []
• dataloader_drop_last: False
• dataloader_num_workers: 4
• dataloader_prefetch_factor: None
• past_index: -1
• disable_tqdm: False
• remove_unused_columns: True
• label_names: None
• load_best_model_at_end: True
• ignore_data_skip: False
• fsdp: []
• fsdp_min_num_params: 0
• fsdp_config: {'min_num_params': 0, 'xla': False, 'xla_fsdp_v2': False, 'xla_fsdp_grad_ckpt': False}
• fsdp_transformer_layer_cls_to_wrap: None
• accelerator_config: {'split_batches': False, 'dispatch_batches': None, 'even_batches': True, 'use_seedable_sampler': True, 'non_blocking': False, 'gradient_accumulation_kwargs': None}
• parallelism_config: None
• deepspeed: None
• label_smoothing_factor: 0.0
• optim: adamw_torch_fused
• optim_args: None
• adafactor: False
• group_by_length: False
• length_column_name: length
• project: huggingface
• trackio_space_id: trackio
• ddp_find_unused_parameters: False
• ddp_bucket_cap_mb: None
• ddp_broadcast_buffers: False
• dataloader_pin_memory: True
• dataloader_persistent_workers: False
• skip_memory_metrics: True
• use_legacy_prediction_loop: False
• push_to_hub: False
• resume_from_checkpoint: None
• hub_model_id: None
• hub_strategy: every_save
• hub_private_repo: None
• hub_always_push: False
• hub_revision: None
• gradient_checkpointing: False
• gradient_checkpointing_kwargs: None
• include_inputs_for_metrics: False
• include_for_metrics: []
• eval_do_concat_batches: True
• fp16_backend: auto
• push_to_hub_model_id: None
• push_to_hub_organization: None
• mp_parameters:
• auto_find_batch_size: False
• full_determinism: False
• torchdynamo: None
• ray_scope: last
• ddp_timeout: 1800
• torch_compile: False
• torch_compile_backend: None
• torch_compile_mode: None
• include_tokens_per_second: False
• include_num_input_tokens_seen: no
• neftune_noise_alpha: None
• optim_target_modules: None
• batch_eval_metrics: False
• eval_on_start: False
• use_liger_kernel: False
• liger_kernel_config: None
• eval_use_gather_object: False
• average_tokens_across_devices: True
• prompts: None
• batch_sampler: no_duplicates
• multi_dataset_batch_sampler: proportional
• router_mapping: {}
• learning_rate_mapping: {}

Training Logs

Click to expand

Epoch	Step	Training Loss	Validation Loss	medembed-triplets-dev-300_cosine_accuracy
-1	-1	-	-	0.5767
0.0056	10	2.6703	-	-
0.0111	20	2.5114	-	-
0.0167	30	2.3725	-	-
0.0222	40	2.2162	-	-
0.0278	50	2.0325	-	-
0.0333	60	1.8674	-	-
0.0389	70	1.7283	-	-
0.0445	80	1.5936	-	-
0.0500	90	1.5153	-	-
0.0556	100	1.421	1.3269	0.6367
0.0611	110	1.3861	-	-
0.0667	120	1.2706	-	-
0.0722	130	1.286	-	-
0.0778	140	1.2037	-	-
0.0833	150	1.1301	-	-
0.0889	160	1.1587	-	-
0.0945	170	1.1741	-	-
0.1000	180	1.0915	-	-
0.1056	190	1.0786	-	-
0.1111	200	1.0518	1.0083	0.7767
0.1167	210	1.0562	-	-
0.1222	220	1.042	-	-
0.1278	230	0.9928	-	-
0.1334	240	1.022	-	-
0.1389	250	1.044	-	-
0.1445	260	0.9621	-	-
0.1500	270	0.98	-	-
0.1556	280	1.0073	-	-
0.1611	290	0.9346	-	-
0.1667	300	0.995	0.8794	0.8100
0.1722	310	0.9057	-	-
0.1778	320	1.0023	-	-
0.1834	330	0.8486	-	-
0.1889	340	0.8888	-	-
0.1945	350	0.9358	-	-
0.2000	360	0.8503	-	-
0.2056	370	0.9181	-	-
0.2111	380	0.8965	-	-
0.2167	390	0.8807	-	-
0.2223	400	0.8833	0.8012	0.8267
0.2278	410	0.8564	-	-
0.2334	420	0.838	-	-
0.2389	430	0.8775	-	-
0.2445	440	0.7808	-	-
0.2500	450	0.7795	-	-
0.2556	460	0.8087	-	-
0.2611	470	0.7418	-	-
0.2667	480	0.7921	-	-
0.2723	490	0.7526	-	-
0.2778	500	0.6838	0.7306	0.8567
0.2834	510	0.7908	-	-
0.2889	520	0.8013	-	-
0.2945	530	0.7176	-	-
0.3000	540	0.7432	-	-
0.3056	550	0.7148	-	-
0.3112	560	0.7034	-	-
0.3167	570	0.7405	-	-
0.3223	580	0.7185	-	-
0.3278	590	0.6609	-	-
0.3334	600	0.6246	0.6693	0.8767
0.3389	610	0.6743	-	-
0.3445	620	0.6826	-	-
0.3500	630	0.7143	-	-
0.3556	640	0.706	-	-
0.3612	650	0.6787	-	-
0.3667	660	0.6784	-	-
0.3723	670	0.6938	-	-
0.3778	680	0.6647	-	-
0.3834	690	0.6732	-	-
0.3889	700	0.6409	0.6291	0.9033
0.3945	710	0.6773	-	-
0.4001	720	0.6767	-	-
0.4056	730	0.6656	-	-
0.4112	740	0.6423	-	-
0.4167	750	0.6341	-	-
0.4223	760	0.6369	-	-
0.4278	770	0.5883	-	-
0.4334	780	0.6497	-	-
0.4389	790	0.6106	-	-
0.4445	800	0.6423	0.5550	0.9067
0.4501	810	0.6054	-	-
0.4556	820	0.6238	-	-
0.4612	830	0.5986	-	-
0.4667	840	0.6196	-	-
0.4723	850	0.5767	-	-
0.4778	860	0.6286	-	-
0.4834	870	0.587	-	-
0.4890	880	0.5661	-	-
0.4945	890	0.51	-	-
0.5001	900	0.6048	0.5459	0.9167
0.5056	910	0.5621	-	-
0.5112	920	0.5954	-	-
0.5167	930	0.5542	-	-
0.5223	940	0.5846	-	-
0.5279	950	0.5864	-	-
0.5334	960	0.6152	-	-
0.5390	970	0.5696	-	-
0.5445	980	0.5639	-	-
0.5501	990	0.555	-	-
0.5556	1000	0.5465	0.4860	0.9200
0.5612	1010	0.5618	-	-
0.5667	1020	0.5228	-	-
0.5723	1030	0.5117	-	-
0.5779	1040	0.5305	-	-
0.5834	1050	0.5048	-	-
0.5890	1060	0.5261	-	-
0.5945	1070	0.4963	-	-
0.6001	1080	0.5034	-	-
0.6056	1090	0.4615	-	-
0.6112	1100	0.5192	0.4642	0.9267
0.6168	1110	0.5226	-	-
0.6223	1120	0.5417	-	-
0.6279	1130	0.5214	-	-
0.6334	1140	0.5256	-	-
0.6390	1150	0.5327	-	-
0.6445	1160	0.5413	-	-
0.6501	1170	0.5261	-	-
0.6556	1180	0.4876	-	-
0.6612	1190	0.4694	-	-
0.6668	1200	0.4795	0.4229	0.9500
0.6723	1210	0.4896	-	-
0.6779	1220	0.4796	-	-
0.6834	1230	0.4732	-	-
0.6890	1240	0.486	-	-
0.6945	1250	0.4942	-	-
0.7001	1260	0.4403	-	-
0.7057	1270	0.4581	-	-
0.7112	1280	0.4494	-	-
0.7168	1290	0.4676	-	-
0.7223	1300	0.4575	0.4201	0.9500
0.7279	1310	0.4395	-	-
0.7334	1320	0.4197	-	-
0.7390	1330	0.4471	-	-
0.7445	1340	0.4724	-	-
0.7501	1350	0.43	-	-
0.7557	1360	0.4691	-	-
0.7612	1370	0.425	-	-
0.7668	1380	0.4244	-	-
0.7723	1390	0.417	-	-
0.7779	1400	0.4398	0.3899	0.9467
0.7834	1410	0.452	-	-
0.7890	1420	0.476	-	-
0.7946	1430	0.4433	-	-
0.8001	1440	0.4021	-	-
0.8057	1450	0.461	-	-
0.8112	1460	0.4191	-	-
0.8168	1470	0.4068	-	-
0.8223	1480	0.4274	-	-
0.8279	1490	0.4238	-	-
0.8334	1500	0.4403	0.3585	0.9433
0.8390	1510	0.3875	-	-
0.8446	1520	0.4004	-	-
0.8501	1530	0.4194	-	-
0.8557	1540	0.3883	-	-
0.8612	1550	0.3861	-	-
0.8668	1560	0.4131	-	-
0.8723	1570	0.4006	-	-
0.8779	1580	0.4195	-	-
0.8835	1590	0.4144	-	-
0.8890	1600	0.4454	0.3131	0.9533
0.8946	1610	0.3751	-	-
0.9001	1620	0.3655	-	-
0.9057	1630	0.3875	-	-
0.9112	1640	0.3922	-	-
0.9168	1650	0.4043	-	-
0.9224	1660	0.3686	-	-
0.9279	1670	0.3745	-	-
0.9335	1680	0.3681	-	-
0.9390	1690	0.3858	-	-
0.9446	1700	0.3817	0.3126	0.9533
0.9501	1710	0.3787	-	-
0.9557	1720	0.3849	-	-
0.9612	1730	0.352	-	-
0.9668	1740	0.3549	-	-
0.9724	1750	0.3634	-	-
0.9779	1760	0.3548	-	-
0.9835	1770	0.3662	-	-
0.9890	1780	0.3915	-	-
0.9946	1790	0.3713	-	-
1.0	1800	0.3647	0.3077	0.9533
1.0056	1810	0.3365	-	-
1.0111	1820	0.3084	-	-
1.0167	1830	0.3261	-	-
1.0222	1840	0.3036	-	-
1.0278	1850	0.3054	-	-
1.0333	1860	0.3384	-	-
1.0389	1870	0.3129	-	-
1.0445	1880	0.3092	-	-
1.0500	1890	0.3043	-	-
1.0556	1900	0.2904	0.3005	0.9533
1.0611	1910	0.2917	-	-
1.0667	1920	0.3196	-	-
1.0722	1930	0.3401	-	-
1.0778	1940	0.312	-	-
1.0833	1950	0.321	-	-
1.0889	1960	0.3052	-	-
1.0945	1970	0.3136	-	-
1.1000	1980	0.3243	-	-
1.1056	1990	0.2843	-	-
1.1111	2000	0.2994	0.2905	0.9533
1.1167	2010	0.3144	-	-
1.1222	2020	0.3029	-	-
1.1278	2030	0.3085	-	-
1.1334	2040	0.3053	-	-
1.1389	2050	0.3009	-	-
1.1445	2060	0.3083	-	-
1.1500	2070	0.2993	-	-
1.1556	2080	0.2932	-	-
1.1611	2090	0.2994	-	-
1.1667	2100	0.2796	0.2704	0.9467
1.1722	2110	0.3033	-	-
1.1778	2120	0.3083	-	-
1.1834	2130	0.3118	-	-
1.1889	2140	0.2978	-	-
1.1945	2150	0.2697	-	-
1.2000	2160	0.2918	-	-
1.2056	2170	0.2814	-	-
1.2111	2180	0.2615	-	-
1.2167	2190	0.289	-	-
1.2223	2200	0.2819	0.2454	0.9533
1.2278	2210	0.2751	-	-
1.2334	2220	0.3228	-	-
1.2389	2230	0.2852	-	-
1.2445	2240	0.2909	-	-
1.2500	2250	0.285	-	-
1.2556	2260	0.2936	-	-
1.2611	2270	0.3052	-	-
1.2667	2280	0.2874	-	-
1.2723	2290	0.3008	-	-
1.2778	2300	0.3161	0.2529	0.9567
1.2834	2310	0.2764	-	-
1.2889	2320	0.2779	-	-
1.2945	2330	0.2547	-	-
1.3000	2340	0.2645	-	-
1.3056	2350	0.3015	-	-
1.3112	2360	0.2835	-	-
1.3167	2370	0.279	-	-
1.3223	2380	0.3158	-	-
1.3278	2390	0.2933	-	-
1.3334	2400	0.2791	0.2412	0.9567
1.3389	2410	0.2783	-	-
1.3445	2420	0.2533	-	-
1.3500	2430	0.2943	-	-
1.3556	2440	0.2808	-	-
1.3612	2450	0.2828	-	-
1.3667	2460	0.2789	-	-
1.3723	2470	0.2814	-	-
1.3778	2480	0.2675	-	-
1.3834	2490	0.2746	-	-
1.3889	2500	0.2907	0.2419	0.9567
1.3945	2510	0.278	-	-
1.4001	2520	0.2594	-	-
1.4056	2530	0.2659	-	-
1.4112	2540	0.2859	-	-
1.4167	2550	0.2755	-	-
1.4223	2560	0.2834	-	-
1.4278	2570	0.26	-	-
1.4334	2580	0.2693	-	-
1.4389	2590	0.2678	-	-
1.4445	2600	0.2624	0.2236	0.96
1.4501	2610	0.257	-	-
1.4556	2620	0.2876	-	-
1.4612	2630	0.2263	-	-
1.4667	2640	0.2508	-	-
1.4723	2650	0.2446	-	-
1.4778	2660	0.2532	-	-
1.4834	2670	0.2383	-	-
1.4890	2680	0.2441	-	-
1.4945	2690	0.241	-	-
1.5001	2700	0.2594	0.2242	0.9567
1.5056	2710	0.2463	-	-
1.5112	2720	0.2797	-	-
1.5167	2730	0.2419	-	-
1.5223	2740	0.2933	-	-
1.5279	2750	0.2724	-	-
1.5334	2760	0.2434	-	-
1.5390	2770	0.2635	-	-
1.5445	2780	0.2317	-	-
1.5501	2790	0.2698	-	-
1.5556	2800	0.2645	0.2309	0.9567
1.5612	2810	0.2585	-	-
1.5667	2820	0.2456	-	-
1.5723	2830	0.2211	-	-
1.5779	2840	0.2547	-	-
1.5834	2850	0.2673	-	-
1.5890	2860	0.2569	-	-
1.5945	2870	0.2477	-	-
1.6001	2880	0.2277	-	-
1.6056	2890	0.223	-	-
1.6112	2900	0.2415	0.2254	0.9600
1.6168	2910	0.2412	-	-
1.6223	2920	0.2564	-	-
1.6279	2930	0.2735	-	-
1.6334	2940	0.2589	-	-
1.6390	2950	0.2678	-	-
1.6445	2960	0.2549	-	-
1.6501	2970	0.227	-	-
1.6556	2980	0.2598	-	-
1.6612	2990	0.2593	-	-
1.6668	3000	0.2248	0.2247	0.9533
1.6723	3010	0.2432	-	-
1.6779	3020	0.2469	-	-
1.6834	3030	0.2394	-	-
1.6890	3040	0.2525	-	-
1.6945	3050	0.2436	-	-
1.7001	3060	0.2247	-	-
1.7057	3070	0.2564	-	-
1.7112	3080	0.2267	-	-
1.7168	3090	0.2667	-	-
1.7223	3100	0.2365	0.2198	0.9600
1.7279	3110	0.2354	-	-
1.7334	3120	0.2396	-	-
1.7390	3130	0.2357	-	-
1.7445	3140	0.2397	-	-
1.7501	3150	0.2371	-	-
1.7557	3160	0.2508	-	-
1.7612	3170	0.2247	-	-
1.7668	3180	0.2561	-	-
1.7723	3190	0.2309	-	-
1.7779	3200	0.2275	0.2126	0.9567
1.7834	3210	0.2234	-	-
1.7890	3220	0.2332	-	-
1.7946	3230	0.2611	-	-
1.8001	3240	0.2227	-	-
1.8057	3250	0.2582	-	-
1.8112	3260	0.2574	-	-
1.8168	3270	0.2144	-	-
1.8223	3280	0.2353	-	-
1.8279	3290	0.241	-	-
1.8334	3300	0.2406	0.2087	0.9567
1.8390	3310	0.2536	-	-
1.8446	3320	0.2579	-	-
1.8501	3330	0.2273	-	-
1.8557	3340	0.2662	-	-
1.8612	3350	0.2356	-	-
1.8668	3360	0.2177	-	-
1.8723	3370	0.2268	-	-
1.8779	3380	0.2337	-	-
1.8835	3390	0.2686	-	-
1.8890	3400	0.2239	0.2087	0.9567
1.8946	3410	0.2559	-	-
1.9001	3420	0.2135	-	-
1.9057	3430	0.2456	-	-
1.9112	3440	0.2461	-	-
1.9168	3450	0.2349	-	-
1.9224	3460	0.2227	-	-
1.9279	3470	0.2443	-	-
1.9335	3480	0.2425	-	-
1.9390	3490	0.2427	-	-
1.9446	3500	0.2469	0.2081	0.9600
1.9501	3510	0.2504	-	-
1.9557	3520	0.2518	-	-
1.9612	3530	0.2241	-	-
1.9668	3540	0.241	-	-
1.9724	3550	0.2204	-	-
1.9779	3560	0.2558	-	-
1.9835	3570	0.2109	-	-
1.9890	3580	0.2067	-	-
1.9946	3590	0.2556	-	-
2.0	3600	0.2603	0.2081	0.9600

• The bold row denotes the saved checkpoint.

Framework Versions

• Python: 3.12.7
• Sentence Transformers: 5.1.1
• Transformers: 4.57.1
• PyTorch: 2.9.0+cu128
• Accelerate: 1.10.1
• Datasets: 3.6.0
• Tokenizers: 0.22.1

Citation

BibTeX

Sentence Transformers

@inproceedings{reimers-2019-sentence-bert,
    title = "Sentence-BERT: Sentence Embeddings using Siamese BERT-Networks",
    author = "Reimers, Nils and Gurevych, Iryna",
    booktitle = "Proceedings of the 2019 Conference on Empirical Methods in Natural Language Processing",
    month = "11",
    year = "2019",
    publisher = "Association for Computational Linguistics",
    url = "https://arxiv.org/abs/1908.10084",
}

MultipleNegativesRankingLoss

@misc{henderson2017efficient,
    title={Efficient Natural Language Response Suggestion for Smart Reply},
    author={Matthew Henderson and Rami Al-Rfou and Brian Strope and Yun-hsuan Sung and Laszlo Lukacs and Ruiqi Guo and Sanjiv Kumar and Balint Miklos and Ray Kurzweil},
    year={2017},
    eprint={1705.00652},
    archivePrefix={arXiv},
    primaryClass={cs.CL}
}