Upload folder using huggingface_hub
Browse files- 1_Pooling/config.json +1 -1
- README.md +119 -95
- config.json +12 -10
- model.safetensors +2 -2
- optimizer.pt +2 -2
- rng_state.pth +1 -1
- scheduler.pt +1 -1
- special_tokens_map.json +1 -1
- tokenizer.json +2 -2
- tokenizer_config.json +1 -2
- trainer_state.json +771 -295
- training_args.bin +1 -1
1_Pooling/config.json
CHANGED
|
@@ -1,5 +1,5 @@
|
|
| 1 |
{
|
| 2 |
-
"word_embedding_dimension":
|
| 3 |
"pooling_mode_cls_token": false,
|
| 4 |
"pooling_mode_mean_tokens": true,
|
| 5 |
"pooling_mode_max_tokens": false,
|
|
|
|
| 1 |
{
|
| 2 |
+
"word_embedding_dimension": 768,
|
| 3 |
"pooling_mode_cls_token": false,
|
| 4 |
"pooling_mode_mean_tokens": true,
|
| 5 |
"pooling_mode_max_tokens": false,
|
README.md
CHANGED
|
@@ -6,7 +6,7 @@ tags:
|
|
| 6 |
- generated_from_trainer
|
| 7 |
- dataset_size:276686
|
| 8 |
- loss:MultipleNegativesRankingLoss
|
| 9 |
-
base_model: intfloat/multilingual-e5-
|
| 10 |
widget:
|
| 11 |
- source_sentence: 'query: Печенеги отступали. Они могли запросто убить оставшегося
|
| 12 |
позади князя Владимира, мальчишку и старика, но получили приказ - уходить. Куря
|
|
@@ -2907,7 +2907,7 @@ metrics:
|
|
| 2907 |
- cosine_ap
|
| 2908 |
- cosine_mcc
|
| 2909 |
model-index:
|
| 2910 |
-
- name: SentenceTransformer based on intfloat/multilingual-e5-
|
| 2911 |
results:
|
| 2912 |
- task:
|
| 2913 |
type: binary-classification
|
|
@@ -2917,42 +2917,42 @@ model-index:
|
|
| 2917 |
type: unknown
|
| 2918 |
metrics:
|
| 2919 |
- type: cosine_accuracy
|
| 2920 |
-
value: 0.
|
| 2921 |
name: Cosine Accuracy
|
| 2922 |
- type: cosine_accuracy_threshold
|
| 2923 |
-
value: 0.
|
| 2924 |
name: Cosine Accuracy Threshold
|
| 2925 |
- type: cosine_f1
|
| 2926 |
-
value: 0.
|
| 2927 |
name: Cosine F1
|
| 2928 |
- type: cosine_f1_threshold
|
| 2929 |
-
value: 0.
|
| 2930 |
name: Cosine F1 Threshold
|
| 2931 |
- type: cosine_precision
|
| 2932 |
-
value: 0.
|
| 2933 |
name: Cosine Precision
|
| 2934 |
- type: cosine_recall
|
| 2935 |
-
value: 0.
|
| 2936 |
name: Cosine Recall
|
| 2937 |
- type: cosine_ap
|
| 2938 |
-
value: 0.
|
| 2939 |
name: Cosine Ap
|
| 2940 |
- type: cosine_mcc
|
| 2941 |
-
value: 0.
|
| 2942 |
name: Cosine Mcc
|
| 2943 |
---
|
| 2944 |
|
| 2945 |
-
# SentenceTransformer based on intfloat/multilingual-e5-
|
| 2946 |
|
| 2947 |
-
This is a [sentence-transformers](https://www.SBERT.net) model finetuned from [intfloat/multilingual-e5-
|
| 2948 |
|
| 2949 |
## Model Details
|
| 2950 |
|
| 2951 |
### Model Description
|
| 2952 |
- **Model Type:** Sentence Transformer
|
| 2953 |
-
- **Base model:** [intfloat/multilingual-e5-
|
| 2954 |
- **Maximum Sequence Length:** 512 tokens
|
| 2955 |
-
- **Output Dimensionality:**
|
| 2956 |
- **Similarity Function:** Cosine Similarity
|
| 2957 |
- **Training Dataset:**
|
| 2958 |
- json
|
|
@@ -2969,8 +2969,8 @@ This is a [sentence-transformers](https://www.SBERT.net) model finetuned from [i
|
|
| 2969 |
|
| 2970 |
```
|
| 2971 |
SentenceTransformer(
|
| 2972 |
-
(0): Transformer({'max_seq_length': 512, 'do_lower_case': False}) with Transformer model:
|
| 2973 |
-
(1): Pooling({'word_embedding_dimension':
|
| 2974 |
(2): Normalize()
|
| 2975 |
)
|
| 2976 |
```
|
|
@@ -2999,7 +2999,7 @@ sentences = [
|
|
| 2999 |
]
|
| 3000 |
embeddings = model.encode(sentences)
|
| 3001 |
print(embeddings.shape)
|
| 3002 |
-
# [3,
|
| 3003 |
|
| 3004 |
# Get the similarity scores for the embeddings
|
| 3005 |
similarities = model.similarity(embeddings, embeddings)
|
|
@@ -3041,14 +3041,14 @@ You can finetune this model on your own dataset.
|
|
| 3041 |
|
| 3042 |
| Metric | Value |
|
| 3043 |
|:--------------------------|:-----------|
|
| 3044 |
-
| cosine_accuracy | 0.
|
| 3045 |
-
| cosine_accuracy_threshold | 0.
|
| 3046 |
-
| cosine_f1 | 0.
|
| 3047 |
-
| cosine_f1_threshold | 0.
|
| 3048 |
-
| cosine_precision | 0.
|
| 3049 |
-
| cosine_recall | 0.
|
| 3050 |
-
| **cosine_ap** | **0.
|
| 3051 |
-
| cosine_mcc | 0.
|
| 3052 |
|
| 3053 |
<!--
|
| 3054 |
## Bias, Risks and Limitations
|
|
@@ -3120,8 +3120,8 @@ You can finetune this model on your own dataset.
|
|
| 3120 |
#### Non-Default Hyperparameters
|
| 3121 |
|
| 3122 |
- `eval_strategy`: epoch
|
| 3123 |
-
- `per_device_train_batch_size`:
|
| 3124 |
-
- `per_device_eval_batch_size`:
|
| 3125 |
- `weight_decay`: 0.01
|
| 3126 |
- `num_train_epochs`: 5
|
| 3127 |
- `bf16`: True
|
|
@@ -3135,8 +3135,8 @@ You can finetune this model on your own dataset.
|
|
| 3135 |
- `do_predict`: False
|
| 3136 |
- `eval_strategy`: epoch
|
| 3137 |
- `prediction_loss_only`: True
|
| 3138 |
-
- `per_device_train_batch_size`:
|
| 3139 |
-
- `per_device_eval_batch_size`:
|
| 3140 |
- `per_gpu_train_batch_size`: None
|
| 3141 |
- `per_gpu_eval_batch_size`: None
|
| 3142 |
- `gradient_accumulation_steps`: 1
|
|
@@ -3248,72 +3248,96 @@ You can finetune this model on your own dataset.
|
|
| 3248 |
</details>
|
| 3249 |
|
| 3250 |
### Training Logs
|
| 3251 |
-
| Epoch | Step
|
| 3252 |
-
|:------:|:----:|:-------------:|:---------------:|:---------:|
|
| 3253 |
-
|
|
| 3254 |
-
|
|
| 3255 |
-
|
|
| 3256 |
-
|
|
| 3257 |
-
|
|
| 3258 |
-
|
|
| 3259 |
-
|
|
| 3260 |
-
|
|
| 3261 |
-
|
|
| 3262 |
-
|
|
| 3263 |
-
|
|
| 3264 |
-
|
|
| 3265 |
-
|
|
| 3266 |
-
|
|
| 3267 |
-
|
|
| 3268 |
-
|
|
| 3269 |
-
|
|
| 3270 |
-
|
|
| 3271 |
-
|
|
| 3272 |
-
|
|
| 3273 |
-
| 1.
|
| 3274 |
-
| 1.
|
| 3275 |
-
| 1.
|
| 3276 |
-
| 1.
|
| 3277 |
-
| 1.
|
| 3278 |
-
| 1.
|
| 3279 |
-
| 1.
|
| 3280 |
-
| 1.
|
| 3281 |
-
| 1.
|
| 3282 |
-
| 1.
|
| 3283 |
-
| 1.
|
| 3284 |
-
| 1.
|
| 3285 |
-
| 1.
|
| 3286 |
-
| 1.
|
| 3287 |
-
| 1.
|
| 3288 |
-
| 1.
|
| 3289 |
-
| 1.
|
| 3290 |
-
| 1.
|
| 3291 |
-
| 1.
|
| 3292 |
-
| 1.
|
| 3293 |
-
| 1.
|
| 3294 |
-
|
|
| 3295 |
-
|
|
| 3296 |
-
| 2.
|
| 3297 |
-
| 2.
|
| 3298 |
-
| 2.
|
| 3299 |
-
| 2.
|
| 3300 |
-
| 2.
|
| 3301 |
-
| 2.
|
| 3302 |
-
| 2.
|
| 3303 |
-
| 2.
|
| 3304 |
-
| 2.
|
| 3305 |
-
| 2.
|
| 3306 |
-
| 2.
|
| 3307 |
-
| 2.
|
| 3308 |
-
| 2.
|
| 3309 |
-
| 2.
|
| 3310 |
-
| 2.
|
| 3311 |
-
| 2.
|
| 3312 |
-
| 2.
|
| 3313 |
-
| 2.
|
| 3314 |
-
| 2.
|
| 3315 |
-
| 2.
|
| 3316 |
-
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 3317 |
|
| 3318 |
|
| 3319 |
### Framework Versions
|
|
|
|
| 6 |
- generated_from_trainer
|
| 7 |
- dataset_size:276686
|
| 8 |
- loss:MultipleNegativesRankingLoss
|
| 9 |
+
base_model: intfloat/multilingual-e5-base
|
| 10 |
widget:
|
| 11 |
- source_sentence: 'query: Печенеги отступали. Они могли запросто убить оставшегося
|
| 12 |
позади князя Владимира, мальчишку и старика, но получили приказ - уходить. Куря
|
|
|
|
| 2907 |
- cosine_ap
|
| 2908 |
- cosine_mcc
|
| 2909 |
model-index:
|
| 2910 |
+
- name: SentenceTransformer based on intfloat/multilingual-e5-base
|
| 2911 |
results:
|
| 2912 |
- task:
|
| 2913 |
type: binary-classification
|
|
|
|
| 2917 |
type: unknown
|
| 2918 |
metrics:
|
| 2919 |
- type: cosine_accuracy
|
| 2920 |
+
value: 0.9225280326197758
|
| 2921 |
name: Cosine Accuracy
|
| 2922 |
- type: cosine_accuracy_threshold
|
| 2923 |
+
value: 0.7901061773300171
|
| 2924 |
name: Cosine Accuracy Threshold
|
| 2925 |
- type: cosine_f1
|
| 2926 |
+
value: 0.7559554803436604
|
| 2927 |
name: Cosine F1
|
| 2928 |
- type: cosine_f1_threshold
|
| 2929 |
+
value: 0.7817596793174744
|
| 2930 |
name: Cosine F1 Threshold
|
| 2931 |
- type: cosine_precision
|
| 2932 |
+
value: 0.756201575623413
|
| 2933 |
name: Cosine Precision
|
| 2934 |
- type: cosine_recall
|
| 2935 |
+
value: 0.7557095451883662
|
| 2936 |
name: Cosine Recall
|
| 2937 |
- type: cosine_ap
|
| 2938 |
+
value: 0.8478615501518483
|
| 2939 |
name: Cosine Ap
|
| 2940 |
- type: cosine_mcc
|
| 2941 |
+
value: 0.7071656901034916
|
| 2942 |
name: Cosine Mcc
|
| 2943 |
---
|
| 2944 |
|
| 2945 |
+
# SentenceTransformer based on intfloat/multilingual-e5-base
|
| 2946 |
|
| 2947 |
+
This is a [sentence-transformers](https://www.SBERT.net) model finetuned from [intfloat/multilingual-e5-base](https://huggingface.co/intfloat/multilingual-e5-base) on the json dataset. It maps sentences & paragraphs to a 768-dimensional dense vector space and can be used for semantic textual similarity, semantic search, paraphrase mining, text classification, clustering, and more.
|
| 2948 |
|
| 2949 |
## Model Details
|
| 2950 |
|
| 2951 |
### Model Description
|
| 2952 |
- **Model Type:** Sentence Transformer
|
| 2953 |
+
- **Base model:** [intfloat/multilingual-e5-base](https://huggingface.co/intfloat/multilingual-e5-base) <!-- at revision 835193815a3936a24a0ee7dc9e3d48c1fbb19c55 -->
|
| 2954 |
- **Maximum Sequence Length:** 512 tokens
|
| 2955 |
+
- **Output Dimensionality:** 768 dimensions
|
| 2956 |
- **Similarity Function:** Cosine Similarity
|
| 2957 |
- **Training Dataset:**
|
| 2958 |
- json
|
|
|
|
| 2969 |
|
| 2970 |
```
|
| 2971 |
SentenceTransformer(
|
| 2972 |
+
(0): Transformer({'max_seq_length': 512, 'do_lower_case': False}) with Transformer model: XLMRobertaModel
|
| 2973 |
+
(1): Pooling({'word_embedding_dimension': 768, 'pooling_mode_cls_token': False, 'pooling_mode_mean_tokens': True, 'pooling_mode_max_tokens': False, 'pooling_mode_mean_sqrt_len_tokens': False, 'pooling_mode_weightedmean_tokens': False, 'pooling_mode_lasttoken': False, 'include_prompt': True})
|
| 2974 |
(2): Normalize()
|
| 2975 |
)
|
| 2976 |
```
|
|
|
|
| 2999 |
]
|
| 3000 |
embeddings = model.encode(sentences)
|
| 3001 |
print(embeddings.shape)
|
| 3002 |
+
# [3, 768]
|
| 3003 |
|
| 3004 |
# Get the similarity scores for the embeddings
|
| 3005 |
similarities = model.similarity(embeddings, embeddings)
|
|
|
|
| 3041 |
|
| 3042 |
| Metric | Value |
|
| 3043 |
|:--------------------------|:-----------|
|
| 3044 |
+
| cosine_accuracy | 0.9225 |
|
| 3045 |
+
| cosine_accuracy_threshold | 0.7901 |
|
| 3046 |
+
| cosine_f1 | 0.756 |
|
| 3047 |
+
| cosine_f1_threshold | 0.7818 |
|
| 3048 |
+
| cosine_precision | 0.7562 |
|
| 3049 |
+
| cosine_recall | 0.7557 |
|
| 3050 |
+
| **cosine_ap** | **0.8479** |
|
| 3051 |
+
| cosine_mcc | 0.7072 |
|
| 3052 |
|
| 3053 |
<!--
|
| 3054 |
## Bias, Risks and Limitations
|
|
|
|
| 3120 |
#### Non-Default Hyperparameters
|
| 3121 |
|
| 3122 |
- `eval_strategy`: epoch
|
| 3123 |
+
- `per_device_train_batch_size`: 64
|
| 3124 |
+
- `per_device_eval_batch_size`: 64
|
| 3125 |
- `weight_decay`: 0.01
|
| 3126 |
- `num_train_epochs`: 5
|
| 3127 |
- `bf16`: True
|
|
|
|
| 3135 |
- `do_predict`: False
|
| 3136 |
- `eval_strategy`: epoch
|
| 3137 |
- `prediction_loss_only`: True
|
| 3138 |
+
- `per_device_train_batch_size`: 64
|
| 3139 |
+
- `per_device_eval_batch_size`: 64
|
| 3140 |
- `per_gpu_train_batch_size`: None
|
| 3141 |
- `per_gpu_eval_batch_size`: None
|
| 3142 |
- `gradient_accumulation_steps`: 1
|
|
|
|
| 3248 |
</details>
|
| 3249 |
|
| 3250 |
### Training Logs
|
| 3251 |
+
| Epoch | Step | Training Loss | Validation Loss | cosine_ap |
|
| 3252 |
+
|:------:|:-----:|:-------------:|:---------------:|:---------:|
|
| 3253 |
+
| 1.0176 | 4400 | 1.4186 | - | - |
|
| 3254 |
+
| 1.0407 | 4500 | 1.4075 | - | - |
|
| 3255 |
+
| 1.0638 | 4600 | 1.3934 | - | - |
|
| 3256 |
+
| 1.0870 | 4700 | 1.3799 | - | - |
|
| 3257 |
+
| 1.1101 | 4800 | 1.3597 | - | - |
|
| 3258 |
+
| 1.1332 | 4900 | 1.3351 | - | - |
|
| 3259 |
+
| 1.1563 | 5000 | 1.3082 | - | - |
|
| 3260 |
+
| 1.1795 | 5100 | 1.3105 | - | - |
|
| 3261 |
+
| 1.2026 | 5200 | 1.2948 | - | - |
|
| 3262 |
+
| 1.2257 | 5300 | 1.3486 | - | - |
|
| 3263 |
+
| 1.2488 | 5400 | 1.3155 | - | - |
|
| 3264 |
+
| 1.2720 | 5500 | 1.2761 | - | - |
|
| 3265 |
+
| 1.2951 | 5600 | 1.2541 | - | - |
|
| 3266 |
+
| 1.3182 | 5700 | 1.2346 | - | - |
|
| 3267 |
+
| 1.3414 | 5800 | 1.2285 | - | - |
|
| 3268 |
+
| 1.3645 | 5900 | 1.2013 | - | - |
|
| 3269 |
+
| 1.3876 | 6000 | 1.1986 | - | - |
|
| 3270 |
+
| 1.4107 | 6100 | 1.1755 | - | - |
|
| 3271 |
+
| 1.4339 | 6200 | 1.1937 | - | - |
|
| 3272 |
+
| 1.4570 | 6300 | 1.202 | - | - |
|
| 3273 |
+
| 1.4801 | 6400 | 1.1607 | - | - |
|
| 3274 |
+
| 1.5032 | 6500 | 1.2116 | - | - |
|
| 3275 |
+
| 1.5264 | 6600 | 1.1797 | - | - |
|
| 3276 |
+
| 1.5495 | 6700 | 1.1571 | - | - |
|
| 3277 |
+
| 1.5726 | 6800 | 1.1526 | - | - |
|
| 3278 |
+
| 1.5957 | 6900 | 1.1438 | - | - |
|
| 3279 |
+
| 1.6189 | 7000 | 1.1634 | - | - |
|
| 3280 |
+
| 1.6420 | 7100 | 1.1367 | - | - |
|
| 3281 |
+
| 1.6651 | 7200 | 1.1133 | - | - |
|
| 3282 |
+
| 1.6883 | 7300 | 1.1156 | - | - |
|
| 3283 |
+
| 1.7114 | 7400 | 1.1102 | - | - |
|
| 3284 |
+
| 1.7345 | 7500 | 1.1123 | - | - |
|
| 3285 |
+
| 1.7576 | 7600 | 1.1066 | - | - |
|
| 3286 |
+
| 1.7808 | 7700 | 1.1291 | - | - |
|
| 3287 |
+
| 1.8039 | 7800 | 1.1094 | - | - |
|
| 3288 |
+
| 1.8270 | 7900 | 1.094 | - | - |
|
| 3289 |
+
| 1.8501 | 8000 | 1.1585 | - | - |
|
| 3290 |
+
| 1.8733 | 8100 | 1.077 | - | - |
|
| 3291 |
+
| 1.8964 | 8200 | 1.108 | - | - |
|
| 3292 |
+
| 1.9195 | 8300 | 1.1431 | - | - |
|
| 3293 |
+
| 1.9426 | 8400 | 1.0784 | - | - |
|
| 3294 |
+
| 1.9658 | 8500 | 1.0834 | - | - |
|
| 3295 |
+
| 1.9889 | 8600 | 1.1268 | - | - |
|
| 3296 |
+
| 2.0 | 8648 | - | 9.6992 | 0.8450 |
|
| 3297 |
+
| 2.0120 | 8700 | 1.0443 | - | - |
|
| 3298 |
+
| 2.0352 | 8800 | 0.9715 | - | - |
|
| 3299 |
+
| 2.0583 | 8900 | 0.957 | - | - |
|
| 3300 |
+
| 2.0814 | 9000 | 0.9784 | - | - |
|
| 3301 |
+
| 2.1045 | 9100 | 0.9581 | - | - |
|
| 3302 |
+
| 2.1277 | 9200 | 0.9569 | - | - |
|
| 3303 |
+
| 2.1508 | 9300 | 0.9518 | - | - |
|
| 3304 |
+
| 2.1739 | 9400 | 0.9485 | - | - |
|
| 3305 |
+
| 2.1970 | 9500 | 0.9433 | - | - |
|
| 3306 |
+
| 2.2202 | 9600 | 0.9392 | - | - |
|
| 3307 |
+
| 2.2433 | 9700 | 0.9248 | - | - |
|
| 3308 |
+
| 2.2664 | 9800 | 0.9105 | - | - |
|
| 3309 |
+
| 2.2895 | 9900 | 0.9769 | - | - |
|
| 3310 |
+
| 2.3127 | 10000 | 0.9502 | - | - |
|
| 3311 |
+
| 2.3358 | 10100 | 0.9604 | - | - |
|
| 3312 |
+
| 2.3589 | 10200 | 0.9291 | - | - |
|
| 3313 |
+
| 2.3821 | 10300 | 0.9552 | - | - |
|
| 3314 |
+
| 2.4052 | 10400 | 0.9621 | - | - |
|
| 3315 |
+
| 2.4283 | 10500 | 0.9357 | - | - |
|
| 3316 |
+
| 2.4514 | 10600 | 0.9323 | - | - |
|
| 3317 |
+
| 2.4746 | 10700 | 0.9327 | - | - |
|
| 3318 |
+
| 2.4977 | 10800 | 0.9067 | - | - |
|
| 3319 |
+
| 2.5208 | 10900 | 0.9411 | - | - |
|
| 3320 |
+
| 2.5439 | 11000 | 0.9305 | - | - |
|
| 3321 |
+
| 2.5671 | 11100 | 0.9378 | - | - |
|
| 3322 |
+
| 2.5902 | 11200 | 0.9171 | - | - |
|
| 3323 |
+
| 2.6133 | 11300 | 0.9074 | - | - |
|
| 3324 |
+
| 2.6364 | 11400 | 0.9262 | - | - |
|
| 3325 |
+
| 2.6596 | 11500 | 0.9063 | - | - |
|
| 3326 |
+
| 2.6827 | 11600 | 0.8814 | - | - |
|
| 3327 |
+
| 2.7058 | 11700 | 0.9089 | - | - |
|
| 3328 |
+
| 2.7290 | 11800 | 0.9048 | - | - |
|
| 3329 |
+
| 2.7521 | 11900 | 0.9268 | - | - |
|
| 3330 |
+
| 2.7752 | 12000 | 0.8913 | - | - |
|
| 3331 |
+
| 2.7983 | 12100 | 0.9064 | - | - |
|
| 3332 |
+
| 2.8215 | 12200 | 0.8585 | - | - |
|
| 3333 |
+
| 2.8446 | 12300 | 0.878 | - | - |
|
| 3334 |
+
| 2.8677 | 12400 | 0.8612 | - | - |
|
| 3335 |
+
| 2.8908 | 12500 | 0.8799 | - | - |
|
| 3336 |
+
| 2.9140 | 12600 | 0.8541 | - | - |
|
| 3337 |
+
| 2.9371 | 12700 | 0.8521 | - | - |
|
| 3338 |
+
| 2.9602 | 12800 | 0.8582 | - | - |
|
| 3339 |
+
| 2.9833 | 12900 | 0.869 | - | - |
|
| 3340 |
+
| 3.0 | 12972 | - | 10.4115 | 0.8479 |
|
| 3341 |
|
| 3342 |
|
| 3343 |
### Framework Versions
|
config.json
CHANGED
|
@@ -1,25 +1,27 @@
|
|
| 1 |
{
|
| 2 |
"architectures": [
|
| 3 |
-
"
|
| 4 |
],
|
| 5 |
"attention_probs_dropout_prob": 0.1,
|
|
|
|
| 6 |
"classifier_dropout": null,
|
|
|
|
| 7 |
"hidden_act": "gelu",
|
| 8 |
"hidden_dropout_prob": 0.1,
|
| 9 |
-
"hidden_size":
|
| 10 |
"initializer_range": 0.02,
|
| 11 |
-
"intermediate_size":
|
| 12 |
-
"layer_norm_eps": 1e-
|
| 13 |
-
"max_position_embeddings":
|
| 14 |
-
"model_type": "
|
| 15 |
"num_attention_heads": 12,
|
| 16 |
"num_hidden_layers": 12,
|
| 17 |
-
"
|
|
|
|
| 18 |
"position_embedding_type": "absolute",
|
| 19 |
-
"tokenizer_class": "XLMRobertaTokenizer",
|
| 20 |
"torch_dtype": "float32",
|
| 21 |
"transformers_version": "4.52.4",
|
| 22 |
-
"type_vocab_size":
|
| 23 |
"use_cache": true,
|
| 24 |
-
"vocab_size":
|
| 25 |
}
|
|
|
|
| 1 |
{
|
| 2 |
"architectures": [
|
| 3 |
+
"XLMRobertaModel"
|
| 4 |
],
|
| 5 |
"attention_probs_dropout_prob": 0.1,
|
| 6 |
+
"bos_token_id": 0,
|
| 7 |
"classifier_dropout": null,
|
| 8 |
+
"eos_token_id": 2,
|
| 9 |
"hidden_act": "gelu",
|
| 10 |
"hidden_dropout_prob": 0.1,
|
| 11 |
+
"hidden_size": 768,
|
| 12 |
"initializer_range": 0.02,
|
| 13 |
+
"intermediate_size": 3072,
|
| 14 |
+
"layer_norm_eps": 1e-05,
|
| 15 |
+
"max_position_embeddings": 514,
|
| 16 |
+
"model_type": "xlm-roberta",
|
| 17 |
"num_attention_heads": 12,
|
| 18 |
"num_hidden_layers": 12,
|
| 19 |
+
"output_past": true,
|
| 20 |
+
"pad_token_id": 1,
|
| 21 |
"position_embedding_type": "absolute",
|
|
|
|
| 22 |
"torch_dtype": "float32",
|
| 23 |
"transformers_version": "4.52.4",
|
| 24 |
+
"type_vocab_size": 1,
|
| 25 |
"use_cache": true,
|
| 26 |
+
"vocab_size": 250002
|
| 27 |
}
|
model.safetensors
CHANGED
|
@@ -1,3 +1,3 @@
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
-
oid sha256:
|
| 3 |
-
size
|
|
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:7224289aa386ecff5a6a47a5878ba9a3a59c124abf579c211a99dcaf14cf2455
|
| 3 |
+
size 1112197096
|
optimizer.pt
CHANGED
|
@@ -1,3 +1,3 @@
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
-
oid sha256:
|
| 3 |
-
size
|
|
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:0e2051e1304f9cdfe43309a1548502446d35d8bb0867f9968f7c9b2b18658dc1
|
| 3 |
+
size 2219789707
|
rng_state.pth
CHANGED
|
@@ -1,3 +1,3 @@
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
-
oid sha256:
|
| 3 |
size 14645
|
|
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:50e4a16b58969404282a68dcf95227f3a6dbfa441535e172c6a1f5d3ab248361
|
| 3 |
size 14645
|
scheduler.pt
CHANGED
|
@@ -1,3 +1,3 @@
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
-
oid sha256:
|
| 3 |
size 1465
|
|
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:8c2cc6b245aaf5dedc3da5a396ac5028b3d52e53a67e380ea18bc62dc40b3e4e
|
| 3 |
size 1465
|
special_tokens_map.json
CHANGED
|
@@ -22,7 +22,7 @@
|
|
| 22 |
},
|
| 23 |
"mask_token": {
|
| 24 |
"content": "<mask>",
|
| 25 |
-
"lstrip":
|
| 26 |
"normalized": false,
|
| 27 |
"rstrip": false,
|
| 28 |
"single_word": false
|
|
|
|
| 22 |
},
|
| 23 |
"mask_token": {
|
| 24 |
"content": "<mask>",
|
| 25 |
+
"lstrip": true,
|
| 26 |
"normalized": false,
|
| 27 |
"rstrip": false,
|
| 28 |
"single_word": false
|
tokenizer.json
CHANGED
|
@@ -1,3 +1,3 @@
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
-
oid sha256:
|
| 3 |
-
size
|
|
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:883b037111086fd4dfebbbc9b7cee11e1517b5e0c0514879478661440f137085
|
| 3 |
+
size 17082987
|
tokenizer_config.json
CHANGED
|
@@ -34,7 +34,7 @@
|
|
| 34 |
},
|
| 35 |
"250001": {
|
| 36 |
"content": "<mask>",
|
| 37 |
-
"lstrip":
|
| 38 |
"normalized": false,
|
| 39 |
"rstrip": false,
|
| 40 |
"single_word": false,
|
|
@@ -50,7 +50,6 @@
|
|
| 50 |
"model_max_length": 512,
|
| 51 |
"pad_token": "<pad>",
|
| 52 |
"sep_token": "</s>",
|
| 53 |
-
"sp_model_kwargs": {},
|
| 54 |
"tokenizer_class": "XLMRobertaTokenizer",
|
| 55 |
"unk_token": "<unk>"
|
| 56 |
}
|
|
|
|
| 34 |
},
|
| 35 |
"250001": {
|
| 36 |
"content": "<mask>",
|
| 37 |
+
"lstrip": true,
|
| 38 |
"normalized": false,
|
| 39 |
"rstrip": false,
|
| 40 |
"single_word": false,
|
|
|
|
| 50 |
"model_max_length": 512,
|
| 51 |
"pad_token": "<pad>",
|
| 52 |
"sep_token": "</s>",
|
|
|
|
| 53 |
"tokenizer_class": "XLMRobertaTokenizer",
|
| 54 |
"unk_token": "<unk>"
|
| 55 |
}
|
trainer_state.json
CHANGED
|
@@ -1,492 +1,968 @@
|
|
| 1 |
{
|
| 2 |
-
"best_global_step":
|
| 3 |
-
"best_metric": 8.
|
| 4 |
-
"best_model_checkpoint": "printing_press/author-paraphrase/models/intfloat/multilingual-e5-
|
| 5 |
"epoch": 3.0,
|
| 6 |
"eval_steps": 500,
|
| 7 |
-
"global_step":
|
| 8 |
"is_hyper_param_search": false,
|
| 9 |
"is_local_process_zero": true,
|
| 10 |
"is_world_process_zero": true,
|
| 11 |
"log_history": [
|
| 12 |
{
|
| 13 |
-
"epoch": 0.
|
| 14 |
-
"grad_norm":
|
| 15 |
-
"learning_rate": 4.
|
| 16 |
-
"loss":
|
| 17 |
"step": 100
|
| 18 |
},
|
| 19 |
{
|
| 20 |
-
"epoch": 0.
|
| 21 |
-
"grad_norm":
|
| 22 |
-
"learning_rate": 4.
|
| 23 |
-
"loss": 2.
|
| 24 |
"step": 200
|
| 25 |
},
|
| 26 |
{
|
| 27 |
-
"epoch": 0.
|
| 28 |
-
"grad_norm":
|
| 29 |
-
"learning_rate": 4.
|
| 30 |
-
"loss": 2.
|
| 31 |
"step": 300
|
| 32 |
},
|
| 33 |
{
|
| 34 |
-
"epoch": 0.
|
| 35 |
-
"grad_norm":
|
| 36 |
-
"learning_rate": 4.
|
| 37 |
-
"loss": 2.
|
| 38 |
"step": 400
|
| 39 |
},
|
| 40 |
{
|
| 41 |
-
"epoch": 0.
|
| 42 |
-
"grad_norm":
|
| 43 |
-
"learning_rate": 4.
|
| 44 |
-
"loss":
|
| 45 |
"step": 500
|
| 46 |
},
|
| 47 |
{
|
| 48 |
-
"epoch": 0.
|
| 49 |
-
"grad_norm":
|
| 50 |
-
"learning_rate": 4.
|
| 51 |
-
"loss":
|
| 52 |
"step": 600
|
| 53 |
},
|
| 54 |
{
|
| 55 |
-
"epoch": 0.
|
| 56 |
-
"grad_norm":
|
| 57 |
-
"learning_rate": 4.
|
| 58 |
-
"loss":
|
| 59 |
"step": 700
|
| 60 |
},
|
| 61 |
{
|
| 62 |
-
"epoch": 0.
|
| 63 |
-
"grad_norm":
|
| 64 |
-
"learning_rate": 4.
|
| 65 |
-
"loss":
|
| 66 |
"step": 800
|
| 67 |
},
|
| 68 |
{
|
| 69 |
-
"epoch": 0.
|
| 70 |
-
"grad_norm":
|
| 71 |
-
"learning_rate": 4.
|
| 72 |
-
"loss":
|
| 73 |
"step": 900
|
| 74 |
},
|
| 75 |
{
|
| 76 |
-
"epoch": 0.
|
| 77 |
-
"grad_norm":
|
| 78 |
-
"learning_rate": 4.
|
| 79 |
-
"loss":
|
| 80 |
"step": 1000
|
| 81 |
},
|
| 82 |
{
|
| 83 |
-
"epoch": 0.
|
| 84 |
-
"grad_norm":
|
| 85 |
-
"learning_rate": 4.
|
| 86 |
-
"loss":
|
| 87 |
"step": 1100
|
| 88 |
},
|
| 89 |
{
|
| 90 |
-
"epoch": 0.
|
| 91 |
-
"grad_norm":
|
| 92 |
-
"learning_rate": 4.
|
| 93 |
-
"loss":
|
| 94 |
"step": 1200
|
| 95 |
},
|
| 96 |
{
|
| 97 |
-
"epoch": 0.
|
| 98 |
-
"grad_norm":
|
| 99 |
-
"learning_rate": 4.
|
| 100 |
-
"loss":
|
| 101 |
"step": 1300
|
| 102 |
},
|
| 103 |
{
|
| 104 |
-
"epoch": 0.
|
| 105 |
-
"grad_norm":
|
| 106 |
-
"learning_rate": 4.
|
| 107 |
-
"loss":
|
| 108 |
"step": 1400
|
| 109 |
},
|
| 110 |
{
|
| 111 |
-
"epoch": 0.
|
| 112 |
-
"grad_norm":
|
| 113 |
-
"learning_rate": 4.
|
| 114 |
-
"loss":
|
| 115 |
"step": 1500
|
| 116 |
},
|
| 117 |
{
|
| 118 |
-
"epoch": 0.
|
| 119 |
-
"grad_norm":
|
| 120 |
-
"learning_rate": 4.
|
| 121 |
-
"loss":
|
| 122 |
"step": 1600
|
| 123 |
},
|
| 124 |
{
|
| 125 |
-
"epoch": 0.
|
| 126 |
-
"grad_norm":
|
| 127 |
-
"learning_rate": 4.
|
| 128 |
-
"loss":
|
| 129 |
"step": 1700
|
| 130 |
},
|
| 131 |
{
|
| 132 |
-
"epoch": 0.
|
| 133 |
-
"grad_norm":
|
| 134 |
-
"learning_rate": 4.
|
| 135 |
-
"loss":
|
| 136 |
"step": 1800
|
| 137 |
},
|
| 138 |
{
|
| 139 |
-
"epoch": 0.
|
| 140 |
-
"grad_norm":
|
| 141 |
-
"learning_rate": 4.
|
| 142 |
-
"loss":
|
| 143 |
"step": 1900
|
| 144 |
},
|
| 145 |
{
|
| 146 |
-
"epoch": 0.
|
| 147 |
-
"grad_norm":
|
| 148 |
-
"learning_rate": 4.
|
| 149 |
-
"loss":
|
| 150 |
"step": 2000
|
| 151 |
},
|
| 152 |
{
|
| 153 |
-
"epoch":
|
| 154 |
-
"
|
| 155 |
-
"
|
| 156 |
-
"
|
| 157 |
-
"eval_cosine_f1": 0.7428981208338088,
|
| 158 |
-
"eval_cosine_f1_threshold": 0.8662768006324768,
|
| 159 |
-
"eval_cosine_mcc": 0.6915836823172079,
|
| 160 |
-
"eval_cosine_precision": 0.7443417485874784,
|
| 161 |
-
"eval_cosine_recall": 0.741460081983213,
|
| 162 |
-
"eval_loss": 8.790307998657227,
|
| 163 |
-
"eval_runtime": 1127.8164,
|
| 164 |
-
"eval_samples_per_second": 163.527,
|
| 165 |
-
"eval_steps_per_second": 1.203,
|
| 166 |
-
"step": 2035
|
| 167 |
-
},
|
| 168 |
-
{
|
| 169 |
-
"epoch": 1.031941031941032,
|
| 170 |
-
"grad_norm": 5.7056660652160645,
|
| 171 |
-
"learning_rate": 3.968550368550369e-05,
|
| 172 |
-
"loss": 2.1558,
|
| 173 |
"step": 2100
|
| 174 |
},
|
| 175 |
{
|
| 176 |
-
"epoch":
|
| 177 |
-
"grad_norm":
|
| 178 |
-
"learning_rate":
|
| 179 |
-
"loss":
|
| 180 |
"step": 2200
|
| 181 |
},
|
| 182 |
{
|
| 183 |
-
"epoch":
|
| 184 |
-
"grad_norm":
|
| 185 |
-
"learning_rate":
|
| 186 |
-
"loss":
|
| 187 |
"step": 2300
|
| 188 |
},
|
| 189 |
{
|
| 190 |
-
"epoch":
|
| 191 |
-
"grad_norm":
|
| 192 |
-
"learning_rate":
|
| 193 |
-
"loss":
|
| 194 |
"step": 2400
|
| 195 |
},
|
| 196 |
{
|
| 197 |
-
"epoch":
|
| 198 |
-
"grad_norm":
|
| 199 |
-
"learning_rate":
|
| 200 |
-
"loss":
|
| 201 |
"step": 2500
|
| 202 |
},
|
| 203 |
{
|
| 204 |
-
"epoch":
|
| 205 |
-
"grad_norm":
|
| 206 |
-
"learning_rate":
|
| 207 |
-
"loss":
|
| 208 |
"step": 2600
|
| 209 |
},
|
| 210 |
{
|
| 211 |
-
"epoch":
|
| 212 |
-
"grad_norm":
|
| 213 |
-
"learning_rate":
|
| 214 |
-
"loss":
|
| 215 |
"step": 2700
|
| 216 |
},
|
| 217 |
{
|
| 218 |
-
"epoch":
|
| 219 |
-
"grad_norm":
|
| 220 |
-
"learning_rate":
|
| 221 |
-
"loss":
|
| 222 |
"step": 2800
|
| 223 |
},
|
| 224 |
{
|
| 225 |
-
"epoch":
|
| 226 |
-
"grad_norm":
|
| 227 |
-
"learning_rate":
|
| 228 |
-
"loss":
|
| 229 |
"step": 2900
|
| 230 |
},
|
| 231 |
{
|
| 232 |
-
"epoch":
|
| 233 |
-
"grad_norm":
|
| 234 |
-
"learning_rate":
|
| 235 |
-
"loss":
|
| 236 |
"step": 3000
|
| 237 |
},
|
| 238 |
{
|
| 239 |
-
"epoch":
|
| 240 |
-
"grad_norm":
|
| 241 |
-
"learning_rate":
|
| 242 |
-
"loss":
|
| 243 |
"step": 3100
|
| 244 |
},
|
| 245 |
{
|
| 246 |
-
"epoch":
|
| 247 |
-
"grad_norm":
|
| 248 |
-
"learning_rate":
|
| 249 |
-
"loss":
|
| 250 |
"step": 3200
|
| 251 |
},
|
| 252 |
{
|
| 253 |
-
"epoch":
|
| 254 |
-
"grad_norm":
|
| 255 |
-
"learning_rate":
|
| 256 |
-
"loss": 1.
|
| 257 |
"step": 3300
|
| 258 |
},
|
| 259 |
{
|
| 260 |
-
"epoch":
|
| 261 |
-
"grad_norm":
|
| 262 |
-
"learning_rate":
|
| 263 |
-
"loss": 1.
|
| 264 |
"step": 3400
|
| 265 |
},
|
| 266 |
{
|
| 267 |
-
"epoch":
|
| 268 |
-
"grad_norm":
|
| 269 |
-
"learning_rate":
|
| 270 |
-
"loss": 1.
|
| 271 |
"step": 3500
|
| 272 |
},
|
| 273 |
{
|
| 274 |
-
"epoch":
|
| 275 |
-
"grad_norm":
|
| 276 |
-
"learning_rate":
|
| 277 |
-
"loss": 1.
|
| 278 |
"step": 3600
|
| 279 |
},
|
| 280 |
{
|
| 281 |
-
"epoch":
|
| 282 |
-
"grad_norm":
|
| 283 |
-
"learning_rate":
|
| 284 |
-
"loss": 1.
|
| 285 |
"step": 3700
|
| 286 |
},
|
| 287 |
{
|
| 288 |
-
"epoch":
|
| 289 |
-
"grad_norm":
|
| 290 |
-
"learning_rate":
|
| 291 |
-
"loss": 1.
|
| 292 |
"step": 3800
|
| 293 |
},
|
| 294 |
{
|
| 295 |
-
"epoch":
|
| 296 |
-
"grad_norm":
|
| 297 |
-
"learning_rate":
|
| 298 |
-
"loss": 1.
|
| 299 |
"step": 3900
|
| 300 |
},
|
| 301 |
{
|
| 302 |
-
"epoch":
|
| 303 |
-
"grad_norm":
|
| 304 |
-
"learning_rate":
|
| 305 |
-
"loss": 1.
|
| 306 |
"step": 4000
|
| 307 |
},
|
| 308 |
{
|
| 309 |
-
"epoch":
|
| 310 |
-
"
|
| 311 |
-
"
|
| 312 |
-
"
|
| 313 |
-
"eval_cosine_f1": 0.7448358673957842,
|
| 314 |
-
"eval_cosine_f1_threshold": 0.8511393070220947,
|
| 315 |
-
"eval_cosine_mcc": 0.6936559183741586,
|
| 316 |
-
"eval_cosine_precision": 0.7430153128945579,
|
| 317 |
-
"eval_cosine_recall": 0.7466653653458261,
|
| 318 |
-
"eval_loss": 9.22366714477539,
|
| 319 |
-
"eval_runtime": 1127.8509,
|
| 320 |
-
"eval_samples_per_second": 163.522,
|
| 321 |
-
"eval_steps_per_second": 1.203,
|
| 322 |
-
"step": 4070
|
| 323 |
-
},
|
| 324 |
-
{
|
| 325 |
-
"epoch": 2.0147420147420148,
|
| 326 |
-
"grad_norm": 7.566407680511475,
|
| 327 |
-
"learning_rate": 2.9857493857493856e-05,
|
| 328 |
-
"loss": 1.9059,
|
| 329 |
"step": 4100
|
| 330 |
},
|
| 331 |
{
|
| 332 |
-
"epoch":
|
| 333 |
-
"grad_norm":
|
| 334 |
-
"learning_rate":
|
| 335 |
-
"loss": 1.
|
| 336 |
"step": 4200
|
| 337 |
},
|
| 338 |
{
|
| 339 |
-
"epoch":
|
| 340 |
-
"grad_norm":
|
| 341 |
-
"learning_rate":
|
| 342 |
-
"loss": 1.
|
| 343 |
"step": 4300
|
| 344 |
},
|
| 345 |
{
|
| 346 |
-
"epoch":
|
| 347 |
-
"
|
| 348 |
-
"
|
| 349 |
-
"
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 350 |
"step": 4400
|
| 351 |
},
|
| 352 |
{
|
| 353 |
-
"epoch":
|
| 354 |
-
"grad_norm":
|
| 355 |
-
"learning_rate":
|
| 356 |
-
"loss": 1.
|
| 357 |
"step": 4500
|
| 358 |
},
|
| 359 |
{
|
| 360 |
-
"epoch":
|
| 361 |
-
"grad_norm":
|
| 362 |
-
"learning_rate":
|
| 363 |
-
"loss": 1.
|
| 364 |
"step": 4600
|
| 365 |
},
|
| 366 |
{
|
| 367 |
-
"epoch":
|
| 368 |
-
"grad_norm":
|
| 369 |
-
"learning_rate":
|
| 370 |
-
"loss": 1.
|
| 371 |
"step": 4700
|
| 372 |
},
|
| 373 |
{
|
| 374 |
-
"epoch":
|
| 375 |
-
"grad_norm":
|
| 376 |
-
"learning_rate":
|
| 377 |
-
"loss": 1.
|
| 378 |
"step": 4800
|
| 379 |
},
|
| 380 |
{
|
| 381 |
-
"epoch":
|
| 382 |
-
"grad_norm":
|
| 383 |
-
"learning_rate":
|
| 384 |
-
"loss": 1.
|
| 385 |
"step": 4900
|
| 386 |
},
|
| 387 |
{
|
| 388 |
-
"epoch":
|
| 389 |
-
"grad_norm":
|
| 390 |
-
"learning_rate":
|
| 391 |
-
"loss": 1.
|
| 392 |
"step": 5000
|
| 393 |
},
|
| 394 |
{
|
| 395 |
-
"epoch":
|
| 396 |
-
"grad_norm":
|
| 397 |
-
"learning_rate":
|
| 398 |
-
"loss": 1.
|
| 399 |
"step": 5100
|
| 400 |
},
|
| 401 |
{
|
| 402 |
-
"epoch":
|
| 403 |
-
"grad_norm":
|
| 404 |
-
"learning_rate":
|
| 405 |
-
"loss": 1.
|
| 406 |
"step": 5200
|
| 407 |
},
|
| 408 |
{
|
| 409 |
-
"epoch":
|
| 410 |
-
"grad_norm":
|
| 411 |
-
"learning_rate":
|
| 412 |
-
"loss": 1.
|
| 413 |
"step": 5300
|
| 414 |
},
|
| 415 |
{
|
| 416 |
-
"epoch":
|
| 417 |
-
"grad_norm":
|
| 418 |
-
"learning_rate":
|
| 419 |
-
"loss": 1.
|
| 420 |
"step": 5400
|
| 421 |
},
|
| 422 |
{
|
| 423 |
-
"epoch":
|
| 424 |
-
"grad_norm":
|
| 425 |
-
"learning_rate":
|
| 426 |
-
"loss": 1.
|
| 427 |
"step": 5500
|
| 428 |
},
|
| 429 |
{
|
| 430 |
-
"epoch":
|
| 431 |
-
"grad_norm":
|
| 432 |
-
"learning_rate":
|
| 433 |
-
"loss": 1.
|
| 434 |
"step": 5600
|
| 435 |
},
|
| 436 |
{
|
| 437 |
-
"epoch":
|
| 438 |
-
"grad_norm":
|
| 439 |
-
"learning_rate":
|
| 440 |
-
"loss": 1.
|
| 441 |
"step": 5700
|
| 442 |
},
|
| 443 |
{
|
| 444 |
-
"epoch":
|
| 445 |
-
"grad_norm":
|
| 446 |
-
"learning_rate":
|
| 447 |
-
"loss": 1.
|
| 448 |
"step": 5800
|
| 449 |
},
|
| 450 |
{
|
| 451 |
-
"epoch":
|
| 452 |
-
"grad_norm":
|
| 453 |
-
"learning_rate":
|
| 454 |
-
"loss": 1.
|
| 455 |
"step": 5900
|
| 456 |
},
|
| 457 |
{
|
| 458 |
-
"epoch":
|
| 459 |
-
"grad_norm":
|
| 460 |
-
"learning_rate":
|
| 461 |
-
"loss": 1.
|
| 462 |
"step": 6000
|
| 463 |
},
|
| 464 |
{
|
| 465 |
-
"epoch":
|
| 466 |
-
"grad_norm":
|
| 467 |
-
"learning_rate":
|
| 468 |
-
"loss": 1.
|
| 469 |
"step": 6100
|
| 470 |
},
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 471 |
{
|
| 472 |
"epoch": 3.0,
|
| 473 |
-
"eval_cosine_accuracy": 0.
|
| 474 |
-
"eval_cosine_accuracy_threshold": 0.
|
| 475 |
-
"eval_cosine_ap": 0.
|
| 476 |
-
"eval_cosine_f1": 0.
|
| 477 |
-
"eval_cosine_f1_threshold": 0.
|
| 478 |
-
"eval_cosine_mcc": 0.
|
| 479 |
-
"eval_cosine_precision": 0.
|
| 480 |
-
"eval_cosine_recall": 0.
|
| 481 |
-
"eval_loss":
|
| 482 |
-
"eval_runtime":
|
| 483 |
-
"eval_samples_per_second":
|
| 484 |
-
"eval_steps_per_second": 1.
|
| 485 |
-
"step":
|
| 486 |
}
|
| 487 |
],
|
| 488 |
"logging_steps": 100,
|
| 489 |
-
"max_steps":
|
| 490 |
"num_input_tokens_seen": 0,
|
| 491 |
"num_train_epochs": 5,
|
| 492 |
"save_steps": 500,
|
|
@@ -503,7 +979,7 @@
|
|
| 503 |
}
|
| 504 |
},
|
| 505 |
"total_flos": 0.0,
|
| 506 |
-
"train_batch_size":
|
| 507 |
"trial_name": null,
|
| 508 |
"trial_params": null
|
| 509 |
}
|
|
|
|
| 1 |
{
|
| 2 |
+
"best_global_step": 4324,
|
| 3 |
+
"best_metric": 8.351966857910156,
|
| 4 |
+
"best_model_checkpoint": "printing_press/author-paraphrase/models/intfloat/multilingual-e5-base/checkpoint-4324",
|
| 5 |
"epoch": 3.0,
|
| 6 |
"eval_steps": 500,
|
| 7 |
+
"global_step": 12972,
|
| 8 |
"is_hyper_param_search": false,
|
| 9 |
"is_local_process_zero": true,
|
| 10 |
"is_world_process_zero": true,
|
| 11 |
"log_history": [
|
| 12 |
{
|
| 13 |
+
"epoch": 0.02312673450508788,
|
| 14 |
+
"grad_norm": 10.87702751159668,
|
| 15 |
+
"learning_rate": 4.977104532839963e-05,
|
| 16 |
+
"loss": 2.4824,
|
| 17 |
"step": 100
|
| 18 |
},
|
| 19 |
{
|
| 20 |
+
"epoch": 0.04625346901017576,
|
| 21 |
+
"grad_norm": 11.214370727539062,
|
| 22 |
+
"learning_rate": 4.953977798334876e-05,
|
| 23 |
+
"loss": 2.2607,
|
| 24 |
"step": 200
|
| 25 |
},
|
| 26 |
{
|
| 27 |
+
"epoch": 0.06938020351526364,
|
| 28 |
+
"grad_norm": 9.89275074005127,
|
| 29 |
+
"learning_rate": 4.930851063829787e-05,
|
| 30 |
+
"loss": 2.1716,
|
| 31 |
"step": 300
|
| 32 |
},
|
| 33 |
{
|
| 34 |
+
"epoch": 0.09250693802035152,
|
| 35 |
+
"grad_norm": 8.298206329345703,
|
| 36 |
+
"learning_rate": 4.9077243293247e-05,
|
| 37 |
+
"loss": 2.0981,
|
| 38 |
"step": 400
|
| 39 |
},
|
| 40 |
{
|
| 41 |
+
"epoch": 0.11563367252543941,
|
| 42 |
+
"grad_norm": 9.424484252929688,
|
| 43 |
+
"learning_rate": 4.8845975948196116e-05,
|
| 44 |
+
"loss": 1.9617,
|
| 45 |
"step": 500
|
| 46 |
},
|
| 47 |
{
|
| 48 |
+
"epoch": 0.13876040703052728,
|
| 49 |
+
"grad_norm": 8.752665519714355,
|
| 50 |
+
"learning_rate": 4.8614708603145235e-05,
|
| 51 |
+
"loss": 1.987,
|
| 52 |
"step": 600
|
| 53 |
},
|
| 54 |
{
|
| 55 |
+
"epoch": 0.16188714153561518,
|
| 56 |
+
"grad_norm": 9.41073989868164,
|
| 57 |
+
"learning_rate": 4.838344125809436e-05,
|
| 58 |
+
"loss": 1.9429,
|
| 59 |
"step": 700
|
| 60 |
},
|
| 61 |
{
|
| 62 |
+
"epoch": 0.18501387604070305,
|
| 63 |
+
"grad_norm": 11.025175094604492,
|
| 64 |
+
"learning_rate": 4.815217391304348e-05,
|
| 65 |
+
"loss": 1.9398,
|
| 66 |
"step": 800
|
| 67 |
},
|
| 68 |
{
|
| 69 |
+
"epoch": 0.20814061054579094,
|
| 70 |
+
"grad_norm": 9.1474027633667,
|
| 71 |
+
"learning_rate": 4.79209065679926e-05,
|
| 72 |
+
"loss": 1.8745,
|
| 73 |
"step": 900
|
| 74 |
},
|
| 75 |
{
|
| 76 |
+
"epoch": 0.23126734505087881,
|
| 77 |
+
"grad_norm": 8.528326034545898,
|
| 78 |
+
"learning_rate": 4.7689639222941726e-05,
|
| 79 |
+
"loss": 1.8484,
|
| 80 |
"step": 1000
|
| 81 |
},
|
| 82 |
{
|
| 83 |
+
"epoch": 0.2543940795559667,
|
| 84 |
+
"grad_norm": 9.510236740112305,
|
| 85 |
+
"learning_rate": 4.7458371877890846e-05,
|
| 86 |
+
"loss": 1.846,
|
| 87 |
"step": 1100
|
| 88 |
},
|
| 89 |
{
|
| 90 |
+
"epoch": 0.27752081406105455,
|
| 91 |
+
"grad_norm": 10.6268892288208,
|
| 92 |
+
"learning_rate": 4.7227104532839965e-05,
|
| 93 |
+
"loss": 1.7953,
|
| 94 |
"step": 1200
|
| 95 |
},
|
| 96 |
{
|
| 97 |
+
"epoch": 0.30064754856614245,
|
| 98 |
+
"grad_norm": 9.342752456665039,
|
| 99 |
+
"learning_rate": 4.699583718778909e-05,
|
| 100 |
+
"loss": 1.8189,
|
| 101 |
"step": 1300
|
| 102 |
},
|
| 103 |
{
|
| 104 |
+
"epoch": 0.32377428307123035,
|
| 105 |
+
"grad_norm": 8.43892765045166,
|
| 106 |
+
"learning_rate": 4.676456984273821e-05,
|
| 107 |
+
"loss": 1.7775,
|
| 108 |
"step": 1400
|
| 109 |
},
|
| 110 |
{
|
| 111 |
+
"epoch": 0.34690101757631825,
|
| 112 |
+
"grad_norm": 12.730334281921387,
|
| 113 |
+
"learning_rate": 4.653330249768733e-05,
|
| 114 |
+
"loss": 1.76,
|
| 115 |
"step": 1500
|
| 116 |
},
|
| 117 |
{
|
| 118 |
+
"epoch": 0.3700277520814061,
|
| 119 |
+
"grad_norm": 27.328882217407227,
|
| 120 |
+
"learning_rate": 4.6302035152636456e-05,
|
| 121 |
+
"loss": 1.7638,
|
| 122 |
"step": 1600
|
| 123 |
},
|
| 124 |
{
|
| 125 |
+
"epoch": 0.393154486586494,
|
| 126 |
+
"grad_norm": 9.745387077331543,
|
| 127 |
+
"learning_rate": 4.607076780758557e-05,
|
| 128 |
+
"loss": 1.7039,
|
| 129 |
"step": 1700
|
| 130 |
},
|
| 131 |
{
|
| 132 |
+
"epoch": 0.4162812210915819,
|
| 133 |
+
"grad_norm": 9.242189407348633,
|
| 134 |
+
"learning_rate": 4.583950046253469e-05,
|
| 135 |
+
"loss": 1.706,
|
| 136 |
"step": 1800
|
| 137 |
},
|
| 138 |
{
|
| 139 |
+
"epoch": 0.43940795559666973,
|
| 140 |
+
"grad_norm": 9.079887390136719,
|
| 141 |
+
"learning_rate": 4.5608233117483814e-05,
|
| 142 |
+
"loss": 1.7255,
|
| 143 |
"step": 1900
|
| 144 |
},
|
| 145 |
{
|
| 146 |
+
"epoch": 0.46253469010175763,
|
| 147 |
+
"grad_norm": 8.81753158569336,
|
| 148 |
+
"learning_rate": 4.537696577243293e-05,
|
| 149 |
+
"loss": 1.705,
|
| 150 |
"step": 2000
|
| 151 |
},
|
| 152 |
{
|
| 153 |
+
"epoch": 0.4856614246068455,
|
| 154 |
+
"grad_norm": 10.990808486938477,
|
| 155 |
+
"learning_rate": 4.514569842738205e-05,
|
| 156 |
+
"loss": 1.6823,
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 157 |
"step": 2100
|
| 158 |
},
|
| 159 |
{
|
| 160 |
+
"epoch": 0.5087881591119334,
|
| 161 |
+
"grad_norm": 9.342223167419434,
|
| 162 |
+
"learning_rate": 4.491443108233118e-05,
|
| 163 |
+
"loss": 1.6921,
|
| 164 |
"step": 2200
|
| 165 |
},
|
| 166 |
{
|
| 167 |
+
"epoch": 0.5319148936170213,
|
| 168 |
+
"grad_norm": 9.898025512695312,
|
| 169 |
+
"learning_rate": 4.46831637372803e-05,
|
| 170 |
+
"loss": 1.6801,
|
| 171 |
"step": 2300
|
| 172 |
},
|
| 173 |
{
|
| 174 |
+
"epoch": 0.5550416281221091,
|
| 175 |
+
"grad_norm": 8.3154296875,
|
| 176 |
+
"learning_rate": 4.445189639222942e-05,
|
| 177 |
+
"loss": 1.6547,
|
| 178 |
"step": 2400
|
| 179 |
},
|
| 180 |
{
|
| 181 |
+
"epoch": 0.5781683626271971,
|
| 182 |
+
"grad_norm": 8.5817232131958,
|
| 183 |
+
"learning_rate": 4.422062904717854e-05,
|
| 184 |
+
"loss": 1.6512,
|
| 185 |
"step": 2500
|
| 186 |
},
|
| 187 |
{
|
| 188 |
+
"epoch": 0.6012950971322849,
|
| 189 |
+
"grad_norm": 8.931965827941895,
|
| 190 |
+
"learning_rate": 4.398936170212766e-05,
|
| 191 |
+
"loss": 1.6466,
|
| 192 |
"step": 2600
|
| 193 |
},
|
| 194 |
{
|
| 195 |
+
"epoch": 0.6244218316373727,
|
| 196 |
+
"grad_norm": 10.271759033203125,
|
| 197 |
+
"learning_rate": 4.375809435707678e-05,
|
| 198 |
+
"loss": 1.6545,
|
| 199 |
"step": 2700
|
| 200 |
},
|
| 201 |
{
|
| 202 |
+
"epoch": 0.6475485661424607,
|
| 203 |
+
"grad_norm": 8.621456146240234,
|
| 204 |
+
"learning_rate": 4.352682701202591e-05,
|
| 205 |
+
"loss": 1.5985,
|
| 206 |
"step": 2800
|
| 207 |
},
|
| 208 |
{
|
| 209 |
+
"epoch": 0.6706753006475485,
|
| 210 |
+
"grad_norm": 9.487942695617676,
|
| 211 |
+
"learning_rate": 4.329555966697503e-05,
|
| 212 |
+
"loss": 1.5941,
|
| 213 |
"step": 2900
|
| 214 |
},
|
| 215 |
{
|
| 216 |
+
"epoch": 0.6938020351526365,
|
| 217 |
+
"grad_norm": 9.156566619873047,
|
| 218 |
+
"learning_rate": 4.306429232192415e-05,
|
| 219 |
+
"loss": 1.6178,
|
| 220 |
"step": 3000
|
| 221 |
},
|
| 222 |
{
|
| 223 |
+
"epoch": 0.7169287696577243,
|
| 224 |
+
"grad_norm": 11.126986503601074,
|
| 225 |
+
"learning_rate": 4.2833024976873266e-05,
|
| 226 |
+
"loss": 1.6035,
|
| 227 |
"step": 3100
|
| 228 |
},
|
| 229 |
{
|
| 230 |
+
"epoch": 0.7400555041628122,
|
| 231 |
+
"grad_norm": 10.304364204406738,
|
| 232 |
+
"learning_rate": 4.2601757631822385e-05,
|
| 233 |
+
"loss": 1.568,
|
| 234 |
"step": 3200
|
| 235 |
},
|
| 236 |
{
|
| 237 |
+
"epoch": 0.7631822386679001,
|
| 238 |
+
"grad_norm": 9.863882064819336,
|
| 239 |
+
"learning_rate": 4.2370490286771505e-05,
|
| 240 |
+
"loss": 1.5733,
|
| 241 |
"step": 3300
|
| 242 |
},
|
| 243 |
{
|
| 244 |
+
"epoch": 0.786308973172988,
|
| 245 |
+
"grad_norm": 10.20065689086914,
|
| 246 |
+
"learning_rate": 4.213922294172063e-05,
|
| 247 |
+
"loss": 1.5841,
|
| 248 |
"step": 3400
|
| 249 |
},
|
| 250 |
{
|
| 251 |
+
"epoch": 0.8094357076780758,
|
| 252 |
+
"grad_norm": 9.240017890930176,
|
| 253 |
+
"learning_rate": 4.190795559666975e-05,
|
| 254 |
+
"loss": 1.5869,
|
| 255 |
"step": 3500
|
| 256 |
},
|
| 257 |
{
|
| 258 |
+
"epoch": 0.8325624421831638,
|
| 259 |
+
"grad_norm": 9.452214241027832,
|
| 260 |
+
"learning_rate": 4.167668825161887e-05,
|
| 261 |
+
"loss": 1.5806,
|
| 262 |
"step": 3600
|
| 263 |
},
|
| 264 |
{
|
| 265 |
+
"epoch": 0.8556891766882516,
|
| 266 |
+
"grad_norm": 10.292157173156738,
|
| 267 |
+
"learning_rate": 4.1445420906567996e-05,
|
| 268 |
+
"loss": 1.5682,
|
| 269 |
"step": 3700
|
| 270 |
},
|
| 271 |
{
|
| 272 |
+
"epoch": 0.8788159111933395,
|
| 273 |
+
"grad_norm": 12.675426483154297,
|
| 274 |
+
"learning_rate": 4.1214153561517115e-05,
|
| 275 |
+
"loss": 1.553,
|
| 276 |
"step": 3800
|
| 277 |
},
|
| 278 |
{
|
| 279 |
+
"epoch": 0.9019426456984274,
|
| 280 |
+
"grad_norm": 11.213452339172363,
|
| 281 |
+
"learning_rate": 4.0982886216466234e-05,
|
| 282 |
+
"loss": 1.5564,
|
| 283 |
"step": 3900
|
| 284 |
},
|
| 285 |
{
|
| 286 |
+
"epoch": 0.9250693802035153,
|
| 287 |
+
"grad_norm": 11.688211441040039,
|
| 288 |
+
"learning_rate": 4.075161887141536e-05,
|
| 289 |
+
"loss": 1.5389,
|
| 290 |
"step": 4000
|
| 291 |
},
|
| 292 |
{
|
| 293 |
+
"epoch": 0.9481961147086031,
|
| 294 |
+
"grad_norm": 19.469762802124023,
|
| 295 |
+
"learning_rate": 4.052035152636448e-05,
|
| 296 |
+
"loss": 1.5091,
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 297 |
"step": 4100
|
| 298 |
},
|
| 299 |
{
|
| 300 |
+
"epoch": 0.971322849213691,
|
| 301 |
+
"grad_norm": 11.896566390991211,
|
| 302 |
+
"learning_rate": 4.02890841813136e-05,
|
| 303 |
+
"loss": 1.5358,
|
| 304 |
"step": 4200
|
| 305 |
},
|
| 306 |
{
|
| 307 |
+
"epoch": 0.9944495837187789,
|
| 308 |
+
"grad_norm": 26.442358016967773,
|
| 309 |
+
"learning_rate": 4.0057816836262725e-05,
|
| 310 |
+
"loss": 1.512,
|
| 311 |
"step": 4300
|
| 312 |
},
|
| 313 |
{
|
| 314 |
+
"epoch": 1.0,
|
| 315 |
+
"eval_cosine_accuracy": 0.9215032424577613,
|
| 316 |
+
"eval_cosine_accuracy_threshold": 0.8687731027603149,
|
| 317 |
+
"eval_cosine_ap": 0.844612185583288,
|
| 318 |
+
"eval_cosine_f1": 0.7527066450567261,
|
| 319 |
+
"eval_cosine_f1_threshold": 0.8616929054260254,
|
| 320 |
+
"eval_cosine_mcc": 0.7030390290848069,
|
| 321 |
+
"eval_cosine_precision": 0.7499838511724048,
|
| 322 |
+
"eval_cosine_recall": 0.7554492810202356,
|
| 323 |
+
"eval_loss": 8.351966857910156,
|
| 324 |
+
"eval_runtime": 1467.219,
|
| 325 |
+
"eval_samples_per_second": 125.699,
|
| 326 |
+
"eval_steps_per_second": 1.964,
|
| 327 |
+
"step": 4324
|
| 328 |
+
},
|
| 329 |
+
{
|
| 330 |
+
"epoch": 1.0175763182238668,
|
| 331 |
+
"grad_norm": 10.624114990234375,
|
| 332 |
+
"learning_rate": 3.9826549491211844e-05,
|
| 333 |
+
"loss": 1.4186,
|
| 334 |
"step": 4400
|
| 335 |
},
|
| 336 |
{
|
| 337 |
+
"epoch": 1.0407030527289547,
|
| 338 |
+
"grad_norm": 10.404448509216309,
|
| 339 |
+
"learning_rate": 3.9595282146160964e-05,
|
| 340 |
+
"loss": 1.4075,
|
| 341 |
"step": 4500
|
| 342 |
},
|
| 343 |
{
|
| 344 |
+
"epoch": 1.0638297872340425,
|
| 345 |
+
"grad_norm": 9.519079208374023,
|
| 346 |
+
"learning_rate": 3.936401480111008e-05,
|
| 347 |
+
"loss": 1.3934,
|
| 348 |
"step": 4600
|
| 349 |
},
|
| 350 |
{
|
| 351 |
+
"epoch": 1.0869565217391304,
|
| 352 |
+
"grad_norm": 9.599648475646973,
|
| 353 |
+
"learning_rate": 3.91327474560592e-05,
|
| 354 |
+
"loss": 1.3799,
|
| 355 |
"step": 4700
|
| 356 |
},
|
| 357 |
{
|
| 358 |
+
"epoch": 1.1100832562442182,
|
| 359 |
+
"grad_norm": 14.585043907165527,
|
| 360 |
+
"learning_rate": 3.890148011100833e-05,
|
| 361 |
+
"loss": 1.3597,
|
| 362 |
"step": 4800
|
| 363 |
},
|
| 364 |
{
|
| 365 |
+
"epoch": 1.1332099907493063,
|
| 366 |
+
"grad_norm": 11.30241584777832,
|
| 367 |
+
"learning_rate": 3.867021276595745e-05,
|
| 368 |
+
"loss": 1.3351,
|
| 369 |
"step": 4900
|
| 370 |
},
|
| 371 |
{
|
| 372 |
+
"epoch": 1.1563367252543941,
|
| 373 |
+
"grad_norm": 10.888731002807617,
|
| 374 |
+
"learning_rate": 3.843894542090657e-05,
|
| 375 |
+
"loss": 1.3082,
|
| 376 |
"step": 5000
|
| 377 |
},
|
| 378 |
{
|
| 379 |
+
"epoch": 1.179463459759482,
|
| 380 |
+
"grad_norm": 9.09694766998291,
|
| 381 |
+
"learning_rate": 3.820767807585569e-05,
|
| 382 |
+
"loss": 1.3105,
|
| 383 |
"step": 5100
|
| 384 |
},
|
| 385 |
{
|
| 386 |
+
"epoch": 1.2025901942645698,
|
| 387 |
+
"grad_norm": 13.162363052368164,
|
| 388 |
+
"learning_rate": 3.797641073080481e-05,
|
| 389 |
+
"loss": 1.2948,
|
| 390 |
"step": 5200
|
| 391 |
},
|
| 392 |
{
|
| 393 |
+
"epoch": 1.2257169287696577,
|
| 394 |
+
"grad_norm": 26.262020111083984,
|
| 395 |
+
"learning_rate": 3.774514338575393e-05,
|
| 396 |
+
"loss": 1.3486,
|
| 397 |
"step": 5300
|
| 398 |
},
|
| 399 |
{
|
| 400 |
+
"epoch": 1.2488436632747457,
|
| 401 |
+
"grad_norm": 21.114444732666016,
|
| 402 |
+
"learning_rate": 3.751387604070306e-05,
|
| 403 |
+
"loss": 1.3155,
|
| 404 |
"step": 5400
|
| 405 |
},
|
| 406 |
{
|
| 407 |
+
"epoch": 1.2719703977798336,
|
| 408 |
+
"grad_norm": 10.668533325195312,
|
| 409 |
+
"learning_rate": 3.728260869565218e-05,
|
| 410 |
+
"loss": 1.2761,
|
| 411 |
"step": 5500
|
| 412 |
},
|
| 413 |
{
|
| 414 |
+
"epoch": 1.2950971322849214,
|
| 415 |
+
"grad_norm": 12.978555679321289,
|
| 416 |
+
"learning_rate": 3.70513413506013e-05,
|
| 417 |
+
"loss": 1.2541,
|
| 418 |
"step": 5600
|
| 419 |
},
|
| 420 |
{
|
| 421 |
+
"epoch": 1.3182238667900092,
|
| 422 |
+
"grad_norm": 13.066253662109375,
|
| 423 |
+
"learning_rate": 3.682007400555042e-05,
|
| 424 |
+
"loss": 1.2346,
|
| 425 |
"step": 5700
|
| 426 |
},
|
| 427 |
{
|
| 428 |
+
"epoch": 1.341350601295097,
|
| 429 |
+
"grad_norm": 11.974534034729004,
|
| 430 |
+
"learning_rate": 3.658880666049954e-05,
|
| 431 |
+
"loss": 1.2285,
|
| 432 |
"step": 5800
|
| 433 |
},
|
| 434 |
{
|
| 435 |
+
"epoch": 1.364477335800185,
|
| 436 |
+
"grad_norm": 12.733871459960938,
|
| 437 |
+
"learning_rate": 3.6357539315448655e-05,
|
| 438 |
+
"loss": 1.2013,
|
| 439 |
"step": 5900
|
| 440 |
},
|
| 441 |
{
|
| 442 |
+
"epoch": 1.3876040703052728,
|
| 443 |
+
"grad_norm": 13.184294700622559,
|
| 444 |
+
"learning_rate": 3.612627197039778e-05,
|
| 445 |
+
"loss": 1.1986,
|
| 446 |
"step": 6000
|
| 447 |
},
|
| 448 |
{
|
| 449 |
+
"epoch": 1.4107308048103608,
|
| 450 |
+
"grad_norm": 10.159133911132812,
|
| 451 |
+
"learning_rate": 3.58950046253469e-05,
|
| 452 |
+
"loss": 1.1755,
|
| 453 |
"step": 6100
|
| 454 |
},
|
| 455 |
+
{
|
| 456 |
+
"epoch": 1.4338575393154487,
|
| 457 |
+
"grad_norm": 12.212769508361816,
|
| 458 |
+
"learning_rate": 3.566373728029602e-05,
|
| 459 |
+
"loss": 1.1937,
|
| 460 |
+
"step": 6200
|
| 461 |
+
},
|
| 462 |
+
{
|
| 463 |
+
"epoch": 1.4569842738205365,
|
| 464 |
+
"grad_norm": 15.231562614440918,
|
| 465 |
+
"learning_rate": 3.5432469935245146e-05,
|
| 466 |
+
"loss": 1.202,
|
| 467 |
+
"step": 6300
|
| 468 |
+
},
|
| 469 |
+
{
|
| 470 |
+
"epoch": 1.4801110083256244,
|
| 471 |
+
"grad_norm": 14.845349311828613,
|
| 472 |
+
"learning_rate": 3.5201202590194265e-05,
|
| 473 |
+
"loss": 1.1607,
|
| 474 |
+
"step": 6400
|
| 475 |
+
},
|
| 476 |
+
{
|
| 477 |
+
"epoch": 1.5032377428307124,
|
| 478 |
+
"grad_norm": 12.119277954101562,
|
| 479 |
+
"learning_rate": 3.4969935245143384e-05,
|
| 480 |
+
"loss": 1.2116,
|
| 481 |
+
"step": 6500
|
| 482 |
+
},
|
| 483 |
+
{
|
| 484 |
+
"epoch": 1.5263644773358003,
|
| 485 |
+
"grad_norm": 12.19071102142334,
|
| 486 |
+
"learning_rate": 3.473866790009251e-05,
|
| 487 |
+
"loss": 1.1797,
|
| 488 |
+
"step": 6600
|
| 489 |
+
},
|
| 490 |
+
{
|
| 491 |
+
"epoch": 1.5494912118408881,
|
| 492 |
+
"grad_norm": 12.373006820678711,
|
| 493 |
+
"learning_rate": 3.450740055504163e-05,
|
| 494 |
+
"loss": 1.1571,
|
| 495 |
+
"step": 6700
|
| 496 |
+
},
|
| 497 |
+
{
|
| 498 |
+
"epoch": 1.572617946345976,
|
| 499 |
+
"grad_norm": 14.413765907287598,
|
| 500 |
+
"learning_rate": 3.427613320999075e-05,
|
| 501 |
+
"loss": 1.1526,
|
| 502 |
+
"step": 6800
|
| 503 |
+
},
|
| 504 |
+
{
|
| 505 |
+
"epoch": 1.5957446808510638,
|
| 506 |
+
"grad_norm": 12.39973258972168,
|
| 507 |
+
"learning_rate": 3.4044865864939875e-05,
|
| 508 |
+
"loss": 1.1438,
|
| 509 |
+
"step": 6900
|
| 510 |
+
},
|
| 511 |
+
{
|
| 512 |
+
"epoch": 1.6188714153561516,
|
| 513 |
+
"grad_norm": 15.030281066894531,
|
| 514 |
+
"learning_rate": 3.3813598519888994e-05,
|
| 515 |
+
"loss": 1.1634,
|
| 516 |
+
"step": 7000
|
| 517 |
+
},
|
| 518 |
+
{
|
| 519 |
+
"epoch": 1.6419981498612395,
|
| 520 |
+
"grad_norm": 15.445487022399902,
|
| 521 |
+
"learning_rate": 3.3582331174838114e-05,
|
| 522 |
+
"loss": 1.1367,
|
| 523 |
+
"step": 7100
|
| 524 |
+
},
|
| 525 |
+
{
|
| 526 |
+
"epoch": 1.6651248843663273,
|
| 527 |
+
"grad_norm": 15.726140022277832,
|
| 528 |
+
"learning_rate": 3.335106382978724e-05,
|
| 529 |
+
"loss": 1.1133,
|
| 530 |
+
"step": 7200
|
| 531 |
+
},
|
| 532 |
+
{
|
| 533 |
+
"epoch": 1.6882516188714154,
|
| 534 |
+
"grad_norm": 12.940756797790527,
|
| 535 |
+
"learning_rate": 3.311979648473636e-05,
|
| 536 |
+
"loss": 1.1156,
|
| 537 |
+
"step": 7300
|
| 538 |
+
},
|
| 539 |
+
{
|
| 540 |
+
"epoch": 1.7113783533765032,
|
| 541 |
+
"grad_norm": 14.717055320739746,
|
| 542 |
+
"learning_rate": 3.288852913968548e-05,
|
| 543 |
+
"loss": 1.1102,
|
| 544 |
+
"step": 7400
|
| 545 |
+
},
|
| 546 |
+
{
|
| 547 |
+
"epoch": 1.734505087881591,
|
| 548 |
+
"grad_norm": 12.775771141052246,
|
| 549 |
+
"learning_rate": 3.26572617946346e-05,
|
| 550 |
+
"loss": 1.1123,
|
| 551 |
+
"step": 7500
|
| 552 |
+
},
|
| 553 |
+
{
|
| 554 |
+
"epoch": 1.7576318223866791,
|
| 555 |
+
"grad_norm": 12.529901504516602,
|
| 556 |
+
"learning_rate": 3.242599444958372e-05,
|
| 557 |
+
"loss": 1.1066,
|
| 558 |
+
"step": 7600
|
| 559 |
+
},
|
| 560 |
+
{
|
| 561 |
+
"epoch": 1.780758556891767,
|
| 562 |
+
"grad_norm": 12.506126403808594,
|
| 563 |
+
"learning_rate": 3.219472710453284e-05,
|
| 564 |
+
"loss": 1.1291,
|
| 565 |
+
"step": 7700
|
| 566 |
+
},
|
| 567 |
+
{
|
| 568 |
+
"epoch": 1.8038852913968548,
|
| 569 |
+
"grad_norm": 16.9326114654541,
|
| 570 |
+
"learning_rate": 3.196345975948196e-05,
|
| 571 |
+
"loss": 1.1094,
|
| 572 |
+
"step": 7800
|
| 573 |
+
},
|
| 574 |
+
{
|
| 575 |
+
"epoch": 1.8270120259019427,
|
| 576 |
+
"grad_norm": 12.257229804992676,
|
| 577 |
+
"learning_rate": 3.173219241443108e-05,
|
| 578 |
+
"loss": 1.094,
|
| 579 |
+
"step": 7900
|
| 580 |
+
},
|
| 581 |
+
{
|
| 582 |
+
"epoch": 1.8501387604070305,
|
| 583 |
+
"grad_norm": 15.346363067626953,
|
| 584 |
+
"learning_rate": 3.150092506938021e-05,
|
| 585 |
+
"loss": 1.1585,
|
| 586 |
+
"step": 8000
|
| 587 |
+
},
|
| 588 |
+
{
|
| 589 |
+
"epoch": 1.8732654949121184,
|
| 590 |
+
"grad_norm": 12.800529479980469,
|
| 591 |
+
"learning_rate": 3.126965772432933e-05,
|
| 592 |
+
"loss": 1.077,
|
| 593 |
+
"step": 8100
|
| 594 |
+
},
|
| 595 |
+
{
|
| 596 |
+
"epoch": 1.8963922294172062,
|
| 597 |
+
"grad_norm": 14.989642143249512,
|
| 598 |
+
"learning_rate": 3.103839037927845e-05,
|
| 599 |
+
"loss": 1.108,
|
| 600 |
+
"step": 8200
|
| 601 |
+
},
|
| 602 |
+
{
|
| 603 |
+
"epoch": 1.919518963922294,
|
| 604 |
+
"grad_norm": 14.930624008178711,
|
| 605 |
+
"learning_rate": 3.080712303422757e-05,
|
| 606 |
+
"loss": 1.1431,
|
| 607 |
+
"step": 8300
|
| 608 |
+
},
|
| 609 |
+
{
|
| 610 |
+
"epoch": 1.942645698427382,
|
| 611 |
+
"grad_norm": 13.488320350646973,
|
| 612 |
+
"learning_rate": 3.057585568917669e-05,
|
| 613 |
+
"loss": 1.0784,
|
| 614 |
+
"step": 8400
|
| 615 |
+
},
|
| 616 |
+
{
|
| 617 |
+
"epoch": 1.96577243293247,
|
| 618 |
+
"grad_norm": 12.30305004119873,
|
| 619 |
+
"learning_rate": 3.034458834412581e-05,
|
| 620 |
+
"loss": 1.0834,
|
| 621 |
+
"step": 8500
|
| 622 |
+
},
|
| 623 |
+
{
|
| 624 |
+
"epoch": 1.9888991674375578,
|
| 625 |
+
"grad_norm": 13.152817726135254,
|
| 626 |
+
"learning_rate": 3.0113320999074934e-05,
|
| 627 |
+
"loss": 1.1268,
|
| 628 |
+
"step": 8600
|
| 629 |
+
},
|
| 630 |
+
{
|
| 631 |
+
"epoch": 2.0,
|
| 632 |
+
"eval_cosine_accuracy": 0.9209068037391286,
|
| 633 |
+
"eval_cosine_accuracy_threshold": 0.8271753191947937,
|
| 634 |
+
"eval_cosine_ap": 0.8450278341834471,
|
| 635 |
+
"eval_cosine_f1": 0.7521474761123443,
|
| 636 |
+
"eval_cosine_f1_threshold": 0.8189181089401245,
|
| 637 |
+
"eval_cosine_mcc": 0.701977809831327,
|
| 638 |
+
"eval_cosine_precision": 0.7438907980145093,
|
| 639 |
+
"eval_cosine_recall": 0.7605894983408159,
|
| 640 |
+
"eval_loss": 9.69921588897705,
|
| 641 |
+
"eval_runtime": 1472.4233,
|
| 642 |
+
"eval_samples_per_second": 125.255,
|
| 643 |
+
"eval_steps_per_second": 1.957,
|
| 644 |
+
"step": 8648
|
| 645 |
+
},
|
| 646 |
+
{
|
| 647 |
+
"epoch": 2.012025901942646,
|
| 648 |
+
"grad_norm": 15.71314811706543,
|
| 649 |
+
"learning_rate": 2.9882053654024057e-05,
|
| 650 |
+
"loss": 1.0443,
|
| 651 |
+
"step": 8700
|
| 652 |
+
},
|
| 653 |
+
{
|
| 654 |
+
"epoch": 2.0351526364477337,
|
| 655 |
+
"grad_norm": 14.261523246765137,
|
| 656 |
+
"learning_rate": 2.9650786308973173e-05,
|
| 657 |
+
"loss": 0.9715,
|
| 658 |
+
"step": 8800
|
| 659 |
+
},
|
| 660 |
+
{
|
| 661 |
+
"epoch": 2.0582793709528215,
|
| 662 |
+
"grad_norm": 13.405384063720703,
|
| 663 |
+
"learning_rate": 2.9419518963922292e-05,
|
| 664 |
+
"loss": 0.957,
|
| 665 |
+
"step": 8900
|
| 666 |
+
},
|
| 667 |
+
{
|
| 668 |
+
"epoch": 2.0814061054579094,
|
| 669 |
+
"grad_norm": 13.5853853225708,
|
| 670 |
+
"learning_rate": 2.9188251618871415e-05,
|
| 671 |
+
"loss": 0.9784,
|
| 672 |
+
"step": 9000
|
| 673 |
+
},
|
| 674 |
+
{
|
| 675 |
+
"epoch": 2.1045328399629972,
|
| 676 |
+
"grad_norm": 14.918572425842285,
|
| 677 |
+
"learning_rate": 2.8956984273820538e-05,
|
| 678 |
+
"loss": 0.9581,
|
| 679 |
+
"step": 9100
|
| 680 |
+
},
|
| 681 |
+
{
|
| 682 |
+
"epoch": 2.127659574468085,
|
| 683 |
+
"grad_norm": 13.354079246520996,
|
| 684 |
+
"learning_rate": 2.8725716928769657e-05,
|
| 685 |
+
"loss": 0.9569,
|
| 686 |
+
"step": 9200
|
| 687 |
+
},
|
| 688 |
+
{
|
| 689 |
+
"epoch": 2.150786308973173,
|
| 690 |
+
"grad_norm": 15.024975776672363,
|
| 691 |
+
"learning_rate": 2.849444958371878e-05,
|
| 692 |
+
"loss": 0.9518,
|
| 693 |
+
"step": 9300
|
| 694 |
+
},
|
| 695 |
+
{
|
| 696 |
+
"epoch": 2.1739130434782608,
|
| 697 |
+
"grad_norm": 13.5723876953125,
|
| 698 |
+
"learning_rate": 2.8263182238667902e-05,
|
| 699 |
+
"loss": 0.9485,
|
| 700 |
+
"step": 9400
|
| 701 |
+
},
|
| 702 |
+
{
|
| 703 |
+
"epoch": 2.1970397779833486,
|
| 704 |
+
"grad_norm": 12.383338928222656,
|
| 705 |
+
"learning_rate": 2.8031914893617022e-05,
|
| 706 |
+
"loss": 0.9433,
|
| 707 |
+
"step": 9500
|
| 708 |
+
},
|
| 709 |
+
{
|
| 710 |
+
"epoch": 2.2201665124884364,
|
| 711 |
+
"grad_norm": 13.54079532623291,
|
| 712 |
+
"learning_rate": 2.7800647548566144e-05,
|
| 713 |
+
"loss": 0.9392,
|
| 714 |
+
"step": 9600
|
| 715 |
+
},
|
| 716 |
+
{
|
| 717 |
+
"epoch": 2.2432932469935247,
|
| 718 |
+
"grad_norm": 15.325400352478027,
|
| 719 |
+
"learning_rate": 2.7569380203515267e-05,
|
| 720 |
+
"loss": 0.9248,
|
| 721 |
+
"step": 9700
|
| 722 |
+
},
|
| 723 |
+
{
|
| 724 |
+
"epoch": 2.2664199814986126,
|
| 725 |
+
"grad_norm": 15.651702880859375,
|
| 726 |
+
"learning_rate": 2.7338112858464387e-05,
|
| 727 |
+
"loss": 0.9105,
|
| 728 |
+
"step": 9800
|
| 729 |
+
},
|
| 730 |
+
{
|
| 731 |
+
"epoch": 2.2895467160037004,
|
| 732 |
+
"grad_norm": 15.196932792663574,
|
| 733 |
+
"learning_rate": 2.710684551341351e-05,
|
| 734 |
+
"loss": 0.9769,
|
| 735 |
+
"step": 9900
|
| 736 |
+
},
|
| 737 |
+
{
|
| 738 |
+
"epoch": 2.3126734505087883,
|
| 739 |
+
"grad_norm": 13.820411682128906,
|
| 740 |
+
"learning_rate": 2.6875578168362632e-05,
|
| 741 |
+
"loss": 0.9502,
|
| 742 |
+
"step": 10000
|
| 743 |
+
},
|
| 744 |
+
{
|
| 745 |
+
"epoch": 2.335800185013876,
|
| 746 |
+
"grad_norm": 15.181200981140137,
|
| 747 |
+
"learning_rate": 2.664431082331175e-05,
|
| 748 |
+
"loss": 0.9604,
|
| 749 |
+
"step": 10100
|
| 750 |
+
},
|
| 751 |
+
{
|
| 752 |
+
"epoch": 2.358926919518964,
|
| 753 |
+
"grad_norm": 17.202287673950195,
|
| 754 |
+
"learning_rate": 2.6413043478260867e-05,
|
| 755 |
+
"loss": 0.9291,
|
| 756 |
+
"step": 10200
|
| 757 |
+
},
|
| 758 |
+
{
|
| 759 |
+
"epoch": 2.382053654024052,
|
| 760 |
+
"grad_norm": 16.69881248474121,
|
| 761 |
+
"learning_rate": 2.618177613320999e-05,
|
| 762 |
+
"loss": 0.9552,
|
| 763 |
+
"step": 10300
|
| 764 |
+
},
|
| 765 |
+
{
|
| 766 |
+
"epoch": 2.4051803885291396,
|
| 767 |
+
"grad_norm": 16.424978256225586,
|
| 768 |
+
"learning_rate": 2.5950508788159113e-05,
|
| 769 |
+
"loss": 0.9621,
|
| 770 |
+
"step": 10400
|
| 771 |
+
},
|
| 772 |
+
{
|
| 773 |
+
"epoch": 2.4283071230342275,
|
| 774 |
+
"grad_norm": 17.192264556884766,
|
| 775 |
+
"learning_rate": 2.5719241443108232e-05,
|
| 776 |
+
"loss": 0.9357,
|
| 777 |
+
"step": 10500
|
| 778 |
+
},
|
| 779 |
+
{
|
| 780 |
+
"epoch": 2.4514338575393153,
|
| 781 |
+
"grad_norm": 16.9178524017334,
|
| 782 |
+
"learning_rate": 2.5487974098057355e-05,
|
| 783 |
+
"loss": 0.9323,
|
| 784 |
+
"step": 10600
|
| 785 |
+
},
|
| 786 |
+
{
|
| 787 |
+
"epoch": 2.474560592044403,
|
| 788 |
+
"grad_norm": 11.07499885559082,
|
| 789 |
+
"learning_rate": 2.5256706753006477e-05,
|
| 790 |
+
"loss": 0.9327,
|
| 791 |
+
"step": 10700
|
| 792 |
+
},
|
| 793 |
+
{
|
| 794 |
+
"epoch": 2.4976873265494914,
|
| 795 |
+
"grad_norm": 28.121694564819336,
|
| 796 |
+
"learning_rate": 2.5025439407955597e-05,
|
| 797 |
+
"loss": 0.9067,
|
| 798 |
+
"step": 10800
|
| 799 |
+
},
|
| 800 |
+
{
|
| 801 |
+
"epoch": 2.520814061054579,
|
| 802 |
+
"grad_norm": 15.262548446655273,
|
| 803 |
+
"learning_rate": 2.479417206290472e-05,
|
| 804 |
+
"loss": 0.9411,
|
| 805 |
+
"step": 10900
|
| 806 |
+
},
|
| 807 |
+
{
|
| 808 |
+
"epoch": 2.543940795559667,
|
| 809 |
+
"grad_norm": 16.59368133544922,
|
| 810 |
+
"learning_rate": 2.4562904717853842e-05,
|
| 811 |
+
"loss": 0.9305,
|
| 812 |
+
"step": 11000
|
| 813 |
+
},
|
| 814 |
+
{
|
| 815 |
+
"epoch": 2.567067530064755,
|
| 816 |
+
"grad_norm": 12.39183235168457,
|
| 817 |
+
"learning_rate": 2.433163737280296e-05,
|
| 818 |
+
"loss": 0.9378,
|
| 819 |
+
"step": 11100
|
| 820 |
+
},
|
| 821 |
+
{
|
| 822 |
+
"epoch": 2.590194264569843,
|
| 823 |
+
"grad_norm": 19.034120559692383,
|
| 824 |
+
"learning_rate": 2.4100370027752084e-05,
|
| 825 |
+
"loss": 0.9171,
|
| 826 |
+
"step": 11200
|
| 827 |
+
},
|
| 828 |
+
{
|
| 829 |
+
"epoch": 2.6133209990749307,
|
| 830 |
+
"grad_norm": 16.727380752563477,
|
| 831 |
+
"learning_rate": 2.3869102682701204e-05,
|
| 832 |
+
"loss": 0.9074,
|
| 833 |
+
"step": 11300
|
| 834 |
+
},
|
| 835 |
+
{
|
| 836 |
+
"epoch": 2.6364477335800185,
|
| 837 |
+
"grad_norm": 17.257230758666992,
|
| 838 |
+
"learning_rate": 2.3637835337650323e-05,
|
| 839 |
+
"loss": 0.9262,
|
| 840 |
+
"step": 11400
|
| 841 |
+
},
|
| 842 |
+
{
|
| 843 |
+
"epoch": 2.6595744680851063,
|
| 844 |
+
"grad_norm": 18.956735610961914,
|
| 845 |
+
"learning_rate": 2.3406567992599446e-05,
|
| 846 |
+
"loss": 0.9063,
|
| 847 |
+
"step": 11500
|
| 848 |
+
},
|
| 849 |
+
{
|
| 850 |
+
"epoch": 2.682701202590194,
|
| 851 |
+
"grad_norm": 15.463052749633789,
|
| 852 |
+
"learning_rate": 2.317530064754857e-05,
|
| 853 |
+
"loss": 0.8814,
|
| 854 |
+
"step": 11600
|
| 855 |
+
},
|
| 856 |
+
{
|
| 857 |
+
"epoch": 2.705827937095282,
|
| 858 |
+
"grad_norm": 13.263724327087402,
|
| 859 |
+
"learning_rate": 2.2944033302497688e-05,
|
| 860 |
+
"loss": 0.9089,
|
| 861 |
+
"step": 11700
|
| 862 |
+
},
|
| 863 |
+
{
|
| 864 |
+
"epoch": 2.72895467160037,
|
| 865 |
+
"grad_norm": 15.596953392028809,
|
| 866 |
+
"learning_rate": 2.271276595744681e-05,
|
| 867 |
+
"loss": 0.9048,
|
| 868 |
+
"step": 11800
|
| 869 |
+
},
|
| 870 |
+
{
|
| 871 |
+
"epoch": 2.752081406105458,
|
| 872 |
+
"grad_norm": 16.61760711669922,
|
| 873 |
+
"learning_rate": 2.2481498612395933e-05,
|
| 874 |
+
"loss": 0.9268,
|
| 875 |
+
"step": 11900
|
| 876 |
+
},
|
| 877 |
+
{
|
| 878 |
+
"epoch": 2.7752081406105455,
|
| 879 |
+
"grad_norm": 16.409652709960938,
|
| 880 |
+
"learning_rate": 2.2250231267345052e-05,
|
| 881 |
+
"loss": 0.8913,
|
| 882 |
+
"step": 12000
|
| 883 |
+
},
|
| 884 |
+
{
|
| 885 |
+
"epoch": 2.798334875115634,
|
| 886 |
+
"grad_norm": 14.509052276611328,
|
| 887 |
+
"learning_rate": 2.2018963922294172e-05,
|
| 888 |
+
"loss": 0.9064,
|
| 889 |
+
"step": 12100
|
| 890 |
+
},
|
| 891 |
+
{
|
| 892 |
+
"epoch": 2.8214616096207217,
|
| 893 |
+
"grad_norm": 13.914097785949707,
|
| 894 |
+
"learning_rate": 2.1787696577243295e-05,
|
| 895 |
+
"loss": 0.8585,
|
| 896 |
+
"step": 12200
|
| 897 |
+
},
|
| 898 |
+
{
|
| 899 |
+
"epoch": 2.8445883441258095,
|
| 900 |
+
"grad_norm": 13.852704048156738,
|
| 901 |
+
"learning_rate": 2.1556429232192414e-05,
|
| 902 |
+
"loss": 0.878,
|
| 903 |
+
"step": 12300
|
| 904 |
+
},
|
| 905 |
+
{
|
| 906 |
+
"epoch": 2.8677150786308974,
|
| 907 |
+
"grad_norm": 15.329246520996094,
|
| 908 |
+
"learning_rate": 2.1325161887141537e-05,
|
| 909 |
+
"loss": 0.8612,
|
| 910 |
+
"step": 12400
|
| 911 |
+
},
|
| 912 |
+
{
|
| 913 |
+
"epoch": 2.890841813135985,
|
| 914 |
+
"grad_norm": 15.046784400939941,
|
| 915 |
+
"learning_rate": 2.109389454209066e-05,
|
| 916 |
+
"loss": 0.8799,
|
| 917 |
+
"step": 12500
|
| 918 |
+
},
|
| 919 |
+
{
|
| 920 |
+
"epoch": 2.913968547641073,
|
| 921 |
+
"grad_norm": 15.824640274047852,
|
| 922 |
+
"learning_rate": 2.086262719703978e-05,
|
| 923 |
+
"loss": 0.8541,
|
| 924 |
+
"step": 12600
|
| 925 |
+
},
|
| 926 |
+
{
|
| 927 |
+
"epoch": 2.937095282146161,
|
| 928 |
+
"grad_norm": 15.991775512695312,
|
| 929 |
+
"learning_rate": 2.0631359851988898e-05,
|
| 930 |
+
"loss": 0.8521,
|
| 931 |
+
"step": 12700
|
| 932 |
+
},
|
| 933 |
+
{
|
| 934 |
+
"epoch": 2.9602220166512487,
|
| 935 |
+
"grad_norm": 16.13051986694336,
|
| 936 |
+
"learning_rate": 2.040009250693802e-05,
|
| 937 |
+
"loss": 0.8582,
|
| 938 |
+
"step": 12800
|
| 939 |
+
},
|
| 940 |
+
{
|
| 941 |
+
"epoch": 2.9833487511563366,
|
| 942 |
+
"grad_norm": 14.818473815917969,
|
| 943 |
+
"learning_rate": 2.0168825161887143e-05,
|
| 944 |
+
"loss": 0.869,
|
| 945 |
+
"step": 12900
|
| 946 |
+
},
|
| 947 |
{
|
| 948 |
"epoch": 3.0,
|
| 949 |
+
"eval_cosine_accuracy": 0.9225280326197758,
|
| 950 |
+
"eval_cosine_accuracy_threshold": 0.7901061773300171,
|
| 951 |
+
"eval_cosine_ap": 0.8478615501518483,
|
| 952 |
+
"eval_cosine_f1": 0.7559554803436604,
|
| 953 |
+
"eval_cosine_f1_threshold": 0.7817596793174744,
|
| 954 |
+
"eval_cosine_mcc": 0.7071656901034916,
|
| 955 |
+
"eval_cosine_precision": 0.756201575623413,
|
| 956 |
+
"eval_cosine_recall": 0.7557095451883662,
|
| 957 |
+
"eval_loss": 10.411548614501953,
|
| 958 |
+
"eval_runtime": 1486.9605,
|
| 959 |
+
"eval_samples_per_second": 124.03,
|
| 960 |
+
"eval_steps_per_second": 1.938,
|
| 961 |
+
"step": 12972
|
| 962 |
}
|
| 963 |
],
|
| 964 |
"logging_steps": 100,
|
| 965 |
+
"max_steps": 21620,
|
| 966 |
"num_input_tokens_seen": 0,
|
| 967 |
"num_train_epochs": 5,
|
| 968 |
"save_steps": 500,
|
|
|
|
| 979 |
}
|
| 980 |
},
|
| 981 |
"total_flos": 0.0,
|
| 982 |
+
"train_batch_size": 64,
|
| 983 |
"trial_name": null,
|
| 984 |
"trial_params": null
|
| 985 |
}
|
training_args.bin
CHANGED
|
@@ -1,3 +1,3 @@
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
-
oid sha256:
|
| 3 |
size 6097
|
|
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:05569020741fe60e0a967a0b69b87b45af12fa4c5d912666a335c2b0eea34235
|
| 3 |
size 6097
|