Training in progress, step 107, checkpoint
Browse files- checkpoint-107/README.md +273 -93
- checkpoint-107/optimizer.pt +1 -1
- checkpoint-107/pytorch_model.bin +1 -1
- checkpoint-107/rng_state.pth +1 -1
- checkpoint-107/scheduler.pt +1 -1
- checkpoint-107/trainer_state.json +455 -455
- checkpoint-107/training_args.bin +1 -1
checkpoint-107/README.md
CHANGED
|
@@ -164,34 +164,34 @@ model-index:
|
|
| 164 |
type: sts-test
|
| 165 |
metrics:
|
| 166 |
- type: pearson_cosine
|
| 167 |
-
value: 0.
|
| 168 |
name: Pearson Cosine
|
| 169 |
- type: spearman_cosine
|
| 170 |
-
value: 0.
|
| 171 |
name: Spearman Cosine
|
| 172 |
- type: pearson_manhattan
|
| 173 |
-
value: 0.
|
| 174 |
name: Pearson Manhattan
|
| 175 |
- type: spearman_manhattan
|
| 176 |
-
value: 0.
|
| 177 |
name: Spearman Manhattan
|
| 178 |
- type: pearson_euclidean
|
| 179 |
-
value: 0.
|
| 180 |
name: Pearson Euclidean
|
| 181 |
- type: spearman_euclidean
|
| 182 |
-
value: 0.
|
| 183 |
name: Spearman Euclidean
|
| 184 |
- type: pearson_dot
|
| 185 |
-
value: 0.
|
| 186 |
name: Pearson Dot
|
| 187 |
- type: spearman_dot
|
| 188 |
-
value: 0.
|
| 189 |
name: Spearman Dot
|
| 190 |
- type: pearson_max
|
| 191 |
-
value: 0.
|
| 192 |
name: Pearson Max
|
| 193 |
- type: spearman_max
|
| 194 |
-
value: 0.
|
| 195 |
name: Spearman Max
|
| 196 |
- task:
|
| 197 |
type: triplet
|
|
@@ -223,109 +223,109 @@ model-index:
|
|
| 223 |
type: VitaminC
|
| 224 |
metrics:
|
| 225 |
- type: cosine_accuracy
|
| 226 |
-
value: 0.
|
| 227 |
name: Cosine Accuracy
|
| 228 |
- type: cosine_accuracy_threshold
|
| 229 |
-
value: 0.
|
| 230 |
name: Cosine Accuracy Threshold
|
| 231 |
- type: cosine_f1
|
| 232 |
-
value: 0.
|
| 233 |
name: Cosine F1
|
| 234 |
- type: cosine_f1_threshold
|
| 235 |
-
value: 0.
|
| 236 |
name: Cosine F1 Threshold
|
| 237 |
- type: cosine_precision
|
| 238 |
-
value: 0.
|
| 239 |
name: Cosine Precision
|
| 240 |
- type: cosine_recall
|
| 241 |
value: 1.0
|
| 242 |
name: Cosine Recall
|
| 243 |
- type: cosine_ap
|
| 244 |
-
value: 0.
|
| 245 |
name: Cosine Ap
|
| 246 |
- type: dot_accuracy
|
| 247 |
-
value: 0.
|
| 248 |
name: Dot Accuracy
|
| 249 |
- type: dot_accuracy_threshold
|
| 250 |
-
value:
|
| 251 |
name: Dot Accuracy Threshold
|
| 252 |
- type: dot_f1
|
| 253 |
-
value: 0.
|
| 254 |
name: Dot F1
|
| 255 |
- type: dot_f1_threshold
|
| 256 |
-
value:
|
| 257 |
name: Dot F1 Threshold
|
| 258 |
- type: dot_precision
|
| 259 |
-
value: 0.
|
| 260 |
name: Dot Precision
|
| 261 |
- type: dot_recall
|
| 262 |
value: 1.0
|
| 263 |
name: Dot Recall
|
| 264 |
- type: dot_ap
|
| 265 |
-
value: 0.
|
| 266 |
name: Dot Ap
|
| 267 |
- type: manhattan_accuracy
|
| 268 |
-
value: 0.
|
| 269 |
name: Manhattan Accuracy
|
| 270 |
- type: manhattan_accuracy_threshold
|
| 271 |
-
value:
|
| 272 |
name: Manhattan Accuracy Threshold
|
| 273 |
- type: manhattan_f1
|
| 274 |
-
value: 0.
|
| 275 |
name: Manhattan F1
|
| 276 |
- type: manhattan_f1_threshold
|
| 277 |
-
value:
|
| 278 |
name: Manhattan F1 Threshold
|
| 279 |
- type: manhattan_precision
|
| 280 |
-
value: 0.
|
| 281 |
name: Manhattan Precision
|
| 282 |
- type: manhattan_recall
|
| 283 |
-
value:
|
| 284 |
name: Manhattan Recall
|
| 285 |
- type: manhattan_ap
|
| 286 |
-
value: 0.
|
| 287 |
name: Manhattan Ap
|
| 288 |
- type: euclidean_accuracy
|
| 289 |
-
value: 0.
|
| 290 |
name: Euclidean Accuracy
|
| 291 |
- type: euclidean_accuracy_threshold
|
| 292 |
-
value:
|
| 293 |
name: Euclidean Accuracy Threshold
|
| 294 |
- type: euclidean_f1
|
| 295 |
-
value: 0.
|
| 296 |
name: Euclidean F1
|
| 297 |
- type: euclidean_f1_threshold
|
| 298 |
-
value:
|
| 299 |
name: Euclidean F1 Threshold
|
| 300 |
- type: euclidean_precision
|
| 301 |
-
value: 0.
|
| 302 |
name: Euclidean Precision
|
| 303 |
- type: euclidean_recall
|
| 304 |
value: 1.0
|
| 305 |
name: Euclidean Recall
|
| 306 |
- type: euclidean_ap
|
| 307 |
-
value: 0.
|
| 308 |
name: Euclidean Ap
|
| 309 |
- type: max_accuracy
|
| 310 |
-
value: 0.
|
| 311 |
name: Max Accuracy
|
| 312 |
- type: max_accuracy_threshold
|
| 313 |
-
value:
|
| 314 |
name: Max Accuracy Threshold
|
| 315 |
- type: max_f1
|
| 316 |
-
value: 0.
|
| 317 |
name: Max F1
|
| 318 |
- type: max_f1_threshold
|
| 319 |
-
value:
|
| 320 |
name: Max F1 Threshold
|
| 321 |
- type: max_precision
|
| 322 |
-
value: 0.
|
| 323 |
name: Max Precision
|
| 324 |
- type: max_recall
|
| 325 |
value: 1.0
|
| 326 |
name: Max Recall
|
| 327 |
- type: max_ap
|
| 328 |
-
value: 0.
|
| 329 |
name: Max Ap
|
| 330 |
---
|
| 331 |
|
|
@@ -388,7 +388,7 @@ Then you can load this model and run inference.
|
|
| 388 |
from sentence_transformers import SentenceTransformer
|
| 389 |
|
| 390 |
# Download from the 🤗 Hub
|
| 391 |
-
model = SentenceTransformer("bobox/DeBERTa-small-ST-v1-toytest
|
| 392 |
# Run inference
|
| 393 |
sentences = [
|
| 394 |
'who did ben assault in home and away',
|
|
@@ -439,16 +439,16 @@ You can finetune this model on your own dataset.
|
|
| 439 |
|
| 440 |
| Metric | Value |
|
| 441 |
|:--------------------|:-----------|
|
| 442 |
-
| pearson_cosine | 0.
|
| 443 |
-
| **spearman_cosine** | **0.
|
| 444 |
-
| pearson_manhattan | 0.
|
| 445 |
-
| spearman_manhattan | 0.
|
| 446 |
-
| pearson_euclidean | 0.
|
| 447 |
-
| spearman_euclidean | 0.
|
| 448 |
-
| pearson_dot | 0.
|
| 449 |
-
| spearman_dot | 0.
|
| 450 |
-
| pearson_max | 0.
|
| 451 |
-
| spearman_max | 0.
|
| 452 |
|
| 453 |
#### Triplet
|
| 454 |
* Dataset: `NLI-v2`
|
|
@@ -468,41 +468,41 @@ You can finetune this model on your own dataset.
|
|
| 468 |
|
| 469 |
| Metric | Value |
|
| 470 |
|:-----------------------------|:-----------|
|
| 471 |
-
| cosine_accuracy | 0.
|
| 472 |
-
| cosine_accuracy_threshold | 0.
|
| 473 |
-
| cosine_f1 | 0.
|
| 474 |
-
| cosine_f1_threshold | 0.
|
| 475 |
-
| cosine_precision | 0.
|
| 476 |
| cosine_recall | 1.0 |
|
| 477 |
-
| cosine_ap | 0.
|
| 478 |
-
| dot_accuracy | 0.
|
| 479 |
-
| dot_accuracy_threshold |
|
| 480 |
-
| dot_f1 | 0.
|
| 481 |
-
| dot_f1_threshold |
|
| 482 |
-
| dot_precision | 0.
|
| 483 |
| dot_recall | 1.0 |
|
| 484 |
-
| dot_ap | 0.
|
| 485 |
-
| manhattan_accuracy | 0.
|
| 486 |
-
| manhattan_accuracy_threshold |
|
| 487 |
-
| manhattan_f1 | 0.
|
| 488 |
-
| manhattan_f1_threshold |
|
| 489 |
-
| manhattan_precision | 0.
|
| 490 |
-
| manhattan_recall |
|
| 491 |
-
| manhattan_ap | 0.
|
| 492 |
-
| euclidean_accuracy | 0.
|
| 493 |
-
| euclidean_accuracy_threshold |
|
| 494 |
-
| euclidean_f1 | 0.
|
| 495 |
-
| euclidean_f1_threshold |
|
| 496 |
-
| euclidean_precision | 0.
|
| 497 |
| euclidean_recall | 1.0 |
|
| 498 |
-
| euclidean_ap | 0.
|
| 499 |
-
| max_accuracy | 0.
|
| 500 |
-
| max_accuracy_threshold |
|
| 501 |
-
| max_f1 | 0.
|
| 502 |
-
| max_f1_threshold |
|
| 503 |
-
| max_precision | 0.
|
| 504 |
| max_recall | 1.0 |
|
| 505 |
-
| **max_ap** | **0.
|
| 506 |
|
| 507 |
<!--
|
| 508 |
## Bias, Risks and Limitations
|
|
@@ -1151,14 +1151,14 @@ You can finetune this model on your own dataset.
|
|
| 1151 |
#### Non-Default Hyperparameters
|
| 1152 |
|
| 1153 |
- `eval_strategy`: steps
|
| 1154 |
-
- `per_device_train_batch_size`:
|
| 1155 |
- `per_device_eval_batch_size`: 64
|
| 1156 |
-
- `gradient_accumulation_steps`:
|
| 1157 |
- `learning_rate`: 4e-05
|
| 1158 |
-
- `weight_decay`:
|
| 1159 |
- `lr_scheduler_type`: cosine_with_min_lr
|
| 1160 |
-
- `lr_scheduler_kwargs`: {'num_cycles': 0.5, 'min_lr':
|
| 1161 |
-
- `warmup_ratio`: 0.
|
| 1162 |
- `save_safetensors`: False
|
| 1163 |
- `fp16`: True
|
| 1164 |
- `push_to_hub`: True
|
|
@@ -1173,14 +1173,14 @@ You can finetune this model on your own dataset.
|
|
| 1173 |
- `do_predict`: False
|
| 1174 |
- `eval_strategy`: steps
|
| 1175 |
- `prediction_loss_only`: True
|
| 1176 |
-
- `per_device_train_batch_size`:
|
| 1177 |
- `per_device_eval_batch_size`: 64
|
| 1178 |
- `per_gpu_train_batch_size`: None
|
| 1179 |
- `per_gpu_eval_batch_size`: None
|
| 1180 |
-
- `gradient_accumulation_steps`:
|
| 1181 |
- `eval_accumulation_steps`: None
|
| 1182 |
- `learning_rate`: 4e-05
|
| 1183 |
-
- `weight_decay`:
|
| 1184 |
- `adam_beta1`: 0.9
|
| 1185 |
- `adam_beta2`: 0.999
|
| 1186 |
- `adam_epsilon`: 1e-08
|
|
@@ -1188,8 +1188,8 @@ You can finetune this model on your own dataset.
|
|
| 1188 |
- `num_train_epochs`: 3
|
| 1189 |
- `max_steps`: -1
|
| 1190 |
- `lr_scheduler_type`: cosine_with_min_lr
|
| 1191 |
-
- `lr_scheduler_kwargs`: {'num_cycles': 0.5, 'min_lr':
|
| 1192 |
-
- `warmup_ratio`: 0.
|
| 1193 |
- `warmup_steps`: 0
|
| 1194 |
- `log_level`: passive
|
| 1195 |
- `log_level_replica`: warning
|
|
@@ -1282,6 +1282,8 @@ You can finetune this model on your own dataset.
|
|
| 1282 |
</details>
|
| 1283 |
|
| 1284 |
### Training Logs
|
|
|
|
|
|
|
| 1285 |
| Epoch | Step | Training Loss | vitaminc-pairs loss | trivia pairs loss | xsum-pairs loss | paws-pos loss | sciq pairs loss | msmarco pairs loss | openbookqa pairs loss | gooaq pairs loss | nq pairs loss | scitail-pairs-pos loss | qasc pairs loss | negation-triplets loss | NLI-v2_max_accuracy | VitaminC_max_ap | sts-test_spearman_cosine |
|
| 1286 |
|:------:|:----:|:-------------:|:-------------------:|:-----------------:|:---------------:|:-------------:|:---------------:|:------------------:|:---------------------:|:----------------:|:-------------:|:----------------------:|:---------------:|:----------------------:|:-------------------:|:---------------:|:------------------------:|
|
| 1287 |
| 0.0169 | 3 | 7.2372 | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - |
|
|
@@ -1319,7 +1321,185 @@ You can finetune this model on your own dataset.
|
|
| 1319 |
| 0.5589 | 99 | 2.1857 | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - |
|
| 1320 |
| 0.5759 | 102 | 1.8881 | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - |
|
| 1321 |
| 0.5928 | 105 | 2.2699 | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - |
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1322 |
|
|
|
|
| 1323 |
|
| 1324 |
### Framework Versions
|
| 1325 |
- Python: 3.10.13
|
|
|
|
| 164 |
type: sts-test
|
| 165 |
metrics:
|
| 166 |
- type: pearson_cosine
|
| 167 |
+
value: 0.8483316632682467
|
| 168 |
name: Pearson Cosine
|
| 169 |
- type: spearman_cosine
|
| 170 |
+
value: 0.8903503892346
|
| 171 |
name: Spearman Cosine
|
| 172 |
- type: pearson_manhattan
|
| 173 |
+
value: 0.8815226866327923
|
| 174 |
name: Pearson Manhattan
|
| 175 |
- type: spearman_manhattan
|
| 176 |
+
value: 0.8865568876827619
|
| 177 |
name: Spearman Manhattan
|
| 178 |
- type: pearson_euclidean
|
| 179 |
+
value: 0.8814283057813619
|
| 180 |
name: Pearson Euclidean
|
| 181 |
- type: spearman_euclidean
|
| 182 |
+
value: 0.8851830636663006
|
| 183 |
name: Spearman Euclidean
|
| 184 |
- type: pearson_dot
|
| 185 |
+
value: 0.8392403098680445
|
| 186 |
name: Pearson Dot
|
| 187 |
- type: spearman_dot
|
| 188 |
+
value: 0.857844431199042
|
| 189 |
name: Spearman Dot
|
| 190 |
- type: pearson_max
|
| 191 |
+
value: 0.8815226866327923
|
| 192 |
name: Pearson Max
|
| 193 |
- type: spearman_max
|
| 194 |
+
value: 0.8903503892346
|
| 195 |
name: Spearman Max
|
| 196 |
- task:
|
| 197 |
type: triplet
|
|
|
|
| 223 |
type: VitaminC
|
| 224 |
metrics:
|
| 225 |
- type: cosine_accuracy
|
| 226 |
+
value: 0.578125
|
| 227 |
name: Cosine Accuracy
|
| 228 |
- type: cosine_accuracy_threshold
|
| 229 |
+
value: 0.7859437465667725
|
| 230 |
name: Cosine Accuracy Threshold
|
| 231 |
- type: cosine_f1
|
| 232 |
+
value: 0.6595174262734584
|
| 233 |
name: Cosine F1
|
| 234 |
- type: cosine_f1_threshold
|
| 235 |
+
value: 0.3211573362350464
|
| 236 |
name: Cosine F1 Threshold
|
| 237 |
- type: cosine_precision
|
| 238 |
+
value: 0.492
|
| 239 |
name: Cosine Precision
|
| 240 |
- type: cosine_recall
|
| 241 |
value: 1.0
|
| 242 |
name: Cosine Recall
|
| 243 |
- type: cosine_ap
|
| 244 |
+
value: 0.5557444337961499
|
| 245 |
name: Cosine Ap
|
| 246 |
- type: dot_accuracy
|
| 247 |
+
value: 0.578125
|
| 248 |
name: Dot Accuracy
|
| 249 |
- type: dot_accuracy_threshold
|
| 250 |
+
value: 315.9444580078125
|
| 251 |
name: Dot Accuracy Threshold
|
| 252 |
- type: dot_f1
|
| 253 |
+
value: 0.6595174262734584
|
| 254 |
name: Dot F1
|
| 255 |
- type: dot_f1_threshold
|
| 256 |
+
value: 129.88558959960938
|
| 257 |
name: Dot F1 Threshold
|
| 258 |
- type: dot_precision
|
| 259 |
+
value: 0.492
|
| 260 |
name: Dot Precision
|
| 261 |
- type: dot_recall
|
| 262 |
value: 1.0
|
| 263 |
name: Dot Recall
|
| 264 |
- type: dot_ap
|
| 265 |
+
value: 0.5539524528858992
|
| 266 |
name: Dot Ap
|
| 267 |
- type: manhattan_accuracy
|
| 268 |
+
value: 0.578125
|
| 269 |
name: Manhattan Accuracy
|
| 270 |
- type: manhattan_accuracy_threshold
|
| 271 |
+
value: 276.40142822265625
|
| 272 |
name: Manhattan Accuracy Threshold
|
| 273 |
- type: manhattan_f1
|
| 274 |
+
value: 0.6576819407008085
|
| 275 |
name: Manhattan F1
|
| 276 |
- type: manhattan_f1_threshold
|
| 277 |
+
value: 469.7353515625
|
| 278 |
name: Manhattan F1 Threshold
|
| 279 |
- type: manhattan_precision
|
| 280 |
+
value: 0.49193548387096775
|
| 281 |
name: Manhattan Precision
|
| 282 |
- type: manhattan_recall
|
| 283 |
+
value: 0.991869918699187
|
| 284 |
name: Manhattan Recall
|
| 285 |
- type: manhattan_ap
|
| 286 |
+
value: 0.5429240708188645
|
| 287 |
name: Manhattan Ap
|
| 288 |
- type: euclidean_accuracy
|
| 289 |
+
value: 0.58203125
|
| 290 |
name: Euclidean Accuracy
|
| 291 |
- type: euclidean_accuracy_threshold
|
| 292 |
+
value: 13.113249778747559
|
| 293 |
name: Euclidean Accuracy Threshold
|
| 294 |
- type: euclidean_f1
|
| 295 |
+
value: 0.6577540106951871
|
| 296 |
name: Euclidean F1
|
| 297 |
- type: euclidean_f1_threshold
|
| 298 |
+
value: 23.90462303161621
|
| 299 |
name: Euclidean F1 Threshold
|
| 300 |
- type: euclidean_precision
|
| 301 |
+
value: 0.4900398406374502
|
| 302 |
name: Euclidean Precision
|
| 303 |
- type: euclidean_recall
|
| 304 |
value: 1.0
|
| 305 |
name: Euclidean Recall
|
| 306 |
- type: euclidean_ap
|
| 307 |
+
value: 0.5510190217865811
|
| 308 |
name: Euclidean Ap
|
| 309 |
- type: max_accuracy
|
| 310 |
+
value: 0.58203125
|
| 311 |
name: Max Accuracy
|
| 312 |
- type: max_accuracy_threshold
|
| 313 |
+
value: 315.9444580078125
|
| 314 |
name: Max Accuracy Threshold
|
| 315 |
- type: max_f1
|
| 316 |
+
value: 0.6595174262734584
|
| 317 |
name: Max F1
|
| 318 |
- type: max_f1_threshold
|
| 319 |
+
value: 469.7353515625
|
| 320 |
name: Max F1 Threshold
|
| 321 |
- type: max_precision
|
| 322 |
+
value: 0.492
|
| 323 |
name: Max Precision
|
| 324 |
- type: max_recall
|
| 325 |
value: 1.0
|
| 326 |
name: Max Recall
|
| 327 |
- type: max_ap
|
| 328 |
+
value: 0.5557444337961499
|
| 329 |
name: Max Ap
|
| 330 |
---
|
| 331 |
|
|
|
|
| 388 |
from sentence_transformers import SentenceTransformer
|
| 389 |
|
| 390 |
# Download from the 🤗 Hub
|
| 391 |
+
model = SentenceTransformer("bobox/DeBERTa-small-ST-v1-toytest")
|
| 392 |
# Run inference
|
| 393 |
sentences = [
|
| 394 |
'who did ben assault in home and away',
|
|
|
|
| 439 |
|
| 440 |
| Metric | Value |
|
| 441 |
|:--------------------|:-----------|
|
| 442 |
+
| pearson_cosine | 0.8483 |
|
| 443 |
+
| **spearman_cosine** | **0.8904** |
|
| 444 |
+
| pearson_manhattan | 0.8815 |
|
| 445 |
+
| spearman_manhattan | 0.8866 |
|
| 446 |
+
| pearson_euclidean | 0.8814 |
|
| 447 |
+
| spearman_euclidean | 0.8852 |
|
| 448 |
+
| pearson_dot | 0.8392 |
|
| 449 |
+
| spearman_dot | 0.8578 |
|
| 450 |
+
| pearson_max | 0.8815 |
|
| 451 |
+
| spearman_max | 0.8904 |
|
| 452 |
|
| 453 |
#### Triplet
|
| 454 |
* Dataset: `NLI-v2`
|
|
|
|
| 468 |
|
| 469 |
| Metric | Value |
|
| 470 |
|:-----------------------------|:-----------|
|
| 471 |
+
| cosine_accuracy | 0.5781 |
|
| 472 |
+
| cosine_accuracy_threshold | 0.7859 |
|
| 473 |
+
| cosine_f1 | 0.6595 |
|
| 474 |
+
| cosine_f1_threshold | 0.3212 |
|
| 475 |
+
| cosine_precision | 0.492 |
|
| 476 |
| cosine_recall | 1.0 |
|
| 477 |
+
| cosine_ap | 0.5557 |
|
| 478 |
+
| dot_accuracy | 0.5781 |
|
| 479 |
+
| dot_accuracy_threshold | 315.9445 |
|
| 480 |
+
| dot_f1 | 0.6595 |
|
| 481 |
+
| dot_f1_threshold | 129.8856 |
|
| 482 |
+
| dot_precision | 0.492 |
|
| 483 |
| dot_recall | 1.0 |
|
| 484 |
+
| dot_ap | 0.554 |
|
| 485 |
+
| manhattan_accuracy | 0.5781 |
|
| 486 |
+
| manhattan_accuracy_threshold | 276.4014 |
|
| 487 |
+
| manhattan_f1 | 0.6577 |
|
| 488 |
+
| manhattan_f1_threshold | 469.7354 |
|
| 489 |
+
| manhattan_precision | 0.4919 |
|
| 490 |
+
| manhattan_recall | 0.9919 |
|
| 491 |
+
| manhattan_ap | 0.5429 |
|
| 492 |
+
| euclidean_accuracy | 0.582 |
|
| 493 |
+
| euclidean_accuracy_threshold | 13.1132 |
|
| 494 |
+
| euclidean_f1 | 0.6578 |
|
| 495 |
+
| euclidean_f1_threshold | 23.9046 |
|
| 496 |
+
| euclidean_precision | 0.49 |
|
| 497 |
| euclidean_recall | 1.0 |
|
| 498 |
+
| euclidean_ap | 0.551 |
|
| 499 |
+
| max_accuracy | 0.582 |
|
| 500 |
+
| max_accuracy_threshold | 315.9445 |
|
| 501 |
+
| max_f1 | 0.6595 |
|
| 502 |
+
| max_f1_threshold | 469.7354 |
|
| 503 |
+
| max_precision | 0.492 |
|
| 504 |
| max_recall | 1.0 |
|
| 505 |
+
| **max_ap** | **0.5557** |
|
| 506 |
|
| 507 |
<!--
|
| 508 |
## Bias, Risks and Limitations
|
|
|
|
| 1151 |
#### Non-Default Hyperparameters
|
| 1152 |
|
| 1153 |
- `eval_strategy`: steps
|
| 1154 |
+
- `per_device_train_batch_size`: 320
|
| 1155 |
- `per_device_eval_batch_size`: 64
|
| 1156 |
+
- `gradient_accumulation_steps`: 4
|
| 1157 |
- `learning_rate`: 4e-05
|
| 1158 |
+
- `weight_decay`: 5e-05
|
| 1159 |
- `lr_scheduler_type`: cosine_with_min_lr
|
| 1160 |
+
- `lr_scheduler_kwargs`: {'num_cycles': 0.5, 'min_lr': 1e-05}
|
| 1161 |
+
- `warmup_ratio`: 0.15
|
| 1162 |
- `save_safetensors`: False
|
| 1163 |
- `fp16`: True
|
| 1164 |
- `push_to_hub`: True
|
|
|
|
| 1173 |
- `do_predict`: False
|
| 1174 |
- `eval_strategy`: steps
|
| 1175 |
- `prediction_loss_only`: True
|
| 1176 |
+
- `per_device_train_batch_size`: 320
|
| 1177 |
- `per_device_eval_batch_size`: 64
|
| 1178 |
- `per_gpu_train_batch_size`: None
|
| 1179 |
- `per_gpu_eval_batch_size`: None
|
| 1180 |
+
- `gradient_accumulation_steps`: 4
|
| 1181 |
- `eval_accumulation_steps`: None
|
| 1182 |
- `learning_rate`: 4e-05
|
| 1183 |
+
- `weight_decay`: 5e-05
|
| 1184 |
- `adam_beta1`: 0.9
|
| 1185 |
- `adam_beta2`: 0.999
|
| 1186 |
- `adam_epsilon`: 1e-08
|
|
|
|
| 1188 |
- `num_train_epochs`: 3
|
| 1189 |
- `max_steps`: -1
|
| 1190 |
- `lr_scheduler_type`: cosine_with_min_lr
|
| 1191 |
+
- `lr_scheduler_kwargs`: {'num_cycles': 0.5, 'min_lr': 1e-05}
|
| 1192 |
+
- `warmup_ratio`: 0.15
|
| 1193 |
- `warmup_steps`: 0
|
| 1194 |
- `log_level`: passive
|
| 1195 |
- `log_level_replica`: warning
|
|
|
|
| 1282 |
</details>
|
| 1283 |
|
| 1284 |
### Training Logs
|
| 1285 |
+
<details><summary>Click to expand</summary>
|
| 1286 |
+
|
| 1287 |
| Epoch | Step | Training Loss | vitaminc-pairs loss | trivia pairs loss | xsum-pairs loss | paws-pos loss | sciq pairs loss | msmarco pairs loss | openbookqa pairs loss | gooaq pairs loss | nq pairs loss | scitail-pairs-pos loss | qasc pairs loss | negation-triplets loss | NLI-v2_max_accuracy | VitaminC_max_ap | sts-test_spearman_cosine |
|
| 1288 |
|:------:|:----:|:-------------:|:-------------------:|:-----------------:|:---------------:|:-------------:|:---------------:|:------------------:|:---------------------:|:----------------:|:-------------:|:----------------------:|:---------------:|:----------------------:|:-------------------:|:---------------:|:------------------------:|
|
| 1289 |
| 0.0169 | 3 | 7.2372 | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - |
|
|
|
|
| 1321 |
| 0.5589 | 99 | 2.1857 | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - |
|
| 1322 |
| 0.5759 | 102 | 1.8881 | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - |
|
| 1323 |
| 0.5928 | 105 | 2.2699 | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - |
|
| 1324 |
+
| 0.6097 | 108 | 2.1425 | 2.7217 | 1.7080 | 1.2066 | 0.0800 | 0.0949 | 1.6446 | 1.5739 | 1.7924 | 2.3649 | 0.2329 | 0.8462 | 2.3389 | 1.0 | 0.5323 | 0.7806 |
|
| 1325 |
+
| 0.6267 | 111 | 2.1276 | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - |
|
| 1326 |
+
| 0.6436 | 114 | 1.7531 | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - |
|
| 1327 |
+
| 0.6606 | 117 | 2.0179 | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - |
|
| 1328 |
+
| 0.6775 | 120 | 1.5305 | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - |
|
| 1329 |
+
| 0.6944 | 123 | 1.6925 | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - |
|
| 1330 |
+
| 0.7114 | 126 | 1.5248 | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - |
|
| 1331 |
+
| 0.7283 | 129 | 1.523 | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - |
|
| 1332 |
+
| 0.7452 | 132 | 1.5474 | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - |
|
| 1333 |
+
| 0.7622 | 135 | 1.7221 | 2.8521 | 1.4495 | 0.7707 | 0.0601 | 0.0751 | 1.1524 | 1.4015 | 1.3955 | 1.7769 | 0.2150 | 0.6356 | 2.0742 | 1.0 | 0.5327 | 0.8315 |
|
| 1334 |
+
| 0.7791 | 138 | 1.5366 | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - |
|
| 1335 |
+
| 0.7960 | 141 | 1.3045 | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - |
|
| 1336 |
+
| 0.8130 | 144 | 1.1999 | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - |
|
| 1337 |
+
| 0.8299 | 147 | 1.3483 | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - |
|
| 1338 |
+
| 0.8469 | 150 | 1.2009 | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - |
|
| 1339 |
+
| 0.8638 | 153 | 1.4495 | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - |
|
| 1340 |
+
| 0.8807 | 156 | 1.2329 | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - |
|
| 1341 |
+
| 0.8977 | 159 | 1.1905 | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - |
|
| 1342 |
+
| 0.9146 | 162 | 1.277 | 2.7764 | 1.2929 | 0.5587 | 0.0525 | 0.0604 | 0.8656 | 1.1903 | 1.1581 | 1.1554 | 0.1988 | 0.4943 | 2.0055 | 1.0 | 0.5311 | 0.8548 |
|
| 1343 |
+
| 0.9315 | 165 | 1.339 | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - |
|
| 1344 |
+
| 0.9485 | 168 | 1.1535 | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - |
|
| 1345 |
+
| 0.9654 | 171 | 1.1643 | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - |
|
| 1346 |
+
| 0.9824 | 174 | 1.2221 | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - |
|
| 1347 |
+
| 0.9993 | 177 | 1.0974 | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - |
|
| 1348 |
+
| 1.0162 | 180 | 1.0984 | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - |
|
| 1349 |
+
| 1.0332 | 183 | 1.0543 | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - |
|
| 1350 |
+
| 1.0501 | 186 | 1.0994 | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - |
|
| 1351 |
+
| 1.0670 | 189 | 1.0621 | 2.6755 | 1.2004 | 0.3837 | 0.0421 | 0.0556 | 0.6897 | 1.0837 | 1.0353 | 0.9604 | 0.1854 | 0.4047 | 1.9071 | 1.0 | 0.5420 | 0.8680 |
|
| 1352 |
+
| 1.0840 | 192 | 0.8724 | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - |
|
| 1353 |
+
| 1.1009 | 195 | 0.9381 | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - |
|
| 1354 |
+
| 1.1179 | 198 | 0.9617 | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - |
|
| 1355 |
+
| 1.1348 | 201 | 1.0139 | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - |
|
| 1356 |
+
| 1.1517 | 204 | 1.1073 | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - |
|
| 1357 |
+
| 1.1687 | 207 | 0.8365 | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - |
|
| 1358 |
+
| 1.1856 | 210 | 1.1012 | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - |
|
| 1359 |
+
| 1.2025 | 213 | 1.0016 | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - |
|
| 1360 |
+
| 1.2195 | 216 | 1.0957 | 2.5466 | 1.1412 | 0.3591 | 0.0395 | 0.0517 | 0.5819 | 0.9366 | 0.9686 | 0.8172 | 0.1901 | 0.3075 | 1.9161 | 1.0 | 0.5385 | 0.8656 |
|
| 1361 |
+
| 1.2364 | 219 | 1.1273 | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - |
|
| 1362 |
+
| 1.2534 | 222 | 1.2568 | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - |
|
| 1363 |
+
| 1.2703 | 225 | 0.873 | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - |
|
| 1364 |
+
| 1.2872 | 228 | 1.0003 | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - |
|
| 1365 |
+
| 1.3042 | 231 | 1.142 | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - |
|
| 1366 |
+
| 1.3211 | 234 | 0.807 | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - |
|
| 1367 |
+
| 1.3380 | 237 | 1.0231 | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - |
|
| 1368 |
+
| 1.3550 | 240 | 0.797 | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - |
|
| 1369 |
+
| 1.3719 | 243 | 0.8473 | 2.5140 | 1.1067 | 0.2802 | 0.0343 | 0.0467 | 0.5559 | 0.8562 | 0.8929 | 0.7435 | 0.1750 | 0.2355 | 1.8629 | 1.0 | 0.5508 | 0.8687 |
|
| 1370 |
+
| 1.3888 | 246 | 0.9531 | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - |
|
| 1371 |
+
| 1.4058 | 249 | 0.9023 | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - |
|
| 1372 |
+
| 1.4227 | 252 | 0.8922 | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - |
|
| 1373 |
+
| 1.4397 | 255 | 0.9874 | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - |
|
| 1374 |
+
| 1.4566 | 258 | 0.8508 | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - |
|
| 1375 |
+
| 1.4735 | 261 | 0.7149 | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - |
|
| 1376 |
+
| 1.4905 | 264 | 0.894 | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - |
|
| 1377 |
+
| 1.5074 | 267 | 0.867 | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - |
|
| 1378 |
+
| 1.5243 | 270 | 0.7493 | 2.5574 | 1.0634 | 0.2217 | 0.0319 | 0.0435 | 0.5027 | 0.7999 | 0.8005 | 0.6530 | 0.1693 | 0.2443 | 1.8535 | 1.0 | 0.5499 | 0.8716 |
|
| 1379 |
+
| 1.5413 | 273 | 0.7974 | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - |
|
| 1380 |
+
| 1.5582 | 276 | 0.797 | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - |
|
| 1381 |
+
| 1.5752 | 279 | 0.6749 | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - |
|
| 1382 |
+
| 1.5921 | 282 | 0.9325 | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - |
|
| 1383 |
+
| 1.6090 | 285 | 0.8418 | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - |
|
| 1384 |
+
| 1.6260 | 288 | 1.0135 | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - |
|
| 1385 |
+
| 1.6429 | 291 | 0.6961 | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - |
|
| 1386 |
+
| 1.6598 | 294 | 0.9361 | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - |
|
| 1387 |
+
| 1.6768 | 297 | 0.6747 | 2.4871 | 0.9762 | 0.2242 | 0.0291 | 0.0396 | 0.5025 | 0.7668 | 0.7546 | 0.6427 | 0.1596 | 0.1963 | 1.7349 | 1.0 | 0.5461 | 0.8787 |
|
| 1388 |
+
| 1.6937 | 300 | 0.7786 | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - |
|
| 1389 |
+
| 1.7107 | 303 | 0.7171 | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - |
|
| 1390 |
+
| 1.7276 | 306 | 0.6627 | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - |
|
| 1391 |
+
| 1.7445 | 309 | 0.6711 | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - |
|
| 1392 |
+
| 1.7615 | 312 | 0.9076 | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - |
|
| 1393 |
+
| 1.7784 | 315 | 0.7414 | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - |
|
| 1394 |
+
| 1.7953 | 318 | 0.582 | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - |
|
| 1395 |
+
| 1.8123 | 321 | 0.6068 | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - |
|
| 1396 |
+
| 1.8292 | 324 | 0.6219 | 2.5197 | 1.0206 | 0.1630 | 0.0273 | 0.0383 | 0.4859 | 0.7109 | 0.7736 | 0.5533 | 0.1535 | 0.2044 | 1.7016 | 1.0 | 0.5532 | 0.8807 |
|
| 1397 |
+
| 1.8462 | 327 | 0.5862 | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - |
|
| 1398 |
+
| 1.8631 | 330 | 0.678 | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - |
|
| 1399 |
+
| 1.8800 | 333 | 0.6272 | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - |
|
| 1400 |
+
| 1.8970 | 336 | 0.5048 | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - |
|
| 1401 |
+
| 1.9139 | 339 | 0.7653 | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - |
|
| 1402 |
+
| 1.9308 | 342 | 0.6613 | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - |
|
| 1403 |
+
| 1.9478 | 345 | 0.6122 | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - |
|
| 1404 |
+
| 1.9647 | 348 | 0.5939 | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - |
|
| 1405 |
+
| 1.9817 | 351 | 0.6923 | 2.4379 | 0.9582 | 0.1464 | 0.0264 | 0.0382 | 0.4348 | 0.7554 | 0.7220 | 0.5432 | 0.1481 | 0.1640 | 1.7345 | 1.0 | 0.5560 | 0.8837 |
|
| 1406 |
+
| 1.9986 | 354 | 0.5712 | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - |
|
| 1407 |
+
| 2.0155 | 357 | 0.5969 | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - |
|
| 1408 |
+
| 2.0325 | 360 | 0.5881 | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - |
|
| 1409 |
+
| 2.0494 | 363 | 0.6005 | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - |
|
| 1410 |
+
| 2.0663 | 366 | 0.6066 | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - |
|
| 1411 |
+
| 2.0833 | 369 | 0.4921 | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - |
|
| 1412 |
+
| 2.1002 | 372 | 0.5354 | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - |
|
| 1413 |
+
| 2.1171 | 375 | 0.5602 | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - |
|
| 1414 |
+
| 2.1341 | 378 | 0.5686 | 2.3908 | 0.9614 | 0.1454 | 0.0271 | 0.0374 | 0.4246 | 0.7796 | 0.6965 | 0.5298 | 0.1401 | 0.1604 | 1.7678 | 1.0 | 0.5539 | 0.8804 |
|
| 1415 |
+
| 2.1510 | 381 | 0.6496 | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - |
|
| 1416 |
+
| 2.1680 | 384 | 0.4713 | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - |
|
| 1417 |
+
| 2.1849 | 387 | 0.6345 | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - |
|
| 1418 |
+
| 2.2018 | 390 | 0.5994 | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - |
|
| 1419 |
+
| 2.2188 | 393 | 0.6763 | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - |
|
| 1420 |
+
| 2.2357 | 396 | 0.7254 | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - |
|
| 1421 |
+
| 2.2526 | 399 | 0.8032 | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - |
|
| 1422 |
+
| 2.2696 | 402 | 0.4914 | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - |
|
| 1423 |
+
| 2.2865 | 405 | 0.6307 | 2.4388 | 0.9862 | 0.1308 | 0.0262 | 0.0379 | 0.3928 | 0.7434 | 0.6976 | 0.4998 | 0.1192 | 0.1466 | 1.7093 | 1.0 | 0.5533 | 0.8859 |
|
| 1424 |
+
| 2.3035 | 408 | 0.7493 | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - |
|
| 1425 |
+
| 2.3204 | 411 | 0.5139 | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - |
|
| 1426 |
+
| 2.3373 | 414 | 0.6364 | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - |
|
| 1427 |
+
| 2.3543 | 417 | 0.4763 | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - |
|
| 1428 |
+
| 2.3712 | 420 | 0.583 | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - |
|
| 1429 |
+
| 2.3881 | 423 | 0.5912 | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - |
|
| 1430 |
+
| 2.4051 | 426 | 0.5936 | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - |
|
| 1431 |
+
| 2.4220 | 429 | 0.5959 | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - |
|
| 1432 |
+
| 2.4390 | 432 | 0.676 | 2.4265 | 0.9634 | 0.1220 | 0.0260 | 0.0362 | 0.4292 | 0.7433 | 0.6771 | 0.4752 | 0.1282 | 0.1304 | 1.6943 | 1.0 | 0.5532 | 0.8878 |
|
| 1433 |
+
| 2.4559 | 435 | 0.5622 | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - |
|
| 1434 |
+
| 2.4728 | 438 | 0.4633 | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - |
|
| 1435 |
+
| 2.4898 | 441 | 0.5955 | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - |
|
| 1436 |
+
| 2.5067 | 444 | 0.6271 | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - |
|
| 1437 |
+
| 2.5236 | 447 | 0.4988 | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - |
|
| 1438 |
+
| 2.5406 | 450 | 0.519 | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - |
|
| 1439 |
+
| 2.5575 | 453 | 0.5538 | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - |
|
| 1440 |
+
| 2.5745 | 456 | 0.4826 | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - |
|
| 1441 |
+
| 2.5914 | 459 | 0.6322 | 2.4541 | 0.9231 | 0.1224 | 0.0253 | 0.0345 | 0.4048 | 0.7595 | 0.6607 | 0.4713 | 0.1168 | 0.1323 | 1.7024 | 1.0 | 0.5557 | 0.8868 |
|
| 1442 |
+
| 2.6083 | 462 | 0.6342 | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - |
|
| 1443 |
+
| 2.6253 | 465 | 0.7012 | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - |
|
| 1444 |
+
| 2.6422 | 468 | 0.4175 | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - |
|
| 1445 |
+
| 2.6591 | 471 | 0.7575 | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - |
|
| 1446 |
+
| 2.6761 | 474 | 0.4687 | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - |
|
| 1447 |
+
| 2.6930 | 477 | 0.5907 | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - |
|
| 1448 |
+
| 2.7100 | 480 | 0.4796 | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - |
|
| 1449 |
+
| 2.7269 | 483 | 0.4809 | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - |
|
| 1450 |
+
| 2.7438 | 486 | 0.4696 | 2.4899 | 0.9546 | 0.1169 | 0.0247 | 0.0343 | 0.4138 | 0.7444 | 0.6688 | 0.4838 | 0.1166 | 0.1279 | 1.6605 | 1.0 | 0.5527 | 0.8883 |
|
| 1451 |
+
| 2.7608 | 489 | 0.6588 | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - |
|
| 1452 |
+
| 2.7777 | 492 | 0.5675 | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - |
|
| 1453 |
+
| 2.7946 | 495 | 0.4007 | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - |
|
| 1454 |
+
| 2.8116 | 498 | 0.4476 | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - |
|
| 1455 |
+
| 2.8285 | 501 | 0.433 | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - |
|
| 1456 |
+
| 2.8454 | 504 | 0.4154 | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - |
|
| 1457 |
+
| 2.8624 | 507 | 0.5416 | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - |
|
| 1458 |
+
| 2.8793 | 510 | 0.4546 | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - |
|
| 1459 |
+
| 2.8963 | 513 | 0.3326 | 2.4924 | 0.9493 | 0.1071 | 0.0248 | 0.0344 | 0.4033 | 0.7376 | 0.6558 | 0.4478 | 0.1148 | 0.1219 | 1.6918 | 1.0 | 0.5534 | 0.8907 |
|
| 1460 |
+
| 2.9132 | 516 | 0.594 | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - |
|
| 1461 |
+
| 2.9301 | 519 | 0.4727 | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - |
|
| 1462 |
+
| 2.9471 | 522 | 0.4701 | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - |
|
| 1463 |
+
| 2.9640 | 525 | 0.4606 | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - |
|
| 1464 |
+
| 2.9809 | 528 | 0.5025 | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - |
|
| 1465 |
+
| 2.9979 | 531 | 0.4314 | 2.4532 | 0.9270 | 0.1131 | 0.0247 | 0.0344 | 0.3951 | 0.7123 | 0.6345 | 0.4383 | 0.1143 | 0.1159 | 1.7003 | 1.0 | 0.5539 | 0.8904 |
|
| 1466 |
+
| 0.0169 | 3 | 0.6012 | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - |
|
| 1467 |
+
| 0.0337 | 6 | 0.7573 | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - |
|
| 1468 |
+
| 0.0506 | 9 | 0.9212 | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - |
|
| 1469 |
+
| 0.0674 | 12 | 0.6117 | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - |
|
| 1470 |
+
| 0.0843 | 15 | 0.8545 | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - |
|
| 1471 |
+
| 0.1011 | 18 | 0.6515 | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - |
|
| 1472 |
+
| 0.1180 | 21 | 0.7159 | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - |
|
| 1473 |
+
| 0.1348 | 24 | 0.7019 | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - |
|
| 1474 |
+
| 0.1517 | 27 | 0.4411 | 2.4659 | 0.9318 | 0.1117 | 0.0249 | 0.0345 | 0.3955 | 0.7092 | 0.6506 | 0.4205 | 0.1150 | 0.1110 | 1.7311 | 1.0 | 0.5512 | 0.8906 |
|
| 1475 |
+
| 0.1685 | 30 | 0.5125 | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - |
|
| 1476 |
+
| 0.1854 | 33 | 0.6885 | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - |
|
| 1477 |
+
| 0.2022 | 36 | 0.6435 | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - |
|
| 1478 |
+
| 0.2191 | 39 | 0.753 | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - |
|
| 1479 |
+
| 0.2360 | 42 | 0.7427 | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - |
|
| 1480 |
+
| 0.2528 | 45 | 0.5083 | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - |
|
| 1481 |
+
| 0.2697 | 48 | 0.7454 | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - |
|
| 1482 |
+
| 0.2865 | 51 | 0.8356 | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - |
|
| 1483 |
+
| 0.3034 | 54 | 0.8864 | 2.4545 | 0.9158 | 0.1009 | 0.0252 | 0.0347 | 0.3809 | 0.7240 | 0.6208 | 0.4417 | 0.1117 | 0.1055 | 1.7278 | 1.0 | 0.5499 | 0.8877 |
|
| 1484 |
+
| 0.3202 | 57 | 0.6015 | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - |
|
| 1485 |
+
| 0.3371 | 60 | 0.9482 | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - |
|
| 1486 |
+
| 0.3539 | 63 | 0.5404 | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - |
|
| 1487 |
+
| 0.3708 | 66 | 0.805 | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - |
|
| 1488 |
+
| 0.3876 | 69 | 0.7184 | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - |
|
| 1489 |
+
| 0.4045 | 72 | 0.8708 | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - |
|
| 1490 |
+
| 0.4213 | 75 | 0.8327 | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - |
|
| 1491 |
+
| 0.4382 | 78 | 0.5025 | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - |
|
| 1492 |
+
| 0.4551 | 81 | 0.6517 | 2.3539 | 0.9324 | 0.0842 | 0.0244 | 0.0348 | 0.3454 | 0.7161 | 0.6094 | 0.4443 | 0.1182 | 0.1060 | 1.6492 | 1.0 | 0.5557 | 0.8904 |
|
| 1493 |
+
| 0.4719 | 84 | 0.5801 | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - |
|
| 1494 |
+
| 0.4888 | 87 | 0.791 | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - |
|
| 1495 |
+
| 0.5056 | 90 | 0.6042 | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - |
|
| 1496 |
+
| 0.5225 | 93 | 0.7559 | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - |
|
| 1497 |
+
| 0.5393 | 96 | 0.6258 | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - |
|
| 1498 |
+
| 0.5562 | 99 | 0.8853 | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - |
|
| 1499 |
+
| 0.5730 | 102 | 0.5947 | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - |
|
| 1500 |
+
| 0.5899 | 105 | 0.644 | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - |
|
| 1501 |
|
| 1502 |
+
</details>
|
| 1503 |
|
| 1504 |
### Framework Versions
|
| 1505 |
- Python: 3.10.13
|
checkpoint-107/optimizer.pt
CHANGED
|
@@ -1,3 +1,3 @@
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
-
oid sha256:
|
| 3 |
size 1130520122
|
|
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:c4b52b19c1b7ba5dc357d641807fc458a5edfc8d957fa781f3b74046c4c35266
|
| 3 |
size 1130520122
|
checkpoint-107/pytorch_model.bin
CHANGED
|
@@ -1,3 +1,3 @@
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
-
oid sha256:
|
| 3 |
size 565251810
|
|
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:2573f8d44707a08d0f8ac75dc82a2db81d33d5f56b9ff43e4b8bd36b6360356e
|
| 3 |
size 565251810
|
checkpoint-107/rng_state.pth
CHANGED
|
@@ -1,3 +1,3 @@
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
-
oid sha256:
|
| 3 |
size 14244
|
|
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:9ee4a49a4489e711ad7e57eb8006fdb27dea4dd2f04c074a42f884b3f1874718
|
| 3 |
size 14244
|
checkpoint-107/scheduler.pt
CHANGED
|
@@ -1,3 +1,3 @@
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
-
oid sha256:
|
| 3 |
size 1064
|
|
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:672e608d66e62675fe93e3516bd09994e5b9413ad5901cbe4ab81e1e4e26683a
|
| 3 |
size 1064
|
checkpoint-107/trainer_state.json
CHANGED
|
@@ -1,7 +1,7 @@
|
|
| 1 |
{
|
| 2 |
"best_metric": null,
|
| 3 |
"best_model_checkpoint": null,
|
| 4 |
-
"epoch": 0.
|
| 5 |
"eval_steps": 27,
|
| 6 |
"global_step": 107,
|
| 7 |
"is_hyper_param_search": false,
|
|
@@ -9,694 +9,694 @@
|
|
| 9 |
"is_world_process_zero": true,
|
| 10 |
"log_history": [
|
| 11 |
{
|
| 12 |
-
"epoch": 0.
|
| 13 |
-
"grad_norm":
|
| 14 |
-
"learning_rate":
|
| 15 |
-
"loss":
|
| 16 |
"step": 3
|
| 17 |
},
|
| 18 |
{
|
| 19 |
-
"epoch": 0.
|
| 20 |
-
"grad_norm":
|
| 21 |
-
"learning_rate":
|
| 22 |
-
"loss":
|
| 23 |
"step": 6
|
| 24 |
},
|
| 25 |
{
|
| 26 |
-
"epoch": 0.
|
| 27 |
-
"grad_norm":
|
| 28 |
-
"learning_rate":
|
| 29 |
-
"loss":
|
| 30 |
"step": 9
|
| 31 |
},
|
| 32 |
{
|
| 33 |
-
"epoch": 0.
|
| 34 |
-
"grad_norm":
|
| 35 |
-
"learning_rate":
|
| 36 |
-
"loss":
|
| 37 |
"step": 12
|
| 38 |
},
|
| 39 |
{
|
| 40 |
-
"epoch": 0.
|
| 41 |
-
"grad_norm":
|
| 42 |
-
"learning_rate":
|
| 43 |
-
"loss":
|
| 44 |
"step": 15
|
| 45 |
},
|
| 46 |
{
|
| 47 |
-
"epoch": 0.
|
| 48 |
-
"grad_norm":
|
| 49 |
-
"learning_rate":
|
| 50 |
-
"loss":
|
| 51 |
"step": 18
|
| 52 |
},
|
| 53 |
{
|
| 54 |
-
"epoch": 0.
|
| 55 |
-
"grad_norm":
|
| 56 |
-
"learning_rate":
|
| 57 |
-
"loss":
|
| 58 |
"step": 21
|
| 59 |
},
|
| 60 |
{
|
| 61 |
-
"epoch": 0.
|
| 62 |
-
"grad_norm":
|
| 63 |
-
"learning_rate":
|
| 64 |
-
"loss":
|
| 65 |
"step": 24
|
| 66 |
},
|
| 67 |
{
|
| 68 |
-
"epoch": 0.
|
| 69 |
-
"grad_norm":
|
| 70 |
-
"learning_rate":
|
| 71 |
-
"loss":
|
| 72 |
"step": 27
|
| 73 |
},
|
| 74 |
{
|
| 75 |
-
"epoch": 0.
|
| 76 |
"eval_NLI-v2_cosine_accuracy": 1.0,
|
| 77 |
-
"eval_NLI-v2_dot_accuracy": 0.
|
| 78 |
"eval_NLI-v2_euclidean_accuracy": 1.0,
|
| 79 |
"eval_NLI-v2_manhattan_accuracy": 1.0,
|
| 80 |
"eval_NLI-v2_max_accuracy": 1.0,
|
| 81 |
-
"eval_VitaminC_cosine_accuracy": 0.
|
| 82 |
-
"eval_VitaminC_cosine_accuracy_threshold": 0.
|
| 83 |
-
"eval_VitaminC_cosine_ap": 0.
|
| 84 |
-
"eval_VitaminC_cosine_f1": 0.
|
| 85 |
-
"eval_VitaminC_cosine_f1_threshold": 0.
|
| 86 |
-
"eval_VitaminC_cosine_precision": 0.
|
| 87 |
"eval_VitaminC_cosine_recall": 1.0,
|
| 88 |
-
"eval_VitaminC_dot_accuracy": 0.
|
| 89 |
-
"eval_VitaminC_dot_accuracy_threshold":
|
| 90 |
-
"eval_VitaminC_dot_ap": 0.
|
| 91 |
-
"eval_VitaminC_dot_f1": 0.
|
| 92 |
-
"eval_VitaminC_dot_f1_threshold":
|
| 93 |
-
"eval_VitaminC_dot_precision": 0.
|
| 94 |
"eval_VitaminC_dot_recall": 1.0,
|
| 95 |
-
"eval_VitaminC_euclidean_accuracy": 0.
|
| 96 |
-
"eval_VitaminC_euclidean_accuracy_threshold":
|
| 97 |
-
"eval_VitaminC_euclidean_ap": 0.
|
| 98 |
-
"eval_VitaminC_euclidean_f1": 0.
|
| 99 |
-
"eval_VitaminC_euclidean_f1_threshold":
|
| 100 |
-
"eval_VitaminC_euclidean_precision": 0.
|
| 101 |
-
"eval_VitaminC_euclidean_recall": 0
|
| 102 |
-
"eval_VitaminC_manhattan_accuracy": 0.
|
| 103 |
-
"eval_VitaminC_manhattan_accuracy_threshold":
|
| 104 |
-
"eval_VitaminC_manhattan_ap": 0.
|
| 105 |
-
"eval_VitaminC_manhattan_f1": 0.
|
| 106 |
-
"eval_VitaminC_manhattan_f1_threshold":
|
| 107 |
-
"eval_VitaminC_manhattan_precision": 0.
|
| 108 |
"eval_VitaminC_manhattan_recall": 1.0,
|
| 109 |
-
"eval_VitaminC_max_accuracy": 0.
|
| 110 |
-
"eval_VitaminC_max_accuracy_threshold":
|
| 111 |
-
"eval_VitaminC_max_ap": 0.
|
| 112 |
-
"eval_VitaminC_max_f1": 0.
|
| 113 |
-
"eval_VitaminC_max_f1_threshold":
|
| 114 |
-
"eval_VitaminC_max_precision": 0.
|
| 115 |
"eval_VitaminC_max_recall": 1.0,
|
| 116 |
-
"eval_sequential_score": 0.
|
| 117 |
-
"eval_sts-test_pearson_cosine": 0.
|
| 118 |
-
"eval_sts-test_pearson_dot": 0.
|
| 119 |
-
"eval_sts-test_pearson_euclidean": 0.
|
| 120 |
-
"eval_sts-test_pearson_manhattan": 0.
|
| 121 |
-
"eval_sts-test_pearson_max": 0.
|
| 122 |
-
"eval_sts-test_spearman_cosine": 0.
|
| 123 |
-
"eval_sts-test_spearman_dot": 0.
|
| 124 |
-
"eval_sts-test_spearman_euclidean": 0.
|
| 125 |
-
"eval_sts-test_spearman_manhattan": 0.
|
| 126 |
-
"eval_sts-test_spearman_max": 0.
|
| 127 |
-
"eval_vitaminc-pairs_loss": 2.
|
| 128 |
-
"eval_vitaminc-pairs_runtime": 1.
|
| 129 |
-
"eval_vitaminc-pairs_samples_per_second": 73.
|
| 130 |
-
"eval_vitaminc-pairs_steps_per_second": 1.
|
| 131 |
"step": 27
|
| 132 |
},
|
| 133 |
{
|
| 134 |
-
"epoch": 0.
|
| 135 |
-
"eval_negation-triplets_loss":
|
| 136 |
-
"eval_negation-triplets_runtime": 0.
|
| 137 |
-
"eval_negation-triplets_samples_per_second":
|
| 138 |
-
"eval_negation-triplets_steps_per_second": 3.
|
| 139 |
"step": 27
|
| 140 |
},
|
| 141 |
{
|
| 142 |
-
"epoch": 0.
|
| 143 |
-
"eval_scitail-pairs-pos_loss":
|
| 144 |
-
"eval_scitail-pairs-pos_runtime": 0.
|
| 145 |
-
"eval_scitail-pairs-pos_samples_per_second":
|
| 146 |
-
"eval_scitail-pairs-pos_steps_per_second": 2.
|
| 147 |
"step": 27
|
| 148 |
},
|
| 149 |
{
|
| 150 |
-
"epoch": 0.
|
| 151 |
-
"eval_xsum-pairs_loss":
|
| 152 |
-
"eval_xsum-pairs_runtime": 3.
|
| 153 |
-
"eval_xsum-pairs_samples_per_second": 40.
|
| 154 |
-
"eval_xsum-pairs_steps_per_second": 0.
|
| 155 |
"step": 27
|
| 156 |
},
|
| 157 |
{
|
| 158 |
-
"epoch": 0.
|
| 159 |
-
"eval_sciq_pairs_loss": 0.
|
| 160 |
-
"eval_sciq_pairs_runtime": 3.
|
| 161 |
-
"eval_sciq_pairs_samples_per_second":
|
| 162 |
-
"eval_sciq_pairs_steps_per_second": 0.
|
| 163 |
"step": 27
|
| 164 |
},
|
| 165 |
{
|
| 166 |
-
"epoch": 0.
|
| 167 |
-
"eval_qasc_pairs_loss":
|
| 168 |
-
"eval_qasc_pairs_runtime": 0.
|
| 169 |
-
"eval_qasc_pairs_samples_per_second":
|
| 170 |
-
"eval_qasc_pairs_steps_per_second": 3.
|
| 171 |
"step": 27
|
| 172 |
},
|
| 173 |
{
|
| 174 |
-
"epoch": 0.
|
| 175 |
-
"eval_openbookqa_pairs_loss":
|
| 176 |
-
"eval_openbookqa_pairs_runtime": 0.
|
| 177 |
-
"eval_openbookqa_pairs_samples_per_second":
|
| 178 |
-
"eval_openbookqa_pairs_steps_per_second": 3.
|
| 179 |
"step": 27
|
| 180 |
},
|
| 181 |
{
|
| 182 |
-
"epoch": 0.
|
| 183 |
-
"eval_msmarco_pairs_loss":
|
| 184 |
-
"eval_msmarco_pairs_runtime": 1.
|
| 185 |
-
"eval_msmarco_pairs_samples_per_second":
|
| 186 |
-
"eval_msmarco_pairs_steps_per_second": 1.
|
| 187 |
"step": 27
|
| 188 |
},
|
| 189 |
{
|
| 190 |
-
"epoch": 0.
|
| 191 |
-
"eval_nq_pairs_loss":
|
| 192 |
-
"eval_nq_pairs_runtime": 2.
|
| 193 |
-
"eval_nq_pairs_samples_per_second": 53.
|
| 194 |
-
"eval_nq_pairs_steps_per_second": 0.
|
| 195 |
"step": 27
|
| 196 |
},
|
| 197 |
{
|
| 198 |
-
"epoch": 0.
|
| 199 |
-
"eval_trivia_pairs_loss":
|
| 200 |
-
"eval_trivia_pairs_runtime": 4.
|
| 201 |
-
"eval_trivia_pairs_samples_per_second": 28.
|
| 202 |
-
"eval_trivia_pairs_steps_per_second": 0.
|
| 203 |
"step": 27
|
| 204 |
},
|
| 205 |
{
|
| 206 |
-
"epoch": 0.
|
| 207 |
-
"eval_gooaq_pairs_loss":
|
| 208 |
-
"eval_gooaq_pairs_runtime": 0.
|
| 209 |
-
"eval_gooaq_pairs_samples_per_second":
|
| 210 |
-
"eval_gooaq_pairs_steps_per_second": 2.
|
| 211 |
"step": 27
|
| 212 |
},
|
| 213 |
{
|
| 214 |
-
"epoch": 0.
|
| 215 |
-
"eval_paws-pos_loss":
|
| 216 |
-
"eval_paws-pos_runtime": 0.
|
| 217 |
-
"eval_paws-pos_samples_per_second":
|
| 218 |
-
"eval_paws-pos_steps_per_second": 2.
|
| 219 |
"step": 27
|
| 220 |
},
|
| 221 |
{
|
| 222 |
-
"epoch": 0.
|
| 223 |
-
"grad_norm":
|
| 224 |
-
"learning_rate":
|
| 225 |
-
"loss":
|
| 226 |
"step": 30
|
| 227 |
},
|
| 228 |
{
|
| 229 |
-
"epoch": 0.
|
| 230 |
-
"grad_norm":
|
| 231 |
-
"learning_rate":
|
| 232 |
-
"loss":
|
| 233 |
"step": 33
|
| 234 |
},
|
| 235 |
{
|
| 236 |
-
"epoch": 0.
|
| 237 |
-
"grad_norm":
|
| 238 |
-
"learning_rate":
|
| 239 |
-
"loss":
|
| 240 |
"step": 36
|
| 241 |
},
|
| 242 |
{
|
| 243 |
-
"epoch": 0.
|
| 244 |
-
"grad_norm":
|
| 245 |
-
"learning_rate":
|
| 246 |
-
"loss":
|
| 247 |
"step": 39
|
| 248 |
},
|
| 249 |
{
|
| 250 |
-
"epoch": 0.
|
| 251 |
-
"grad_norm":
|
| 252 |
-
"learning_rate":
|
| 253 |
-
"loss":
|
| 254 |
"step": 42
|
| 255 |
},
|
| 256 |
{
|
| 257 |
-
"epoch": 0.
|
| 258 |
-
"grad_norm":
|
| 259 |
-
"learning_rate":
|
| 260 |
-
"loss":
|
| 261 |
"step": 45
|
| 262 |
},
|
| 263 |
{
|
| 264 |
-
"epoch": 0.
|
| 265 |
-
"grad_norm":
|
| 266 |
-
"learning_rate":
|
| 267 |
-
"loss":
|
| 268 |
"step": 48
|
| 269 |
},
|
| 270 |
{
|
| 271 |
-
"epoch": 0.
|
| 272 |
-
"grad_norm":
|
| 273 |
-
"learning_rate":
|
| 274 |
-
"loss":
|
| 275 |
"step": 51
|
| 276 |
},
|
| 277 |
{
|
| 278 |
-
"epoch": 0.
|
| 279 |
-
"grad_norm":
|
| 280 |
-
"learning_rate":
|
| 281 |
-
"loss":
|
| 282 |
"step": 54
|
| 283 |
},
|
| 284 |
{
|
| 285 |
-
"epoch": 0.
|
| 286 |
"eval_NLI-v2_cosine_accuracy": 1.0,
|
| 287 |
-
"eval_NLI-v2_dot_accuracy": 0.
|
| 288 |
"eval_NLI-v2_euclidean_accuracy": 1.0,
|
| 289 |
"eval_NLI-v2_manhattan_accuracy": 1.0,
|
| 290 |
"eval_NLI-v2_max_accuracy": 1.0,
|
| 291 |
-
"eval_VitaminC_cosine_accuracy": 0.
|
| 292 |
-
"eval_VitaminC_cosine_accuracy_threshold": 0.
|
| 293 |
-
"eval_VitaminC_cosine_ap": 0.
|
| 294 |
-
"eval_VitaminC_cosine_f1": 0.
|
| 295 |
-
"eval_VitaminC_cosine_f1_threshold": 0.
|
| 296 |
-
"eval_VitaminC_cosine_precision": 0.
|
| 297 |
-
"eval_VitaminC_cosine_recall": 0
|
| 298 |
-
"eval_VitaminC_dot_accuracy": 0.
|
| 299 |
-
"eval_VitaminC_dot_accuracy_threshold":
|
| 300 |
-
"eval_VitaminC_dot_ap": 0.
|
| 301 |
-
"eval_VitaminC_dot_f1": 0.
|
| 302 |
-
"eval_VitaminC_dot_f1_threshold":
|
| 303 |
-
"eval_VitaminC_dot_precision": 0.
|
| 304 |
-
"eval_VitaminC_dot_recall": 0
|
| 305 |
-
"eval_VitaminC_euclidean_accuracy": 0.
|
| 306 |
-
"eval_VitaminC_euclidean_accuracy_threshold":
|
| 307 |
-
"eval_VitaminC_euclidean_ap": 0.
|
| 308 |
-
"eval_VitaminC_euclidean_f1": 0.
|
| 309 |
-
"eval_VitaminC_euclidean_f1_threshold":
|
| 310 |
-
"eval_VitaminC_euclidean_precision": 0.
|
| 311 |
-
"eval_VitaminC_euclidean_recall": 0
|
| 312 |
-
"eval_VitaminC_manhattan_accuracy": 0.
|
| 313 |
-
"eval_VitaminC_manhattan_accuracy_threshold":
|
| 314 |
-
"eval_VitaminC_manhattan_ap": 0.
|
| 315 |
-
"eval_VitaminC_manhattan_f1": 0.
|
| 316 |
-
"eval_VitaminC_manhattan_f1_threshold":
|
| 317 |
-
"eval_VitaminC_manhattan_precision": 0.
|
| 318 |
-
"eval_VitaminC_manhattan_recall": 0
|
| 319 |
-
"eval_VitaminC_max_accuracy": 0.
|
| 320 |
-
"eval_VitaminC_max_accuracy_threshold":
|
| 321 |
-
"eval_VitaminC_max_ap": 0.
|
| 322 |
-
"eval_VitaminC_max_f1": 0.
|
| 323 |
-
"eval_VitaminC_max_f1_threshold":
|
| 324 |
-
"eval_VitaminC_max_precision": 0.
|
| 325 |
-
"eval_VitaminC_max_recall": 0
|
| 326 |
-
"eval_sequential_score": 0.
|
| 327 |
-
"eval_sts-test_pearson_cosine": 0.
|
| 328 |
-
"eval_sts-test_pearson_dot": 0.
|
| 329 |
-
"eval_sts-test_pearson_euclidean": 0.
|
| 330 |
-
"eval_sts-test_pearson_manhattan": 0.
|
| 331 |
-
"eval_sts-test_pearson_max": 0.
|
| 332 |
-
"eval_sts-test_spearman_cosine": 0.
|
| 333 |
-
"eval_sts-test_spearman_dot": 0.
|
| 334 |
-
"eval_sts-test_spearman_euclidean": 0.
|
| 335 |
-
"eval_sts-test_spearman_manhattan": 0.
|
| 336 |
-
"eval_sts-test_spearman_max": 0.
|
| 337 |
-
"eval_vitaminc-pairs_loss": 2.
|
| 338 |
-
"eval_vitaminc-pairs_runtime": 1.
|
| 339 |
-
"eval_vitaminc-pairs_samples_per_second": 74.
|
| 340 |
-
"eval_vitaminc-pairs_steps_per_second": 1.
|
| 341 |
"step": 54
|
| 342 |
},
|
| 343 |
{
|
| 344 |
-
"epoch": 0.
|
| 345 |
-
"eval_negation-triplets_loss":
|
| 346 |
-
"eval_negation-triplets_runtime": 0.
|
| 347 |
-
"eval_negation-triplets_samples_per_second":
|
| 348 |
-
"eval_negation-triplets_steps_per_second": 3.
|
| 349 |
"step": 54
|
| 350 |
},
|
| 351 |
{
|
| 352 |
-
"epoch": 0.
|
| 353 |
-
"eval_scitail-pairs-pos_loss":
|
| 354 |
-
"eval_scitail-pairs-pos_runtime": 0.
|
| 355 |
-
"eval_scitail-pairs-pos_samples_per_second":
|
| 356 |
-
"eval_scitail-pairs-pos_steps_per_second": 2.
|
| 357 |
"step": 54
|
| 358 |
},
|
| 359 |
{
|
| 360 |
-
"epoch": 0.
|
| 361 |
-
"eval_xsum-pairs_loss":
|
| 362 |
-
"eval_xsum-pairs_runtime": 3.
|
| 363 |
-
"eval_xsum-pairs_samples_per_second": 40.
|
| 364 |
-
"eval_xsum-pairs_steps_per_second": 0.
|
| 365 |
"step": 54
|
| 366 |
},
|
| 367 |
{
|
| 368 |
-
"epoch": 0.
|
| 369 |
-
"eval_sciq_pairs_loss": 0.
|
| 370 |
-
"eval_sciq_pairs_runtime": 3.
|
| 371 |
-
"eval_sciq_pairs_samples_per_second":
|
| 372 |
-
"eval_sciq_pairs_steps_per_second": 0.
|
| 373 |
"step": 54
|
| 374 |
},
|
| 375 |
{
|
| 376 |
-
"epoch": 0.
|
| 377 |
-
"eval_qasc_pairs_loss":
|
| 378 |
-
"eval_qasc_pairs_runtime": 0.
|
| 379 |
-
"eval_qasc_pairs_samples_per_second": 204.
|
| 380 |
-
"eval_qasc_pairs_steps_per_second": 3.
|
| 381 |
"step": 54
|
| 382 |
},
|
| 383 |
{
|
| 384 |
-
"epoch": 0.
|
| 385 |
-
"eval_openbookqa_pairs_loss":
|
| 386 |
-
"eval_openbookqa_pairs_runtime": 0.
|
| 387 |
-
"eval_openbookqa_pairs_samples_per_second":
|
| 388 |
-
"eval_openbookqa_pairs_steps_per_second": 3.
|
| 389 |
"step": 54
|
| 390 |
},
|
| 391 |
{
|
| 392 |
-
"epoch": 0.
|
| 393 |
-
"eval_msmarco_pairs_loss":
|
| 394 |
-
"eval_msmarco_pairs_runtime": 1.
|
| 395 |
-
"eval_msmarco_pairs_samples_per_second": 99.
|
| 396 |
-
"eval_msmarco_pairs_steps_per_second": 1.
|
| 397 |
"step": 54
|
| 398 |
},
|
| 399 |
{
|
| 400 |
-
"epoch": 0.
|
| 401 |
-
"eval_nq_pairs_loss":
|
| 402 |
-
"eval_nq_pairs_runtime": 2.
|
| 403 |
-
"eval_nq_pairs_samples_per_second": 53.
|
| 404 |
-
"eval_nq_pairs_steps_per_second": 0.
|
| 405 |
"step": 54
|
| 406 |
},
|
| 407 |
{
|
| 408 |
-
"epoch": 0.
|
| 409 |
-
"eval_trivia_pairs_loss":
|
| 410 |
-
"eval_trivia_pairs_runtime": 4.
|
| 411 |
-
"eval_trivia_pairs_samples_per_second": 28.
|
| 412 |
-
"eval_trivia_pairs_steps_per_second": 0.
|
| 413 |
"step": 54
|
| 414 |
},
|
| 415 |
{
|
| 416 |
-
"epoch": 0.
|
| 417 |
-
"eval_gooaq_pairs_loss":
|
| 418 |
-
"eval_gooaq_pairs_runtime": 0.
|
| 419 |
-
"eval_gooaq_pairs_samples_per_second":
|
| 420 |
-
"eval_gooaq_pairs_steps_per_second": 2.
|
| 421 |
"step": 54
|
| 422 |
},
|
| 423 |
{
|
| 424 |
-
"epoch": 0.
|
| 425 |
-
"eval_paws-pos_loss": 0.
|
| 426 |
-
"eval_paws-pos_runtime": 0.
|
| 427 |
-
"eval_paws-pos_samples_per_second":
|
| 428 |
-
"eval_paws-pos_steps_per_second": 2.
|
| 429 |
"step": 54
|
| 430 |
},
|
| 431 |
{
|
| 432 |
-
"epoch": 0.
|
| 433 |
-
"grad_norm":
|
| 434 |
-
"learning_rate":
|
| 435 |
-
"loss":
|
| 436 |
"step": 57
|
| 437 |
},
|
| 438 |
{
|
| 439 |
-
"epoch": 0.
|
| 440 |
-
"grad_norm":
|
| 441 |
-
"learning_rate":
|
| 442 |
-
"loss":
|
| 443 |
"step": 60
|
| 444 |
},
|
| 445 |
{
|
| 446 |
-
"epoch": 0.
|
| 447 |
-
"grad_norm":
|
| 448 |
-
"learning_rate":
|
| 449 |
-
"loss":
|
| 450 |
"step": 63
|
| 451 |
},
|
| 452 |
{
|
| 453 |
-
"epoch": 0.
|
| 454 |
-
"grad_norm":
|
| 455 |
-
"learning_rate":
|
| 456 |
-
"loss":
|
| 457 |
"step": 66
|
| 458 |
},
|
| 459 |
{
|
| 460 |
-
"epoch": 0.
|
| 461 |
-
"grad_norm":
|
| 462 |
-
"learning_rate":
|
| 463 |
-
"loss":
|
| 464 |
"step": 69
|
| 465 |
},
|
| 466 |
{
|
| 467 |
-
"epoch": 0.
|
| 468 |
-
"grad_norm":
|
| 469 |
-
"learning_rate":
|
| 470 |
-
"loss":
|
| 471 |
"step": 72
|
| 472 |
},
|
| 473 |
{
|
| 474 |
-
"epoch": 0.
|
| 475 |
-
"grad_norm":
|
| 476 |
-
"learning_rate":
|
| 477 |
-
"loss":
|
| 478 |
"step": 75
|
| 479 |
},
|
| 480 |
{
|
| 481 |
-
"epoch": 0.
|
| 482 |
-
"grad_norm":
|
| 483 |
-
"learning_rate":
|
| 484 |
-
"loss":
|
| 485 |
"step": 78
|
| 486 |
},
|
| 487 |
{
|
| 488 |
-
"epoch": 0.
|
| 489 |
-
"grad_norm":
|
| 490 |
-
"learning_rate":
|
| 491 |
-
"loss":
|
| 492 |
"step": 81
|
| 493 |
},
|
| 494 |
{
|
| 495 |
-
"epoch": 0.
|
| 496 |
"eval_NLI-v2_cosine_accuracy": 1.0,
|
| 497 |
"eval_NLI-v2_dot_accuracy": 0.0,
|
| 498 |
"eval_NLI-v2_euclidean_accuracy": 1.0,
|
| 499 |
"eval_NLI-v2_manhattan_accuracy": 1.0,
|
| 500 |
"eval_NLI-v2_max_accuracy": 1.0,
|
| 501 |
-
"eval_VitaminC_cosine_accuracy": 0.
|
| 502 |
-
"eval_VitaminC_cosine_accuracy_threshold": 0.
|
| 503 |
-
"eval_VitaminC_cosine_ap": 0.
|
| 504 |
-
"eval_VitaminC_cosine_f1": 0.
|
| 505 |
-
"eval_VitaminC_cosine_f1_threshold": 0.
|
| 506 |
-
"eval_VitaminC_cosine_precision": 0.
|
| 507 |
"eval_VitaminC_cosine_recall": 1.0,
|
| 508 |
-
"eval_VitaminC_dot_accuracy": 0.
|
| 509 |
-
"eval_VitaminC_dot_accuracy_threshold":
|
| 510 |
-
"eval_VitaminC_dot_ap": 0.
|
| 511 |
-
"eval_VitaminC_dot_f1": 0.
|
| 512 |
-
"eval_VitaminC_dot_f1_threshold":
|
| 513 |
-
"eval_VitaminC_dot_precision": 0.
|
| 514 |
"eval_VitaminC_dot_recall": 1.0,
|
| 515 |
-
"eval_VitaminC_euclidean_accuracy": 0.
|
| 516 |
-
"eval_VitaminC_euclidean_accuracy_threshold":
|
| 517 |
-
"eval_VitaminC_euclidean_ap": 0.
|
| 518 |
-
"eval_VitaminC_euclidean_f1": 0.
|
| 519 |
-
"eval_VitaminC_euclidean_f1_threshold":
|
| 520 |
-
"eval_VitaminC_euclidean_precision": 0.
|
| 521 |
"eval_VitaminC_euclidean_recall": 1.0,
|
| 522 |
-
"eval_VitaminC_manhattan_accuracy": 0.
|
| 523 |
-
"eval_VitaminC_manhattan_accuracy_threshold":
|
| 524 |
-
"eval_VitaminC_manhattan_ap": 0.
|
| 525 |
-
"eval_VitaminC_manhattan_f1": 0.
|
| 526 |
-
"eval_VitaminC_manhattan_f1_threshold":
|
| 527 |
-
"eval_VitaminC_manhattan_precision": 0.
|
| 528 |
-
"eval_VitaminC_manhattan_recall":
|
| 529 |
-
"eval_VitaminC_max_accuracy": 0.
|
| 530 |
-
"eval_VitaminC_max_accuracy_threshold":
|
| 531 |
-
"eval_VitaminC_max_ap": 0.
|
| 532 |
-
"eval_VitaminC_max_f1": 0.
|
| 533 |
-
"eval_VitaminC_max_f1_threshold":
|
| 534 |
-
"eval_VitaminC_max_precision": 0.
|
| 535 |
"eval_VitaminC_max_recall": 1.0,
|
| 536 |
-
"eval_sequential_score": 0.
|
| 537 |
-
"eval_sts-test_pearson_cosine": 0.
|
| 538 |
-
"eval_sts-test_pearson_dot": 0.
|
| 539 |
-
"eval_sts-test_pearson_euclidean": 0.
|
| 540 |
-
"eval_sts-test_pearson_manhattan": 0.
|
| 541 |
-
"eval_sts-test_pearson_max": 0.
|
| 542 |
-
"eval_sts-test_spearman_cosine": 0.
|
| 543 |
-
"eval_sts-test_spearman_dot": 0.
|
| 544 |
-
"eval_sts-test_spearman_euclidean": 0.
|
| 545 |
-
"eval_sts-test_spearman_manhattan": 0.
|
| 546 |
-
"eval_sts-test_spearman_max": 0.
|
| 547 |
-
"eval_vitaminc-pairs_loss": 2.
|
| 548 |
-
"eval_vitaminc-pairs_runtime": 1.
|
| 549 |
-
"eval_vitaminc-pairs_samples_per_second": 73.
|
| 550 |
-
"eval_vitaminc-pairs_steps_per_second": 1.
|
| 551 |
"step": 81
|
| 552 |
},
|
| 553 |
{
|
| 554 |
-
"epoch": 0.
|
| 555 |
-
"eval_negation-triplets_loss":
|
| 556 |
-
"eval_negation-triplets_runtime": 0.
|
| 557 |
-
"eval_negation-triplets_samples_per_second":
|
| 558 |
-
"eval_negation-triplets_steps_per_second": 3.
|
| 559 |
"step": 81
|
| 560 |
},
|
| 561 |
{
|
| 562 |
-
"epoch": 0.
|
| 563 |
-
"eval_scitail-pairs-pos_loss": 0.
|
| 564 |
-
"eval_scitail-pairs-pos_runtime": 0.
|
| 565 |
-
"eval_scitail-pairs-pos_samples_per_second":
|
| 566 |
-
"eval_scitail-pairs-pos_steps_per_second": 2.
|
| 567 |
"step": 81
|
| 568 |
},
|
| 569 |
{
|
| 570 |
-
"epoch": 0.
|
| 571 |
-
"eval_xsum-pairs_loss":
|
| 572 |
-
"eval_xsum-pairs_runtime": 3.
|
| 573 |
-
"eval_xsum-pairs_samples_per_second": 40.
|
| 574 |
-
"eval_xsum-pairs_steps_per_second": 0.
|
| 575 |
"step": 81
|
| 576 |
},
|
| 577 |
{
|
| 578 |
-
"epoch": 0.
|
| 579 |
-
"eval_sciq_pairs_loss": 0.
|
| 580 |
-
"eval_sciq_pairs_runtime": 3.
|
| 581 |
-
"eval_sciq_pairs_samples_per_second": 39.
|
| 582 |
-
"eval_sciq_pairs_steps_per_second": 0.
|
| 583 |
"step": 81
|
| 584 |
},
|
| 585 |
{
|
| 586 |
-
"epoch": 0.
|
| 587 |
-
"eval_qasc_pairs_loss":
|
| 588 |
-
"eval_qasc_pairs_runtime": 0.
|
| 589 |
-
"eval_qasc_pairs_samples_per_second":
|
| 590 |
-
"eval_qasc_pairs_steps_per_second": 3.
|
| 591 |
"step": 81
|
| 592 |
},
|
| 593 |
{
|
| 594 |
-
"epoch": 0.
|
| 595 |
-
"eval_openbookqa_pairs_loss":
|
| 596 |
-
"eval_openbookqa_pairs_runtime": 0.
|
| 597 |
-
"eval_openbookqa_pairs_samples_per_second":
|
| 598 |
-
"eval_openbookqa_pairs_steps_per_second": 3.
|
| 599 |
"step": 81
|
| 600 |
},
|
| 601 |
{
|
| 602 |
-
"epoch": 0.
|
| 603 |
-
"eval_msmarco_pairs_loss":
|
| 604 |
-
"eval_msmarco_pairs_runtime": 1.
|
| 605 |
-
"eval_msmarco_pairs_samples_per_second":
|
| 606 |
-
"eval_msmarco_pairs_steps_per_second": 1.
|
| 607 |
"step": 81
|
| 608 |
},
|
| 609 |
{
|
| 610 |
-
"epoch": 0.
|
| 611 |
-
"eval_nq_pairs_loss":
|
| 612 |
-
"eval_nq_pairs_runtime": 2.
|
| 613 |
-
"eval_nq_pairs_samples_per_second": 53.
|
| 614 |
-
"eval_nq_pairs_steps_per_second": 0.
|
| 615 |
"step": 81
|
| 616 |
},
|
| 617 |
{
|
| 618 |
-
"epoch": 0.
|
| 619 |
-
"eval_trivia_pairs_loss":
|
| 620 |
-
"eval_trivia_pairs_runtime": 4.
|
| 621 |
-
"eval_trivia_pairs_samples_per_second": 28.
|
| 622 |
-
"eval_trivia_pairs_steps_per_second": 0.
|
| 623 |
"step": 81
|
| 624 |
},
|
| 625 |
{
|
| 626 |
-
"epoch": 0.
|
| 627 |
-
"eval_gooaq_pairs_loss":
|
| 628 |
-
"eval_gooaq_pairs_runtime": 0.
|
| 629 |
-
"eval_gooaq_pairs_samples_per_second": 146.
|
| 630 |
-
"eval_gooaq_pairs_steps_per_second": 2.
|
| 631 |
"step": 81
|
| 632 |
},
|
| 633 |
{
|
| 634 |
-
"epoch": 0.
|
| 635 |
-
"eval_paws-pos_loss": 0.
|
| 636 |
-
"eval_paws-pos_runtime": 0.
|
| 637 |
-
"eval_paws-pos_samples_per_second":
|
| 638 |
-
"eval_paws-pos_steps_per_second": 2.
|
| 639 |
"step": 81
|
| 640 |
},
|
| 641 |
{
|
| 642 |
-
"epoch": 0.
|
| 643 |
-
"grad_norm":
|
| 644 |
-
"learning_rate":
|
| 645 |
-
"loss":
|
| 646 |
"step": 84
|
| 647 |
},
|
| 648 |
{
|
| 649 |
-
"epoch": 0.
|
| 650 |
-
"grad_norm":
|
| 651 |
-
"learning_rate":
|
| 652 |
-
"loss":
|
| 653 |
"step": 87
|
| 654 |
},
|
| 655 |
{
|
| 656 |
-
"epoch": 0.
|
| 657 |
-
"grad_norm":
|
| 658 |
-
"learning_rate":
|
| 659 |
-
"loss":
|
| 660 |
"step": 90
|
| 661 |
},
|
| 662 |
{
|
| 663 |
-
"epoch": 0.
|
| 664 |
-
"grad_norm":
|
| 665 |
-
"learning_rate":
|
| 666 |
-
"loss":
|
| 667 |
"step": 93
|
| 668 |
},
|
| 669 |
{
|
| 670 |
-
"epoch": 0.
|
| 671 |
-
"grad_norm":
|
| 672 |
-
"learning_rate":
|
| 673 |
-
"loss":
|
| 674 |
"step": 96
|
| 675 |
},
|
| 676 |
{
|
| 677 |
-
"epoch": 0.
|
| 678 |
-
"grad_norm":
|
| 679 |
-
"learning_rate":
|
| 680 |
-
"loss":
|
| 681 |
"step": 99
|
| 682 |
},
|
| 683 |
{
|
| 684 |
-
"epoch": 0.
|
| 685 |
-
"grad_norm":
|
| 686 |
-
"learning_rate":
|
| 687 |
-
"loss":
|
| 688 |
"step": 102
|
| 689 |
},
|
| 690 |
{
|
| 691 |
-
"epoch": 0.
|
| 692 |
-
"grad_norm":
|
| 693 |
-
"learning_rate":
|
| 694 |
-
"loss":
|
| 695 |
"step": 105
|
| 696 |
}
|
| 697 |
],
|
| 698 |
"logging_steps": 3,
|
| 699 |
-
"max_steps":
|
| 700 |
"num_input_tokens_seen": 0,
|
| 701 |
"num_train_epochs": 3,
|
| 702 |
"save_steps": 107,
|
|
@@ -713,7 +713,7 @@
|
|
| 713 |
}
|
| 714 |
},
|
| 715 |
"total_flos": 0.0,
|
| 716 |
-
"train_batch_size":
|
| 717 |
"trial_name": null,
|
| 718 |
"trial_params": null
|
| 719 |
}
|
|
|
|
| 1 |
{
|
| 2 |
"best_metric": null,
|
| 3 |
"best_model_checkpoint": null,
|
| 4 |
+
"epoch": 0.601123595505618,
|
| 5 |
"eval_steps": 27,
|
| 6 |
"global_step": 107,
|
| 7 |
"is_hyper_param_search": false,
|
|
|
|
| 9 |
"is_world_process_zero": true,
|
| 10 |
"log_history": [
|
| 11 |
{
|
| 12 |
+
"epoch": 0.016853932584269662,
|
| 13 |
+
"grad_norm": 2.9885776042938232,
|
| 14 |
+
"learning_rate": 1.4814814814814815e-06,
|
| 15 |
+
"loss": 0.6012,
|
| 16 |
"step": 3
|
| 17 |
},
|
| 18 |
{
|
| 19 |
+
"epoch": 0.033707865168539325,
|
| 20 |
+
"grad_norm": 3.184929132461548,
|
| 21 |
+
"learning_rate": 2.962962962962963e-06,
|
| 22 |
+
"loss": 0.7573,
|
| 23 |
"step": 6
|
| 24 |
},
|
| 25 |
{
|
| 26 |
+
"epoch": 0.05056179775280899,
|
| 27 |
+
"grad_norm": 3.256159782409668,
|
| 28 |
+
"learning_rate": 4.444444444444444e-06,
|
| 29 |
+
"loss": 0.9212,
|
| 30 |
"step": 9
|
| 31 |
},
|
| 32 |
{
|
| 33 |
+
"epoch": 0.06741573033707865,
|
| 34 |
+
"grad_norm": 2.833339214324951,
|
| 35 |
+
"learning_rate": 5.925925925925926e-06,
|
| 36 |
+
"loss": 0.6117,
|
| 37 |
"step": 12
|
| 38 |
},
|
| 39 |
{
|
| 40 |
+
"epoch": 0.08426966292134831,
|
| 41 |
+
"grad_norm": 3.08292818069458,
|
| 42 |
+
"learning_rate": 7.4074074074074075e-06,
|
| 43 |
+
"loss": 0.8545,
|
| 44 |
"step": 15
|
| 45 |
},
|
| 46 |
{
|
| 47 |
+
"epoch": 0.10112359550561797,
|
| 48 |
+
"grad_norm": 2.317431688308716,
|
| 49 |
+
"learning_rate": 8.888888888888888e-06,
|
| 50 |
+
"loss": 0.6515,
|
| 51 |
"step": 18
|
| 52 |
},
|
| 53 |
{
|
| 54 |
+
"epoch": 0.11797752808988764,
|
| 55 |
+
"grad_norm": 2.9611644744873047,
|
| 56 |
+
"learning_rate": 1.037037037037037e-05,
|
| 57 |
+
"loss": 0.7159,
|
| 58 |
"step": 21
|
| 59 |
},
|
| 60 |
{
|
| 61 |
+
"epoch": 0.1348314606741573,
|
| 62 |
+
"grad_norm": 2.698537826538086,
|
| 63 |
+
"learning_rate": 1.1851851851851852e-05,
|
| 64 |
+
"loss": 0.7019,
|
| 65 |
"step": 24
|
| 66 |
},
|
| 67 |
{
|
| 68 |
+
"epoch": 0.15168539325842698,
|
| 69 |
+
"grad_norm": 2.222154378890991,
|
| 70 |
+
"learning_rate": 1.3333333333333333e-05,
|
| 71 |
+
"loss": 0.4411,
|
| 72 |
"step": 27
|
| 73 |
},
|
| 74 |
{
|
| 75 |
+
"epoch": 0.15168539325842698,
|
| 76 |
"eval_NLI-v2_cosine_accuracy": 1.0,
|
| 77 |
+
"eval_NLI-v2_dot_accuracy": 0.0,
|
| 78 |
"eval_NLI-v2_euclidean_accuracy": 1.0,
|
| 79 |
"eval_NLI-v2_manhattan_accuracy": 1.0,
|
| 80 |
"eval_NLI-v2_max_accuracy": 1.0,
|
| 81 |
+
"eval_VitaminC_cosine_accuracy": 0.578125,
|
| 82 |
+
"eval_VitaminC_cosine_accuracy_threshold": 0.7817381620407104,
|
| 83 |
+
"eval_VitaminC_cosine_ap": 0.5507972943944112,
|
| 84 |
+
"eval_VitaminC_cosine_f1": 0.6595174262734584,
|
| 85 |
+
"eval_VitaminC_cosine_f1_threshold": 0.28573715686798096,
|
| 86 |
+
"eval_VitaminC_cosine_precision": 0.492,
|
| 87 |
"eval_VitaminC_cosine_recall": 1.0,
|
| 88 |
+
"eval_VitaminC_dot_accuracy": 0.5703125,
|
| 89 |
+
"eval_VitaminC_dot_accuracy_threshold": 316.7283020019531,
|
| 90 |
+
"eval_VitaminC_dot_ap": 0.5511866185449577,
|
| 91 |
+
"eval_VitaminC_dot_f1": 0.6577540106951871,
|
| 92 |
+
"eval_VitaminC_dot_f1_threshold": 106.75863647460938,
|
| 93 |
+
"eval_VitaminC_dot_precision": 0.4900398406374502,
|
| 94 |
"eval_VitaminC_dot_recall": 1.0,
|
| 95 |
+
"eval_VitaminC_euclidean_accuracy": 0.578125,
|
| 96 |
+
"eval_VitaminC_euclidean_accuracy_threshold": 13.298419952392578,
|
| 97 |
+
"eval_VitaminC_euclidean_ap": 0.5476323986807207,
|
| 98 |
+
"eval_VitaminC_euclidean_f1": 0.6577540106951871,
|
| 99 |
+
"eval_VitaminC_euclidean_f1_threshold": 23.83933448791504,
|
| 100 |
+
"eval_VitaminC_euclidean_precision": 0.4900398406374502,
|
| 101 |
+
"eval_VitaminC_euclidean_recall": 1.0,
|
| 102 |
+
"eval_VitaminC_manhattan_accuracy": 0.578125,
|
| 103 |
+
"eval_VitaminC_manhattan_accuracy_threshold": 279.69085693359375,
|
| 104 |
+
"eval_VitaminC_manhattan_ap": 0.5412538781107805,
|
| 105 |
+
"eval_VitaminC_manhattan_f1": 0.6577540106951871,
|
| 106 |
+
"eval_VitaminC_manhattan_f1_threshold": 499.8836364746094,
|
| 107 |
+
"eval_VitaminC_manhattan_precision": 0.4900398406374502,
|
| 108 |
"eval_VitaminC_manhattan_recall": 1.0,
|
| 109 |
+
"eval_VitaminC_max_accuracy": 0.578125,
|
| 110 |
+
"eval_VitaminC_max_accuracy_threshold": 316.7283020019531,
|
| 111 |
+
"eval_VitaminC_max_ap": 0.5511866185449577,
|
| 112 |
+
"eval_VitaminC_max_f1": 0.6595174262734584,
|
| 113 |
+
"eval_VitaminC_max_f1_threshold": 499.8836364746094,
|
| 114 |
+
"eval_VitaminC_max_precision": 0.492,
|
| 115 |
"eval_VitaminC_max_recall": 1.0,
|
| 116 |
+
"eval_sequential_score": 0.5511866185449577,
|
| 117 |
+
"eval_sts-test_pearson_cosine": 0.8488243436029344,
|
| 118 |
+
"eval_sts-test_pearson_dot": 0.8480167969551653,
|
| 119 |
+
"eval_sts-test_pearson_euclidean": 0.8800283985117625,
|
| 120 |
+
"eval_sts-test_pearson_manhattan": 0.880588311422627,
|
| 121 |
+
"eval_sts-test_pearson_max": 0.880588311422627,
|
| 122 |
+
"eval_sts-test_spearman_cosine": 0.8905659331642088,
|
| 123 |
+
"eval_sts-test_spearman_dot": 0.8692084657204004,
|
| 124 |
+
"eval_sts-test_spearman_euclidean": 0.8809566840232712,
|
| 125 |
+
"eval_sts-test_spearman_manhattan": 0.883434007028195,
|
| 126 |
+
"eval_sts-test_spearman_max": 0.8905659331642088,
|
| 127 |
+
"eval_vitaminc-pairs_loss": 2.465860366821289,
|
| 128 |
+
"eval_vitaminc-pairs_runtime": 1.4615,
|
| 129 |
+
"eval_vitaminc-pairs_samples_per_second": 73.899,
|
| 130 |
+
"eval_vitaminc-pairs_steps_per_second": 1.368,
|
| 131 |
"step": 27
|
| 132 |
},
|
| 133 |
{
|
| 134 |
+
"epoch": 0.15168539325842698,
|
| 135 |
+
"eval_negation-triplets_loss": 1.7310789823532104,
|
| 136 |
+
"eval_negation-triplets_runtime": 0.3009,
|
| 137 |
+
"eval_negation-triplets_samples_per_second": 212.692,
|
| 138 |
+
"eval_negation-triplets_steps_per_second": 3.323,
|
| 139 |
"step": 27
|
| 140 |
},
|
| 141 |
{
|
| 142 |
+
"epoch": 0.15168539325842698,
|
| 143 |
+
"eval_scitail-pairs-pos_loss": 0.1150394082069397,
|
| 144 |
+
"eval_scitail-pairs-pos_runtime": 0.3739,
|
| 145 |
+
"eval_scitail-pairs-pos_samples_per_second": 144.431,
|
| 146 |
+
"eval_scitail-pairs-pos_steps_per_second": 2.675,
|
| 147 |
"step": 27
|
| 148 |
},
|
| 149 |
{
|
| 150 |
+
"epoch": 0.15168539325842698,
|
| 151 |
+
"eval_xsum-pairs_loss": 0.11168850213289261,
|
| 152 |
+
"eval_xsum-pairs_runtime": 3.1697,
|
| 153 |
+
"eval_xsum-pairs_samples_per_second": 40.382,
|
| 154 |
+
"eval_xsum-pairs_steps_per_second": 0.631,
|
| 155 |
"step": 27
|
| 156 |
},
|
| 157 |
{
|
| 158 |
+
"epoch": 0.15168539325842698,
|
| 159 |
+
"eval_sciq_pairs_loss": 0.03450964391231537,
|
| 160 |
+
"eval_sciq_pairs_runtime": 3.3283,
|
| 161 |
+
"eval_sciq_pairs_samples_per_second": 38.459,
|
| 162 |
+
"eval_sciq_pairs_steps_per_second": 0.601,
|
| 163 |
"step": 27
|
| 164 |
},
|
| 165 |
{
|
| 166 |
+
"epoch": 0.15168539325842698,
|
| 167 |
+
"eval_qasc_pairs_loss": 0.11095743626356125,
|
| 168 |
+
"eval_qasc_pairs_runtime": 0.6261,
|
| 169 |
+
"eval_qasc_pairs_samples_per_second": 204.45,
|
| 170 |
+
"eval_qasc_pairs_steps_per_second": 3.195,
|
| 171 |
"step": 27
|
| 172 |
},
|
| 173 |
{
|
| 174 |
+
"epoch": 0.15168539325842698,
|
| 175 |
+
"eval_openbookqa_pairs_loss": 0.7092063426971436,
|
| 176 |
+
"eval_openbookqa_pairs_runtime": 0.5866,
|
| 177 |
+
"eval_openbookqa_pairs_samples_per_second": 218.19,
|
| 178 |
+
"eval_openbookqa_pairs_steps_per_second": 3.409,
|
| 179 |
"step": 27
|
| 180 |
},
|
| 181 |
{
|
| 182 |
+
"epoch": 0.15168539325842698,
|
| 183 |
+
"eval_msmarco_pairs_loss": 0.3955218493938446,
|
| 184 |
+
"eval_msmarco_pairs_runtime": 1.2942,
|
| 185 |
+
"eval_msmarco_pairs_samples_per_second": 98.902,
|
| 186 |
+
"eval_msmarco_pairs_steps_per_second": 1.545,
|
| 187 |
"step": 27
|
| 188 |
},
|
| 189 |
{
|
| 190 |
+
"epoch": 0.15168539325842698,
|
| 191 |
+
"eval_nq_pairs_loss": 0.42051073908805847,
|
| 192 |
+
"eval_nq_pairs_runtime": 2.3875,
|
| 193 |
+
"eval_nq_pairs_samples_per_second": 53.612,
|
| 194 |
+
"eval_nq_pairs_steps_per_second": 0.838,
|
| 195 |
"step": 27
|
| 196 |
},
|
| 197 |
{
|
| 198 |
+
"epoch": 0.15168539325842698,
|
| 199 |
+
"eval_trivia_pairs_loss": 0.93178790807724,
|
| 200 |
+
"eval_trivia_pairs_runtime": 4.4363,
|
| 201 |
+
"eval_trivia_pairs_samples_per_second": 28.853,
|
| 202 |
+
"eval_trivia_pairs_steps_per_second": 0.451,
|
| 203 |
"step": 27
|
| 204 |
},
|
| 205 |
{
|
| 206 |
+
"epoch": 0.15168539325842698,
|
| 207 |
+
"eval_gooaq_pairs_loss": 0.6505913138389587,
|
| 208 |
+
"eval_gooaq_pairs_runtime": 0.8826,
|
| 209 |
+
"eval_gooaq_pairs_samples_per_second": 145.027,
|
| 210 |
+
"eval_gooaq_pairs_steps_per_second": 2.266,
|
| 211 |
"step": 27
|
| 212 |
},
|
| 213 |
{
|
| 214 |
+
"epoch": 0.15168539325842698,
|
| 215 |
+
"eval_paws-pos_loss": 0.024931101128458977,
|
| 216 |
+
"eval_paws-pos_runtime": 0.6852,
|
| 217 |
+
"eval_paws-pos_samples_per_second": 186.805,
|
| 218 |
+
"eval_paws-pos_steps_per_second": 2.919,
|
| 219 |
"step": 27
|
| 220 |
},
|
| 221 |
{
|
| 222 |
+
"epoch": 0.16853932584269662,
|
| 223 |
+
"grad_norm": 2.826900005340576,
|
| 224 |
+
"learning_rate": 1.4814814814814815e-05,
|
| 225 |
+
"loss": 0.5125,
|
| 226 |
"step": 30
|
| 227 |
},
|
| 228 |
{
|
| 229 |
+
"epoch": 0.1853932584269663,
|
| 230 |
+
"grad_norm": 2.9938910007476807,
|
| 231 |
+
"learning_rate": 1.6296296296296297e-05,
|
| 232 |
+
"loss": 0.6885,
|
| 233 |
"step": 33
|
| 234 |
},
|
| 235 |
{
|
| 236 |
+
"epoch": 0.20224719101123595,
|
| 237 |
+
"grad_norm": 3.3046395778656006,
|
| 238 |
+
"learning_rate": 1.7777777777777777e-05,
|
| 239 |
+
"loss": 0.6435,
|
| 240 |
"step": 36
|
| 241 |
},
|
| 242 |
{
|
| 243 |
+
"epoch": 0.21910112359550563,
|
| 244 |
+
"grad_norm": 2.4184651374816895,
|
| 245 |
+
"learning_rate": 1.925925925925926e-05,
|
| 246 |
+
"loss": 0.753,
|
| 247 |
"step": 39
|
| 248 |
},
|
| 249 |
{
|
| 250 |
+
"epoch": 0.23595505617977527,
|
| 251 |
+
"grad_norm": 2.9905433654785156,
|
| 252 |
+
"learning_rate": 2.074074074074074e-05,
|
| 253 |
+
"loss": 0.7427,
|
| 254 |
"step": 42
|
| 255 |
},
|
| 256 |
{
|
| 257 |
+
"epoch": 0.25280898876404495,
|
| 258 |
+
"grad_norm": 2.745820999145508,
|
| 259 |
+
"learning_rate": 2.2222222222222227e-05,
|
| 260 |
+
"loss": 0.5083,
|
| 261 |
"step": 45
|
| 262 |
},
|
| 263 |
{
|
| 264 |
+
"epoch": 0.2696629213483146,
|
| 265 |
+
"grad_norm": 2.6370577812194824,
|
| 266 |
+
"learning_rate": 2.3703703703703703e-05,
|
| 267 |
+
"loss": 0.7454,
|
| 268 |
"step": 48
|
| 269 |
},
|
| 270 |
{
|
| 271 |
+
"epoch": 0.28651685393258425,
|
| 272 |
+
"grad_norm": 3.044011116027832,
|
| 273 |
+
"learning_rate": 2.5185185185185187e-05,
|
| 274 |
+
"loss": 0.8356,
|
| 275 |
"step": 51
|
| 276 |
},
|
| 277 |
{
|
| 278 |
+
"epoch": 0.30337078651685395,
|
| 279 |
+
"grad_norm": 3.718804121017456,
|
| 280 |
+
"learning_rate": 2.6666666666666667e-05,
|
| 281 |
+
"loss": 0.8864,
|
| 282 |
"step": 54
|
| 283 |
},
|
| 284 |
{
|
| 285 |
+
"epoch": 0.30337078651685395,
|
| 286 |
"eval_NLI-v2_cosine_accuracy": 1.0,
|
| 287 |
+
"eval_NLI-v2_dot_accuracy": 0.0,
|
| 288 |
"eval_NLI-v2_euclidean_accuracy": 1.0,
|
| 289 |
"eval_NLI-v2_manhattan_accuracy": 1.0,
|
| 290 |
"eval_NLI-v2_max_accuracy": 1.0,
|
| 291 |
+
"eval_VitaminC_cosine_accuracy": 0.57421875,
|
| 292 |
+
"eval_VitaminC_cosine_accuracy_threshold": 0.7991844415664673,
|
| 293 |
+
"eval_VitaminC_cosine_ap": 0.5485498837322925,
|
| 294 |
+
"eval_VitaminC_cosine_f1": 0.6595174262734584,
|
| 295 |
+
"eval_VitaminC_cosine_f1_threshold": 0.3160865008831024,
|
| 296 |
+
"eval_VitaminC_cosine_precision": 0.492,
|
| 297 |
+
"eval_VitaminC_cosine_recall": 1.0,
|
| 298 |
+
"eval_VitaminC_dot_accuracy": 0.578125,
|
| 299 |
+
"eval_VitaminC_dot_accuracy_threshold": 327.0416564941406,
|
| 300 |
+
"eval_VitaminC_dot_ap": 0.54993134882601,
|
| 301 |
+
"eval_VitaminC_dot_f1": 0.6595174262734584,
|
| 302 |
+
"eval_VitaminC_dot_f1_threshold": 117.44181060791016,
|
| 303 |
+
"eval_VitaminC_dot_precision": 0.492,
|
| 304 |
+
"eval_VitaminC_dot_recall": 1.0,
|
| 305 |
+
"eval_VitaminC_euclidean_accuracy": 0.57421875,
|
| 306 |
+
"eval_VitaminC_euclidean_accuracy_threshold": 13.019258499145508,
|
| 307 |
+
"eval_VitaminC_euclidean_ap": 0.5435066540334542,
|
| 308 |
+
"eval_VitaminC_euclidean_f1": 0.6577540106951871,
|
| 309 |
+
"eval_VitaminC_euclidean_f1_threshold": 23.688644409179688,
|
| 310 |
+
"eval_VitaminC_euclidean_precision": 0.4900398406374502,
|
| 311 |
+
"eval_VitaminC_euclidean_recall": 1.0,
|
| 312 |
+
"eval_VitaminC_manhattan_accuracy": 0.57421875,
|
| 313 |
+
"eval_VitaminC_manhattan_accuracy_threshold": 283.876220703125,
|
| 314 |
+
"eval_VitaminC_manhattan_ap": 0.5416615397828658,
|
| 315 |
+
"eval_VitaminC_manhattan_f1": 0.6559999999999999,
|
| 316 |
+
"eval_VitaminC_manhattan_f1_threshold": 514.0216064453125,
|
| 317 |
+
"eval_VitaminC_manhattan_precision": 0.4880952380952381,
|
| 318 |
+
"eval_VitaminC_manhattan_recall": 1.0,
|
| 319 |
+
"eval_VitaminC_max_accuracy": 0.578125,
|
| 320 |
+
"eval_VitaminC_max_accuracy_threshold": 327.0416564941406,
|
| 321 |
+
"eval_VitaminC_max_ap": 0.54993134882601,
|
| 322 |
+
"eval_VitaminC_max_f1": 0.6595174262734584,
|
| 323 |
+
"eval_VitaminC_max_f1_threshold": 514.0216064453125,
|
| 324 |
+
"eval_VitaminC_max_precision": 0.492,
|
| 325 |
+
"eval_VitaminC_max_recall": 1.0,
|
| 326 |
+
"eval_sequential_score": 0.54993134882601,
|
| 327 |
+
"eval_sts-test_pearson_cosine": 0.8452615878553369,
|
| 328 |
+
"eval_sts-test_pearson_dot": 0.8404858620687519,
|
| 329 |
+
"eval_sts-test_pearson_euclidean": 0.8780527810910925,
|
| 330 |
+
"eval_sts-test_pearson_manhattan": 0.878916157345712,
|
| 331 |
+
"eval_sts-test_pearson_max": 0.878916157345712,
|
| 332 |
+
"eval_sts-test_spearman_cosine": 0.8876915367075635,
|
| 333 |
+
"eval_sts-test_spearman_dot": 0.8608104875327304,
|
| 334 |
+
"eval_sts-test_spearman_euclidean": 0.8804138856889071,
|
| 335 |
+
"eval_sts-test_spearman_manhattan": 0.8822803815444743,
|
| 336 |
+
"eval_sts-test_spearman_max": 0.8876915367075635,
|
| 337 |
+
"eval_vitaminc-pairs_loss": 2.454524040222168,
|
| 338 |
+
"eval_vitaminc-pairs_runtime": 1.4583,
|
| 339 |
+
"eval_vitaminc-pairs_samples_per_second": 74.057,
|
| 340 |
+
"eval_vitaminc-pairs_steps_per_second": 1.371,
|
| 341 |
"step": 54
|
| 342 |
},
|
| 343 |
{
|
| 344 |
+
"epoch": 0.30337078651685395,
|
| 345 |
+
"eval_negation-triplets_loss": 1.7277792692184448,
|
| 346 |
+
"eval_negation-triplets_runtime": 0.3027,
|
| 347 |
+
"eval_negation-triplets_samples_per_second": 211.436,
|
| 348 |
+
"eval_negation-triplets_steps_per_second": 3.304,
|
| 349 |
"step": 54
|
| 350 |
},
|
| 351 |
{
|
| 352 |
+
"epoch": 0.30337078651685395,
|
| 353 |
+
"eval_scitail-pairs-pos_loss": 0.11168555170297623,
|
| 354 |
+
"eval_scitail-pairs-pos_runtime": 0.3726,
|
| 355 |
+
"eval_scitail-pairs-pos_samples_per_second": 144.911,
|
| 356 |
+
"eval_scitail-pairs-pos_steps_per_second": 2.684,
|
| 357 |
"step": 54
|
| 358 |
},
|
| 359 |
{
|
| 360 |
+
"epoch": 0.30337078651685395,
|
| 361 |
+
"eval_xsum-pairs_loss": 0.10087604075670242,
|
| 362 |
+
"eval_xsum-pairs_runtime": 3.1701,
|
| 363 |
+
"eval_xsum-pairs_samples_per_second": 40.377,
|
| 364 |
+
"eval_xsum-pairs_steps_per_second": 0.631,
|
| 365 |
"step": 54
|
| 366 |
},
|
| 367 |
{
|
| 368 |
+
"epoch": 0.30337078651685395,
|
| 369 |
+
"eval_sciq_pairs_loss": 0.03466618433594704,
|
| 370 |
+
"eval_sciq_pairs_runtime": 3.3778,
|
| 371 |
+
"eval_sciq_pairs_samples_per_second": 37.895,
|
| 372 |
+
"eval_sciq_pairs_steps_per_second": 0.592,
|
| 373 |
"step": 54
|
| 374 |
},
|
| 375 |
{
|
| 376 |
+
"epoch": 0.30337078651685395,
|
| 377 |
+
"eval_qasc_pairs_loss": 0.10551701486110687,
|
| 378 |
+
"eval_qasc_pairs_runtime": 0.6271,
|
| 379 |
+
"eval_qasc_pairs_samples_per_second": 204.125,
|
| 380 |
+
"eval_qasc_pairs_steps_per_second": 3.189,
|
| 381 |
"step": 54
|
| 382 |
},
|
| 383 |
{
|
| 384 |
+
"epoch": 0.30337078651685395,
|
| 385 |
+
"eval_openbookqa_pairs_loss": 0.7239958643913269,
|
| 386 |
+
"eval_openbookqa_pairs_runtime": 0.5811,
|
| 387 |
+
"eval_openbookqa_pairs_samples_per_second": 220.255,
|
| 388 |
+
"eval_openbookqa_pairs_steps_per_second": 3.441,
|
| 389 |
"step": 54
|
| 390 |
},
|
| 391 |
{
|
| 392 |
+
"epoch": 0.30337078651685395,
|
| 393 |
+
"eval_msmarco_pairs_loss": 0.3808779716491699,
|
| 394 |
+
"eval_msmarco_pairs_runtime": 1.2919,
|
| 395 |
+
"eval_msmarco_pairs_samples_per_second": 99.082,
|
| 396 |
+
"eval_msmarco_pairs_steps_per_second": 1.548,
|
| 397 |
"step": 54
|
| 398 |
},
|
| 399 |
{
|
| 400 |
+
"epoch": 0.30337078651685395,
|
| 401 |
+
"eval_nq_pairs_loss": 0.44170400500297546,
|
| 402 |
+
"eval_nq_pairs_runtime": 2.3835,
|
| 403 |
+
"eval_nq_pairs_samples_per_second": 53.703,
|
| 404 |
+
"eval_nq_pairs_steps_per_second": 0.839,
|
| 405 |
"step": 54
|
| 406 |
},
|
| 407 |
{
|
| 408 |
+
"epoch": 0.30337078651685395,
|
| 409 |
+
"eval_trivia_pairs_loss": 0.9158428907394409,
|
| 410 |
+
"eval_trivia_pairs_runtime": 4.4326,
|
| 411 |
+
"eval_trivia_pairs_samples_per_second": 28.877,
|
| 412 |
+
"eval_trivia_pairs_steps_per_second": 0.451,
|
| 413 |
"step": 54
|
| 414 |
},
|
| 415 |
{
|
| 416 |
+
"epoch": 0.30337078651685395,
|
| 417 |
+
"eval_gooaq_pairs_loss": 0.6208247542381287,
|
| 418 |
+
"eval_gooaq_pairs_runtime": 0.8797,
|
| 419 |
+
"eval_gooaq_pairs_samples_per_second": 145.497,
|
| 420 |
+
"eval_gooaq_pairs_steps_per_second": 2.273,
|
| 421 |
"step": 54
|
| 422 |
},
|
| 423 |
{
|
| 424 |
+
"epoch": 0.30337078651685395,
|
| 425 |
+
"eval_paws-pos_loss": 0.02517784759402275,
|
| 426 |
+
"eval_paws-pos_runtime": 0.694,
|
| 427 |
+
"eval_paws-pos_samples_per_second": 184.442,
|
| 428 |
+
"eval_paws-pos_steps_per_second": 2.882,
|
| 429 |
"step": 54
|
| 430 |
},
|
| 431 |
{
|
| 432 |
+
"epoch": 0.3202247191011236,
|
| 433 |
+
"grad_norm": 2.173736572265625,
|
| 434 |
+
"learning_rate": 2.814814814814815e-05,
|
| 435 |
+
"loss": 0.6015,
|
| 436 |
"step": 57
|
| 437 |
},
|
| 438 |
{
|
| 439 |
+
"epoch": 0.33707865168539325,
|
| 440 |
+
"grad_norm": 3.8964712619781494,
|
| 441 |
+
"learning_rate": 2.962962962962963e-05,
|
| 442 |
+
"loss": 0.9482,
|
| 443 |
"step": 60
|
| 444 |
},
|
| 445 |
{
|
| 446 |
+
"epoch": 0.3539325842696629,
|
| 447 |
+
"grad_norm": 2.659498691558838,
|
| 448 |
+
"learning_rate": 3.111111111111112e-05,
|
| 449 |
+
"loss": 0.5404,
|
| 450 |
"step": 63
|
| 451 |
},
|
| 452 |
{
|
| 453 |
+
"epoch": 0.3707865168539326,
|
| 454 |
+
"grad_norm": 3.3499844074249268,
|
| 455 |
+
"learning_rate": 3.259259259259259e-05,
|
| 456 |
+
"loss": 0.805,
|
| 457 |
"step": 66
|
| 458 |
},
|
| 459 |
{
|
| 460 |
+
"epoch": 0.38764044943820225,
|
| 461 |
+
"grad_norm": 3.770142078399658,
|
| 462 |
+
"learning_rate": 3.4074074074074077e-05,
|
| 463 |
+
"loss": 0.7184,
|
| 464 |
"step": 69
|
| 465 |
},
|
| 466 |
{
|
| 467 |
+
"epoch": 0.4044943820224719,
|
| 468 |
+
"grad_norm": 3.740880012512207,
|
| 469 |
+
"learning_rate": 3.555555555555555e-05,
|
| 470 |
+
"loss": 0.8708,
|
| 471 |
"step": 72
|
| 472 |
},
|
| 473 |
{
|
| 474 |
+
"epoch": 0.42134831460674155,
|
| 475 |
+
"grad_norm": 2.981106996536255,
|
| 476 |
+
"learning_rate": 3.703703703703704e-05,
|
| 477 |
+
"loss": 0.8327,
|
| 478 |
"step": 75
|
| 479 |
},
|
| 480 |
{
|
| 481 |
+
"epoch": 0.43820224719101125,
|
| 482 |
+
"grad_norm": 2.3469011783599854,
|
| 483 |
+
"learning_rate": 3.851851851851852e-05,
|
| 484 |
+
"loss": 0.5025,
|
| 485 |
"step": 78
|
| 486 |
},
|
| 487 |
{
|
| 488 |
+
"epoch": 0.4550561797752809,
|
| 489 |
+
"grad_norm": 3.296035051345825,
|
| 490 |
+
"learning_rate": 4e-05,
|
| 491 |
+
"loss": 0.6517,
|
| 492 |
"step": 81
|
| 493 |
},
|
| 494 |
{
|
| 495 |
+
"epoch": 0.4550561797752809,
|
| 496 |
"eval_NLI-v2_cosine_accuracy": 1.0,
|
| 497 |
"eval_NLI-v2_dot_accuracy": 0.0,
|
| 498 |
"eval_NLI-v2_euclidean_accuracy": 1.0,
|
| 499 |
"eval_NLI-v2_manhattan_accuracy": 1.0,
|
| 500 |
"eval_NLI-v2_max_accuracy": 1.0,
|
| 501 |
+
"eval_VitaminC_cosine_accuracy": 0.578125,
|
| 502 |
+
"eval_VitaminC_cosine_accuracy_threshold": 0.7859437465667725,
|
| 503 |
+
"eval_VitaminC_cosine_ap": 0.5557444337961499,
|
| 504 |
+
"eval_VitaminC_cosine_f1": 0.6595174262734584,
|
| 505 |
+
"eval_VitaminC_cosine_f1_threshold": 0.3211573362350464,
|
| 506 |
+
"eval_VitaminC_cosine_precision": 0.492,
|
| 507 |
"eval_VitaminC_cosine_recall": 1.0,
|
| 508 |
+
"eval_VitaminC_dot_accuracy": 0.578125,
|
| 509 |
+
"eval_VitaminC_dot_accuracy_threshold": 315.9444580078125,
|
| 510 |
+
"eval_VitaminC_dot_ap": 0.5539524528858992,
|
| 511 |
+
"eval_VitaminC_dot_f1": 0.6595174262734584,
|
| 512 |
+
"eval_VitaminC_dot_f1_threshold": 129.88558959960938,
|
| 513 |
+
"eval_VitaminC_dot_precision": 0.492,
|
| 514 |
"eval_VitaminC_dot_recall": 1.0,
|
| 515 |
+
"eval_VitaminC_euclidean_accuracy": 0.58203125,
|
| 516 |
+
"eval_VitaminC_euclidean_accuracy_threshold": 13.113249778747559,
|
| 517 |
+
"eval_VitaminC_euclidean_ap": 0.5510190217865811,
|
| 518 |
+
"eval_VitaminC_euclidean_f1": 0.6577540106951871,
|
| 519 |
+
"eval_VitaminC_euclidean_f1_threshold": 23.90462303161621,
|
| 520 |
+
"eval_VitaminC_euclidean_precision": 0.4900398406374502,
|
| 521 |
"eval_VitaminC_euclidean_recall": 1.0,
|
| 522 |
+
"eval_VitaminC_manhattan_accuracy": 0.578125,
|
| 523 |
+
"eval_VitaminC_manhattan_accuracy_threshold": 276.40142822265625,
|
| 524 |
+
"eval_VitaminC_manhattan_ap": 0.5429240708188645,
|
| 525 |
+
"eval_VitaminC_manhattan_f1": 0.6576819407008085,
|
| 526 |
+
"eval_VitaminC_manhattan_f1_threshold": 469.7353515625,
|
| 527 |
+
"eval_VitaminC_manhattan_precision": 0.49193548387096775,
|
| 528 |
+
"eval_VitaminC_manhattan_recall": 0.991869918699187,
|
| 529 |
+
"eval_VitaminC_max_accuracy": 0.58203125,
|
| 530 |
+
"eval_VitaminC_max_accuracy_threshold": 315.9444580078125,
|
| 531 |
+
"eval_VitaminC_max_ap": 0.5557444337961499,
|
| 532 |
+
"eval_VitaminC_max_f1": 0.6595174262734584,
|
| 533 |
+
"eval_VitaminC_max_f1_threshold": 469.7353515625,
|
| 534 |
+
"eval_VitaminC_max_precision": 0.492,
|
| 535 |
"eval_VitaminC_max_recall": 1.0,
|
| 536 |
+
"eval_sequential_score": 0.5557444337961499,
|
| 537 |
+
"eval_sts-test_pearson_cosine": 0.8483316632682467,
|
| 538 |
+
"eval_sts-test_pearson_dot": 0.8392403098680445,
|
| 539 |
+
"eval_sts-test_pearson_euclidean": 0.8814283057813619,
|
| 540 |
+
"eval_sts-test_pearson_manhattan": 0.8815226866327923,
|
| 541 |
+
"eval_sts-test_pearson_max": 0.8815226866327923,
|
| 542 |
+
"eval_sts-test_spearman_cosine": 0.8903503892346,
|
| 543 |
+
"eval_sts-test_spearman_dot": 0.857844431199042,
|
| 544 |
+
"eval_sts-test_spearman_euclidean": 0.8851830636663006,
|
| 545 |
+
"eval_sts-test_spearman_manhattan": 0.8865568876827619,
|
| 546 |
+
"eval_sts-test_spearman_max": 0.8903503892346,
|
| 547 |
+
"eval_vitaminc-pairs_loss": 2.3538782596588135,
|
| 548 |
+
"eval_vitaminc-pairs_runtime": 1.4618,
|
| 549 |
+
"eval_vitaminc-pairs_samples_per_second": 73.88,
|
| 550 |
+
"eval_vitaminc-pairs_steps_per_second": 1.368,
|
| 551 |
"step": 81
|
| 552 |
},
|
| 553 |
{
|
| 554 |
+
"epoch": 0.4550561797752809,
|
| 555 |
+
"eval_negation-triplets_loss": 1.649215579032898,
|
| 556 |
+
"eval_negation-triplets_runtime": 0.3081,
|
| 557 |
+
"eval_negation-triplets_samples_per_second": 207.723,
|
| 558 |
+
"eval_negation-triplets_steps_per_second": 3.246,
|
| 559 |
"step": 81
|
| 560 |
},
|
| 561 |
{
|
| 562 |
+
"epoch": 0.4550561797752809,
|
| 563 |
+
"eval_scitail-pairs-pos_loss": 0.11823470890522003,
|
| 564 |
+
"eval_scitail-pairs-pos_runtime": 0.376,
|
| 565 |
+
"eval_scitail-pairs-pos_samples_per_second": 143.616,
|
| 566 |
+
"eval_scitail-pairs-pos_steps_per_second": 2.66,
|
| 567 |
"step": 81
|
| 568 |
},
|
| 569 |
{
|
| 570 |
+
"epoch": 0.4550561797752809,
|
| 571 |
+
"eval_xsum-pairs_loss": 0.08420603722333908,
|
| 572 |
+
"eval_xsum-pairs_runtime": 3.1576,
|
| 573 |
+
"eval_xsum-pairs_samples_per_second": 40.538,
|
| 574 |
+
"eval_xsum-pairs_steps_per_second": 0.633,
|
| 575 |
"step": 81
|
| 576 |
},
|
| 577 |
{
|
| 578 |
+
"epoch": 0.4550561797752809,
|
| 579 |
+
"eval_sciq_pairs_loss": 0.034781794995069504,
|
| 580 |
+
"eval_sciq_pairs_runtime": 3.2597,
|
| 581 |
+
"eval_sciq_pairs_samples_per_second": 39.267,
|
| 582 |
+
"eval_sciq_pairs_steps_per_second": 0.614,
|
| 583 |
"step": 81
|
| 584 |
},
|
| 585 |
{
|
| 586 |
+
"epoch": 0.4550561797752809,
|
| 587 |
+
"eval_qasc_pairs_loss": 0.10597346723079681,
|
| 588 |
+
"eval_qasc_pairs_runtime": 0.6245,
|
| 589 |
+
"eval_qasc_pairs_samples_per_second": 204.979,
|
| 590 |
+
"eval_qasc_pairs_steps_per_second": 3.203,
|
| 591 |
"step": 81
|
| 592 |
},
|
| 593 |
{
|
| 594 |
+
"epoch": 0.4550561797752809,
|
| 595 |
+
"eval_openbookqa_pairs_loss": 0.7160983681678772,
|
| 596 |
+
"eval_openbookqa_pairs_runtime": 0.5767,
|
| 597 |
+
"eval_openbookqa_pairs_samples_per_second": 221.961,
|
| 598 |
+
"eval_openbookqa_pairs_steps_per_second": 3.468,
|
| 599 |
"step": 81
|
| 600 |
},
|
| 601 |
{
|
| 602 |
+
"epoch": 0.4550561797752809,
|
| 603 |
+
"eval_msmarco_pairs_loss": 0.3454173803329468,
|
| 604 |
+
"eval_msmarco_pairs_runtime": 1.2912,
|
| 605 |
+
"eval_msmarco_pairs_samples_per_second": 99.134,
|
| 606 |
+
"eval_msmarco_pairs_steps_per_second": 1.549,
|
| 607 |
"step": 81
|
| 608 |
},
|
| 609 |
{
|
| 610 |
+
"epoch": 0.4550561797752809,
|
| 611 |
+
"eval_nq_pairs_loss": 0.4442503750324249,
|
| 612 |
+
"eval_nq_pairs_runtime": 2.3854,
|
| 613 |
+
"eval_nq_pairs_samples_per_second": 53.659,
|
| 614 |
+
"eval_nq_pairs_steps_per_second": 0.838,
|
| 615 |
"step": 81
|
| 616 |
},
|
| 617 |
{
|
| 618 |
+
"epoch": 0.4550561797752809,
|
| 619 |
+
"eval_trivia_pairs_loss": 0.9324482679367065,
|
| 620 |
+
"eval_trivia_pairs_runtime": 4.4251,
|
| 621 |
+
"eval_trivia_pairs_samples_per_second": 28.926,
|
| 622 |
+
"eval_trivia_pairs_steps_per_second": 0.452,
|
| 623 |
"step": 81
|
| 624 |
},
|
| 625 |
{
|
| 626 |
+
"epoch": 0.4550561797752809,
|
| 627 |
+
"eval_gooaq_pairs_loss": 0.6094165444374084,
|
| 628 |
+
"eval_gooaq_pairs_runtime": 0.8751,
|
| 629 |
+
"eval_gooaq_pairs_samples_per_second": 146.261,
|
| 630 |
+
"eval_gooaq_pairs_steps_per_second": 2.285,
|
| 631 |
"step": 81
|
| 632 |
},
|
| 633 |
{
|
| 634 |
+
"epoch": 0.4550561797752809,
|
| 635 |
+
"eval_paws-pos_loss": 0.024421451613307,
|
| 636 |
+
"eval_paws-pos_runtime": 0.6865,
|
| 637 |
+
"eval_paws-pos_samples_per_second": 186.444,
|
| 638 |
+
"eval_paws-pos_steps_per_second": 2.913,
|
| 639 |
"step": 81
|
| 640 |
},
|
| 641 |
{
|
| 642 |
+
"epoch": 0.47191011235955055,
|
| 643 |
+
"grad_norm": 3.1395561695098877,
|
| 644 |
+
"learning_rate": 3.999675367909485e-05,
|
| 645 |
+
"loss": 0.5801,
|
| 646 |
"step": 84
|
| 647 |
},
|
| 648 |
{
|
| 649 |
+
"epoch": 0.4887640449438202,
|
| 650 |
+
"grad_norm": 2.7977917194366455,
|
| 651 |
+
"learning_rate": 3.998701612152597e-05,
|
| 652 |
+
"loss": 0.791,
|
| 653 |
"step": 87
|
| 654 |
},
|
| 655 |
{
|
| 656 |
+
"epoch": 0.5056179775280899,
|
| 657 |
+
"grad_norm": 2.3682048320770264,
|
| 658 |
+
"learning_rate": 3.997079154212493e-05,
|
| 659 |
+
"loss": 0.6042,
|
| 660 |
"step": 90
|
| 661 |
},
|
| 662 |
{
|
| 663 |
+
"epoch": 0.5224719101123596,
|
| 664 |
+
"grad_norm": 2.843482255935669,
|
| 665 |
+
"learning_rate": 3.99480869635839e-05,
|
| 666 |
+
"loss": 0.7559,
|
| 667 |
"step": 93
|
| 668 |
},
|
| 669 |
{
|
| 670 |
+
"epoch": 0.5393258426966292,
|
| 671 |
+
"grad_norm": 2.7346785068511963,
|
| 672 |
+
"learning_rate": 3.9918912213415936e-05,
|
| 673 |
+
"loss": 0.6258,
|
| 674 |
"step": 96
|
| 675 |
},
|
| 676 |
{
|
| 677 |
+
"epoch": 0.5561797752808989,
|
| 678 |
+
"grad_norm": 3.149007558822632,
|
| 679 |
+
"learning_rate": 3.9883279919701226e-05,
|
| 680 |
+
"loss": 0.8853,
|
| 681 |
"step": 99
|
| 682 |
},
|
| 683 |
{
|
| 684 |
+
"epoch": 0.5730337078651685,
|
| 685 |
+
"grad_norm": 3.3424761295318604,
|
| 686 |
+
"learning_rate": 3.9841205505621106e-05,
|
| 687 |
+
"loss": 0.5947,
|
| 688 |
"step": 102
|
| 689 |
},
|
| 690 |
{
|
| 691 |
+
"epoch": 0.5898876404494382,
|
| 692 |
+
"grad_norm": 2.6377146244049072,
|
| 693 |
+
"learning_rate": 3.979270718278224e-05,
|
| 694 |
+
"loss": 0.644,
|
| 695 |
"step": 105
|
| 696 |
}
|
| 697 |
],
|
| 698 |
"logging_steps": 3,
|
| 699 |
+
"max_steps": 534,
|
| 700 |
"num_input_tokens_seen": 0,
|
| 701 |
"num_train_epochs": 3,
|
| 702 |
"save_steps": 107,
|
|
|
|
| 713 |
}
|
| 714 |
},
|
| 715 |
"total_flos": 0.0,
|
| 716 |
+
"train_batch_size": 320,
|
| 717 |
"trial_name": null,
|
| 718 |
"trial_params": null
|
| 719 |
}
|
checkpoint-107/training_args.bin
CHANGED
|
@@ -1,3 +1,3 @@
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
-
oid sha256:
|
| 3 |
size 5688
|
|
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:72d6ebbf0ffc45e3199e7e67afe865d0f054853a38220ea09a039bd30fc6a761
|
| 3 |
size 5688
|