End of training
Browse files- README.md +1 -1
- all_results.json +6 -6
- runs/Jan12_17-57-19_GAN-SVR/events.out.tfevents.1736694740.GAN-SVR.3870842.4 +2 -2
- test_results.json +6 -6
- trainer_state.json +780 -309
README.md
CHANGED
|
@@ -18,7 +18,7 @@ should probably proofread and complete it, then remove this comment. -->
|
|
| 18 |
|
| 19 |
This model is a fine-tuned version of [MCG-NJU/videomae-base](https://huggingface.co/MCG-NJU/videomae-base) on an unknown dataset.
|
| 20 |
It achieves the following results on the evaluation set:
|
| 21 |
-
- Loss: 2.
|
| 22 |
- Accuracy: 0.4883
|
| 23 |
|
| 24 |
## Model description
|
|
|
|
| 18 |
|
| 19 |
This model is a fine-tuned version of [MCG-NJU/videomae-base](https://huggingface.co/MCG-NJU/videomae-base) on an unknown dataset.
|
| 20 |
It achieves the following results on the evaluation set:
|
| 21 |
+
- Loss: 2.1806
|
| 22 |
- Accuracy: 0.4883
|
| 23 |
|
| 24 |
## Model description
|
all_results.json
CHANGED
|
@@ -1,8 +1,8 @@
|
|
| 1 |
{
|
| 2 |
-
"epoch":
|
| 3 |
-
"eval_accuracy": 0.
|
| 4 |
-
"eval_loss":
|
| 5 |
-
"eval_runtime":
|
| 6 |
-
"eval_samples_per_second": 1.
|
| 7 |
-
"eval_steps_per_second": 0.
|
| 8 |
}
|
|
|
|
| 1 |
{
|
| 2 |
+
"epoch": 30.03125,
|
| 3 |
+
"eval_accuracy": 0.4882943143812709,
|
| 4 |
+
"eval_loss": 2.180596113204956,
|
| 5 |
+
"eval_runtime": 179.6038,
|
| 6 |
+
"eval_samples_per_second": 1.665,
|
| 7 |
+
"eval_steps_per_second": 0.056
|
| 8 |
}
|
runs/Jan12_17-57-19_GAN-SVR/events.out.tfevents.1736694740.GAN-SVR.3870842.4
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 sha256:8a4152e639af0b4173a211351cdff00a09ef6d111370e2b8acf891f431281470
|
| 3 |
+
size 734
|
test_results.json
CHANGED
|
@@ -1,8 +1,8 @@
|
|
| 1 |
{
|
| 2 |
-
"epoch":
|
| 3 |
-
"eval_accuracy": 0.
|
| 4 |
-
"eval_loss":
|
| 5 |
-
"eval_runtime":
|
| 6 |
-
"eval_samples_per_second": 1.
|
| 7 |
-
"eval_steps_per_second": 0.
|
| 8 |
}
|
|
|
|
| 1 |
{
|
| 2 |
+
"epoch": 30.03125,
|
| 3 |
+
"eval_accuracy": 0.4882943143812709,
|
| 4 |
+
"eval_loss": 2.180596113204956,
|
| 5 |
+
"eval_runtime": 179.6038,
|
| 6 |
+
"eval_samples_per_second": 1.665,
|
| 7 |
+
"eval_steps_per_second": 0.056
|
| 8 |
}
|
trainer_state.json
CHANGED
|
@@ -1,523 +1,994 @@
|
|
| 1 |
{
|
| 2 |
-
"best_metric": 0.
|
| 3 |
-
"best_model_checkpoint": "videomae-base-finetuned-ucf101-subset/checkpoint-
|
| 4 |
-
"epoch":
|
| 5 |
"eval_steps": 500,
|
| 6 |
-
"global_step":
|
| 7 |
"is_hyper_param_search": false,
|
| 8 |
"is_local_process_zero": true,
|
| 9 |
"is_world_process_zero": true,
|
| 10 |
"log_history": [
|
| 11 |
{
|
| 12 |
-
"epoch": 0.
|
| 13 |
-
"grad_norm": 4.
|
| 14 |
-
"learning_rate":
|
| 15 |
-
"loss": 4.
|
| 16 |
"step": 10
|
| 17 |
},
|
| 18 |
{
|
| 19 |
-
"epoch": 0.
|
| 20 |
-
"grad_norm":
|
| 21 |
-
"learning_rate":
|
| 22 |
-
"loss": 4.
|
| 23 |
"step": 20
|
| 24 |
},
|
| 25 |
{
|
| 26 |
-
"epoch": 0.
|
| 27 |
-
"grad_norm":
|
| 28 |
-
"learning_rate":
|
| 29 |
-
"loss": 4.
|
| 30 |
"step": 30
|
| 31 |
},
|
| 32 |
{
|
| 33 |
-
"epoch": 0.
|
| 34 |
"eval_accuracy": 0.0033444816053511705,
|
| 35 |
-
"eval_loss": 4.
|
| 36 |
-
"eval_runtime":
|
| 37 |
-
"eval_samples_per_second": 1.
|
| 38 |
-
"eval_steps_per_second": 0.
|
| 39 |
"step": 31
|
| 40 |
},
|
| 41 |
{
|
| 42 |
-
"epoch": 1.
|
| 43 |
-
"grad_norm":
|
| 44 |
-
"learning_rate":
|
| 45 |
-
"loss": 4.
|
| 46 |
"step": 40
|
| 47 |
},
|
| 48 |
{
|
| 49 |
-
"epoch": 1.
|
| 50 |
-
"grad_norm":
|
| 51 |
-
"learning_rate":
|
| 52 |
-
"loss": 4.
|
| 53 |
"step": 50
|
| 54 |
},
|
| 55 |
{
|
| 56 |
-
"epoch": 1.
|
| 57 |
-
"grad_norm": 3.
|
| 58 |
-
"learning_rate":
|
| 59 |
-
"loss": 4.
|
| 60 |
"step": 60
|
| 61 |
},
|
| 62 |
{
|
| 63 |
-
"epoch": 1.
|
| 64 |
-
"eval_accuracy": 0.
|
| 65 |
-
"eval_loss": 4.
|
| 66 |
-
"eval_runtime":
|
| 67 |
-
"eval_samples_per_second": 1.
|
| 68 |
-
"eval_steps_per_second": 0.
|
| 69 |
"step": 62
|
| 70 |
},
|
| 71 |
{
|
| 72 |
-
"epoch": 2.
|
| 73 |
-
"grad_norm": 3.
|
| 74 |
-
"learning_rate":
|
| 75 |
-
"loss": 4.
|
| 76 |
"step": 70
|
| 77 |
},
|
| 78 |
{
|
| 79 |
-
"epoch": 2.
|
| 80 |
-
"grad_norm": 3.
|
| 81 |
-
"learning_rate": 4.
|
| 82 |
-
"loss": 4.
|
| 83 |
"step": 80
|
| 84 |
},
|
| 85 |
{
|
| 86 |
-
"epoch": 2.
|
| 87 |
-
"grad_norm": 3.
|
| 88 |
-
"learning_rate": 4.
|
| 89 |
-
"loss": 4.
|
| 90 |
"step": 90
|
| 91 |
},
|
| 92 |
{
|
| 93 |
-
"epoch": 2.
|
| 94 |
-
"eval_accuracy": 0.
|
| 95 |
-
"eval_loss": 4.
|
| 96 |
-
"eval_runtime":
|
| 97 |
-
"eval_samples_per_second": 1.
|
| 98 |
-
"eval_steps_per_second": 0.
|
| 99 |
"step": 93
|
| 100 |
},
|
| 101 |
{
|
| 102 |
-
"epoch": 3.
|
| 103 |
-
"grad_norm":
|
| 104 |
-
"learning_rate": 4.
|
| 105 |
-
"loss": 4.
|
| 106 |
"step": 100
|
| 107 |
},
|
| 108 |
{
|
| 109 |
-
"epoch": 3.
|
| 110 |
-
"grad_norm":
|
| 111 |
-
"learning_rate": 4.
|
| 112 |
-
"loss": 4.
|
| 113 |
"step": 110
|
| 114 |
},
|
| 115 |
{
|
| 116 |
-
"epoch": 3.
|
| 117 |
-
"grad_norm": 3.
|
| 118 |
-
"learning_rate": 4.
|
| 119 |
"loss": 4.2445,
|
| 120 |
"step": 120
|
| 121 |
},
|
| 122 |
{
|
| 123 |
-
"epoch": 3.
|
| 124 |
"eval_accuracy": 0.006688963210702341,
|
| 125 |
-
"eval_loss": 4.
|
| 126 |
-
"eval_runtime":
|
| 127 |
-
"eval_samples_per_second": 1.
|
| 128 |
-
"eval_steps_per_second": 0.
|
| 129 |
"step": 124
|
| 130 |
},
|
| 131 |
{
|
| 132 |
-
"epoch": 4.
|
| 133 |
-
"grad_norm":
|
| 134 |
-
"learning_rate": 4.
|
| 135 |
-
"loss": 4.
|
| 136 |
"step": 130
|
| 137 |
},
|
| 138 |
{
|
| 139 |
-
"epoch": 4.
|
| 140 |
-
"grad_norm":
|
| 141 |
-
"learning_rate":
|
| 142 |
-
"loss": 4.
|
| 143 |
"step": 140
|
| 144 |
},
|
| 145 |
{
|
| 146 |
-
"epoch": 4.
|
| 147 |
-
"grad_norm": 2.
|
| 148 |
-
"learning_rate":
|
| 149 |
-
"loss": 4.
|
| 150 |
"step": 150
|
| 151 |
},
|
| 152 |
{
|
| 153 |
-
"epoch": 4.
|
| 154 |
"eval_accuracy": 0.010033444816053512,
|
| 155 |
-
"eval_loss": 4.
|
| 156 |
-
"eval_runtime":
|
| 157 |
-
"eval_samples_per_second": 1.
|
| 158 |
-
"eval_steps_per_second": 0.
|
| 159 |
"step": 155
|
| 160 |
},
|
| 161 |
{
|
| 162 |
-
"epoch": 5.
|
| 163 |
-
"grad_norm":
|
| 164 |
-
"learning_rate":
|
| 165 |
-
"loss": 4.
|
| 166 |
"step": 160
|
| 167 |
},
|
| 168 |
{
|
| 169 |
-
"epoch": 5.
|
| 170 |
-
"grad_norm":
|
| 171 |
-
"learning_rate":
|
| 172 |
-
"loss": 4.
|
| 173 |
"step": 170
|
| 174 |
},
|
| 175 |
{
|
| 176 |
-
"epoch": 5.
|
| 177 |
-
"grad_norm": 3.
|
| 178 |
-
"learning_rate":
|
| 179 |
-
"loss": 4.
|
| 180 |
"step": 180
|
| 181 |
},
|
| 182 |
{
|
| 183 |
-
"epoch": 5.
|
| 184 |
-
"eval_accuracy": 0.
|
| 185 |
-
"eval_loss": 4.
|
| 186 |
-
"eval_runtime":
|
| 187 |
-
"eval_samples_per_second": 1.
|
| 188 |
-
"eval_steps_per_second": 0.
|
| 189 |
"step": 186
|
| 190 |
},
|
| 191 |
{
|
| 192 |
-
"epoch": 6.
|
| 193 |
-
"grad_norm":
|
| 194 |
-
"learning_rate":
|
| 195 |
-
"loss": 4.
|
| 196 |
"step": 190
|
| 197 |
},
|
| 198 |
{
|
| 199 |
-
"epoch": 6.
|
| 200 |
-
"grad_norm": 3.
|
| 201 |
-
"learning_rate":
|
| 202 |
-
"loss": 4.
|
| 203 |
"step": 200
|
| 204 |
},
|
| 205 |
{
|
| 206 |
-
"epoch": 6.
|
| 207 |
-
"grad_norm": 3.
|
| 208 |
-
"learning_rate":
|
| 209 |
-
"loss": 4.
|
| 210 |
"step": 210
|
| 211 |
},
|
| 212 |
{
|
| 213 |
-
"epoch": 6.
|
| 214 |
-
"eval_accuracy": 0.
|
| 215 |
-
"eval_loss": 4.
|
| 216 |
-
"eval_runtime":
|
| 217 |
-
"eval_samples_per_second": 1.
|
| 218 |
-
"eval_steps_per_second": 0.
|
| 219 |
"step": 217
|
| 220 |
},
|
| 221 |
{
|
| 222 |
-
"epoch": 7.
|
| 223 |
-
"grad_norm":
|
| 224 |
-
"learning_rate":
|
| 225 |
-
"loss": 4.
|
| 226 |
"step": 220
|
| 227 |
},
|
| 228 |
{
|
| 229 |
-
"epoch": 7.
|
| 230 |
-
"grad_norm":
|
| 231 |
-
"learning_rate":
|
| 232 |
-
"loss": 4.
|
| 233 |
"step": 230
|
| 234 |
},
|
| 235 |
{
|
| 236 |
-
"epoch": 7.
|
| 237 |
-
"grad_norm":
|
| 238 |
-
"learning_rate":
|
| 239 |
-
"loss":
|
| 240 |
"step": 240
|
| 241 |
},
|
| 242 |
{
|
| 243 |
-
"epoch": 7.
|
| 244 |
-
"eval_accuracy": 0.
|
| 245 |
-
"eval_loss":
|
| 246 |
-
"eval_runtime":
|
| 247 |
-
"eval_samples_per_second": 1.
|
| 248 |
-
"eval_steps_per_second": 0.
|
| 249 |
"step": 248
|
| 250 |
},
|
| 251 |
{
|
| 252 |
-
"epoch": 8.
|
| 253 |
-
"grad_norm":
|
| 254 |
-
"learning_rate":
|
| 255 |
-
"loss":
|
| 256 |
"step": 250
|
| 257 |
},
|
| 258 |
{
|
| 259 |
-
"epoch": 8.
|
| 260 |
-
"grad_norm": 4.
|
| 261 |
-
"learning_rate":
|
| 262 |
-
"loss":
|
| 263 |
"step": 260
|
| 264 |
},
|
| 265 |
{
|
| 266 |
-
"epoch": 8.
|
| 267 |
-
"grad_norm":
|
| 268 |
-
"learning_rate":
|
| 269 |
-
"loss":
|
| 270 |
"step": 270
|
| 271 |
},
|
| 272 |
{
|
| 273 |
-
"epoch": 8.
|
| 274 |
-
"eval_accuracy": 0.
|
| 275 |
-
"eval_loss": 3.
|
| 276 |
-
"eval_runtime":
|
| 277 |
-
"eval_samples_per_second": 1.
|
| 278 |
-
"eval_steps_per_second": 0.
|
| 279 |
"step": 279
|
| 280 |
},
|
| 281 |
{
|
| 282 |
-
"epoch": 9.
|
| 283 |
-
"grad_norm":
|
| 284 |
-
"learning_rate":
|
| 285 |
-
"loss": 3.
|
| 286 |
"step": 280
|
| 287 |
},
|
| 288 |
{
|
| 289 |
-
"epoch": 9.
|
| 290 |
-
"grad_norm": 5.
|
| 291 |
-
"learning_rate":
|
| 292 |
-
"loss": 3.
|
| 293 |
"step": 290
|
| 294 |
},
|
| 295 |
{
|
| 296 |
-
"epoch": 9.
|
| 297 |
-
"grad_norm":
|
| 298 |
-
"learning_rate":
|
| 299 |
-
"loss": 3.
|
| 300 |
"step": 300
|
| 301 |
},
|
| 302 |
{
|
| 303 |
-
"epoch": 9.
|
| 304 |
-
"grad_norm":
|
| 305 |
-
"learning_rate":
|
| 306 |
-
"loss": 3.
|
| 307 |
"step": 310
|
| 308 |
},
|
| 309 |
{
|
| 310 |
-
"epoch": 9.
|
| 311 |
"eval_accuracy": 0.10702341137123746,
|
| 312 |
-
"eval_loss": 3.
|
| 313 |
-
"eval_runtime":
|
| 314 |
-
"eval_samples_per_second": 1.
|
| 315 |
-
"eval_steps_per_second": 0.
|
| 316 |
"step": 310
|
| 317 |
},
|
| 318 |
{
|
| 319 |
-
"epoch": 10.
|
| 320 |
-
"grad_norm":
|
| 321 |
-
"learning_rate":
|
| 322 |
-
"loss": 3.
|
| 323 |
"step": 320
|
| 324 |
},
|
| 325 |
{
|
| 326 |
-
"epoch": 10.
|
| 327 |
-
"grad_norm":
|
| 328 |
-
"learning_rate":
|
| 329 |
-
"loss": 3.
|
| 330 |
"step": 330
|
| 331 |
},
|
| 332 |
{
|
| 333 |
-
"epoch": 10.
|
| 334 |
-
"grad_norm": 6.
|
| 335 |
-
"learning_rate":
|
| 336 |
-
"loss": 3.
|
| 337 |
"step": 340
|
| 338 |
},
|
| 339 |
{
|
| 340 |
-
"epoch": 10.
|
| 341 |
-
"eval_accuracy": 0.
|
| 342 |
-
"eval_loss": 3.
|
| 343 |
-
"eval_runtime":
|
| 344 |
-
"eval_samples_per_second": 1.
|
| 345 |
-
"eval_steps_per_second": 0.
|
| 346 |
"step": 341
|
| 347 |
},
|
| 348 |
{
|
| 349 |
-
"epoch": 11.
|
| 350 |
-
"grad_norm": 6.
|
| 351 |
-
"learning_rate":
|
| 352 |
-
"loss":
|
| 353 |
"step": 350
|
| 354 |
},
|
| 355 |
{
|
| 356 |
-
"epoch": 11.
|
| 357 |
-
"grad_norm":
|
| 358 |
-
"learning_rate":
|
| 359 |
-
"loss":
|
| 360 |
"step": 360
|
| 361 |
},
|
| 362 |
{
|
| 363 |
-
"epoch": 11.
|
| 364 |
-
"grad_norm":
|
| 365 |
-
"learning_rate":
|
| 366 |
-
"loss":
|
| 367 |
"step": 370
|
| 368 |
},
|
| 369 |
{
|
| 370 |
-
"epoch": 11.
|
| 371 |
-
"eval_accuracy": 0.
|
| 372 |
-
"eval_loss": 3.
|
| 373 |
-
"eval_runtime":
|
| 374 |
-
"eval_samples_per_second": 1.
|
| 375 |
-
"eval_steps_per_second": 0.
|
| 376 |
"step": 372
|
| 377 |
},
|
| 378 |
{
|
| 379 |
-
"epoch": 12.
|
| 380 |
-
"grad_norm":
|
| 381 |
-
"learning_rate":
|
| 382 |
-
"loss":
|
| 383 |
"step": 380
|
| 384 |
},
|
| 385 |
{
|
| 386 |
-
"epoch": 12.
|
| 387 |
-
"grad_norm": 6.
|
| 388 |
-
"learning_rate":
|
| 389 |
-
"loss":
|
| 390 |
"step": 390
|
| 391 |
},
|
| 392 |
{
|
| 393 |
-
"epoch": 12.
|
| 394 |
-
"grad_norm": 7.
|
| 395 |
-
"learning_rate":
|
| 396 |
-
"loss": 2.
|
| 397 |
"step": 400
|
| 398 |
},
|
| 399 |
{
|
| 400 |
-
"epoch": 12.
|
| 401 |
-
"eval_accuracy": 0.
|
| 402 |
-
"eval_loss": 3.
|
| 403 |
-
"eval_runtime":
|
| 404 |
-
"eval_samples_per_second": 1.
|
| 405 |
-
"eval_steps_per_second": 0.
|
| 406 |
"step": 403
|
| 407 |
},
|
| 408 |
{
|
| 409 |
-
"epoch": 13.
|
| 410 |
-
"grad_norm": 6.
|
| 411 |
-
"learning_rate":
|
| 412 |
-
"loss": 2.
|
| 413 |
"step": 410
|
| 414 |
},
|
| 415 |
{
|
| 416 |
-
"epoch": 13.
|
| 417 |
-
"grad_norm":
|
| 418 |
-
"learning_rate":
|
| 419 |
-
"loss": 2.
|
| 420 |
"step": 420
|
| 421 |
},
|
| 422 |
{
|
| 423 |
-
"epoch": 13.
|
| 424 |
-
"grad_norm":
|
| 425 |
-
"learning_rate":
|
| 426 |
-
"loss": 2.
|
| 427 |
"step": 430
|
| 428 |
},
|
| 429 |
{
|
| 430 |
-
"epoch": 13.
|
| 431 |
-
"eval_accuracy": 0.
|
| 432 |
-
"eval_loss": 3.
|
| 433 |
-
"eval_runtime":
|
| 434 |
-
"eval_samples_per_second": 1.
|
| 435 |
-
"eval_steps_per_second": 0.
|
| 436 |
"step": 434
|
| 437 |
},
|
| 438 |
{
|
| 439 |
-
"epoch": 14.
|
| 440 |
-
"grad_norm":
|
| 441 |
-
"learning_rate":
|
| 442 |
-
"loss": 2.
|
| 443 |
"step": 440
|
| 444 |
},
|
| 445 |
{
|
| 446 |
-
"epoch": 14.
|
| 447 |
-
"grad_norm":
|
| 448 |
-
"learning_rate":
|
| 449 |
-
"loss": 2.
|
| 450 |
"step": 450
|
| 451 |
},
|
| 452 |
{
|
| 453 |
-
"epoch": 14.
|
| 454 |
-
"grad_norm": 6.
|
| 455 |
-
"learning_rate": 2.
|
| 456 |
-
"loss": 2.
|
| 457 |
"step": 460
|
| 458 |
},
|
| 459 |
{
|
| 460 |
-
"epoch": 14.
|
| 461 |
-
"eval_accuracy": 0.
|
| 462 |
-
"eval_loss": 3.
|
| 463 |
-
"eval_runtime": 155.
|
| 464 |
-
"eval_samples_per_second": 1.
|
| 465 |
-
"eval_steps_per_second": 0.
|
| 466 |
"step": 465
|
| 467 |
},
|
| 468 |
{
|
| 469 |
-
"epoch": 15.
|
| 470 |
-
"grad_norm":
|
| 471 |
-
"learning_rate":
|
| 472 |
-
"loss": 2.
|
| 473 |
"step": 470
|
| 474 |
},
|
| 475 |
{
|
| 476 |
-
"epoch": 15.
|
| 477 |
-
"grad_norm":
|
| 478 |
-
"learning_rate":
|
| 479 |
-
"loss": 2.
|
| 480 |
"step": 480
|
| 481 |
},
|
| 482 |
{
|
| 483 |
-
"epoch": 15.
|
| 484 |
-
"
|
| 485 |
-
"
|
| 486 |
-
"
|
| 487 |
-
"
|
| 488 |
-
"eval_steps_per_second": 0.053,
|
| 489 |
-
"step": 480
|
| 490 |
},
|
| 491 |
{
|
| 492 |
-
"epoch": 15.
|
| 493 |
-
"
|
| 494 |
-
"
|
| 495 |
-
"
|
| 496 |
-
"
|
| 497 |
-
"
|
| 498 |
-
"
|
| 499 |
},
|
| 500 |
{
|
| 501 |
-
"epoch":
|
| 502 |
-
"
|
| 503 |
-
"
|
| 504 |
-
"
|
| 505 |
-
"
|
| 506 |
-
"eval_steps_per_second": 0.063,
|
| 507 |
-
"step": 480
|
| 508 |
},
|
| 509 |
{
|
| 510 |
-
"epoch":
|
| 511 |
-
"
|
| 512 |
-
"
|
| 513 |
-
"
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 514 |
"eval_samples_per_second": 1.894,
|
| 515 |
"eval_steps_per_second": 0.063,
|
| 516 |
-
"step":
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 517 |
}
|
| 518 |
],
|
| 519 |
"logging_steps": 10,
|
| 520 |
-
"max_steps":
|
| 521 |
"num_input_tokens_seen": 0,
|
| 522 |
"num_train_epochs": 9223372036854775807,
|
| 523 |
"save_steps": 500,
|
|
@@ -533,7 +1004,7 @@
|
|
| 533 |
"attributes": {}
|
| 534 |
}
|
| 535 |
},
|
| 536 |
-
"total_flos":
|
| 537 |
"train_batch_size": 32,
|
| 538 |
"trial_name": null,
|
| 539 |
"trial_params": null
|
|
|
|
| 1 |
{
|
| 2 |
+
"best_metric": 0.4882943143812709,
|
| 3 |
+
"best_model_checkpoint": "videomae-base-finetuned-ucf101-subset/checkpoint-806",
|
| 4 |
+
"epoch": 30.03125,
|
| 5 |
"eval_steps": 500,
|
| 6 |
+
"global_step": 960,
|
| 7 |
"is_hyper_param_search": false,
|
| 8 |
"is_local_process_zero": true,
|
| 9 |
"is_world_process_zero": true,
|
| 10 |
"log_history": [
|
| 11 |
{
|
| 12 |
+
"epoch": 0.010416666666666666,
|
| 13 |
+
"grad_norm": 4.111547470092773,
|
| 14 |
+
"learning_rate": 5.208333333333334e-06,
|
| 15 |
+
"loss": 4.2127,
|
| 16 |
"step": 10
|
| 17 |
},
|
| 18 |
{
|
| 19 |
+
"epoch": 0.020833333333333332,
|
| 20 |
+
"grad_norm": 4.218038082122803,
|
| 21 |
+
"learning_rate": 1.0416666666666668e-05,
|
| 22 |
+
"loss": 4.2247,
|
| 23 |
"step": 20
|
| 24 |
},
|
| 25 |
{
|
| 26 |
+
"epoch": 0.03125,
|
| 27 |
+
"grad_norm": 3.766047954559326,
|
| 28 |
+
"learning_rate": 1.5625e-05,
|
| 29 |
+
"loss": 4.2427,
|
| 30 |
"step": 30
|
| 31 |
},
|
| 32 |
{
|
| 33 |
+
"epoch": 0.03229166666666667,
|
| 34 |
"eval_accuracy": 0.0033444816053511705,
|
| 35 |
+
"eval_loss": 4.22645378112793,
|
| 36 |
+
"eval_runtime": 157.4916,
|
| 37 |
+
"eval_samples_per_second": 1.899,
|
| 38 |
+
"eval_steps_per_second": 0.063,
|
| 39 |
"step": 31
|
| 40 |
},
|
| 41 |
{
|
| 42 |
+
"epoch": 1.009375,
|
| 43 |
+
"grad_norm": 4.1204304695129395,
|
| 44 |
+
"learning_rate": 2.0833333333333336e-05,
|
| 45 |
+
"loss": 4.2374,
|
| 46 |
"step": 40
|
| 47 |
},
|
| 48 |
{
|
| 49 |
+
"epoch": 1.0197916666666667,
|
| 50 |
+
"grad_norm": 4.202085971832275,
|
| 51 |
+
"learning_rate": 2.604166666666667e-05,
|
| 52 |
+
"loss": 4.228,
|
| 53 |
"step": 50
|
| 54 |
},
|
| 55 |
{
|
| 56 |
+
"epoch": 1.0302083333333334,
|
| 57 |
+
"grad_norm": 3.5194547176361084,
|
| 58 |
+
"learning_rate": 3.125e-05,
|
| 59 |
+
"loss": 4.2321,
|
| 60 |
"step": 60
|
| 61 |
},
|
| 62 |
{
|
| 63 |
+
"epoch": 1.0322916666666666,
|
| 64 |
+
"eval_accuracy": 0.010033444816053512,
|
| 65 |
+
"eval_loss": 4.223534107208252,
|
| 66 |
+
"eval_runtime": 156.7976,
|
| 67 |
+
"eval_samples_per_second": 1.907,
|
| 68 |
+
"eval_steps_per_second": 0.064,
|
| 69 |
"step": 62
|
| 70 |
},
|
| 71 |
{
|
| 72 |
+
"epoch": 2.0083333333333333,
|
| 73 |
+
"grad_norm": 3.131030559539795,
|
| 74 |
+
"learning_rate": 3.6458333333333336e-05,
|
| 75 |
+
"loss": 4.2293,
|
| 76 |
"step": 70
|
| 77 |
},
|
| 78 |
{
|
| 79 |
+
"epoch": 2.01875,
|
| 80 |
+
"grad_norm": 3.120128631591797,
|
| 81 |
+
"learning_rate": 4.166666666666667e-05,
|
| 82 |
+
"loss": 4.2347,
|
| 83 |
"step": 80
|
| 84 |
},
|
| 85 |
{
|
| 86 |
+
"epoch": 2.029166666666667,
|
| 87 |
+
"grad_norm": 3.0988192558288574,
|
| 88 |
+
"learning_rate": 4.6875e-05,
|
| 89 |
+
"loss": 4.24,
|
| 90 |
"step": 90
|
| 91 |
},
|
| 92 |
{
|
| 93 |
+
"epoch": 2.0322916666666666,
|
| 94 |
+
"eval_accuracy": 0.010033444816053512,
|
| 95 |
+
"eval_loss": 4.228224754333496,
|
| 96 |
+
"eval_runtime": 156.5504,
|
| 97 |
+
"eval_samples_per_second": 1.91,
|
| 98 |
+
"eval_steps_per_second": 0.064,
|
| 99 |
"step": 93
|
| 100 |
},
|
| 101 |
{
|
| 102 |
+
"epoch": 3.0072916666666667,
|
| 103 |
+
"grad_norm": 3.102827310562134,
|
| 104 |
+
"learning_rate": 4.976851851851852e-05,
|
| 105 |
+
"loss": 4.2195,
|
| 106 |
"step": 100
|
| 107 |
},
|
| 108 |
{
|
| 109 |
+
"epoch": 3.017708333333333,
|
| 110 |
+
"grad_norm": 2.8366568088531494,
|
| 111 |
+
"learning_rate": 4.9189814814814815e-05,
|
| 112 |
+
"loss": 4.2354,
|
| 113 |
"step": 110
|
| 114 |
},
|
| 115 |
{
|
| 116 |
+
"epoch": 3.028125,
|
| 117 |
+
"grad_norm": 3.0085482597351074,
|
| 118 |
+
"learning_rate": 4.8611111111111115e-05,
|
| 119 |
"loss": 4.2445,
|
| 120 |
"step": 120
|
| 121 |
},
|
| 122 |
{
|
| 123 |
+
"epoch": 3.0322916666666666,
|
| 124 |
"eval_accuracy": 0.006688963210702341,
|
| 125 |
+
"eval_loss": 4.224982261657715,
|
| 126 |
+
"eval_runtime": 151.1437,
|
| 127 |
+
"eval_samples_per_second": 1.978,
|
| 128 |
+
"eval_steps_per_second": 0.066,
|
| 129 |
"step": 124
|
| 130 |
},
|
| 131 |
{
|
| 132 |
+
"epoch": 4.00625,
|
| 133 |
+
"grad_norm": 3.300618886947632,
|
| 134 |
+
"learning_rate": 4.803240740740741e-05,
|
| 135 |
+
"loss": 4.2226,
|
| 136 |
"step": 130
|
| 137 |
},
|
| 138 |
{
|
| 139 |
+
"epoch": 4.016666666666667,
|
| 140 |
+
"grad_norm": 3.202220916748047,
|
| 141 |
+
"learning_rate": 4.745370370370371e-05,
|
| 142 |
+
"loss": 4.2265,
|
| 143 |
"step": 140
|
| 144 |
},
|
| 145 |
{
|
| 146 |
+
"epoch": 4.027083333333334,
|
| 147 |
+
"grad_norm": 2.977271556854248,
|
| 148 |
+
"learning_rate": 4.6875e-05,
|
| 149 |
+
"loss": 4.2327,
|
| 150 |
"step": 150
|
| 151 |
},
|
| 152 |
{
|
| 153 |
+
"epoch": 4.032291666666667,
|
| 154 |
"eval_accuracy": 0.010033444816053512,
|
| 155 |
+
"eval_loss": 4.2244367599487305,
|
| 156 |
+
"eval_runtime": 156.5602,
|
| 157 |
+
"eval_samples_per_second": 1.91,
|
| 158 |
+
"eval_steps_per_second": 0.064,
|
| 159 |
"step": 155
|
| 160 |
},
|
| 161 |
{
|
| 162 |
+
"epoch": 5.005208333333333,
|
| 163 |
+
"grad_norm": 3.7291171550750732,
|
| 164 |
+
"learning_rate": 4.62962962962963e-05,
|
| 165 |
+
"loss": 4.2106,
|
| 166 |
"step": 160
|
| 167 |
},
|
| 168 |
{
|
| 169 |
+
"epoch": 5.015625,
|
| 170 |
+
"grad_norm": 3.581210136413574,
|
| 171 |
+
"learning_rate": 4.5717592592592594e-05,
|
| 172 |
+
"loss": 4.2112,
|
| 173 |
"step": 170
|
| 174 |
},
|
| 175 |
{
|
| 176 |
+
"epoch": 5.026041666666667,
|
| 177 |
+
"grad_norm": 3.7153217792510986,
|
| 178 |
+
"learning_rate": 4.5138888888888894e-05,
|
| 179 |
+
"loss": 4.2104,
|
| 180 |
"step": 180
|
| 181 |
},
|
| 182 |
{
|
| 183 |
+
"epoch": 5.032291666666667,
|
| 184 |
+
"eval_accuracy": 0.020066889632107024,
|
| 185 |
+
"eval_loss": 4.2100114822387695,
|
| 186 |
+
"eval_runtime": 150.442,
|
| 187 |
+
"eval_samples_per_second": 1.987,
|
| 188 |
+
"eval_steps_per_second": 0.066,
|
| 189 |
"step": 186
|
| 190 |
},
|
| 191 |
{
|
| 192 |
+
"epoch": 6.004166666666666,
|
| 193 |
+
"grad_norm": 3.1238365173339844,
|
| 194 |
+
"learning_rate": 4.456018518518519e-05,
|
| 195 |
+
"loss": 4.2205,
|
| 196 |
"step": 190
|
| 197 |
},
|
| 198 |
{
|
| 199 |
+
"epoch": 6.014583333333333,
|
| 200 |
+
"grad_norm": 3.4233925342559814,
|
| 201 |
+
"learning_rate": 4.3981481481481486e-05,
|
| 202 |
+
"loss": 4.2012,
|
| 203 |
"step": 200
|
| 204 |
},
|
| 205 |
{
|
| 206 |
+
"epoch": 6.025,
|
| 207 |
+
"grad_norm": 3.511300563812256,
|
| 208 |
+
"learning_rate": 4.340277777777778e-05,
|
| 209 |
+
"loss": 4.2374,
|
| 210 |
"step": 210
|
| 211 |
},
|
| 212 |
{
|
| 213 |
+
"epoch": 6.032291666666667,
|
| 214 |
+
"eval_accuracy": 0.006688963210702341,
|
| 215 |
+
"eval_loss": 4.2022294998168945,
|
| 216 |
+
"eval_runtime": 153.3689,
|
| 217 |
+
"eval_samples_per_second": 1.95,
|
| 218 |
+
"eval_steps_per_second": 0.065,
|
| 219 |
"step": 217
|
| 220 |
},
|
| 221 |
{
|
| 222 |
+
"epoch": 7.003125,
|
| 223 |
+
"grad_norm": 3.343992233276367,
|
| 224 |
+
"learning_rate": 4.282407407407408e-05,
|
| 225 |
+
"loss": 4.2055,
|
| 226 |
"step": 220
|
| 227 |
},
|
| 228 |
{
|
| 229 |
+
"epoch": 7.013541666666667,
|
| 230 |
+
"grad_norm": 2.911720037460327,
|
| 231 |
+
"learning_rate": 4.224537037037037e-05,
|
| 232 |
+
"loss": 4.1687,
|
| 233 |
"step": 230
|
| 234 |
},
|
| 235 |
{
|
| 236 |
+
"epoch": 7.023958333333334,
|
| 237 |
+
"grad_norm": 3.1043291091918945,
|
| 238 |
+
"learning_rate": 4.166666666666667e-05,
|
| 239 |
+
"loss": 4.1597,
|
| 240 |
"step": 240
|
| 241 |
},
|
| 242 |
{
|
| 243 |
+
"epoch": 7.032291666666667,
|
| 244 |
+
"eval_accuracy": 0.030100334448160536,
|
| 245 |
+
"eval_loss": 4.118756294250488,
|
| 246 |
+
"eval_runtime": 152.3194,
|
| 247 |
+
"eval_samples_per_second": 1.963,
|
| 248 |
+
"eval_steps_per_second": 0.066,
|
| 249 |
"step": 248
|
| 250 |
},
|
| 251 |
{
|
| 252 |
+
"epoch": 8.002083333333333,
|
| 253 |
+
"grad_norm": 3.4208502769470215,
|
| 254 |
+
"learning_rate": 4.1087962962962965e-05,
|
| 255 |
+
"loss": 4.1418,
|
| 256 |
"step": 250
|
| 257 |
},
|
| 258 |
{
|
| 259 |
+
"epoch": 8.0125,
|
| 260 |
+
"grad_norm": 4.338245868682861,
|
| 261 |
+
"learning_rate": 4.0509259259259265e-05,
|
| 262 |
+
"loss": 4.0753,
|
| 263 |
"step": 260
|
| 264 |
},
|
| 265 |
{
|
| 266 |
+
"epoch": 8.022916666666667,
|
| 267 |
+
"grad_norm": 4.86409854888916,
|
| 268 |
+
"learning_rate": 3.993055555555556e-05,
|
| 269 |
+
"loss": 4.0522,
|
| 270 |
"step": 270
|
| 271 |
},
|
| 272 |
{
|
| 273 |
+
"epoch": 8.032291666666667,
|
| 274 |
+
"eval_accuracy": 0.07023411371237458,
|
| 275 |
+
"eval_loss": 3.935123920440674,
|
| 276 |
+
"eval_runtime": 155.145,
|
| 277 |
+
"eval_samples_per_second": 1.927,
|
| 278 |
+
"eval_steps_per_second": 0.064,
|
| 279 |
"step": 279
|
| 280 |
},
|
| 281 |
{
|
| 282 |
+
"epoch": 9.001041666666667,
|
| 283 |
+
"grad_norm": 4.609180927276611,
|
| 284 |
+
"learning_rate": 3.935185185185186e-05,
|
| 285 |
+
"loss": 3.9967,
|
| 286 |
"step": 280
|
| 287 |
},
|
| 288 |
{
|
| 289 |
+
"epoch": 9.011458333333334,
|
| 290 |
+
"grad_norm": 5.407143592834473,
|
| 291 |
+
"learning_rate": 3.877314814814815e-05,
|
| 292 |
+
"loss": 3.8626,
|
| 293 |
"step": 290
|
| 294 |
},
|
| 295 |
{
|
| 296 |
+
"epoch": 9.021875,
|
| 297 |
+
"grad_norm": 4.2119598388671875,
|
| 298 |
+
"learning_rate": 3.8194444444444444e-05,
|
| 299 |
+
"loss": 3.821,
|
| 300 |
"step": 300
|
| 301 |
},
|
| 302 |
{
|
| 303 |
+
"epoch": 9.032291666666667,
|
| 304 |
+
"grad_norm": 14.079527854919434,
|
| 305 |
+
"learning_rate": 3.7615740740740744e-05,
|
| 306 |
+
"loss": 3.768,
|
| 307 |
"step": 310
|
| 308 |
},
|
| 309 |
{
|
| 310 |
+
"epoch": 9.032291666666667,
|
| 311 |
"eval_accuracy": 0.10702341137123746,
|
| 312 |
+
"eval_loss": 3.680009126663208,
|
| 313 |
+
"eval_runtime": 152.6981,
|
| 314 |
+
"eval_samples_per_second": 1.958,
|
| 315 |
+
"eval_steps_per_second": 0.065,
|
| 316 |
"step": 310
|
| 317 |
},
|
| 318 |
{
|
| 319 |
+
"epoch": 10.010416666666666,
|
| 320 |
+
"grad_norm": 4.879744529724121,
|
| 321 |
+
"learning_rate": 3.7037037037037037e-05,
|
| 322 |
+
"loss": 3.54,
|
| 323 |
"step": 320
|
| 324 |
},
|
| 325 |
{
|
| 326 |
+
"epoch": 10.020833333333334,
|
| 327 |
+
"grad_norm": 6.769280910491943,
|
| 328 |
+
"learning_rate": 3.6458333333333336e-05,
|
| 329 |
+
"loss": 3.6435,
|
| 330 |
"step": 330
|
| 331 |
},
|
| 332 |
{
|
| 333 |
+
"epoch": 10.03125,
|
| 334 |
+
"grad_norm": 6.974754810333252,
|
| 335 |
+
"learning_rate": 3.587962962962963e-05,
|
| 336 |
+
"loss": 3.5147,
|
| 337 |
"step": 340
|
| 338 |
},
|
| 339 |
{
|
| 340 |
+
"epoch": 10.032291666666667,
|
| 341 |
+
"eval_accuracy": 0.11036789297658862,
|
| 342 |
+
"eval_loss": 3.541635036468506,
|
| 343 |
+
"eval_runtime": 156.1418,
|
| 344 |
+
"eval_samples_per_second": 1.915,
|
| 345 |
+
"eval_steps_per_second": 0.064,
|
| 346 |
"step": 341
|
| 347 |
},
|
| 348 |
{
|
| 349 |
+
"epoch": 11.009375,
|
| 350 |
+
"grad_norm": 6.004092216491699,
|
| 351 |
+
"learning_rate": 3.530092592592593e-05,
|
| 352 |
+
"loss": 3.2736,
|
| 353 |
"step": 350
|
| 354 |
},
|
| 355 |
{
|
| 356 |
+
"epoch": 11.019791666666666,
|
| 357 |
+
"grad_norm": 4.7057905197143555,
|
| 358 |
+
"learning_rate": 3.472222222222222e-05,
|
| 359 |
+
"loss": 3.2459,
|
| 360 |
"step": 360
|
| 361 |
},
|
| 362 |
{
|
| 363 |
+
"epoch": 11.030208333333333,
|
| 364 |
+
"grad_norm": 7.564170837402344,
|
| 365 |
+
"learning_rate": 3.414351851851852e-05,
|
| 366 |
+
"loss": 3.2878,
|
| 367 |
"step": 370
|
| 368 |
},
|
| 369 |
{
|
| 370 |
+
"epoch": 11.032291666666667,
|
| 371 |
+
"eval_accuracy": 0.07023411371237458,
|
| 372 |
+
"eval_loss": 3.707416534423828,
|
| 373 |
+
"eval_runtime": 151.3197,
|
| 374 |
+
"eval_samples_per_second": 1.976,
|
| 375 |
+
"eval_steps_per_second": 0.066,
|
| 376 |
"step": 372
|
| 377 |
},
|
| 378 |
{
|
| 379 |
+
"epoch": 12.008333333333333,
|
| 380 |
+
"grad_norm": 6.44061279296875,
|
| 381 |
+
"learning_rate": 3.3564814814814815e-05,
|
| 382 |
+
"loss": 3.048,
|
| 383 |
"step": 380
|
| 384 |
},
|
| 385 |
{
|
| 386 |
+
"epoch": 12.01875,
|
| 387 |
+
"grad_norm": 6.303668022155762,
|
| 388 |
+
"learning_rate": 3.2986111111111115e-05,
|
| 389 |
+
"loss": 3.0619,
|
| 390 |
"step": 390
|
| 391 |
},
|
| 392 |
{
|
| 393 |
+
"epoch": 12.029166666666667,
|
| 394 |
+
"grad_norm": 7.074206352233887,
|
| 395 |
+
"learning_rate": 3.240740740740741e-05,
|
| 396 |
+
"loss": 2.9491,
|
| 397 |
"step": 400
|
| 398 |
},
|
| 399 |
{
|
| 400 |
+
"epoch": 12.032291666666667,
|
| 401 |
+
"eval_accuracy": 0.10702341137123746,
|
| 402 |
+
"eval_loss": 3.3954155445098877,
|
| 403 |
+
"eval_runtime": 154.7501,
|
| 404 |
+
"eval_samples_per_second": 1.932,
|
| 405 |
+
"eval_steps_per_second": 0.065,
|
| 406 |
"step": 403
|
| 407 |
},
|
| 408 |
{
|
| 409 |
+
"epoch": 13.007291666666667,
|
| 410 |
+
"grad_norm": 6.961267471313477,
|
| 411 |
+
"learning_rate": 3.182870370370371e-05,
|
| 412 |
+
"loss": 2.8801,
|
| 413 |
"step": 410
|
| 414 |
},
|
| 415 |
{
|
| 416 |
+
"epoch": 13.017708333333333,
|
| 417 |
+
"grad_norm": 5.816298007965088,
|
| 418 |
+
"learning_rate": 3.125e-05,
|
| 419 |
+
"loss": 2.9025,
|
| 420 |
"step": 420
|
| 421 |
},
|
| 422 |
{
|
| 423 |
+
"epoch": 13.028125,
|
| 424 |
+
"grad_norm": 10.059369087219238,
|
| 425 |
+
"learning_rate": 3.06712962962963e-05,
|
| 426 |
+
"loss": 2.806,
|
| 427 |
"step": 430
|
| 428 |
},
|
| 429 |
{
|
| 430 |
+
"epoch": 13.032291666666667,
|
| 431 |
+
"eval_accuracy": 0.1705685618729097,
|
| 432 |
+
"eval_loss": 3.2551913261413574,
|
| 433 |
+
"eval_runtime": 154.5283,
|
| 434 |
+
"eval_samples_per_second": 1.935,
|
| 435 |
+
"eval_steps_per_second": 0.065,
|
| 436 |
"step": 434
|
| 437 |
},
|
| 438 |
{
|
| 439 |
+
"epoch": 14.00625,
|
| 440 |
+
"grad_norm": 6.142374515533447,
|
| 441 |
+
"learning_rate": 3.0092592592592593e-05,
|
| 442 |
+
"loss": 2.6302,
|
| 443 |
"step": 440
|
| 444 |
},
|
| 445 |
{
|
| 446 |
+
"epoch": 14.016666666666667,
|
| 447 |
+
"grad_norm": 5.806746959686279,
|
| 448 |
+
"learning_rate": 2.951388888888889e-05,
|
| 449 |
+
"loss": 2.5778,
|
| 450 |
"step": 450
|
| 451 |
},
|
| 452 |
{
|
| 453 |
+
"epoch": 14.027083333333334,
|
| 454 |
+
"grad_norm": 6.683168888092041,
|
| 455 |
+
"learning_rate": 2.8935185185185186e-05,
|
| 456 |
+
"loss": 2.4568,
|
| 457 |
"step": 460
|
| 458 |
},
|
| 459 |
{
|
| 460 |
+
"epoch": 14.032291666666667,
|
| 461 |
+
"eval_accuracy": 0.2040133779264214,
|
| 462 |
+
"eval_loss": 3.0654280185699463,
|
| 463 |
+
"eval_runtime": 155.0,
|
| 464 |
+
"eval_samples_per_second": 1.929,
|
| 465 |
+
"eval_steps_per_second": 0.065,
|
| 466 |
"step": 465
|
| 467 |
},
|
| 468 |
{
|
| 469 |
+
"epoch": 15.005208333333334,
|
| 470 |
+
"grad_norm": 8.81658935546875,
|
| 471 |
+
"learning_rate": 2.8356481481481483e-05,
|
| 472 |
+
"loss": 2.4185,
|
| 473 |
"step": 470
|
| 474 |
},
|
| 475 |
{
|
| 476 |
+
"epoch": 15.015625,
|
| 477 |
+
"grad_norm": 17.862884521484375,
|
| 478 |
+
"learning_rate": 2.777777777777778e-05,
|
| 479 |
+
"loss": 2.3865,
|
| 480 |
"step": 480
|
| 481 |
},
|
| 482 |
{
|
| 483 |
+
"epoch": 15.026041666666666,
|
| 484 |
+
"grad_norm": 7.982626438140869,
|
| 485 |
+
"learning_rate": 2.7199074074074076e-05,
|
| 486 |
+
"loss": 2.3102,
|
| 487 |
+
"step": 490
|
|
|
|
|
|
|
| 488 |
},
|
| 489 |
{
|
| 490 |
+
"epoch": 15.032291666666667,
|
| 491 |
+
"eval_accuracy": 0.3010033444816054,
|
| 492 |
+
"eval_loss": 2.7440292835235596,
|
| 493 |
+
"eval_runtime": 154.0941,
|
| 494 |
+
"eval_samples_per_second": 1.94,
|
| 495 |
+
"eval_steps_per_second": 0.065,
|
| 496 |
+
"step": 496
|
| 497 |
},
|
| 498 |
{
|
| 499 |
+
"epoch": 16.004166666666666,
|
| 500 |
+
"grad_norm": 7.09642219543457,
|
| 501 |
+
"learning_rate": 2.6620370370370372e-05,
|
| 502 |
+
"loss": 2.1208,
|
| 503 |
+
"step": 500
|
|
|
|
|
|
|
| 504 |
},
|
| 505 |
{
|
| 506 |
+
"epoch": 16.014583333333334,
|
| 507 |
+
"grad_norm": 7.21767520904541,
|
| 508 |
+
"learning_rate": 2.604166666666667e-05,
|
| 509 |
+
"loss": 2.1481,
|
| 510 |
+
"step": 510
|
| 511 |
+
},
|
| 512 |
+
{
|
| 513 |
+
"epoch": 16.025,
|
| 514 |
+
"grad_norm": 8.300311088562012,
|
| 515 |
+
"learning_rate": 2.5462962962962965e-05,
|
| 516 |
+
"loss": 2.2079,
|
| 517 |
+
"step": 520
|
| 518 |
+
},
|
| 519 |
+
{
|
| 520 |
+
"epoch": 16.032291666666666,
|
| 521 |
+
"eval_accuracy": 0.31438127090301005,
|
| 522 |
+
"eval_loss": 2.6789305210113525,
|
| 523 |
+
"eval_runtime": 157.8279,
|
| 524 |
"eval_samples_per_second": 1.894,
|
| 525 |
"eval_steps_per_second": 0.063,
|
| 526 |
+
"step": 527
|
| 527 |
+
},
|
| 528 |
+
{
|
| 529 |
+
"epoch": 17.003125,
|
| 530 |
+
"grad_norm": 8.606009483337402,
|
| 531 |
+
"learning_rate": 2.488425925925926e-05,
|
| 532 |
+
"loss": 2.0646,
|
| 533 |
+
"step": 530
|
| 534 |
+
},
|
| 535 |
+
{
|
| 536 |
+
"epoch": 17.013541666666665,
|
| 537 |
+
"grad_norm": 7.775322437286377,
|
| 538 |
+
"learning_rate": 2.4305555555555558e-05,
|
| 539 |
+
"loss": 1.9341,
|
| 540 |
+
"step": 540
|
| 541 |
+
},
|
| 542 |
+
{
|
| 543 |
+
"epoch": 17.023958333333333,
|
| 544 |
+
"grad_norm": 7.849579811096191,
|
| 545 |
+
"learning_rate": 2.3726851851851854e-05,
|
| 546 |
+
"loss": 1.9638,
|
| 547 |
+
"step": 550
|
| 548 |
+
},
|
| 549 |
+
{
|
| 550 |
+
"epoch": 17.032291666666666,
|
| 551 |
+
"eval_accuracy": 0.36789297658862874,
|
| 552 |
+
"eval_loss": 2.5920491218566895,
|
| 553 |
+
"eval_runtime": 157.1641,
|
| 554 |
+
"eval_samples_per_second": 1.902,
|
| 555 |
+
"eval_steps_per_second": 0.064,
|
| 556 |
+
"step": 558
|
| 557 |
+
},
|
| 558 |
+
{
|
| 559 |
+
"epoch": 18.002083333333335,
|
| 560 |
+
"grad_norm": 10.023223876953125,
|
| 561 |
+
"learning_rate": 2.314814814814815e-05,
|
| 562 |
+
"loss": 2.0018,
|
| 563 |
+
"step": 560
|
| 564 |
+
},
|
| 565 |
+
{
|
| 566 |
+
"epoch": 18.0125,
|
| 567 |
+
"grad_norm": 7.027205467224121,
|
| 568 |
+
"learning_rate": 2.2569444444444447e-05,
|
| 569 |
+
"loss": 1.8571,
|
| 570 |
+
"step": 570
|
| 571 |
+
},
|
| 572 |
+
{
|
| 573 |
+
"epoch": 18.022916666666667,
|
| 574 |
+
"grad_norm": 9.41115665435791,
|
| 575 |
+
"learning_rate": 2.1990740740740743e-05,
|
| 576 |
+
"loss": 1.7914,
|
| 577 |
+
"step": 580
|
| 578 |
+
},
|
| 579 |
+
{
|
| 580 |
+
"epoch": 18.032291666666666,
|
| 581 |
+
"eval_accuracy": 0.3377926421404682,
|
| 582 |
+
"eval_loss": 2.6151952743530273,
|
| 583 |
+
"eval_runtime": 162.3006,
|
| 584 |
+
"eval_samples_per_second": 1.842,
|
| 585 |
+
"eval_steps_per_second": 0.062,
|
| 586 |
+
"step": 589
|
| 587 |
+
},
|
| 588 |
+
{
|
| 589 |
+
"epoch": 19.001041666666666,
|
| 590 |
+
"grad_norm": 9.176055908203125,
|
| 591 |
+
"learning_rate": 2.141203703703704e-05,
|
| 592 |
+
"loss": 1.8165,
|
| 593 |
+
"step": 590
|
| 594 |
+
},
|
| 595 |
+
{
|
| 596 |
+
"epoch": 19.011458333333334,
|
| 597 |
+
"grad_norm": 7.396921157836914,
|
| 598 |
+
"learning_rate": 2.0833333333333336e-05,
|
| 599 |
+
"loss": 1.6863,
|
| 600 |
+
"step": 600
|
| 601 |
+
},
|
| 602 |
+
{
|
| 603 |
+
"epoch": 19.021875,
|
| 604 |
+
"grad_norm": 9.088372230529785,
|
| 605 |
+
"learning_rate": 2.0254629629629632e-05,
|
| 606 |
+
"loss": 1.597,
|
| 607 |
+
"step": 610
|
| 608 |
+
},
|
| 609 |
+
{
|
| 610 |
+
"epoch": 19.032291666666666,
|
| 611 |
+
"grad_norm": 16.782529830932617,
|
| 612 |
+
"learning_rate": 1.967592592592593e-05,
|
| 613 |
+
"loss": 1.6925,
|
| 614 |
+
"step": 620
|
| 615 |
+
},
|
| 616 |
+
{
|
| 617 |
+
"epoch": 19.032291666666666,
|
| 618 |
+
"eval_accuracy": 0.34448160535117056,
|
| 619 |
+
"eval_loss": 2.5970685482025146,
|
| 620 |
+
"eval_runtime": 161.5247,
|
| 621 |
+
"eval_samples_per_second": 1.851,
|
| 622 |
+
"eval_steps_per_second": 0.062,
|
| 623 |
+
"step": 620
|
| 624 |
+
},
|
| 625 |
+
{
|
| 626 |
+
"epoch": 20.010416666666668,
|
| 627 |
+
"grad_norm": 10.128305435180664,
|
| 628 |
+
"learning_rate": 1.9097222222222222e-05,
|
| 629 |
+
"loss": 1.6303,
|
| 630 |
+
"step": 630
|
| 631 |
+
},
|
| 632 |
+
{
|
| 633 |
+
"epoch": 20.020833333333332,
|
| 634 |
+
"grad_norm": 8.396921157836914,
|
| 635 |
+
"learning_rate": 1.8518518518518518e-05,
|
| 636 |
+
"loss": 1.6306,
|
| 637 |
+
"step": 640
|
| 638 |
+
},
|
| 639 |
+
{
|
| 640 |
+
"epoch": 20.03125,
|
| 641 |
+
"grad_norm": 8.574676513671875,
|
| 642 |
+
"learning_rate": 1.7939814814814815e-05,
|
| 643 |
+
"loss": 1.5124,
|
| 644 |
+
"step": 650
|
| 645 |
+
},
|
| 646 |
+
{
|
| 647 |
+
"epoch": 20.032291666666666,
|
| 648 |
+
"eval_accuracy": 0.34782608695652173,
|
| 649 |
+
"eval_loss": 2.5766701698303223,
|
| 650 |
+
"eval_runtime": 162.6227,
|
| 651 |
+
"eval_samples_per_second": 1.839,
|
| 652 |
+
"eval_steps_per_second": 0.061,
|
| 653 |
+
"step": 651
|
| 654 |
+
},
|
| 655 |
+
{
|
| 656 |
+
"epoch": 21.009375,
|
| 657 |
+
"grad_norm": 8.482582092285156,
|
| 658 |
+
"learning_rate": 1.736111111111111e-05,
|
| 659 |
+
"loss": 1.5163,
|
| 660 |
+
"step": 660
|
| 661 |
+
},
|
| 662 |
+
{
|
| 663 |
+
"epoch": 21.019791666666666,
|
| 664 |
+
"grad_norm": 7.806921482086182,
|
| 665 |
+
"learning_rate": 1.6782407407407408e-05,
|
| 666 |
+
"loss": 1.4579,
|
| 667 |
+
"step": 670
|
| 668 |
+
},
|
| 669 |
+
{
|
| 670 |
+
"epoch": 21.030208333333334,
|
| 671 |
+
"grad_norm": 10.889120101928711,
|
| 672 |
+
"learning_rate": 1.6203703703703704e-05,
|
| 673 |
+
"loss": 1.4834,
|
| 674 |
+
"step": 680
|
| 675 |
+
},
|
| 676 |
+
{
|
| 677 |
+
"epoch": 21.032291666666666,
|
| 678 |
+
"eval_accuracy": 0.3879598662207358,
|
| 679 |
+
"eval_loss": 2.4438529014587402,
|
| 680 |
+
"eval_runtime": 163.8828,
|
| 681 |
+
"eval_samples_per_second": 1.824,
|
| 682 |
+
"eval_steps_per_second": 0.061,
|
| 683 |
+
"step": 682
|
| 684 |
+
},
|
| 685 |
+
{
|
| 686 |
+
"epoch": 22.008333333333333,
|
| 687 |
+
"grad_norm": 8.548680305480957,
|
| 688 |
+
"learning_rate": 1.5625e-05,
|
| 689 |
+
"loss": 1.3913,
|
| 690 |
+
"step": 690
|
| 691 |
+
},
|
| 692 |
+
{
|
| 693 |
+
"epoch": 22.01875,
|
| 694 |
+
"grad_norm": 10.439925193786621,
|
| 695 |
+
"learning_rate": 1.5046296296296297e-05,
|
| 696 |
+
"loss": 1.4161,
|
| 697 |
+
"step": 700
|
| 698 |
+
},
|
| 699 |
+
{
|
| 700 |
+
"epoch": 22.029166666666665,
|
| 701 |
+
"grad_norm": 8.317498207092285,
|
| 702 |
+
"learning_rate": 1.4467592592592593e-05,
|
| 703 |
+
"loss": 1.4565,
|
| 704 |
+
"step": 710
|
| 705 |
+
},
|
| 706 |
+
{
|
| 707 |
+
"epoch": 22.032291666666666,
|
| 708 |
+
"eval_accuracy": 0.38461538461538464,
|
| 709 |
+
"eval_loss": 2.405748128890991,
|
| 710 |
+
"eval_runtime": 162.732,
|
| 711 |
+
"eval_samples_per_second": 1.837,
|
| 712 |
+
"eval_steps_per_second": 0.061,
|
| 713 |
+
"step": 713
|
| 714 |
+
},
|
| 715 |
+
{
|
| 716 |
+
"epoch": 23.007291666666667,
|
| 717 |
+
"grad_norm": 8.510457038879395,
|
| 718 |
+
"learning_rate": 1.388888888888889e-05,
|
| 719 |
+
"loss": 1.4043,
|
| 720 |
+
"step": 720
|
| 721 |
+
},
|
| 722 |
+
{
|
| 723 |
+
"epoch": 23.017708333333335,
|
| 724 |
+
"grad_norm": 9.268413543701172,
|
| 725 |
+
"learning_rate": 1.3310185185185186e-05,
|
| 726 |
+
"loss": 1.2734,
|
| 727 |
+
"step": 730
|
| 728 |
+
},
|
| 729 |
+
{
|
| 730 |
+
"epoch": 23.028125,
|
| 731 |
+
"grad_norm": 8.000787734985352,
|
| 732 |
+
"learning_rate": 1.2731481481481482e-05,
|
| 733 |
+
"loss": 1.279,
|
| 734 |
+
"step": 740
|
| 735 |
+
},
|
| 736 |
+
{
|
| 737 |
+
"epoch": 23.032291666666666,
|
| 738 |
+
"eval_accuracy": 0.35451505016722407,
|
| 739 |
+
"eval_loss": 2.550072193145752,
|
| 740 |
+
"eval_runtime": 160.4305,
|
| 741 |
+
"eval_samples_per_second": 1.864,
|
| 742 |
+
"eval_steps_per_second": 0.062,
|
| 743 |
+
"step": 744
|
| 744 |
+
},
|
| 745 |
+
{
|
| 746 |
+
"epoch": 24.00625,
|
| 747 |
+
"grad_norm": 8.658973693847656,
|
| 748 |
+
"learning_rate": 1.2152777777777779e-05,
|
| 749 |
+
"loss": 1.1963,
|
| 750 |
+
"step": 750
|
| 751 |
+
},
|
| 752 |
+
{
|
| 753 |
+
"epoch": 24.016666666666666,
|
| 754 |
+
"grad_norm": 7.558023929595947,
|
| 755 |
+
"learning_rate": 1.1574074074074075e-05,
|
| 756 |
+
"loss": 1.2601,
|
| 757 |
+
"step": 760
|
| 758 |
+
},
|
| 759 |
+
{
|
| 760 |
+
"epoch": 24.027083333333334,
|
| 761 |
+
"grad_norm": 9.91243839263916,
|
| 762 |
+
"learning_rate": 1.0995370370370372e-05,
|
| 763 |
+
"loss": 1.1477,
|
| 764 |
+
"step": 770
|
| 765 |
+
},
|
| 766 |
+
{
|
| 767 |
+
"epoch": 24.032291666666666,
|
| 768 |
+
"eval_accuracy": 0.44816053511705684,
|
| 769 |
+
"eval_loss": 2.3246614933013916,
|
| 770 |
+
"eval_runtime": 157.9743,
|
| 771 |
+
"eval_samples_per_second": 1.893,
|
| 772 |
+
"eval_steps_per_second": 0.063,
|
| 773 |
+
"step": 775
|
| 774 |
+
},
|
| 775 |
+
{
|
| 776 |
+
"epoch": 25.005208333333332,
|
| 777 |
+
"grad_norm": 10.049819946289062,
|
| 778 |
+
"learning_rate": 1.0416666666666668e-05,
|
| 779 |
+
"loss": 1.1727,
|
| 780 |
+
"step": 780
|
| 781 |
+
},
|
| 782 |
+
{
|
| 783 |
+
"epoch": 25.015625,
|
| 784 |
+
"grad_norm": 8.463665008544922,
|
| 785 |
+
"learning_rate": 9.837962962962964e-06,
|
| 786 |
+
"loss": 1.1207,
|
| 787 |
+
"step": 790
|
| 788 |
+
},
|
| 789 |
+
{
|
| 790 |
+
"epoch": 25.026041666666668,
|
| 791 |
+
"grad_norm": 7.52623176574707,
|
| 792 |
+
"learning_rate": 9.259259259259259e-06,
|
| 793 |
+
"loss": 1.2573,
|
| 794 |
+
"step": 800
|
| 795 |
+
},
|
| 796 |
+
{
|
| 797 |
+
"epoch": 25.032291666666666,
|
| 798 |
+
"eval_accuracy": 0.4882943143812709,
|
| 799 |
+
"eval_loss": 2.1776490211486816,
|
| 800 |
+
"eval_runtime": 158.1033,
|
| 801 |
+
"eval_samples_per_second": 1.891,
|
| 802 |
+
"eval_steps_per_second": 0.063,
|
| 803 |
+
"step": 806
|
| 804 |
+
},
|
| 805 |
+
{
|
| 806 |
+
"epoch": 26.004166666666666,
|
| 807 |
+
"grad_norm": 9.336162567138672,
|
| 808 |
+
"learning_rate": 8.680555555555556e-06,
|
| 809 |
+
"loss": 1.1791,
|
| 810 |
+
"step": 810
|
| 811 |
+
},
|
| 812 |
+
{
|
| 813 |
+
"epoch": 26.014583333333334,
|
| 814 |
+
"grad_norm": 6.728664875030518,
|
| 815 |
+
"learning_rate": 8.101851851851852e-06,
|
| 816 |
+
"loss": 1.1284,
|
| 817 |
+
"step": 820
|
| 818 |
+
},
|
| 819 |
+
{
|
| 820 |
+
"epoch": 26.025,
|
| 821 |
+
"grad_norm": 7.307468414306641,
|
| 822 |
+
"learning_rate": 7.523148148148148e-06,
|
| 823 |
+
"loss": 1.0825,
|
| 824 |
+
"step": 830
|
| 825 |
+
},
|
| 826 |
+
{
|
| 827 |
+
"epoch": 26.032291666666666,
|
| 828 |
+
"eval_accuracy": 0.4782608695652174,
|
| 829 |
+
"eval_loss": 2.14430832862854,
|
| 830 |
+
"eval_runtime": 156.854,
|
| 831 |
+
"eval_samples_per_second": 1.906,
|
| 832 |
+
"eval_steps_per_second": 0.064,
|
| 833 |
+
"step": 837
|
| 834 |
+
},
|
| 835 |
+
{
|
| 836 |
+
"epoch": 27.003125,
|
| 837 |
+
"grad_norm": 8.843693733215332,
|
| 838 |
+
"learning_rate": 6.944444444444445e-06,
|
| 839 |
+
"loss": 1.1431,
|
| 840 |
+
"step": 840
|
| 841 |
+
},
|
| 842 |
+
{
|
| 843 |
+
"epoch": 27.013541666666665,
|
| 844 |
+
"grad_norm": 9.648551940917969,
|
| 845 |
+
"learning_rate": 6.365740740740741e-06,
|
| 846 |
+
"loss": 1.0873,
|
| 847 |
+
"step": 850
|
| 848 |
+
},
|
| 849 |
+
{
|
| 850 |
+
"epoch": 27.023958333333333,
|
| 851 |
+
"grad_norm": 10.551318168640137,
|
| 852 |
+
"learning_rate": 5.787037037037038e-06,
|
| 853 |
+
"loss": 1.2121,
|
| 854 |
+
"step": 860
|
| 855 |
+
},
|
| 856 |
+
{
|
| 857 |
+
"epoch": 27.032291666666666,
|
| 858 |
+
"eval_accuracy": 0.4782608695652174,
|
| 859 |
+
"eval_loss": 2.149031639099121,
|
| 860 |
+
"eval_runtime": 156.4789,
|
| 861 |
+
"eval_samples_per_second": 1.911,
|
| 862 |
+
"eval_steps_per_second": 0.064,
|
| 863 |
+
"step": 868
|
| 864 |
+
},
|
| 865 |
+
{
|
| 866 |
+
"epoch": 28.002083333333335,
|
| 867 |
+
"grad_norm": 7.037572383880615,
|
| 868 |
+
"learning_rate": 5.208333333333334e-06,
|
| 869 |
+
"loss": 1.0147,
|
| 870 |
+
"step": 870
|
| 871 |
+
},
|
| 872 |
+
{
|
| 873 |
+
"epoch": 28.0125,
|
| 874 |
+
"grad_norm": 11.031582832336426,
|
| 875 |
+
"learning_rate": 4.6296296296296296e-06,
|
| 876 |
+
"loss": 1.0574,
|
| 877 |
+
"step": 880
|
| 878 |
+
},
|
| 879 |
+
{
|
| 880 |
+
"epoch": 28.022916666666667,
|
| 881 |
+
"grad_norm": 7.408577919006348,
|
| 882 |
+
"learning_rate": 4.050925925925926e-06,
|
| 883 |
+
"loss": 1.0887,
|
| 884 |
+
"step": 890
|
| 885 |
+
},
|
| 886 |
+
{
|
| 887 |
+
"epoch": 28.032291666666666,
|
| 888 |
+
"eval_accuracy": 0.47157190635451507,
|
| 889 |
+
"eval_loss": 2.151550054550171,
|
| 890 |
+
"eval_runtime": 157.7417,
|
| 891 |
+
"eval_samples_per_second": 1.896,
|
| 892 |
+
"eval_steps_per_second": 0.063,
|
| 893 |
+
"step": 899
|
| 894 |
+
},
|
| 895 |
+
{
|
| 896 |
+
"epoch": 29.001041666666666,
|
| 897 |
+
"grad_norm": 7.416742324829102,
|
| 898 |
+
"learning_rate": 3.4722222222222224e-06,
|
| 899 |
+
"loss": 1.1448,
|
| 900 |
+
"step": 900
|
| 901 |
+
},
|
| 902 |
+
{
|
| 903 |
+
"epoch": 29.011458333333334,
|
| 904 |
+
"grad_norm": 8.579631805419922,
|
| 905 |
+
"learning_rate": 2.893518518518519e-06,
|
| 906 |
+
"loss": 1.051,
|
| 907 |
+
"step": 910
|
| 908 |
+
},
|
| 909 |
+
{
|
| 910 |
+
"epoch": 29.021875,
|
| 911 |
+
"grad_norm": 7.20076847076416,
|
| 912 |
+
"learning_rate": 2.3148148148148148e-06,
|
| 913 |
+
"loss": 0.9936,
|
| 914 |
+
"step": 920
|
| 915 |
+
},
|
| 916 |
+
{
|
| 917 |
+
"epoch": 29.032291666666666,
|
| 918 |
+
"grad_norm": 20.40438461303711,
|
| 919 |
+
"learning_rate": 1.7361111111111112e-06,
|
| 920 |
+
"loss": 1.1127,
|
| 921 |
+
"step": 930
|
| 922 |
+
},
|
| 923 |
+
{
|
| 924 |
+
"epoch": 29.032291666666666,
|
| 925 |
+
"eval_accuracy": 0.4882943143812709,
|
| 926 |
+
"eval_loss": 2.1050899028778076,
|
| 927 |
+
"eval_runtime": 158.1493,
|
| 928 |
+
"eval_samples_per_second": 1.891,
|
| 929 |
+
"eval_steps_per_second": 0.063,
|
| 930 |
+
"step": 930
|
| 931 |
+
},
|
| 932 |
+
{
|
| 933 |
+
"epoch": 30.010416666666668,
|
| 934 |
+
"grad_norm": 7.520814895629883,
|
| 935 |
+
"learning_rate": 1.1574074074074074e-06,
|
| 936 |
+
"loss": 1.0214,
|
| 937 |
+
"step": 940
|
| 938 |
+
},
|
| 939 |
+
{
|
| 940 |
+
"epoch": 30.020833333333332,
|
| 941 |
+
"grad_norm": 7.830852508544922,
|
| 942 |
+
"learning_rate": 5.787037037037037e-07,
|
| 943 |
+
"loss": 0.9516,
|
| 944 |
+
"step": 950
|
| 945 |
+
},
|
| 946 |
+
{
|
| 947 |
+
"epoch": 30.03125,
|
| 948 |
+
"grad_norm": 8.230799674987793,
|
| 949 |
+
"learning_rate": 0.0,
|
| 950 |
+
"loss": 0.9905,
|
| 951 |
+
"step": 960
|
| 952 |
+
},
|
| 953 |
+
{
|
| 954 |
+
"epoch": 30.03125,
|
| 955 |
+
"eval_accuracy": 0.4816053511705686,
|
| 956 |
+
"eval_loss": 2.117016553878784,
|
| 957 |
+
"eval_runtime": 175.2626,
|
| 958 |
+
"eval_samples_per_second": 1.706,
|
| 959 |
+
"eval_steps_per_second": 0.057,
|
| 960 |
+
"step": 960
|
| 961 |
+
},
|
| 962 |
+
{
|
| 963 |
+
"epoch": 30.03125,
|
| 964 |
+
"step": 960,
|
| 965 |
+
"total_flos": 3.733004678582315e+19,
|
| 966 |
+
"train_loss": 2.590478341778119,
|
| 967 |
+
"train_runtime": 22304.5183,
|
| 968 |
+
"train_samples_per_second": 1.377,
|
| 969 |
+
"train_steps_per_second": 0.043
|
| 970 |
+
},
|
| 971 |
+
{
|
| 972 |
+
"epoch": 30.03125,
|
| 973 |
+
"eval_accuracy": 0.4882943143812709,
|
| 974 |
+
"eval_loss": 2.1814045906066895,
|
| 975 |
+
"eval_runtime": 181.2056,
|
| 976 |
+
"eval_samples_per_second": 1.65,
|
| 977 |
+
"eval_steps_per_second": 0.055,
|
| 978 |
+
"step": 960
|
| 979 |
+
},
|
| 980 |
+
{
|
| 981 |
+
"epoch": 30.03125,
|
| 982 |
+
"eval_accuracy": 0.4882943143812709,
|
| 983 |
+
"eval_loss": 2.180596113204956,
|
| 984 |
+
"eval_runtime": 179.6038,
|
| 985 |
+
"eval_samples_per_second": 1.665,
|
| 986 |
+
"eval_steps_per_second": 0.056,
|
| 987 |
+
"step": 960
|
| 988 |
}
|
| 989 |
],
|
| 990 |
"logging_steps": 10,
|
| 991 |
+
"max_steps": 960,
|
| 992 |
"num_input_tokens_seen": 0,
|
| 993 |
"num_train_epochs": 9223372036854775807,
|
| 994 |
"save_steps": 500,
|
|
|
|
| 1004 |
"attributes": {}
|
| 1005 |
}
|
| 1006 |
},
|
| 1007 |
+
"total_flos": 3.733004678582315e+19,
|
| 1008 |
"train_batch_size": 32,
|
| 1009 |
"trial_name": null,
|
| 1010 |
"trial_params": null
|