Abdulwahab Sahyoun
commited on
Commit
•
06d286f
1
Parent(s):
6bf0d45
End of training
Browse files- all_results.json +10 -10
- eval_results.json +6 -6
- train_results.json +5 -5
- trainer_state.json +379 -22
all_results.json
CHANGED
|
@@ -1,14 +1,14 @@
|
|
| 1 |
{
|
| 2 |
-
"epoch":
|
| 3 |
-
"eval_loss": 0.
|
| 4 |
-
"eval_runtime":
|
| 5 |
"eval_samples": 4899,
|
| 6 |
-
"eval_samples_per_second":
|
| 7 |
-
"eval_steps_per_second": 0.
|
| 8 |
-
"eval_wer": 0.
|
| 9 |
-
"train_loss":
|
| 10 |
-
"train_runtime":
|
| 11 |
"train_samples": 14730,
|
| 12 |
-
"train_samples_per_second":
|
| 13 |
-
"train_steps_per_second": 0.
|
| 14 |
}
|
| 1 |
{
|
| 2 |
+
"epoch": 30.0,
|
| 3 |
+
"eval_loss": 0.8535701632499695,
|
| 4 |
+
"eval_runtime": 201.4101,
|
| 5 |
"eval_samples": 4899,
|
| 6 |
+
"eval_samples_per_second": 24.324,
|
| 7 |
+
"eval_steps_per_second": 0.765,
|
| 8 |
+
"eval_wer": 0.3736551873276748,
|
| 9 |
+
"train_loss": 0.15137778351272363,
|
| 10 |
+
"train_runtime": 18309.2993,
|
| 11 |
"train_samples": 14730,
|
| 12 |
+
"train_samples_per_second": 24.135,
|
| 13 |
+
"train_steps_per_second": 0.377
|
| 14 |
}
|
eval_results.json
CHANGED
|
@@ -1,9 +1,9 @@
|
|
| 1 |
{
|
| 2 |
-
"epoch":
|
| 3 |
-
"eval_loss": 0.
|
| 4 |
-
"eval_runtime":
|
| 5 |
"eval_samples": 4899,
|
| 6 |
-
"eval_samples_per_second":
|
| 7 |
-
"eval_steps_per_second": 0.
|
| 8 |
-
"eval_wer": 0.
|
| 9 |
}
|
| 1 |
{
|
| 2 |
+
"epoch": 30.0,
|
| 3 |
+
"eval_loss": 0.8535701632499695,
|
| 4 |
+
"eval_runtime": 201.4101,
|
| 5 |
"eval_samples": 4899,
|
| 6 |
+
"eval_samples_per_second": 24.324,
|
| 7 |
+
"eval_steps_per_second": 0.765,
|
| 8 |
+
"eval_wer": 0.3736551873276748
|
| 9 |
}
|
train_results.json
CHANGED
|
@@ -1,8 +1,8 @@
|
|
| 1 |
{
|
| 2 |
-
"epoch":
|
| 3 |
-
"train_loss":
|
| 4 |
-
"train_runtime":
|
| 5 |
"train_samples": 14730,
|
| 6 |
-
"train_samples_per_second":
|
| 7 |
-
"train_steps_per_second": 0.
|
| 8 |
}
|
| 1 |
{
|
| 2 |
+
"epoch": 30.0,
|
| 3 |
+
"train_loss": 0.15137778351272363,
|
| 4 |
+
"train_runtime": 18309.2993,
|
| 5 |
"train_samples": 14730,
|
| 6 |
+
"train_samples_per_second": 24.135,
|
| 7 |
+
"train_steps_per_second": 0.377
|
| 8 |
}
|
trainer_state.json
CHANGED
|
@@ -1,8 +1,8 @@
|
|
| 1 |
{
|
| 2 |
"best_metric": null,
|
| 3 |
"best_model_checkpoint": null,
|
| 4 |
-
"epoch":
|
| 5 |
-
"global_step":
|
| 6 |
"is_hyper_param_search": false,
|
| 7 |
"is_local_process_zero": true,
|
| 8 |
"is_world_process_zero": true,
|
|
@@ -333,35 +333,392 @@
|
|
| 333 |
},
|
| 334 |
{
|
| 335 |
"epoch": 14.35,
|
| 336 |
-
"eval_loss": 0.
|
| 337 |
-
"eval_runtime":
|
| 338 |
-
"eval_samples_per_second": 24.
|
| 339 |
-
"eval_steps_per_second": 0.
|
| 340 |
-
"eval_wer": 0.
|
| 341 |
"step": 3300
|
| 342 |
},
|
| 343 |
{
|
| 344 |
"epoch": 14.78,
|
| 345 |
-
"eval_loss": 0.
|
| 346 |
-
"eval_runtime": 201.
|
| 347 |
-
"eval_samples_per_second": 24.
|
| 348 |
-
"eval_steps_per_second": 0.
|
| 349 |
-
"eval_wer": 0.
|
| 350 |
"step": 3400
|
| 351 |
},
|
| 352 |
{
|
| 353 |
-
"epoch": 15.
|
| 354 |
-
"
|
| 355 |
-
"
|
| 356 |
-
"
|
| 357 |
-
|
| 358 |
-
|
| 359 |
-
"
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 360 |
}
|
| 361 |
],
|
| 362 |
-
"max_steps":
|
| 363 |
-
"num_train_epochs":
|
| 364 |
-
"total_flos":
|
| 365 |
"trial_name": null,
|
| 366 |
"trial_params": null
|
| 367 |
}
|
| 1 |
{
|
| 2 |
"best_metric": null,
|
| 3 |
"best_model_checkpoint": null,
|
| 4 |
+
"epoch": 29.997830802603037,
|
| 5 |
+
"global_step": 6900,
|
| 6 |
"is_hyper_param_search": false,
|
| 7 |
"is_local_process_zero": true,
|
| 8 |
"is_world_process_zero": true,
|
| 333 |
},
|
| 334 |
{
|
| 335 |
"epoch": 14.35,
|
| 336 |
+
"eval_loss": 0.7177483439445496,
|
| 337 |
+
"eval_runtime": 201.1511,
|
| 338 |
+
"eval_samples_per_second": 24.355,
|
| 339 |
+
"eval_steps_per_second": 0.766,
|
| 340 |
+
"eval_wer": 0.4009839433421636,
|
| 341 |
"step": 3300
|
| 342 |
},
|
| 343 |
{
|
| 344 |
"epoch": 14.78,
|
| 345 |
+
"eval_loss": 0.6887663006782532,
|
| 346 |
+
"eval_runtime": 201.3869,
|
| 347 |
+
"eval_samples_per_second": 24.326,
|
| 348 |
+
"eval_steps_per_second": 0.765,
|
| 349 |
+
"eval_wer": 0.4042547440125426,
|
| 350 |
"step": 3400
|
| 351 |
},
|
| 352 |
{
|
| 353 |
+
"epoch": 15.22,
|
| 354 |
+
"learning_rate": 0.000159515625,
|
| 355 |
+
"loss": 0.3767,
|
| 356 |
+
"step": 3500
|
| 357 |
+
},
|
| 358 |
+
{
|
| 359 |
+
"epoch": 15.22,
|
| 360 |
+
"eval_loss": 0.7124109268188477,
|
| 361 |
+
"eval_runtime": 201.401,
|
| 362 |
+
"eval_samples_per_second": 24.325,
|
| 363 |
+
"eval_steps_per_second": 0.765,
|
| 364 |
+
"eval_wer": 0.4202303076174515,
|
| 365 |
+
"step": 3500
|
| 366 |
+
},
|
| 367 |
+
{
|
| 368 |
+
"epoch": 15.65,
|
| 369 |
+
"eval_loss": 0.7275710701942444,
|
| 370 |
+
"eval_runtime": 201.0133,
|
| 371 |
+
"eval_samples_per_second": 24.372,
|
| 372 |
+
"eval_steps_per_second": 0.766,
|
| 373 |
+
"eval_wer": 0.41201275882575555,
|
| 374 |
+
"step": 3600
|
| 375 |
+
},
|
| 376 |
+
{
|
| 377 |
+
"epoch": 16.09,
|
| 378 |
+
"eval_loss": 0.7265337109565735,
|
| 379 |
+
"eval_runtime": 201.5776,
|
| 380 |
+
"eval_samples_per_second": 24.303,
|
| 381 |
+
"eval_steps_per_second": 0.764,
|
| 382 |
+
"eval_wer": 0.40336270746607555,
|
| 383 |
+
"step": 3700
|
| 384 |
+
},
|
| 385 |
+
{
|
| 386 |
+
"epoch": 16.52,
|
| 387 |
+
"eval_loss": 0.7392393946647644,
|
| 388 |
+
"eval_runtime": 200.4049,
|
| 389 |
+
"eval_samples_per_second": 24.446,
|
| 390 |
+
"eval_steps_per_second": 0.768,
|
| 391 |
+
"eval_wer": 0.40771476455641453,
|
| 392 |
+
"step": 3800
|
| 393 |
+
},
|
| 394 |
+
{
|
| 395 |
+
"epoch": 16.95,
|
| 396 |
+
"eval_loss": 0.7403403520584106,
|
| 397 |
+
"eval_runtime": 200.8623,
|
| 398 |
+
"eval_samples_per_second": 24.39,
|
| 399 |
+
"eval_steps_per_second": 0.767,
|
| 400 |
+
"eval_wer": 0.39652376060982863,
|
| 401 |
+
"step": 3900
|
| 402 |
+
},
|
| 403 |
+
{
|
| 404 |
+
"epoch": 17.39,
|
| 405 |
+
"learning_rate": 0.000136078125,
|
| 406 |
+
"loss": 0.3603,
|
| 407 |
+
"step": 4000
|
| 408 |
+
},
|
| 409 |
+
{
|
| 410 |
+
"epoch": 17.39,
|
| 411 |
+
"eval_loss": 0.7444584369659424,
|
| 412 |
+
"eval_runtime": 203.8048,
|
| 413 |
+
"eval_samples_per_second": 24.038,
|
| 414 |
+
"eval_steps_per_second": 0.756,
|
| 415 |
+
"eval_wer": 0.4016326971941396,
|
| 416 |
+
"step": 4000
|
| 417 |
+
},
|
| 418 |
+
{
|
| 419 |
+
"epoch": 17.82,
|
| 420 |
+
"eval_loss": 0.757925271987915,
|
| 421 |
+
"eval_runtime": 201.3248,
|
| 422 |
+
"eval_samples_per_second": 24.334,
|
| 423 |
+
"eval_steps_per_second": 0.765,
|
| 424 |
+
"eval_wer": 0.4012272260366546,
|
| 425 |
+
"step": 4100
|
| 426 |
+
},
|
| 427 |
+
{
|
| 428 |
+
"epoch": 18.26,
|
| 429 |
+
"eval_loss": 0.7225211262702942,
|
| 430 |
+
"eval_runtime": 200.7191,
|
| 431 |
+
"eval_samples_per_second": 24.407,
|
| 432 |
+
"eval_steps_per_second": 0.767,
|
| 433 |
+
"eval_wer": 0.3962804779153376,
|
| 434 |
+
"step": 4200
|
| 435 |
+
},
|
| 436 |
+
{
|
| 437 |
+
"epoch": 18.69,
|
| 438 |
+
"eval_loss": 0.7354857325553894,
|
| 439 |
+
"eval_runtime": 200.5207,
|
| 440 |
+
"eval_samples_per_second": 24.431,
|
| 441 |
+
"eval_steps_per_second": 0.768,
|
| 442 |
+
"eval_wer": 0.3951181272638806,
|
| 443 |
+
"step": 4300
|
| 444 |
+
},
|
| 445 |
+
{
|
| 446 |
+
"epoch": 19.13,
|
| 447 |
+
"eval_loss": 0.7482349276542664,
|
| 448 |
+
"eval_runtime": 200.7847,
|
| 449 |
+
"eval_samples_per_second": 24.399,
|
| 450 |
+
"eval_steps_per_second": 0.767,
|
| 451 |
+
"eval_wer": 0.39252311185597666,
|
| 452 |
+
"step": 4400
|
| 453 |
+
},
|
| 454 |
+
{
|
| 455 |
+
"epoch": 19.56,
|
| 456 |
+
"learning_rate": 0.000112640625,
|
| 457 |
+
"loss": 0.3153,
|
| 458 |
+
"step": 4500
|
| 459 |
+
},
|
| 460 |
+
{
|
| 461 |
+
"epoch": 19.56,
|
| 462 |
+
"eval_loss": 0.772259533405304,
|
| 463 |
+
"eval_runtime": 200.8729,
|
| 464 |
+
"eval_samples_per_second": 24.389,
|
| 465 |
+
"eval_steps_per_second": 0.767,
|
| 466 |
+
"eval_wer": 0.3971725144618046,
|
| 467 |
+
"step": 4500
|
| 468 |
+
},
|
| 469 |
+
{
|
| 470 |
+
"epoch": 20.0,
|
| 471 |
+
"eval_loss": 0.7469175457954407,
|
| 472 |
+
"eval_runtime": 200.5892,
|
| 473 |
+
"eval_samples_per_second": 24.423,
|
| 474 |
+
"eval_steps_per_second": 0.768,
|
| 475 |
+
"eval_wer": 0.3897929393955777,
|
| 476 |
+
"step": 4600
|
| 477 |
+
},
|
| 478 |
+
{
|
| 479 |
+
"epoch": 20.43,
|
| 480 |
+
"eval_loss": 0.7799847722053528,
|
| 481 |
+
"eval_runtime": 200.7406,
|
| 482 |
+
"eval_samples_per_second": 24.405,
|
| 483 |
+
"eval_steps_per_second": 0.767,
|
| 484 |
+
"eval_wer": 0.39444234200140565,
|
| 485 |
+
"step": 4700
|
| 486 |
+
},
|
| 487 |
+
{
|
| 488 |
+
"epoch": 20.87,
|
| 489 |
+
"eval_loss": 0.7827086448669434,
|
| 490 |
+
"eval_runtime": 200.8886,
|
| 491 |
+
"eval_samples_per_second": 24.387,
|
| 492 |
+
"eval_steps_per_second": 0.767,
|
| 493 |
+
"eval_wer": 0.3897388765745797,
|
| 494 |
+
"step": 4800
|
| 495 |
+
},
|
| 496 |
+
{
|
| 497 |
+
"epoch": 21.3,
|
| 498 |
+
"eval_loss": 0.7935096621513367,
|
| 499 |
+
"eval_runtime": 204.4398,
|
| 500 |
+
"eval_samples_per_second": 23.963,
|
| 501 |
+
"eval_steps_per_second": 0.753,
|
| 502 |
+
"eval_wer": 0.3913607612045196,
|
| 503 |
+
"step": 4900
|
| 504 |
+
},
|
| 505 |
+
{
|
| 506 |
+
"epoch": 21.74,
|
| 507 |
+
"learning_rate": 8.9203125e-05,
|
| 508 |
+
"loss": 0.286,
|
| 509 |
+
"step": 5000
|
| 510 |
+
},
|
| 511 |
+
{
|
| 512 |
+
"epoch": 21.74,
|
| 513 |
+
"eval_loss": 0.7984351515769958,
|
| 514 |
+
"eval_runtime": 200.4244,
|
| 515 |
+
"eval_samples_per_second": 24.443,
|
| 516 |
+
"eval_steps_per_second": 0.768,
|
| 517 |
+
"eval_wer": 0.37497972644212574,
|
| 518 |
+
"step": 5000
|
| 519 |
+
},
|
| 520 |
+
{
|
| 521 |
+
"epoch": 22.17,
|
| 522 |
+
"eval_loss": 0.7945317625999451,
|
| 523 |
+
"eval_runtime": 200.6266,
|
| 524 |
+
"eval_samples_per_second": 24.418,
|
| 525 |
+
"eval_steps_per_second": 0.768,
|
| 526 |
+
"eval_wer": 0.3830080553603287,
|
| 527 |
+
"step": 5100
|
| 528 |
+
},
|
| 529 |
+
{
|
| 530 |
+
"epoch": 22.61,
|
| 531 |
+
"eval_loss": 0.8010774254798889,
|
| 532 |
+
"eval_runtime": 200.4942,
|
| 533 |
+
"eval_samples_per_second": 24.435,
|
| 534 |
+
"eval_steps_per_second": 0.768,
|
| 535 |
+
"eval_wer": 0.37746661620803373,
|
| 536 |
+
"step": 5200
|
| 537 |
+
},
|
| 538 |
+
{
|
| 539 |
+
"epoch": 23.04,
|
| 540 |
+
"eval_loss": 0.7977942228317261,
|
| 541 |
+
"eval_runtime": 201.0949,
|
| 542 |
+
"eval_samples_per_second": 24.362,
|
| 543 |
+
"eval_steps_per_second": 0.766,
|
| 544 |
+
"eval_wer": 0.3823863329188517,
|
| 545 |
+
"step": 5300
|
| 546 |
+
},
|
| 547 |
+
{
|
| 548 |
+
"epoch": 23.48,
|
| 549 |
+
"eval_loss": 0.8160513043403625,
|
| 550 |
+
"eval_runtime": 201.474,
|
| 551 |
+
"eval_samples_per_second": 24.316,
|
| 552 |
+
"eval_steps_per_second": 0.764,
|
| 553 |
+
"eval_wer": 0.3833324322863167,
|
| 554 |
+
"step": 5400
|
| 555 |
+
},
|
| 556 |
+
{
|
| 557 |
+
"epoch": 23.91,
|
| 558 |
+
"learning_rate": 6.576562499999999e-05,
|
| 559 |
+
"loss": 0.2615,
|
| 560 |
+
"step": 5500
|
| 561 |
+
},
|
| 562 |
+
{
|
| 563 |
+
"epoch": 23.91,
|
| 564 |
+
"eval_loss": 0.7822675704956055,
|
| 565 |
+
"eval_runtime": 200.6337,
|
| 566 |
+
"eval_samples_per_second": 24.418,
|
| 567 |
+
"eval_steps_per_second": 0.768,
|
| 568 |
+
"eval_wer": 0.3857922906417257,
|
| 569 |
+
"step": 5500
|
| 570 |
+
},
|
| 571 |
+
{
|
| 572 |
+
"epoch": 24.35,
|
| 573 |
+
"eval_loss": 0.8311657309532166,
|
| 574 |
+
"eval_runtime": 201.9809,
|
| 575 |
+
"eval_samples_per_second": 24.255,
|
| 576 |
+
"eval_steps_per_second": 0.762,
|
| 577 |
+
"eval_wer": 0.3863329188517057,
|
| 578 |
+
"step": 5600
|
| 579 |
+
},
|
| 580 |
+
{
|
| 581 |
+
"epoch": 24.78,
|
| 582 |
+
"eval_loss": 0.8427298069000244,
|
| 583 |
+
"eval_runtime": 201.3389,
|
| 584 |
+
"eval_samples_per_second": 24.332,
|
| 585 |
+
"eval_steps_per_second": 0.765,
|
| 586 |
+
"eval_wer": 0.3818997675298697,
|
| 587 |
+
"step": 5700
|
| 588 |
+
},
|
| 589 |
+
{
|
| 590 |
+
"epoch": 25.22,
|
| 591 |
+
"eval_loss": 0.8432016968727112,
|
| 592 |
+
"eval_runtime": 205.0165,
|
| 593 |
+
"eval_samples_per_second": 23.896,
|
| 594 |
+
"eval_steps_per_second": 0.751,
|
| 595 |
+
"eval_wer": 0.3801697572579337,
|
| 596 |
+
"step": 5800
|
| 597 |
+
},
|
| 598 |
+
{
|
| 599 |
+
"epoch": 25.65,
|
| 600 |
+
"eval_loss": 0.8285790085792542,
|
| 601 |
+
"eval_runtime": 200.9997,
|
| 602 |
+
"eval_samples_per_second": 24.373,
|
| 603 |
+
"eval_steps_per_second": 0.766,
|
| 604 |
+
"eval_wer": 0.37943990917446074,
|
| 605 |
+
"step": 5900
|
| 606 |
+
},
|
| 607 |
+
{
|
| 608 |
+
"epoch": 26.09,
|
| 609 |
+
"learning_rate": 4.2328124999999994e-05,
|
| 610 |
+
"loss": 0.2408,
|
| 611 |
+
"step": 6000
|
| 612 |
+
},
|
| 613 |
+
{
|
| 614 |
+
"epoch": 26.09,
|
| 615 |
+
"eval_loss": 0.8224411606788635,
|
| 616 |
+
"eval_runtime": 200.5192,
|
| 617 |
+
"eval_samples_per_second": 24.432,
|
| 618 |
+
"eval_steps_per_second": 0.768,
|
| 619 |
+
"eval_wer": 0.3824403957398497,
|
| 620 |
+
"step": 6000
|
| 621 |
+
},
|
| 622 |
+
{
|
| 623 |
+
"epoch": 26.52,
|
| 624 |
+
"eval_loss": 0.8228213787078857,
|
| 625 |
+
"eval_runtime": 201.4468,
|
| 626 |
+
"eval_samples_per_second": 24.319,
|
| 627 |
+
"eval_steps_per_second": 0.764,
|
| 628 |
+
"eval_wer": 0.3823052386873547,
|
| 629 |
+
"step": 6100
|
| 630 |
+
},
|
| 631 |
+
{
|
| 632 |
+
"epoch": 26.95,
|
| 633 |
+
"eval_loss": 0.832374095916748,
|
| 634 |
+
"eval_runtime": 200.428,
|
| 635 |
+
"eval_samples_per_second": 24.443,
|
| 636 |
+
"eval_steps_per_second": 0.768,
|
| 637 |
+
"eval_wer": 0.3794669405849597,
|
| 638 |
+
"step": 6200
|
| 639 |
+
},
|
| 640 |
+
{
|
| 641 |
+
"epoch": 27.39,
|
| 642 |
+
"eval_loss": 0.8564098477363586,
|
| 643 |
+
"eval_runtime": 199.9963,
|
| 644 |
+
"eval_samples_per_second": 24.495,
|
| 645 |
+
"eval_steps_per_second": 0.77,
|
| 646 |
+
"eval_wer": 0.37441206682164674,
|
| 647 |
+
"step": 6300
|
| 648 |
+
},
|
| 649 |
+
{
|
| 650 |
+
"epoch": 27.82,
|
| 651 |
+
"eval_loss": 0.8629336953163147,
|
| 652 |
+
"eval_runtime": 200.1836,
|
| 653 |
+
"eval_samples_per_second": 24.473,
|
| 654 |
+
"eval_steps_per_second": 0.769,
|
| 655 |
+
"eval_wer": 0.37738552197653674,
|
| 656 |
+
"step": 6400
|
| 657 |
+
},
|
| 658 |
+
{
|
| 659 |
+
"epoch": 28.26,
|
| 660 |
+
"learning_rate": 1.8890625e-05,
|
| 661 |
+
"loss": 0.2254,
|
| 662 |
+
"step": 6500
|
| 663 |
+
},
|
| 664 |
+
{
|
| 665 |
+
"epoch": 28.26,
|
| 666 |
+
"eval_loss": 0.8544773459434509,
|
| 667 |
+
"eval_runtime": 200.1232,
|
| 668 |
+
"eval_samples_per_second": 24.48,
|
| 669 |
+
"eval_steps_per_second": 0.77,
|
| 670 |
+
"eval_wer": 0.3778450559550197,
|
| 671 |
+
"step": 6500
|
| 672 |
+
},
|
| 673 |
+
{
|
| 674 |
+
"epoch": 28.69,
|
| 675 |
+
"eval_loss": 0.8491827845573425,
|
| 676 |
+
"eval_runtime": 201.2475,
|
| 677 |
+
"eval_samples_per_second": 24.343,
|
| 678 |
+
"eval_steps_per_second": 0.765,
|
| 679 |
+
"eval_wer": 0.3767097367140617,
|
| 680 |
+
"step": 6600
|
| 681 |
+
},
|
| 682 |
+
{
|
| 683 |
+
"epoch": 29.13,
|
| 684 |
+
"eval_loss": 0.8510707020759583,
|
| 685 |
+
"eval_runtime": 203.9748,
|
| 686 |
+
"eval_samples_per_second": 24.018,
|
| 687 |
+
"eval_steps_per_second": 0.755,
|
| 688 |
+
"eval_wer": 0.37508785208412176,
|
| 689 |
+
"step": 6700
|
| 690 |
+
},
|
| 691 |
+
{
|
| 692 |
+
"epoch": 29.56,
|
| 693 |
+
"eval_loss": 0.8490829467773438,
|
| 694 |
+
"eval_runtime": 200.922,
|
| 695 |
+
"eval_samples_per_second": 24.383,
|
| 696 |
+
"eval_steps_per_second": 0.766,
|
| 697 |
+
"eval_wer": 0.3752770719576147,
|
| 698 |
+
"step": 6800
|
| 699 |
+
},
|
| 700 |
+
{
|
| 701 |
+
"epoch": 30.0,
|
| 702 |
+
"eval_loss": 0.8535701632499695,
|
| 703 |
+
"eval_runtime": 201.4279,
|
| 704 |
+
"eval_samples_per_second": 24.321,
|
| 705 |
+
"eval_steps_per_second": 0.765,
|
| 706 |
+
"eval_wer": 0.3736551873276748,
|
| 707 |
+
"step": 6900
|
| 708 |
+
},
|
| 709 |
+
{
|
| 710 |
+
"epoch": 30.0,
|
| 711 |
+
"step": 6900,
|
| 712 |
+
"total_flos": 5.975239123849969e+19,
|
| 713 |
+
"train_loss": 0.15137778351272363,
|
| 714 |
+
"train_runtime": 18309.2993,
|
| 715 |
+
"train_samples_per_second": 24.135,
|
| 716 |
+
"train_steps_per_second": 0.377
|
| 717 |
}
|
| 718 |
],
|
| 719 |
+
"max_steps": 6900,
|
| 720 |
+
"num_train_epochs": 30,
|
| 721 |
+
"total_flos": 5.975239123849969e+19,
|
| 722 |
"trial_name": null,
|
| 723 |
"trial_params": null
|
| 724 |
}
|