Training in progress, step 100, checkpoint
Browse files
last-checkpoint/adapter_model.safetensors
CHANGED
|
@@ -1,3 +1,3 @@
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
-
oid sha256:
|
| 3 |
size 3653851880
|
|
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:5c9b2364be7f3c84d373b7ef1118db06fe33dc81a615f30a3fa612077f33a019
|
| 3 |
size 3653851880
|
last-checkpoint/optimizer.pt
CHANGED
|
@@ -1,3 +1,3 @@
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
-
oid sha256:
|
| 3 |
size 1856183844
|
|
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:fe079ab42c71d5ee681ef764bdaa65934d9b6485a23efa69d07b25da6a845155
|
| 3 |
size 1856183844
|
last-checkpoint/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 sha256:3a60c7d771c1fd156acee762fba03c724cb41829a3f71df370ecd1d20b134982
|
| 3 |
size 1064
|
last-checkpoint/trainer_state.json
CHANGED
|
@@ -1,9 +1,9 @@
|
|
| 1 |
{
|
| 2 |
"best_metric": null,
|
| 3 |
"best_model_checkpoint": null,
|
| 4 |
-
"epoch": 0.
|
| 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,
|
|
@@ -357,6 +357,356 @@
|
|
| 357 |
"learning_rate": 0.00011736481776669306,
|
| 358 |
"loss": 0.0993,
|
| 359 |
"step": 50
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 360 |
}
|
| 361 |
],
|
| 362 |
"logging_steps": 1,
|
|
@@ -371,12 +721,12 @@
|
|
| 371 |
"should_evaluate": false,
|
| 372 |
"should_log": false,
|
| 373 |
"should_save": true,
|
| 374 |
-
"should_training_stop":
|
| 375 |
},
|
| 376 |
"attributes": {}
|
| 377 |
}
|
| 378 |
},
|
| 379 |
-
"total_flos":
|
| 380 |
"train_batch_size": 4,
|
| 381 |
"trial_name": null,
|
| 382 |
"trial_params": null
|
|
|
|
| 1 |
{
|
| 2 |
"best_metric": null,
|
| 3 |
"best_model_checkpoint": null,
|
| 4 |
+
"epoch": 0.0379066075955365,
|
| 5 |
"eval_steps": 500,
|
| 6 |
+
"global_step": 100,
|
| 7 |
"is_hyper_param_search": false,
|
| 8 |
"is_local_process_zero": true,
|
| 9 |
"is_world_process_zero": true,
|
|
|
|
| 357 |
"learning_rate": 0.00011736481776669306,
|
| 358 |
"loss": 0.0993,
|
| 359 |
"step": 50
|
| 360 |
+
},
|
| 361 |
+
{
|
| 362 |
+
"epoch": 0.019332369873723613,
|
| 363 |
+
"grad_norm": 0.06991136819124222,
|
| 364 |
+
"learning_rate": 0.00011391731009600654,
|
| 365 |
+
"loss": 0.0873,
|
| 366 |
+
"step": 51
|
| 367 |
+
},
|
| 368 |
+
{
|
| 369 |
+
"epoch": 0.019711435949678977,
|
| 370 |
+
"grad_norm": 0.06695719808340073,
|
| 371 |
+
"learning_rate": 0.00011045284632676536,
|
| 372 |
+
"loss": 0.0876,
|
| 373 |
+
"step": 52
|
| 374 |
+
},
|
| 375 |
+
{
|
| 376 |
+
"epoch": 0.020090502025634345,
|
| 377 |
+
"grad_norm": 0.07722793519496918,
|
| 378 |
+
"learning_rate": 0.00010697564737441252,
|
| 379 |
+
"loss": 0.0977,
|
| 380 |
+
"step": 53
|
| 381 |
+
},
|
| 382 |
+
{
|
| 383 |
+
"epoch": 0.02046956810158971,
|
| 384 |
+
"grad_norm": 0.06521150469779968,
|
| 385 |
+
"learning_rate": 0.00010348994967025012,
|
| 386 |
+
"loss": 0.0856,
|
| 387 |
+
"step": 54
|
| 388 |
+
},
|
| 389 |
+
{
|
| 390 |
+
"epoch": 0.020848634177545074,
|
| 391 |
+
"grad_norm": 0.07508374005556107,
|
| 392 |
+
"learning_rate": 0.0001,
|
| 393 |
+
"loss": 0.0876,
|
| 394 |
+
"step": 55
|
| 395 |
+
},
|
| 396 |
+
{
|
| 397 |
+
"epoch": 0.021227700253500438,
|
| 398 |
+
"grad_norm": 0.05650763958692551,
|
| 399 |
+
"learning_rate": 9.651005032974994e-05,
|
| 400 |
+
"loss": 0.0837,
|
| 401 |
+
"step": 56
|
| 402 |
+
},
|
| 403 |
+
{
|
| 404 |
+
"epoch": 0.021606766329455802,
|
| 405 |
+
"grad_norm": 0.07390966266393661,
|
| 406 |
+
"learning_rate": 9.302435262558747e-05,
|
| 407 |
+
"loss": 0.0879,
|
| 408 |
+
"step": 57
|
| 409 |
+
},
|
| 410 |
+
{
|
| 411 |
+
"epoch": 0.021985832405411167,
|
| 412 |
+
"grad_norm": 0.058437567204236984,
|
| 413 |
+
"learning_rate": 8.954715367323468e-05,
|
| 414 |
+
"loss": 0.0773,
|
| 415 |
+
"step": 58
|
| 416 |
+
},
|
| 417 |
+
{
|
| 418 |
+
"epoch": 0.022364898481366535,
|
| 419 |
+
"grad_norm": 0.06248459964990616,
|
| 420 |
+
"learning_rate": 8.608268990399349e-05,
|
| 421 |
+
"loss": 0.0788,
|
| 422 |
+
"step": 59
|
| 423 |
+
},
|
| 424 |
+
{
|
| 425 |
+
"epoch": 0.0227439645573219,
|
| 426 |
+
"grad_norm": 0.0653211697936058,
|
| 427 |
+
"learning_rate": 8.263518223330697e-05,
|
| 428 |
+
"loss": 0.0855,
|
| 429 |
+
"step": 60
|
| 430 |
+
},
|
| 431 |
+
{
|
| 432 |
+
"epoch": 0.023123030633277263,
|
| 433 |
+
"grad_norm": 0.062243081629276276,
|
| 434 |
+
"learning_rate": 7.920883091822408e-05,
|
| 435 |
+
"loss": 0.0886,
|
| 436 |
+
"step": 61
|
| 437 |
+
},
|
| 438 |
+
{
|
| 439 |
+
"epoch": 0.023502096709232628,
|
| 440 |
+
"grad_norm": 0.055676043033599854,
|
| 441 |
+
"learning_rate": 7.580781044003324e-05,
|
| 442 |
+
"loss": 0.0897,
|
| 443 |
+
"step": 62
|
| 444 |
+
},
|
| 445 |
+
{
|
| 446 |
+
"epoch": 0.023881162785187992,
|
| 447 |
+
"grad_norm": 0.05823361128568649,
|
| 448 |
+
"learning_rate": 7.243626441830009e-05,
|
| 449 |
+
"loss": 0.0753,
|
| 450 |
+
"step": 63
|
| 451 |
+
},
|
| 452 |
+
{
|
| 453 |
+
"epoch": 0.02426022886114336,
|
| 454 |
+
"grad_norm": 0.047967329621315,
|
| 455 |
+
"learning_rate": 6.909830056250527e-05,
|
| 456 |
+
"loss": 0.0804,
|
| 457 |
+
"step": 64
|
| 458 |
+
},
|
| 459 |
+
{
|
| 460 |
+
"epoch": 0.024639294937098724,
|
| 461 |
+
"grad_norm": 0.04651477187871933,
|
| 462 |
+
"learning_rate": 6.579798566743314e-05,
|
| 463 |
+
"loss": 0.0737,
|
| 464 |
+
"step": 65
|
| 465 |
+
},
|
| 466 |
+
{
|
| 467 |
+
"epoch": 0.02501836101305409,
|
| 468 |
+
"grad_norm": 0.06440019607543945,
|
| 469 |
+
"learning_rate": 6.25393406584088e-05,
|
| 470 |
+
"loss": 0.0931,
|
| 471 |
+
"step": 66
|
| 472 |
+
},
|
| 473 |
+
{
|
| 474 |
+
"epoch": 0.025397427089009453,
|
| 475 |
+
"grad_norm": 0.0673041045665741,
|
| 476 |
+
"learning_rate": 5.9326335692419995e-05,
|
| 477 |
+
"loss": 0.0926,
|
| 478 |
+
"step": 67
|
| 479 |
+
},
|
| 480 |
+
{
|
| 481 |
+
"epoch": 0.025776493164964817,
|
| 482 |
+
"grad_norm": 0.06460334360599518,
|
| 483 |
+
"learning_rate": 5.616288532109225e-05,
|
| 484 |
+
"loss": 0.1055,
|
| 485 |
+
"step": 68
|
| 486 |
+
},
|
| 487 |
+
{
|
| 488 |
+
"epoch": 0.02615555924092018,
|
| 489 |
+
"grad_norm": 0.06001191958785057,
|
| 490 |
+
"learning_rate": 5.305284372141095e-05,
|
| 491 |
+
"loss": 0.0815,
|
| 492 |
+
"step": 69
|
| 493 |
+
},
|
| 494 |
+
{
|
| 495 |
+
"epoch": 0.02653462531687555,
|
| 496 |
+
"grad_norm": 0.05296599864959717,
|
| 497 |
+
"learning_rate": 5.000000000000002e-05,
|
| 498 |
+
"loss": 0.0837,
|
| 499 |
+
"step": 70
|
| 500 |
+
},
|
| 501 |
+
{
|
| 502 |
+
"epoch": 0.026913691392830914,
|
| 503 |
+
"grad_norm": 0.0675460696220398,
|
| 504 |
+
"learning_rate": 4.700807357667952e-05,
|
| 505 |
+
"loss": 0.0848,
|
| 506 |
+
"step": 71
|
| 507 |
+
},
|
| 508 |
+
{
|
| 509 |
+
"epoch": 0.027292757468786278,
|
| 510 |
+
"grad_norm": 0.05409557744860649,
|
| 511 |
+
"learning_rate": 4.4080709652925336e-05,
|
| 512 |
+
"loss": 0.0779,
|
| 513 |
+
"step": 72
|
| 514 |
+
},
|
| 515 |
+
{
|
| 516 |
+
"epoch": 0.027671823544741642,
|
| 517 |
+
"grad_norm": 0.05015251785516739,
|
| 518 |
+
"learning_rate": 4.12214747707527e-05,
|
| 519 |
+
"loss": 0.0788,
|
| 520 |
+
"step": 73
|
| 521 |
+
},
|
| 522 |
+
{
|
| 523 |
+
"epoch": 0.028050889620697007,
|
| 524 |
+
"grad_norm": 0.05301973596215248,
|
| 525 |
+
"learning_rate": 3.843385246743417e-05,
|
| 526 |
+
"loss": 0.0764,
|
| 527 |
+
"step": 74
|
| 528 |
+
},
|
| 529 |
+
{
|
| 530 |
+
"epoch": 0.02842995569665237,
|
| 531 |
+
"grad_norm": 0.05640785023570061,
|
| 532 |
+
"learning_rate": 3.5721239031346066e-05,
|
| 533 |
+
"loss": 0.0867,
|
| 534 |
+
"step": 75
|
| 535 |
+
},
|
| 536 |
+
{
|
| 537 |
+
"epoch": 0.02880902177260774,
|
| 538 |
+
"grad_norm": 0.055719390511512756,
|
| 539 |
+
"learning_rate": 3.308693936411421e-05,
|
| 540 |
+
"loss": 0.0788,
|
| 541 |
+
"step": 76
|
| 542 |
+
},
|
| 543 |
+
{
|
| 544 |
+
"epoch": 0.029188087848563103,
|
| 545 |
+
"grad_norm": 0.06163398548960686,
|
| 546 |
+
"learning_rate": 3.053416295410026e-05,
|
| 547 |
+
"loss": 0.0807,
|
| 548 |
+
"step": 77
|
| 549 |
+
},
|
| 550 |
+
{
|
| 551 |
+
"epoch": 0.029567153924518468,
|
| 552 |
+
"grad_norm": 0.06289924681186676,
|
| 553 |
+
"learning_rate": 2.8066019966134904e-05,
|
| 554 |
+
"loss": 0.0808,
|
| 555 |
+
"step": 78
|
| 556 |
+
},
|
| 557 |
+
{
|
| 558 |
+
"epoch": 0.029946220000473832,
|
| 559 |
+
"grad_norm": 0.056900832802057266,
|
| 560 |
+
"learning_rate": 2.5685517452260567e-05,
|
| 561 |
+
"loss": 0.0762,
|
| 562 |
+
"step": 79
|
| 563 |
+
},
|
| 564 |
+
{
|
| 565 |
+
"epoch": 0.030325286076429196,
|
| 566 |
+
"grad_norm": 0.06545019149780273,
|
| 567 |
+
"learning_rate": 2.339555568810221e-05,
|
| 568 |
+
"loss": 0.0929,
|
| 569 |
+
"step": 80
|
| 570 |
+
},
|
| 571 |
+
{
|
| 572 |
+
"epoch": 0.030704352152384564,
|
| 573 |
+
"grad_norm": 0.060016706585884094,
|
| 574 |
+
"learning_rate": 2.119892463932781e-05,
|
| 575 |
+
"loss": 0.0752,
|
| 576 |
+
"step": 81
|
| 577 |
+
},
|
| 578 |
+
{
|
| 579 |
+
"epoch": 0.03108341822833993,
|
| 580 |
+
"grad_norm": 0.055878885090351105,
|
| 581 |
+
"learning_rate": 1.9098300562505266e-05,
|
| 582 |
+
"loss": 0.0753,
|
| 583 |
+
"step": 82
|
| 584 |
+
},
|
| 585 |
+
{
|
| 586 |
+
"epoch": 0.03146248430429529,
|
| 587 |
+
"grad_norm": 0.06015196815133095,
|
| 588 |
+
"learning_rate": 1.7096242744495837e-05,
|
| 589 |
+
"loss": 0.086,
|
| 590 |
+
"step": 83
|
| 591 |
+
},
|
| 592 |
+
{
|
| 593 |
+
"epoch": 0.03184155038025066,
|
| 594 |
+
"grad_norm": 0.05178290605545044,
|
| 595 |
+
"learning_rate": 1.5195190384357404e-05,
|
| 596 |
+
"loss": 0.0793,
|
| 597 |
+
"step": 84
|
| 598 |
+
},
|
| 599 |
+
{
|
| 600 |
+
"epoch": 0.032220616456206025,
|
| 601 |
+
"grad_norm": 0.05161258578300476,
|
| 602 |
+
"learning_rate": 1.339745962155613e-05,
|
| 603 |
+
"loss": 0.0809,
|
| 604 |
+
"step": 85
|
| 605 |
+
},
|
| 606 |
+
{
|
| 607 |
+
"epoch": 0.03259968253216139,
|
| 608 |
+
"grad_norm": 0.05504770204424858,
|
| 609 |
+
"learning_rate": 1.1705240714107302e-05,
|
| 610 |
+
"loss": 0.0772,
|
| 611 |
+
"step": 86
|
| 612 |
+
},
|
| 613 |
+
{
|
| 614 |
+
"epoch": 0.032978748608116754,
|
| 615 |
+
"grad_norm": 0.05566380172967911,
|
| 616 |
+
"learning_rate": 1.0120595370083318e-05,
|
| 617 |
+
"loss": 0.0726,
|
| 618 |
+
"step": 87
|
| 619 |
+
},
|
| 620 |
+
{
|
| 621 |
+
"epoch": 0.03335781468407212,
|
| 622 |
+
"grad_norm": 0.05389956384897232,
|
| 623 |
+
"learning_rate": 8.645454235739903e-06,
|
| 624 |
+
"loss": 0.0779,
|
| 625 |
+
"step": 88
|
| 626 |
+
},
|
| 627 |
+
{
|
| 628 |
+
"epoch": 0.03373688076002748,
|
| 629 |
+
"grad_norm": 0.05445749685168266,
|
| 630 |
+
"learning_rate": 7.281614543321269e-06,
|
| 631 |
+
"loss": 0.076,
|
| 632 |
+
"step": 89
|
| 633 |
+
},
|
| 634 |
+
{
|
| 635 |
+
"epoch": 0.03411594683598285,
|
| 636 |
+
"grad_norm": 0.04557236656546593,
|
| 637 |
+
"learning_rate": 6.030737921409169e-06,
|
| 638 |
+
"loss": 0.075,
|
| 639 |
+
"step": 90
|
| 640 |
+
},
|
| 641 |
+
{
|
| 642 |
+
"epoch": 0.03449501291193821,
|
| 643 |
+
"grad_norm": 0.050466809421777725,
|
| 644 |
+
"learning_rate": 4.8943483704846475e-06,
|
| 645 |
+
"loss": 0.0747,
|
| 646 |
+
"step": 91
|
| 647 |
+
},
|
| 648 |
+
{
|
| 649 |
+
"epoch": 0.034874078987893575,
|
| 650 |
+
"grad_norm": 0.04975885897874832,
|
| 651 |
+
"learning_rate": 3.873830406168111e-06,
|
| 652 |
+
"loss": 0.0746,
|
| 653 |
+
"step": 92
|
| 654 |
+
},
|
| 655 |
+
{
|
| 656 |
+
"epoch": 0.03525314506384894,
|
| 657 |
+
"grad_norm": 0.055442556738853455,
|
| 658 |
+
"learning_rate": 2.970427372400353e-06,
|
| 659 |
+
"loss": 0.0807,
|
| 660 |
+
"step": 93
|
| 661 |
+
},
|
| 662 |
+
{
|
| 663 |
+
"epoch": 0.035632211139804304,
|
| 664 |
+
"grad_norm": 0.05043969675898552,
|
| 665 |
+
"learning_rate": 2.1852399266194314e-06,
|
| 666 |
+
"loss": 0.0741,
|
| 667 |
+
"step": 94
|
| 668 |
+
},
|
| 669 |
+
{
|
| 670 |
+
"epoch": 0.036011277215759675,
|
| 671 |
+
"grad_norm": 0.05411124229431152,
|
| 672 |
+
"learning_rate": 1.5192246987791981e-06,
|
| 673 |
+
"loss": 0.0741,
|
| 674 |
+
"step": 95
|
| 675 |
+
},
|
| 676 |
+
{
|
| 677 |
+
"epoch": 0.03639034329171504,
|
| 678 |
+
"grad_norm": 0.06104011833667755,
|
| 679 |
+
"learning_rate": 9.731931258429638e-07,
|
| 680 |
+
"loss": 0.0822,
|
| 681 |
+
"step": 96
|
| 682 |
+
},
|
| 683 |
+
{
|
| 684 |
+
"epoch": 0.036769409367670404,
|
| 685 |
+
"grad_norm": 0.05103430524468422,
|
| 686 |
+
"learning_rate": 5.478104631726711e-07,
|
| 687 |
+
"loss": 0.0716,
|
| 688 |
+
"step": 97
|
| 689 |
+
},
|
| 690 |
+
{
|
| 691 |
+
"epoch": 0.03714847544362577,
|
| 692 |
+
"grad_norm": 0.057701822370290756,
|
| 693 |
+
"learning_rate": 2.4359497401758024e-07,
|
| 694 |
+
"loss": 0.0767,
|
| 695 |
+
"step": 98
|
| 696 |
+
},
|
| 697 |
+
{
|
| 698 |
+
"epoch": 0.03752754151958113,
|
| 699 |
+
"grad_norm": 0.05738105624914169,
|
| 700 |
+
"learning_rate": 6.09172980904238e-08,
|
| 701 |
+
"loss": 0.0837,
|
| 702 |
+
"step": 99
|
| 703 |
+
},
|
| 704 |
+
{
|
| 705 |
+
"epoch": 0.0379066075955365,
|
| 706 |
+
"grad_norm": 0.05406734347343445,
|
| 707 |
+
"learning_rate": 0.0,
|
| 708 |
+
"loss": 0.0856,
|
| 709 |
+
"step": 100
|
| 710 |
}
|
| 711 |
],
|
| 712 |
"logging_steps": 1,
|
|
|
|
| 721 |
"should_evaluate": false,
|
| 722 |
"should_log": false,
|
| 723 |
"should_save": true,
|
| 724 |
+
"should_training_stop": true
|
| 725 |
},
|
| 726 |
"attributes": {}
|
| 727 |
}
|
| 728 |
},
|
| 729 |
+
"total_flos": 3.888051858648392e+17,
|
| 730 |
"train_batch_size": 4,
|
| 731 |
"trial_name": null,
|
| 732 |
"trial_params": null
|