Training in progress, step 100, checkpoint
Browse files
last-checkpoint/optimizer.pt
CHANGED
|
@@ -1,3 +1,3 @@
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
-
oid sha256:
|
| 3 |
size 150486964
|
|
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:2606d7119ba7935995826adc5eb32cd916f154b702f42a138d66a4eae0e01a98
|
| 3 |
size 150486964
|
last-checkpoint/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:ccaa67892548a17ef65aa58dc258dcd4f2fc3b750b343136a710016c6c67c07b
|
| 3 |
size 14244
|
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 sha256:2cdd5e1998f1139b3d2606fc1207245b65164fdf9041e6fa58bd90acc8d38fa5
|
| 3 |
size 1064
|
last-checkpoint/trainer_state.json
CHANGED
|
@@ -1,9 +1,9 @@
|
|
| 1 |
{
|
| 2 |
"best_metric": NaN,
|
| 3 |
"best_model_checkpoint": "miner_id_24/checkpoint-50",
|
| 4 |
-
"epoch": 0.
|
| 5 |
"eval_steps": 50,
|
| 6 |
-
"global_step":
|
| 7 |
"is_hyper_param_search": false,
|
| 8 |
"is_local_process_zero": true,
|
| 9 |
"is_world_process_zero": true,
|
|
@@ -373,6 +373,364 @@
|
|
| 373 |
"eval_samples_per_second": 15.39,
|
| 374 |
"eval_steps_per_second": 15.39,
|
| 375 |
"step": 50
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 376 |
}
|
| 377 |
],
|
| 378 |
"logging_steps": 1,
|
|
@@ -387,7 +745,7 @@
|
|
| 387 |
"early_stopping_threshold": 0.0
|
| 388 |
},
|
| 389 |
"attributes": {
|
| 390 |
-
"early_stopping_patience_counter":
|
| 391 |
}
|
| 392 |
},
|
| 393 |
"TrainerControl": {
|
|
@@ -396,12 +754,12 @@
|
|
| 396 |
"should_evaluate": false,
|
| 397 |
"should_log": false,
|
| 398 |
"should_save": true,
|
| 399 |
-
"should_training_stop":
|
| 400 |
},
|
| 401 |
"attributes": {}
|
| 402 |
}
|
| 403 |
},
|
| 404 |
-
"total_flos":
|
| 405 |
"train_batch_size": 1,
|
| 406 |
"trial_name": null,
|
| 407 |
"trial_params": null
|
|
|
|
| 1 |
{
|
| 2 |
"best_metric": NaN,
|
| 3 |
"best_model_checkpoint": "miner_id_24/checkpoint-50",
|
| 4 |
+
"epoch": 0.21499596882558453,
|
| 5 |
"eval_steps": 50,
|
| 6 |
+
"global_step": 100,
|
| 7 |
"is_hyper_param_search": false,
|
| 8 |
"is_local_process_zero": true,
|
| 9 |
"is_world_process_zero": true,
|
|
|
|
| 373 |
"eval_samples_per_second": 15.39,
|
| 374 |
"eval_steps_per_second": 15.39,
|
| 375 |
"step": 50
|
| 376 |
+
},
|
| 377 |
+
{
|
| 378 |
+
"epoch": 0.10964794410104811,
|
| 379 |
+
"grad_norm": NaN,
|
| 380 |
+
"learning_rate": 7.65e-06,
|
| 381 |
+
"loss": 0.0,
|
| 382 |
+
"step": 51
|
| 383 |
+
},
|
| 384 |
+
{
|
| 385 |
+
"epoch": 0.11179790378930395,
|
| 386 |
+
"grad_norm": NaN,
|
| 387 |
+
"learning_rate": 7.8e-06,
|
| 388 |
+
"loss": 0.0,
|
| 389 |
+
"step": 52
|
| 390 |
+
},
|
| 391 |
+
{
|
| 392 |
+
"epoch": 0.1139478634775598,
|
| 393 |
+
"grad_norm": NaN,
|
| 394 |
+
"learning_rate": 7.95e-06,
|
| 395 |
+
"loss": 0.0,
|
| 396 |
+
"step": 53
|
| 397 |
+
},
|
| 398 |
+
{
|
| 399 |
+
"epoch": 0.11609782316581564,
|
| 400 |
+
"grad_norm": NaN,
|
| 401 |
+
"learning_rate": 8.1e-06,
|
| 402 |
+
"loss": 0.0,
|
| 403 |
+
"step": 54
|
| 404 |
+
},
|
| 405 |
+
{
|
| 406 |
+
"epoch": 0.11824778285407149,
|
| 407 |
+
"grad_norm": NaN,
|
| 408 |
+
"learning_rate": 8.25e-06,
|
| 409 |
+
"loss": 0.0,
|
| 410 |
+
"step": 55
|
| 411 |
+
},
|
| 412 |
+
{
|
| 413 |
+
"epoch": 0.12039774254232734,
|
| 414 |
+
"grad_norm": NaN,
|
| 415 |
+
"learning_rate": 8.400000000000001e-06,
|
| 416 |
+
"loss": 0.0,
|
| 417 |
+
"step": 56
|
| 418 |
+
},
|
| 419 |
+
{
|
| 420 |
+
"epoch": 0.12254770223058317,
|
| 421 |
+
"grad_norm": NaN,
|
| 422 |
+
"learning_rate": 8.55e-06,
|
| 423 |
+
"loss": 0.0,
|
| 424 |
+
"step": 57
|
| 425 |
+
},
|
| 426 |
+
{
|
| 427 |
+
"epoch": 0.12469766191883902,
|
| 428 |
+
"grad_norm": NaN,
|
| 429 |
+
"learning_rate": 8.7e-06,
|
| 430 |
+
"loss": 0.0,
|
| 431 |
+
"step": 58
|
| 432 |
+
},
|
| 433 |
+
{
|
| 434 |
+
"epoch": 0.12684762160709487,
|
| 435 |
+
"grad_norm": NaN,
|
| 436 |
+
"learning_rate": 8.85e-06,
|
| 437 |
+
"loss": 0.0,
|
| 438 |
+
"step": 59
|
| 439 |
+
},
|
| 440 |
+
{
|
| 441 |
+
"epoch": 0.12899758129535072,
|
| 442 |
+
"grad_norm": NaN,
|
| 443 |
+
"learning_rate": 9e-06,
|
| 444 |
+
"loss": 0.0,
|
| 445 |
+
"step": 60
|
| 446 |
+
},
|
| 447 |
+
{
|
| 448 |
+
"epoch": 0.13114754098360656,
|
| 449 |
+
"grad_norm": NaN,
|
| 450 |
+
"learning_rate": 9.15e-06,
|
| 451 |
+
"loss": 0.0,
|
| 452 |
+
"step": 61
|
| 453 |
+
},
|
| 454 |
+
{
|
| 455 |
+
"epoch": 0.1332975006718624,
|
| 456 |
+
"grad_norm": NaN,
|
| 457 |
+
"learning_rate": 9.3e-06,
|
| 458 |
+
"loss": 0.0,
|
| 459 |
+
"step": 62
|
| 460 |
+
},
|
| 461 |
+
{
|
| 462 |
+
"epoch": 0.13544746036011826,
|
| 463 |
+
"grad_norm": NaN,
|
| 464 |
+
"learning_rate": 9.450000000000001e-06,
|
| 465 |
+
"loss": 0.0,
|
| 466 |
+
"step": 63
|
| 467 |
+
},
|
| 468 |
+
{
|
| 469 |
+
"epoch": 0.13759742004837408,
|
| 470 |
+
"grad_norm": NaN,
|
| 471 |
+
"learning_rate": 9.600000000000001e-06,
|
| 472 |
+
"loss": 0.0,
|
| 473 |
+
"step": 64
|
| 474 |
+
},
|
| 475 |
+
{
|
| 476 |
+
"epoch": 0.13974737973662993,
|
| 477 |
+
"grad_norm": NaN,
|
| 478 |
+
"learning_rate": 9.75e-06,
|
| 479 |
+
"loss": 0.0,
|
| 480 |
+
"step": 65
|
| 481 |
+
},
|
| 482 |
+
{
|
| 483 |
+
"epoch": 0.14189733942488578,
|
| 484 |
+
"grad_norm": NaN,
|
| 485 |
+
"learning_rate": 9.9e-06,
|
| 486 |
+
"loss": 0.0,
|
| 487 |
+
"step": 66
|
| 488 |
+
},
|
| 489 |
+
{
|
| 490 |
+
"epoch": 0.14404729911314162,
|
| 491 |
+
"grad_norm": NaN,
|
| 492 |
+
"learning_rate": 1.005e-05,
|
| 493 |
+
"loss": 0.0,
|
| 494 |
+
"step": 67
|
| 495 |
+
},
|
| 496 |
+
{
|
| 497 |
+
"epoch": 0.14619725880139747,
|
| 498 |
+
"grad_norm": NaN,
|
| 499 |
+
"learning_rate": 1.02e-05,
|
| 500 |
+
"loss": 0.0,
|
| 501 |
+
"step": 68
|
| 502 |
+
},
|
| 503 |
+
{
|
| 504 |
+
"epoch": 0.14834721848965332,
|
| 505 |
+
"grad_norm": NaN,
|
| 506 |
+
"learning_rate": 1.035e-05,
|
| 507 |
+
"loss": 0.0,
|
| 508 |
+
"step": 69
|
| 509 |
+
},
|
| 510 |
+
{
|
| 511 |
+
"epoch": 0.15049717817790917,
|
| 512 |
+
"grad_norm": NaN,
|
| 513 |
+
"learning_rate": 1.05e-05,
|
| 514 |
+
"loss": 0.0,
|
| 515 |
+
"step": 70
|
| 516 |
+
},
|
| 517 |
+
{
|
| 518 |
+
"epoch": 0.15264713786616502,
|
| 519 |
+
"grad_norm": NaN,
|
| 520 |
+
"learning_rate": 1.065e-05,
|
| 521 |
+
"loss": 0.0,
|
| 522 |
+
"step": 71
|
| 523 |
+
},
|
| 524 |
+
{
|
| 525 |
+
"epoch": 0.15479709755442086,
|
| 526 |
+
"grad_norm": NaN,
|
| 527 |
+
"learning_rate": 1.08e-05,
|
| 528 |
+
"loss": 0.0,
|
| 529 |
+
"step": 72
|
| 530 |
+
},
|
| 531 |
+
{
|
| 532 |
+
"epoch": 0.1569470572426767,
|
| 533 |
+
"grad_norm": NaN,
|
| 534 |
+
"learning_rate": 1.095e-05,
|
| 535 |
+
"loss": 0.0,
|
| 536 |
+
"step": 73
|
| 537 |
+
},
|
| 538 |
+
{
|
| 539 |
+
"epoch": 0.15909701693093253,
|
| 540 |
+
"grad_norm": NaN,
|
| 541 |
+
"learning_rate": 1.11e-05,
|
| 542 |
+
"loss": 0.0,
|
| 543 |
+
"step": 74
|
| 544 |
+
},
|
| 545 |
+
{
|
| 546 |
+
"epoch": 0.16124697661918838,
|
| 547 |
+
"grad_norm": NaN,
|
| 548 |
+
"learning_rate": 1.125e-05,
|
| 549 |
+
"loss": 0.0,
|
| 550 |
+
"step": 75
|
| 551 |
+
},
|
| 552 |
+
{
|
| 553 |
+
"epoch": 0.16339693630744423,
|
| 554 |
+
"grad_norm": NaN,
|
| 555 |
+
"learning_rate": 1.1400000000000001e-05,
|
| 556 |
+
"loss": 0.0,
|
| 557 |
+
"step": 76
|
| 558 |
+
},
|
| 559 |
+
{
|
| 560 |
+
"epoch": 0.16554689599570008,
|
| 561 |
+
"grad_norm": NaN,
|
| 562 |
+
"learning_rate": 1.1550000000000001e-05,
|
| 563 |
+
"loss": 0.0,
|
| 564 |
+
"step": 77
|
| 565 |
+
},
|
| 566 |
+
{
|
| 567 |
+
"epoch": 0.16769685568395593,
|
| 568 |
+
"grad_norm": NaN,
|
| 569 |
+
"learning_rate": 1.1700000000000001e-05,
|
| 570 |
+
"loss": 0.0,
|
| 571 |
+
"step": 78
|
| 572 |
+
},
|
| 573 |
+
{
|
| 574 |
+
"epoch": 0.16984681537221177,
|
| 575 |
+
"grad_norm": NaN,
|
| 576 |
+
"learning_rate": 1.185e-05,
|
| 577 |
+
"loss": 0.0,
|
| 578 |
+
"step": 79
|
| 579 |
+
},
|
| 580 |
+
{
|
| 581 |
+
"epoch": 0.17199677506046762,
|
| 582 |
+
"grad_norm": NaN,
|
| 583 |
+
"learning_rate": 1.2e-05,
|
| 584 |
+
"loss": 0.0,
|
| 585 |
+
"step": 80
|
| 586 |
+
},
|
| 587 |
+
{
|
| 588 |
+
"epoch": 0.17414673474872347,
|
| 589 |
+
"grad_norm": NaN,
|
| 590 |
+
"learning_rate": 1.215e-05,
|
| 591 |
+
"loss": 0.0,
|
| 592 |
+
"step": 81
|
| 593 |
+
},
|
| 594 |
+
{
|
| 595 |
+
"epoch": 0.17629669443697932,
|
| 596 |
+
"grad_norm": NaN,
|
| 597 |
+
"learning_rate": 1.2299999999999999e-05,
|
| 598 |
+
"loss": 0.0,
|
| 599 |
+
"step": 82
|
| 600 |
+
},
|
| 601 |
+
{
|
| 602 |
+
"epoch": 0.17844665412523517,
|
| 603 |
+
"grad_norm": NaN,
|
| 604 |
+
"learning_rate": 1.245e-05,
|
| 605 |
+
"loss": 0.0,
|
| 606 |
+
"step": 83
|
| 607 |
+
},
|
| 608 |
+
{
|
| 609 |
+
"epoch": 0.18059661381349099,
|
| 610 |
+
"grad_norm": NaN,
|
| 611 |
+
"learning_rate": 1.26e-05,
|
| 612 |
+
"loss": 0.0,
|
| 613 |
+
"step": 84
|
| 614 |
+
},
|
| 615 |
+
{
|
| 616 |
+
"epoch": 0.18274657350174683,
|
| 617 |
+
"grad_norm": NaN,
|
| 618 |
+
"learning_rate": 1.275e-05,
|
| 619 |
+
"loss": 0.0,
|
| 620 |
+
"step": 85
|
| 621 |
+
},
|
| 622 |
+
{
|
| 623 |
+
"epoch": 0.18489653319000268,
|
| 624 |
+
"grad_norm": NaN,
|
| 625 |
+
"learning_rate": 1.29e-05,
|
| 626 |
+
"loss": 0.0,
|
| 627 |
+
"step": 86
|
| 628 |
+
},
|
| 629 |
+
{
|
| 630 |
+
"epoch": 0.18704649287825853,
|
| 631 |
+
"grad_norm": NaN,
|
| 632 |
+
"learning_rate": 1.305e-05,
|
| 633 |
+
"loss": 0.0,
|
| 634 |
+
"step": 87
|
| 635 |
+
},
|
| 636 |
+
{
|
| 637 |
+
"epoch": 0.18919645256651438,
|
| 638 |
+
"grad_norm": NaN,
|
| 639 |
+
"learning_rate": 1.32e-05,
|
| 640 |
+
"loss": 0.0,
|
| 641 |
+
"step": 88
|
| 642 |
+
},
|
| 643 |
+
{
|
| 644 |
+
"epoch": 0.19134641225477023,
|
| 645 |
+
"grad_norm": NaN,
|
| 646 |
+
"learning_rate": 1.3350000000000001e-05,
|
| 647 |
+
"loss": 0.0,
|
| 648 |
+
"step": 89
|
| 649 |
+
},
|
| 650 |
+
{
|
| 651 |
+
"epoch": 0.19349637194302607,
|
| 652 |
+
"grad_norm": NaN,
|
| 653 |
+
"learning_rate": 1.3500000000000001e-05,
|
| 654 |
+
"loss": 0.0,
|
| 655 |
+
"step": 90
|
| 656 |
+
},
|
| 657 |
+
{
|
| 658 |
+
"epoch": 0.19564633163128192,
|
| 659 |
+
"grad_norm": NaN,
|
| 660 |
+
"learning_rate": 1.3650000000000001e-05,
|
| 661 |
+
"loss": 0.0,
|
| 662 |
+
"step": 91
|
| 663 |
+
},
|
| 664 |
+
{
|
| 665 |
+
"epoch": 0.19779629131953777,
|
| 666 |
+
"grad_norm": NaN,
|
| 667 |
+
"learning_rate": 1.3800000000000002e-05,
|
| 668 |
+
"loss": 0.0,
|
| 669 |
+
"step": 92
|
| 670 |
+
},
|
| 671 |
+
{
|
| 672 |
+
"epoch": 0.1999462510077936,
|
| 673 |
+
"grad_norm": NaN,
|
| 674 |
+
"learning_rate": 1.395e-05,
|
| 675 |
+
"loss": 0.0,
|
| 676 |
+
"step": 93
|
| 677 |
+
},
|
| 678 |
+
{
|
| 679 |
+
"epoch": 0.20209621069604944,
|
| 680 |
+
"grad_norm": NaN,
|
| 681 |
+
"learning_rate": 1.4099999999999999e-05,
|
| 682 |
+
"loss": 0.0,
|
| 683 |
+
"step": 94
|
| 684 |
+
},
|
| 685 |
+
{
|
| 686 |
+
"epoch": 0.2042461703843053,
|
| 687 |
+
"grad_norm": NaN,
|
| 688 |
+
"learning_rate": 1.4249999999999999e-05,
|
| 689 |
+
"loss": 0.0,
|
| 690 |
+
"step": 95
|
| 691 |
+
},
|
| 692 |
+
{
|
| 693 |
+
"epoch": 0.20639613007256113,
|
| 694 |
+
"grad_norm": NaN,
|
| 695 |
+
"learning_rate": 1.44e-05,
|
| 696 |
+
"loss": 0.0,
|
| 697 |
+
"step": 96
|
| 698 |
+
},
|
| 699 |
+
{
|
| 700 |
+
"epoch": 0.20854608976081698,
|
| 701 |
+
"grad_norm": NaN,
|
| 702 |
+
"learning_rate": 1.455e-05,
|
| 703 |
+
"loss": 0.0,
|
| 704 |
+
"step": 97
|
| 705 |
+
},
|
| 706 |
+
{
|
| 707 |
+
"epoch": 0.21069604944907283,
|
| 708 |
+
"grad_norm": NaN,
|
| 709 |
+
"learning_rate": 1.47e-05,
|
| 710 |
+
"loss": 0.0,
|
| 711 |
+
"step": 98
|
| 712 |
+
},
|
| 713 |
+
{
|
| 714 |
+
"epoch": 0.21284600913732868,
|
| 715 |
+
"grad_norm": NaN,
|
| 716 |
+
"learning_rate": 1.485e-05,
|
| 717 |
+
"loss": 0.0,
|
| 718 |
+
"step": 99
|
| 719 |
+
},
|
| 720 |
+
{
|
| 721 |
+
"epoch": 0.21499596882558453,
|
| 722 |
+
"grad_norm": NaN,
|
| 723 |
+
"learning_rate": 1.5e-05,
|
| 724 |
+
"loss": 0.0,
|
| 725 |
+
"step": 100
|
| 726 |
+
},
|
| 727 |
+
{
|
| 728 |
+
"epoch": 0.21499596882558453,
|
| 729 |
+
"eval_loss": NaN,
|
| 730 |
+
"eval_runtime": 25.4729,
|
| 731 |
+
"eval_samples_per_second": 15.389,
|
| 732 |
+
"eval_steps_per_second": 15.389,
|
| 733 |
+
"step": 100
|
| 734 |
}
|
| 735 |
],
|
| 736 |
"logging_steps": 1,
|
|
|
|
| 745 |
"early_stopping_threshold": 0.0
|
| 746 |
},
|
| 747 |
"attributes": {
|
| 748 |
+
"early_stopping_patience_counter": 1
|
| 749 |
}
|
| 750 |
},
|
| 751 |
"TrainerControl": {
|
|
|
|
| 754 |
"should_evaluate": false,
|
| 755 |
"should_log": false,
|
| 756 |
"should_save": true,
|
| 757 |
+
"should_training_stop": true
|
| 758 |
},
|
| 759 |
"attributes": {}
|
| 760 |
}
|
| 761 |
},
|
| 762 |
+
"total_flos": 3.17940289241088e+16,
|
| 763 |
"train_batch_size": 1,
|
| 764 |
"trial_name": null,
|
| 765 |
"trial_params": null
|