Training in progress, step 166, 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 216151256
|
|
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:f9e7aff42b36fe14e95ece06193160112474b8a29fc3680ce273c922ca5686f6
|
| 3 |
size 216151256
|
last-checkpoint/optimizer.pt
CHANGED
|
@@ -1,3 +1,3 @@
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
-
oid sha256:
|
| 3 |
size 432640054
|
|
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:2e21f976a284dd81c64396ec6b6206079943029f7c09ac486e503562b06e47e6
|
| 3 |
size 432640054
|
last-checkpoint/rng_state_0.pth
CHANGED
|
@@ -1,3 +1,3 @@
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
-
oid sha256:
|
| 3 |
size 14512
|
|
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:bb3607b5839cda7054779e8f957cbf2db3456879873cc4e34eac04cbf33f5db8
|
| 3 |
size 14512
|
last-checkpoint/rng_state_1.pth
CHANGED
|
@@ -1,3 +1,3 @@
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
-
oid sha256:
|
| 3 |
size 14512
|
|
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:fd635f6ad590a43a7a075b3fb4377adaa95cf2d835f115014607cf181d2b6449
|
| 3 |
size 14512
|
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:e3070c5337425657c2fec031251a5e4e8042c43dd7a5d4d7f77fa453b02282be
|
| 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": 83,
|
| 6 |
-
"global_step":
|
| 7 |
"is_hyper_param_search": false,
|
| 8 |
"is_local_process_zero": true,
|
| 9 |
"is_world_process_zero": true,
|
|
@@ -604,6 +604,595 @@
|
|
| 604 |
"eval_samples_per_second": 6.015,
|
| 605 |
"eval_steps_per_second": 1.506,
|
| 606 |
"step": 83
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 607 |
}
|
| 608 |
],
|
| 609 |
"logging_steps": 1,
|
|
@@ -623,7 +1212,7 @@
|
|
| 623 |
"attributes": {}
|
| 624 |
}
|
| 625 |
},
|
| 626 |
-
"total_flos":
|
| 627 |
"train_batch_size": 2,
|
| 628 |
"trial_name": null,
|
| 629 |
"trial_params": null
|
|
|
|
| 1 |
{
|
| 2 |
"best_metric": null,
|
| 3 |
"best_model_checkpoint": null,
|
| 4 |
+
"epoch": 0.5001883239171375,
|
| 5 |
"eval_steps": 83,
|
| 6 |
+
"global_step": 166,
|
| 7 |
"is_hyper_param_search": false,
|
| 8 |
"is_local_process_zero": true,
|
| 9 |
"is_world_process_zero": true,
|
|
|
|
| 604 |
"eval_samples_per_second": 6.015,
|
| 605 |
"eval_steps_per_second": 1.506,
|
| 606 |
"step": 83
|
| 607 |
+
},
|
| 608 |
+
{
|
| 609 |
+
"epoch": 0.25310734463276835,
|
| 610 |
+
"grad_norm": 1.4990551471710205,
|
| 611 |
+
"learning_rate": 9.643097751841854e-05,
|
| 612 |
+
"loss": 0.5189,
|
| 613 |
+
"step": 84
|
| 614 |
+
},
|
| 615 |
+
{
|
| 616 |
+
"epoch": 0.256120527306968,
|
| 617 |
+
"grad_norm": 1.4080356359481812,
|
| 618 |
+
"learning_rate": 9.622067162048112e-05,
|
| 619 |
+
"loss": 0.5112,
|
| 620 |
+
"step": 85
|
| 621 |
+
},
|
| 622 |
+
{
|
| 623 |
+
"epoch": 0.2591337099811676,
|
| 624 |
+
"grad_norm": 1.3661057949066162,
|
| 625 |
+
"learning_rate": 9.600458850350588e-05,
|
| 626 |
+
"loss": 0.4688,
|
| 627 |
+
"step": 86
|
| 628 |
+
},
|
| 629 |
+
{
|
| 630 |
+
"epoch": 0.2621468926553672,
|
| 631 |
+
"grad_norm": 1.570552945137024,
|
| 632 |
+
"learning_rate": 9.578275517617645e-05,
|
| 633 |
+
"loss": 0.5058,
|
| 634 |
+
"step": 87
|
| 635 |
+
},
|
| 636 |
+
{
|
| 637 |
+
"epoch": 0.26516007532956687,
|
| 638 |
+
"grad_norm": 1.6037708520889282,
|
| 639 |
+
"learning_rate": 9.555519936590738e-05,
|
| 640 |
+
"loss": 0.5201,
|
| 641 |
+
"step": 88
|
| 642 |
+
},
|
| 643 |
+
{
|
| 644 |
+
"epoch": 0.26817325800376646,
|
| 645 |
+
"grad_norm": 1.5268930196762085,
|
| 646 |
+
"learning_rate": 9.532194951537838e-05,
|
| 647 |
+
"loss": 0.4661,
|
| 648 |
+
"step": 89
|
| 649 |
+
},
|
| 650 |
+
{
|
| 651 |
+
"epoch": 0.2711864406779661,
|
| 652 |
+
"grad_norm": 1.7837523221969604,
|
| 653 |
+
"learning_rate": 9.508303477897924e-05,
|
| 654 |
+
"loss": 0.5005,
|
| 655 |
+
"step": 90
|
| 656 |
+
},
|
| 657 |
+
{
|
| 658 |
+
"epoch": 0.27419962335216574,
|
| 659 |
+
"grad_norm": 1.3590326309204102,
|
| 660 |
+
"learning_rate": 9.483848501916578e-05,
|
| 661 |
+
"loss": 0.3866,
|
| 662 |
+
"step": 91
|
| 663 |
+
},
|
| 664 |
+
{
|
| 665 |
+
"epoch": 0.27721280602636533,
|
| 666 |
+
"grad_norm": 1.5031671524047852,
|
| 667 |
+
"learning_rate": 9.458833080272722e-05,
|
| 668 |
+
"loss": 0.3559,
|
| 669 |
+
"step": 92
|
| 670 |
+
},
|
| 671 |
+
{
|
| 672 |
+
"epoch": 0.280225988700565,
|
| 673 |
+
"grad_norm": 1.2212880849838257,
|
| 674 |
+
"learning_rate": 9.433260339696563e-05,
|
| 675 |
+
"loss": 0.3586,
|
| 676 |
+
"step": 93
|
| 677 |
+
},
|
| 678 |
+
{
|
| 679 |
+
"epoch": 0.2832391713747646,
|
| 680 |
+
"grad_norm": 1.8385019302368164,
|
| 681 |
+
"learning_rate": 9.407133476578778e-05,
|
| 682 |
+
"loss": 0.4775,
|
| 683 |
+
"step": 94
|
| 684 |
+
},
|
| 685 |
+
{
|
| 686 |
+
"epoch": 0.2862523540489642,
|
| 687 |
+
"grad_norm": 2.6899161338806152,
|
| 688 |
+
"learning_rate": 9.38045575657098e-05,
|
| 689 |
+
"loss": 0.6809,
|
| 690 |
+
"step": 95
|
| 691 |
+
},
|
| 692 |
+
{
|
| 693 |
+
"epoch": 0.28926553672316385,
|
| 694 |
+
"grad_norm": 3.9981398582458496,
|
| 695 |
+
"learning_rate": 9.353230514177552e-05,
|
| 696 |
+
"loss": 0.8967,
|
| 697 |
+
"step": 96
|
| 698 |
+
},
|
| 699 |
+
{
|
| 700 |
+
"epoch": 0.29227871939736344,
|
| 701 |
+
"grad_norm": 3.7616143226623535,
|
| 702 |
+
"learning_rate": 9.325461152338846e-05,
|
| 703 |
+
"loss": 0.9173,
|
| 704 |
+
"step": 97
|
| 705 |
+
},
|
| 706 |
+
{
|
| 707 |
+
"epoch": 0.2952919020715631,
|
| 708 |
+
"grad_norm": 3.3938989639282227,
|
| 709 |
+
"learning_rate": 9.297151142005851e-05,
|
| 710 |
+
"loss": 0.7849,
|
| 711 |
+
"step": 98
|
| 712 |
+
},
|
| 713 |
+
{
|
| 714 |
+
"epoch": 0.2983050847457627,
|
| 715 |
+
"grad_norm": 3.3373446464538574,
|
| 716 |
+
"learning_rate": 9.268304021706349e-05,
|
| 717 |
+
"loss": 0.6619,
|
| 718 |
+
"step": 99
|
| 719 |
+
},
|
| 720 |
+
{
|
| 721 |
+
"epoch": 0.3013182674199623,
|
| 722 |
+
"grad_norm": 4.476459503173828,
|
| 723 |
+
"learning_rate": 9.23892339710263e-05,
|
| 724 |
+
"loss": 0.7758,
|
| 725 |
+
"step": 100
|
| 726 |
+
},
|
| 727 |
+
{
|
| 728 |
+
"epoch": 0.30433145009416196,
|
| 729 |
+
"grad_norm": 2.75358510017395,
|
| 730 |
+
"learning_rate": 9.209012940540805e-05,
|
| 731 |
+
"loss": 0.7565,
|
| 732 |
+
"step": 101
|
| 733 |
+
},
|
| 734 |
+
{
|
| 735 |
+
"epoch": 0.3073446327683616,
|
| 736 |
+
"grad_norm": 2.192662000656128,
|
| 737 |
+
"learning_rate": 9.178576390591802e-05,
|
| 738 |
+
"loss": 0.6634,
|
| 739 |
+
"step": 102
|
| 740 |
+
},
|
| 741 |
+
{
|
| 742 |
+
"epoch": 0.3103578154425612,
|
| 743 |
+
"grad_norm": 2.3334836959838867,
|
| 744 |
+
"learning_rate": 9.147617551584066e-05,
|
| 745 |
+
"loss": 0.6961,
|
| 746 |
+
"step": 103
|
| 747 |
+
},
|
| 748 |
+
{
|
| 749 |
+
"epoch": 0.31337099811676083,
|
| 750 |
+
"grad_norm": 1.9057625532150269,
|
| 751 |
+
"learning_rate": 9.116140293128051e-05,
|
| 752 |
+
"loss": 0.5762,
|
| 753 |
+
"step": 104
|
| 754 |
+
},
|
| 755 |
+
{
|
| 756 |
+
"epoch": 0.3163841807909605,
|
| 757 |
+
"grad_norm": 1.5543274879455566,
|
| 758 |
+
"learning_rate": 9.084148549632547e-05,
|
| 759 |
+
"loss": 0.5249,
|
| 760 |
+
"step": 105
|
| 761 |
+
},
|
| 762 |
+
{
|
| 763 |
+
"epoch": 0.31939736346516007,
|
| 764 |
+
"grad_norm": 1.3116902112960815,
|
| 765 |
+
"learning_rate": 9.051646319812918e-05,
|
| 766 |
+
"loss": 0.4895,
|
| 767 |
+
"step": 106
|
| 768 |
+
},
|
| 769 |
+
{
|
| 770 |
+
"epoch": 0.3224105461393597,
|
| 771 |
+
"grad_norm": 1.6137094497680664,
|
| 772 |
+
"learning_rate": 9.018637666191283e-05,
|
| 773 |
+
"loss": 0.5036,
|
| 774 |
+
"step": 107
|
| 775 |
+
},
|
| 776 |
+
{
|
| 777 |
+
"epoch": 0.3254237288135593,
|
| 778 |
+
"grad_norm": 1.4955766201019287,
|
| 779 |
+
"learning_rate": 8.985126714588738e-05,
|
| 780 |
+
"loss": 0.4571,
|
| 781 |
+
"step": 108
|
| 782 |
+
},
|
| 783 |
+
{
|
| 784 |
+
"epoch": 0.32843691148775894,
|
| 785 |
+
"grad_norm": 1.5371748208999634,
|
| 786 |
+
"learning_rate": 8.951117653609666e-05,
|
| 787 |
+
"loss": 0.4958,
|
| 788 |
+
"step": 109
|
| 789 |
+
},
|
| 790 |
+
{
|
| 791 |
+
"epoch": 0.3314500941619586,
|
| 792 |
+
"grad_norm": 1.2266839742660522,
|
| 793 |
+
"learning_rate": 8.916614734118184e-05,
|
| 794 |
+
"loss": 0.4171,
|
| 795 |
+
"step": 110
|
| 796 |
+
},
|
| 797 |
+
{
|
| 798 |
+
"epoch": 0.3344632768361582,
|
| 799 |
+
"grad_norm": 1.21657133102417,
|
| 800 |
+
"learning_rate": 8.881622268706825e-05,
|
| 801 |
+
"loss": 0.421,
|
| 802 |
+
"step": 111
|
| 803 |
+
},
|
| 804 |
+
{
|
| 805 |
+
"epoch": 0.3374764595103578,
|
| 806 |
+
"grad_norm": 1.2184901237487793,
|
| 807 |
+
"learning_rate": 8.8461446311575e-05,
|
| 808 |
+
"loss": 0.4307,
|
| 809 |
+
"step": 112
|
| 810 |
+
},
|
| 811 |
+
{
|
| 812 |
+
"epoch": 0.34048964218455746,
|
| 813 |
+
"grad_norm": 1.5124021768569946,
|
| 814 |
+
"learning_rate": 8.810186255894803e-05,
|
| 815 |
+
"loss": 0.4865,
|
| 816 |
+
"step": 113
|
| 817 |
+
},
|
| 818 |
+
{
|
| 819 |
+
"epoch": 0.34350282485875705,
|
| 820 |
+
"grad_norm": 1.078994870185852,
|
| 821 |
+
"learning_rate": 8.773751637431748e-05,
|
| 822 |
+
"loss": 0.3592,
|
| 823 |
+
"step": 114
|
| 824 |
+
},
|
| 825 |
+
{
|
| 826 |
+
"epoch": 0.3465160075329567,
|
| 827 |
+
"grad_norm": 1.2173560857772827,
|
| 828 |
+
"learning_rate": 8.736845329807993e-05,
|
| 829 |
+
"loss": 0.3757,
|
| 830 |
+
"step": 115
|
| 831 |
+
},
|
| 832 |
+
{
|
| 833 |
+
"epoch": 0.3495291902071563,
|
| 834 |
+
"grad_norm": 1.4223103523254395,
|
| 835 |
+
"learning_rate": 8.69947194602061e-05,
|
| 836 |
+
"loss": 0.4002,
|
| 837 |
+
"step": 116
|
| 838 |
+
},
|
| 839 |
+
{
|
| 840 |
+
"epoch": 0.3525423728813559,
|
| 841 |
+
"grad_norm": 1.2369580268859863,
|
| 842 |
+
"learning_rate": 8.66163615744751e-05,
|
| 843 |
+
"loss": 0.3891,
|
| 844 |
+
"step": 117
|
| 845 |
+
},
|
| 846 |
+
{
|
| 847 |
+
"epoch": 0.35555555555555557,
|
| 848 |
+
"grad_norm": 1.2306034564971924,
|
| 849 |
+
"learning_rate": 8.623342693263548e-05,
|
| 850 |
+
"loss": 0.3176,
|
| 851 |
+
"step": 118
|
| 852 |
+
},
|
| 853 |
+
{
|
| 854 |
+
"epoch": 0.35856873822975516,
|
| 855 |
+
"grad_norm": 1.20809805393219,
|
| 856 |
+
"learning_rate": 8.584596339849417e-05,
|
| 857 |
+
"loss": 0.3715,
|
| 858 |
+
"step": 119
|
| 859 |
+
},
|
| 860 |
+
{
|
| 861 |
+
"epoch": 0.3615819209039548,
|
| 862 |
+
"grad_norm": 1.59524405002594,
|
| 863 |
+
"learning_rate": 8.545401940193392e-05,
|
| 864 |
+
"loss": 0.4539,
|
| 865 |
+
"step": 120
|
| 866 |
+
},
|
| 867 |
+
{
|
| 868 |
+
"epoch": 0.36459510357815444,
|
| 869 |
+
"grad_norm": 2.4288361072540283,
|
| 870 |
+
"learning_rate": 8.505764393285984e-05,
|
| 871 |
+
"loss": 0.7094,
|
| 872 |
+
"step": 121
|
| 873 |
+
},
|
| 874 |
+
{
|
| 875 |
+
"epoch": 0.36760828625235403,
|
| 876 |
+
"grad_norm": 2.587125778198242,
|
| 877 |
+
"learning_rate": 8.46568865350762e-05,
|
| 878 |
+
"loss": 0.7052,
|
| 879 |
+
"step": 122
|
| 880 |
+
},
|
| 881 |
+
{
|
| 882 |
+
"epoch": 0.3706214689265537,
|
| 883 |
+
"grad_norm": 3.610764980316162,
|
| 884 |
+
"learning_rate": 8.425179730009368e-05,
|
| 885 |
+
"loss": 0.6835,
|
| 886 |
+
"step": 123
|
| 887 |
+
},
|
| 888 |
+
{
|
| 889 |
+
"epoch": 0.3736346516007533,
|
| 890 |
+
"grad_norm": 2.254451274871826,
|
| 891 |
+
"learning_rate": 8.384242686086848e-05,
|
| 892 |
+
"loss": 0.5733,
|
| 893 |
+
"step": 124
|
| 894 |
+
},
|
| 895 |
+
{
|
| 896 |
+
"epoch": 0.3766478342749529,
|
| 897 |
+
"grad_norm": 3.2182092666625977,
|
| 898 |
+
"learning_rate": 8.342882638547351e-05,
|
| 899 |
+
"loss": 0.7416,
|
| 900 |
+
"step": 125
|
| 901 |
+
},
|
| 902 |
+
{
|
| 903 |
+
"epoch": 0.37966101694915255,
|
| 904 |
+
"grad_norm": 2.0895962715148926,
|
| 905 |
+
"learning_rate": 8.301104757070274e-05,
|
| 906 |
+
"loss": 0.611,
|
| 907 |
+
"step": 126
|
| 908 |
+
},
|
| 909 |
+
{
|
| 910 |
+
"epoch": 0.38267419962335214,
|
| 911 |
+
"grad_norm": 1.9307582378387451,
|
| 912 |
+
"learning_rate": 8.258914263560971e-05,
|
| 913 |
+
"loss": 0.6099,
|
| 914 |
+
"step": 127
|
| 915 |
+
},
|
| 916 |
+
{
|
| 917 |
+
"epoch": 0.3856873822975518,
|
| 918 |
+
"grad_norm": 1.7885206937789917,
|
| 919 |
+
"learning_rate": 8.216316431498028e-05,
|
| 920 |
+
"loss": 0.4832,
|
| 921 |
+
"step": 128
|
| 922 |
+
},
|
| 923 |
+
{
|
| 924 |
+
"epoch": 0.3887005649717514,
|
| 925 |
+
"grad_norm": 1.2265185117721558,
|
| 926 |
+
"learning_rate": 8.173316585274145e-05,
|
| 927 |
+
"loss": 0.4042,
|
| 928 |
+
"step": 129
|
| 929 |
+
},
|
| 930 |
+
{
|
| 931 |
+
"epoch": 0.391713747645951,
|
| 932 |
+
"grad_norm": 1.369534969329834,
|
| 933 |
+
"learning_rate": 8.129920099530607e-05,
|
| 934 |
+
"loss": 0.4681,
|
| 935 |
+
"step": 130
|
| 936 |
+
},
|
| 937 |
+
{
|
| 938 |
+
"epoch": 0.39472693032015066,
|
| 939 |
+
"grad_norm": 1.340951681137085,
|
| 940 |
+
"learning_rate": 8.086132398485524e-05,
|
| 941 |
+
"loss": 0.4775,
|
| 942 |
+
"step": 131
|
| 943 |
+
},
|
| 944 |
+
{
|
| 945 |
+
"epoch": 0.3977401129943503,
|
| 946 |
+
"grad_norm": 1.1047234535217285,
|
| 947 |
+
"learning_rate": 8.041958955255814e-05,
|
| 948 |
+
"loss": 0.4508,
|
| 949 |
+
"step": 132
|
| 950 |
+
},
|
| 951 |
+
{
|
| 952 |
+
"epoch": 0.4007532956685499,
|
| 953 |
+
"grad_norm": 1.0403156280517578,
|
| 954 |
+
"learning_rate": 7.99740529117313e-05,
|
| 955 |
+
"loss": 0.4217,
|
| 956 |
+
"step": 133
|
| 957 |
+
},
|
| 958 |
+
{
|
| 959 |
+
"epoch": 0.40376647834274954,
|
| 960 |
+
"grad_norm": 0.9500618577003479,
|
| 961 |
+
"learning_rate": 7.952476975093729e-05,
|
| 962 |
+
"loss": 0.34,
|
| 963 |
+
"step": 134
|
| 964 |
+
},
|
| 965 |
+
{
|
| 966 |
+
"epoch": 0.4067796610169492,
|
| 967 |
+
"grad_norm": 1.1021428108215332,
|
| 968 |
+
"learning_rate": 7.907179622702408e-05,
|
| 969 |
+
"loss": 0.392,
|
| 970 |
+
"step": 135
|
| 971 |
+
},
|
| 972 |
+
{
|
| 973 |
+
"epoch": 0.40979284369114877,
|
| 974 |
+
"grad_norm": 1.2623156309127808,
|
| 975 |
+
"learning_rate": 7.861518895810596e-05,
|
| 976 |
+
"loss": 0.4238,
|
| 977 |
+
"step": 136
|
| 978 |
+
},
|
| 979 |
+
{
|
| 980 |
+
"epoch": 0.4128060263653484,
|
| 981 |
+
"grad_norm": 1.395652413368225,
|
| 982 |
+
"learning_rate": 7.815500501648653e-05,
|
| 983 |
+
"loss": 0.4211,
|
| 984 |
+
"step": 137
|
| 985 |
+
},
|
| 986 |
+
{
|
| 987 |
+
"epoch": 0.415819209039548,
|
| 988 |
+
"grad_norm": 1.3175368309020996,
|
| 989 |
+
"learning_rate": 7.769130192152538e-05,
|
| 990 |
+
"loss": 0.415,
|
| 991 |
+
"step": 138
|
| 992 |
+
},
|
| 993 |
+
{
|
| 994 |
+
"epoch": 0.41883239171374764,
|
| 995 |
+
"grad_norm": 1.3882197141647339,
|
| 996 |
+
"learning_rate": 7.722413763244838e-05,
|
| 997 |
+
"loss": 0.422,
|
| 998 |
+
"step": 139
|
| 999 |
+
},
|
| 1000 |
+
{
|
| 1001 |
+
"epoch": 0.4218455743879473,
|
| 1002 |
+
"grad_norm": 1.396023154258728,
|
| 1003 |
+
"learning_rate": 7.675357054110336e-05,
|
| 1004 |
+
"loss": 0.466,
|
| 1005 |
+
"step": 140
|
| 1006 |
+
},
|
| 1007 |
+
{
|
| 1008 |
+
"epoch": 0.4248587570621469,
|
| 1009 |
+
"grad_norm": 1.0779083967208862,
|
| 1010 |
+
"learning_rate": 7.627965946466166e-05,
|
| 1011 |
+
"loss": 0.3576,
|
| 1012 |
+
"step": 141
|
| 1013 |
+
},
|
| 1014 |
+
{
|
| 1015 |
+
"epoch": 0.4278719397363465,
|
| 1016 |
+
"grad_norm": 1.2511008977890015,
|
| 1017 |
+
"learning_rate": 7.580246363826621e-05,
|
| 1018 |
+
"loss": 0.301,
|
| 1019 |
+
"step": 142
|
| 1020 |
+
},
|
| 1021 |
+
{
|
| 1022 |
+
"epoch": 0.43088512241054616,
|
| 1023 |
+
"grad_norm": 1.13119375705719,
|
| 1024 |
+
"learning_rate": 7.532204270762786e-05,
|
| 1025 |
+
"loss": 0.3332,
|
| 1026 |
+
"step": 143
|
| 1027 |
+
},
|
| 1028 |
+
{
|
| 1029 |
+
"epoch": 0.43389830508474575,
|
| 1030 |
+
"grad_norm": 2.0195682048797607,
|
| 1031 |
+
"learning_rate": 7.483845672156998e-05,
|
| 1032 |
+
"loss": 0.6475,
|
| 1033 |
+
"step": 144
|
| 1034 |
+
},
|
| 1035 |
+
{
|
| 1036 |
+
"epoch": 0.4369114877589454,
|
| 1037 |
+
"grad_norm": 2.429945230484009,
|
| 1038 |
+
"learning_rate": 7.435176612452286e-05,
|
| 1039 |
+
"loss": 0.7177,
|
| 1040 |
+
"step": 145
|
| 1041 |
+
},
|
| 1042 |
+
{
|
| 1043 |
+
"epoch": 0.439924670433145,
|
| 1044 |
+
"grad_norm": 3.0756828784942627,
|
| 1045 |
+
"learning_rate": 7.386203174896872e-05,
|
| 1046 |
+
"loss": 0.741,
|
| 1047 |
+
"step": 146
|
| 1048 |
+
},
|
| 1049 |
+
{
|
| 1050 |
+
"epoch": 0.4429378531073446,
|
| 1051 |
+
"grad_norm": 3.7236998081207275,
|
| 1052 |
+
"learning_rate": 7.336931480783801e-05,
|
| 1053 |
+
"loss": 0.7999,
|
| 1054 |
+
"step": 147
|
| 1055 |
+
},
|
| 1056 |
+
{
|
| 1057 |
+
"epoch": 0.44595103578154427,
|
| 1058 |
+
"grad_norm": 2.7121517658233643,
|
| 1059 |
+
"learning_rate": 7.287367688685835e-05,
|
| 1060 |
+
"loss": 0.6044,
|
| 1061 |
+
"step": 148
|
| 1062 |
+
},
|
| 1063 |
+
{
|
| 1064 |
+
"epoch": 0.44896421845574386,
|
| 1065 |
+
"grad_norm": 3.661588668823242,
|
| 1066 |
+
"learning_rate": 7.237517993685678e-05,
|
| 1067 |
+
"loss": 0.5553,
|
| 1068 |
+
"step": 149
|
| 1069 |
+
},
|
| 1070 |
+
{
|
| 1071 |
+
"epoch": 0.4519774011299435,
|
| 1072 |
+
"grad_norm": 4.68520975112915,
|
| 1073 |
+
"learning_rate": 7.187388626601637e-05,
|
| 1074 |
+
"loss": 0.411,
|
| 1075 |
+
"step": 150
|
| 1076 |
+
},
|
| 1077 |
+
{
|
| 1078 |
+
"epoch": 0.45499058380414314,
|
| 1079 |
+
"grad_norm": 1.866217017173767,
|
| 1080 |
+
"learning_rate": 7.136985853208824e-05,
|
| 1081 |
+
"loss": 0.5442,
|
| 1082 |
+
"step": 151
|
| 1083 |
+
},
|
| 1084 |
+
{
|
| 1085 |
+
"epoch": 0.45800376647834273,
|
| 1086 |
+
"grad_norm": 1.6526014804840088,
|
| 1087 |
+
"learning_rate": 7.086315973455981e-05,
|
| 1088 |
+
"loss": 0.5071,
|
| 1089 |
+
"step": 152
|
| 1090 |
+
},
|
| 1091 |
+
{
|
| 1092 |
+
"epoch": 0.4610169491525424,
|
| 1093 |
+
"grad_norm": 1.3213937282562256,
|
| 1094 |
+
"learning_rate": 7.035385320678036e-05,
|
| 1095 |
+
"loss": 0.4598,
|
| 1096 |
+
"step": 153
|
| 1097 |
+
},
|
| 1098 |
+
{
|
| 1099 |
+
"epoch": 0.464030131826742,
|
| 1100 |
+
"grad_norm": 0.959452211856842,
|
| 1101 |
+
"learning_rate": 6.984200260804484e-05,
|
| 1102 |
+
"loss": 0.3485,
|
| 1103 |
+
"step": 154
|
| 1104 |
+
},
|
| 1105 |
+
{
|
| 1106 |
+
"epoch": 0.4670433145009416,
|
| 1107 |
+
"grad_norm": 1.0355703830718994,
|
| 1108 |
+
"learning_rate": 6.932767191563703e-05,
|
| 1109 |
+
"loss": 0.3648,
|
| 1110 |
+
"step": 155
|
| 1111 |
+
},
|
| 1112 |
+
{
|
| 1113 |
+
"epoch": 0.47005649717514125,
|
| 1114 |
+
"grad_norm": 0.9991386532783508,
|
| 1115 |
+
"learning_rate": 6.881092541683278e-05,
|
| 1116 |
+
"loss": 0.3535,
|
| 1117 |
+
"step": 156
|
| 1118 |
+
},
|
| 1119 |
+
{
|
| 1120 |
+
"epoch": 0.47306967984934084,
|
| 1121 |
+
"grad_norm": 1.0915963649749756,
|
| 1122 |
+
"learning_rate": 6.829182770086474e-05,
|
| 1123 |
+
"loss": 0.3682,
|
| 1124 |
+
"step": 157
|
| 1125 |
+
},
|
| 1126 |
+
{
|
| 1127 |
+
"epoch": 0.4760828625235405,
|
| 1128 |
+
"grad_norm": 0.9837580323219299,
|
| 1129 |
+
"learning_rate": 6.777044365084907e-05,
|
| 1130 |
+
"loss": 0.3703,
|
| 1131 |
+
"step": 158
|
| 1132 |
+
},
|
| 1133 |
+
{
|
| 1134 |
+
"epoch": 0.47909604519774013,
|
| 1135 |
+
"grad_norm": 1.258581280708313,
|
| 1136 |
+
"learning_rate": 6.724683843567568e-05,
|
| 1137 |
+
"loss": 0.4104,
|
| 1138 |
+
"step": 159
|
| 1139 |
+
},
|
| 1140 |
+
{
|
| 1141 |
+
"epoch": 0.4821092278719397,
|
| 1142 |
+
"grad_norm": 0.832224428653717,
|
| 1143 |
+
"learning_rate": 6.672107750186255e-05,
|
| 1144 |
+
"loss": 0.2934,
|
| 1145 |
+
"step": 160
|
| 1146 |
+
},
|
| 1147 |
+
{
|
| 1148 |
+
"epoch": 0.48512241054613936,
|
| 1149 |
+
"grad_norm": 0.881106436252594,
|
| 1150 |
+
"learning_rate": 6.619322656537552e-05,
|
| 1151 |
+
"loss": 0.3127,
|
| 1152 |
+
"step": 161
|
| 1153 |
+
},
|
| 1154 |
+
{
|
| 1155 |
+
"epoch": 0.488135593220339,
|
| 1156 |
+
"grad_norm": 1.257350206375122,
|
| 1157 |
+
"learning_rate": 6.566335160341424e-05,
|
| 1158 |
+
"loss": 0.3804,
|
| 1159 |
+
"step": 162
|
| 1160 |
+
},
|
| 1161 |
+
{
|
| 1162 |
+
"epoch": 0.4911487758945386,
|
| 1163 |
+
"grad_norm": 1.6826411485671997,
|
| 1164 |
+
"learning_rate": 6.513151884616556e-05,
|
| 1165 |
+
"loss": 0.4807,
|
| 1166 |
+
"step": 163
|
| 1167 |
+
},
|
| 1168 |
+
{
|
| 1169 |
+
"epoch": 0.49416195856873824,
|
| 1170 |
+
"grad_norm": 1.31766676902771,
|
| 1171 |
+
"learning_rate": 6.459779476852528e-05,
|
| 1172 |
+
"loss": 0.3872,
|
| 1173 |
+
"step": 164
|
| 1174 |
+
},
|
| 1175 |
+
{
|
| 1176 |
+
"epoch": 0.4971751412429379,
|
| 1177 |
+
"grad_norm": 1.438594102859497,
|
| 1178 |
+
"learning_rate": 6.406224608178932e-05,
|
| 1179 |
+
"loss": 0.3868,
|
| 1180 |
+
"step": 165
|
| 1181 |
+
},
|
| 1182 |
+
{
|
| 1183 |
+
"epoch": 0.5001883239171375,
|
| 1184 |
+
"grad_norm": 1.198364496231079,
|
| 1185 |
+
"learning_rate": 6.352493972531534e-05,
|
| 1186 |
+
"loss": 0.3361,
|
| 1187 |
+
"step": 166
|
| 1188 |
+
},
|
| 1189 |
+
{
|
| 1190 |
+
"epoch": 0.5001883239171375,
|
| 1191 |
+
"eval_loss": NaN,
|
| 1192 |
+
"eval_runtime": 93.1419,
|
| 1193 |
+
"eval_samples_per_second": 6.002,
|
| 1194 |
+
"eval_steps_per_second": 1.503,
|
| 1195 |
+
"step": 166
|
| 1196 |
}
|
| 1197 |
],
|
| 1198 |
"logging_steps": 1,
|
|
|
|
| 1212 |
"attributes": {}
|
| 1213 |
}
|
| 1214 |
},
|
| 1215 |
+
"total_flos": 5.46878716765012e+17,
|
| 1216 |
"train_batch_size": 2,
|
| 1217 |
"trial_name": null,
|
| 1218 |
"trial_params": null
|