Checkpoint for SFT at step 150
Browse files- model.safetensors +1 -1
- optimizer.pt +1 -1
- scheduler.pt +1 -1
- training_state.json +251 -1
model.safetensors
CHANGED
|
@@ -1,3 +1,3 @@
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
-
oid sha256:
|
| 3 |
size 22080496
|
|
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:0290d1dc2193a03b39037ef2a3eb7b6afabee7bd240a3581decb36e85260d239
|
| 3 |
size 22080496
|
optimizer.pt
CHANGED
|
@@ -1,3 +1,3 @@
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
-
oid sha256:
|
| 3 |
size 44193018
|
|
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:d4e2d6fe8d19f03b32584724d889d3ce5aeb6e1d9d3509a6586092621adf2a45
|
| 3 |
size 44193018
|
scheduler.pt
CHANGED
|
@@ -1,3 +1,3 @@
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
-
oid sha256:
|
| 3 |
size 1000
|
|
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:ea7fd64a788bd82addddc94b5c58f0cc158d2b4edf51ac5c62f232f520c28b37
|
| 3 |
size 1000
|
training_state.json
CHANGED
|
@@ -1,7 +1,7 @@
|
|
| 1 |
{
|
| 2 |
"phase": "SFT",
|
| 3 |
"epoch": 0,
|
| 4 |
-
"total_sft_steps":
|
| 5 |
"metrics": [
|
| 6 |
{
|
| 7 |
"Step": 0,
|
|
@@ -507,6 +507,256 @@
|
|
| 507 |
"Step": 100,
|
| 508 |
"Metric": "SFT Loss",
|
| 509 |
"Value": 8.235538482666016
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 510 |
}
|
| 511 |
],
|
| 512 |
"sft_complete": false
|
|
|
|
| 1 |
{
|
| 2 |
"phase": "SFT",
|
| 3 |
"epoch": 0,
|
| 4 |
+
"total_sft_steps": 150,
|
| 5 |
"metrics": [
|
| 6 |
{
|
| 7 |
"Step": 0,
|
|
|
|
| 507 |
"Step": 100,
|
| 508 |
"Metric": "SFT Loss",
|
| 509 |
"Value": 8.235538482666016
|
| 510 |
+
},
|
| 511 |
+
{
|
| 512 |
+
"Step": 101,
|
| 513 |
+
"Metric": "SFT Loss",
|
| 514 |
+
"Value": 8.270655632019043
|
| 515 |
+
},
|
| 516 |
+
{
|
| 517 |
+
"Step": 102,
|
| 518 |
+
"Metric": "SFT Loss",
|
| 519 |
+
"Value": 8.269576072692871
|
| 520 |
+
},
|
| 521 |
+
{
|
| 522 |
+
"Step": 103,
|
| 523 |
+
"Metric": "SFT Loss",
|
| 524 |
+
"Value": 8.274468421936035
|
| 525 |
+
},
|
| 526 |
+
{
|
| 527 |
+
"Step": 104,
|
| 528 |
+
"Metric": "SFT Loss",
|
| 529 |
+
"Value": 8.309279441833496
|
| 530 |
+
},
|
| 531 |
+
{
|
| 532 |
+
"Step": 105,
|
| 533 |
+
"Metric": "SFT Loss",
|
| 534 |
+
"Value": 8.239542961120605
|
| 535 |
+
},
|
| 536 |
+
{
|
| 537 |
+
"Step": 106,
|
| 538 |
+
"Metric": "SFT Loss",
|
| 539 |
+
"Value": 8.188124656677246
|
| 540 |
+
},
|
| 541 |
+
{
|
| 542 |
+
"Step": 107,
|
| 543 |
+
"Metric": "SFT Loss",
|
| 544 |
+
"Value": 8.163637161254883
|
| 545 |
+
},
|
| 546 |
+
{
|
| 547 |
+
"Step": 108,
|
| 548 |
+
"Metric": "SFT Loss",
|
| 549 |
+
"Value": 8.162602424621582
|
| 550 |
+
},
|
| 551 |
+
{
|
| 552 |
+
"Step": 109,
|
| 553 |
+
"Metric": "SFT Loss",
|
| 554 |
+
"Value": 8.176586151123047
|
| 555 |
+
},
|
| 556 |
+
{
|
| 557 |
+
"Step": 110,
|
| 558 |
+
"Metric": "SFT Loss",
|
| 559 |
+
"Value": 8.174212455749512
|
| 560 |
+
},
|
| 561 |
+
{
|
| 562 |
+
"Step": 111,
|
| 563 |
+
"Metric": "SFT Loss",
|
| 564 |
+
"Value": 8.19734001159668
|
| 565 |
+
},
|
| 566 |
+
{
|
| 567 |
+
"Step": 112,
|
| 568 |
+
"Metric": "SFT Loss",
|
| 569 |
+
"Value": 8.076298713684082
|
| 570 |
+
},
|
| 571 |
+
{
|
| 572 |
+
"Step": 113,
|
| 573 |
+
"Metric": "SFT Loss",
|
| 574 |
+
"Value": 8.18896770477295
|
| 575 |
+
},
|
| 576 |
+
{
|
| 577 |
+
"Step": 114,
|
| 578 |
+
"Metric": "SFT Loss",
|
| 579 |
+
"Value": 8.258679389953613
|
| 580 |
+
},
|
| 581 |
+
{
|
| 582 |
+
"Step": 115,
|
| 583 |
+
"Metric": "SFT Loss",
|
| 584 |
+
"Value": 8.18835735321045
|
| 585 |
+
},
|
| 586 |
+
{
|
| 587 |
+
"Step": 116,
|
| 588 |
+
"Metric": "SFT Loss",
|
| 589 |
+
"Value": 8.127300262451172
|
| 590 |
+
},
|
| 591 |
+
{
|
| 592 |
+
"Step": 117,
|
| 593 |
+
"Metric": "SFT Loss",
|
| 594 |
+
"Value": 8.145550727844238
|
| 595 |
+
},
|
| 596 |
+
{
|
| 597 |
+
"Step": 118,
|
| 598 |
+
"Metric": "SFT Loss",
|
| 599 |
+
"Value": 8.09005069732666
|
| 600 |
+
},
|
| 601 |
+
{
|
| 602 |
+
"Step": 119,
|
| 603 |
+
"Metric": "SFT Loss",
|
| 604 |
+
"Value": 7.936030387878418
|
| 605 |
+
},
|
| 606 |
+
{
|
| 607 |
+
"Step": 120,
|
| 608 |
+
"Metric": "SFT Loss",
|
| 609 |
+
"Value": 8.104690551757812
|
| 610 |
+
},
|
| 611 |
+
{
|
| 612 |
+
"Step": 121,
|
| 613 |
+
"Metric": "SFT Loss",
|
| 614 |
+
"Value": 8.058869361877441
|
| 615 |
+
},
|
| 616 |
+
{
|
| 617 |
+
"Step": 122,
|
| 618 |
+
"Metric": "SFT Loss",
|
| 619 |
+
"Value": 8.058721542358398
|
| 620 |
+
},
|
| 621 |
+
{
|
| 622 |
+
"Step": 123,
|
| 623 |
+
"Metric": "SFT Loss",
|
| 624 |
+
"Value": 7.885481357574463
|
| 625 |
+
},
|
| 626 |
+
{
|
| 627 |
+
"Step": 124,
|
| 628 |
+
"Metric": "SFT Loss",
|
| 629 |
+
"Value": 8.024746894836426
|
| 630 |
+
},
|
| 631 |
+
{
|
| 632 |
+
"Step": 125,
|
| 633 |
+
"Metric": "SFT Loss",
|
| 634 |
+
"Value": 8.017570495605469
|
| 635 |
+
},
|
| 636 |
+
{
|
| 637 |
+
"Step": 126,
|
| 638 |
+
"Metric": "SFT Loss",
|
| 639 |
+
"Value": 8.01369571685791
|
| 640 |
+
},
|
| 641 |
+
{
|
| 642 |
+
"Step": 127,
|
| 643 |
+
"Metric": "SFT Loss",
|
| 644 |
+
"Value": 7.956777572631836
|
| 645 |
+
},
|
| 646 |
+
{
|
| 647 |
+
"Step": 128,
|
| 648 |
+
"Metric": "SFT Loss",
|
| 649 |
+
"Value": 8.059258460998535
|
| 650 |
+
},
|
| 651 |
+
{
|
| 652 |
+
"Step": 129,
|
| 653 |
+
"Metric": "SFT Loss",
|
| 654 |
+
"Value": 8.099721908569336
|
| 655 |
+
},
|
| 656 |
+
{
|
| 657 |
+
"Step": 130,
|
| 658 |
+
"Metric": "SFT Loss",
|
| 659 |
+
"Value": 7.995508193969727
|
| 660 |
+
},
|
| 661 |
+
{
|
| 662 |
+
"Step": 131,
|
| 663 |
+
"Metric": "SFT Loss",
|
| 664 |
+
"Value": 7.869641304016113
|
| 665 |
+
},
|
| 666 |
+
{
|
| 667 |
+
"Step": 132,
|
| 668 |
+
"Metric": "SFT Loss",
|
| 669 |
+
"Value": 7.967761993408203
|
| 670 |
+
},
|
| 671 |
+
{
|
| 672 |
+
"Step": 133,
|
| 673 |
+
"Metric": "SFT Loss",
|
| 674 |
+
"Value": 8.04328441619873
|
| 675 |
+
},
|
| 676 |
+
{
|
| 677 |
+
"Step": 134,
|
| 678 |
+
"Metric": "SFT Loss",
|
| 679 |
+
"Value": 7.801205158233643
|
| 680 |
+
},
|
| 681 |
+
{
|
| 682 |
+
"Step": 135,
|
| 683 |
+
"Metric": "SFT Loss",
|
| 684 |
+
"Value": 7.971508979797363
|
| 685 |
+
},
|
| 686 |
+
{
|
| 687 |
+
"Step": 136,
|
| 688 |
+
"Metric": "SFT Loss",
|
| 689 |
+
"Value": 7.872110843658447
|
| 690 |
+
},
|
| 691 |
+
{
|
| 692 |
+
"Step": 137,
|
| 693 |
+
"Metric": "SFT Loss",
|
| 694 |
+
"Value": 8.073846817016602
|
| 695 |
+
},
|
| 696 |
+
{
|
| 697 |
+
"Step": 138,
|
| 698 |
+
"Metric": "SFT Loss",
|
| 699 |
+
"Value": 7.676604270935059
|
| 700 |
+
},
|
| 701 |
+
{
|
| 702 |
+
"Step": 139,
|
| 703 |
+
"Metric": "SFT Loss",
|
| 704 |
+
"Value": 7.913880825042725
|
| 705 |
+
},
|
| 706 |
+
{
|
| 707 |
+
"Step": 140,
|
| 708 |
+
"Metric": "SFT Loss",
|
| 709 |
+
"Value": 8.005086898803711
|
| 710 |
+
},
|
| 711 |
+
{
|
| 712 |
+
"Step": 141,
|
| 713 |
+
"Metric": "SFT Loss",
|
| 714 |
+
"Value": 7.983025074005127
|
| 715 |
+
},
|
| 716 |
+
{
|
| 717 |
+
"Step": 142,
|
| 718 |
+
"Metric": "SFT Loss",
|
| 719 |
+
"Value": 7.844381809234619
|
| 720 |
+
},
|
| 721 |
+
{
|
| 722 |
+
"Step": 143,
|
| 723 |
+
"Metric": "SFT Loss",
|
| 724 |
+
"Value": 7.966534614562988
|
| 725 |
+
},
|
| 726 |
+
{
|
| 727 |
+
"Step": 144,
|
| 728 |
+
"Metric": "SFT Loss",
|
| 729 |
+
"Value": 7.799246311187744
|
| 730 |
+
},
|
| 731 |
+
{
|
| 732 |
+
"Step": 145,
|
| 733 |
+
"Metric": "SFT Loss",
|
| 734 |
+
"Value": 7.769224166870117
|
| 735 |
+
},
|
| 736 |
+
{
|
| 737 |
+
"Step": 146,
|
| 738 |
+
"Metric": "SFT Loss",
|
| 739 |
+
"Value": 7.8659234046936035
|
| 740 |
+
},
|
| 741 |
+
{
|
| 742 |
+
"Step": 147,
|
| 743 |
+
"Metric": "SFT Loss",
|
| 744 |
+
"Value": 7.893615245819092
|
| 745 |
+
},
|
| 746 |
+
{
|
| 747 |
+
"Step": 148,
|
| 748 |
+
"Metric": "SFT Loss",
|
| 749 |
+
"Value": 7.57839822769165
|
| 750 |
+
},
|
| 751 |
+
{
|
| 752 |
+
"Step": 149,
|
| 753 |
+
"Metric": "SFT Loss",
|
| 754 |
+
"Value": 7.796144485473633
|
| 755 |
+
},
|
| 756 |
+
{
|
| 757 |
+
"Step": 150,
|
| 758 |
+
"Metric": "SFT Loss",
|
| 759 |
+
"Value": 7.760519504547119
|
| 760 |
}
|
| 761 |
],
|
| 762 |
"sft_complete": false
|