Commit
•
c300659
1
Parent(s):
c034307
Training in progress, step 42000
Browse files- adapter_model.bin +1 -1
- {checkpoint-38000 → checkpoint-41000/adapter_model}/README.md +0 -0
- {checkpoint-38000 → checkpoint-41000/adapter_model}/adapter_config.json +0 -0
- {checkpoint-38000 → checkpoint-41000/adapter_model}/adapter_model.bin +1 -1
- {checkpoint-38000/adapter_model → checkpoint-42000}/README.md +0 -0
- {checkpoint-38000/adapter_model → checkpoint-42000}/adapter_config.json +0 -0
- {checkpoint-38000/adapter_model → checkpoint-42000}/adapter_model.bin +1 -1
- {checkpoint-38000 → checkpoint-42000}/optimizer.pt +1 -1
- {checkpoint-38000 → checkpoint-42000}/rng_state.pth +1 -1
- {checkpoint-38000 → checkpoint-42000}/scheduler.pt +1 -1
- {checkpoint-38000 → checkpoint-42000}/trainer_state.json +277 -5
- {checkpoint-38000 → checkpoint-42000}/training_args.bin +0 -0
adapter_model.bin
CHANGED
|
@@ -1,3 +1,3 @@
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
-
oid sha256:
|
| 3 |
size 500897101
|
|
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:f723ba3028ed4671b1303b95a8e6ced28139e1673ad79569d94508cdbcf276e6
|
| 3 |
size 500897101
|
{checkpoint-38000 → checkpoint-41000/adapter_model}/README.md
RENAMED
|
File without changes
|
{checkpoint-38000 → checkpoint-41000/adapter_model}/adapter_config.json
RENAMED
|
File without changes
|
{checkpoint-38000 → checkpoint-41000/adapter_model}/adapter_model.bin
RENAMED
|
@@ -1,3 +1,3 @@
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
-
oid sha256:
|
| 3 |
size 500897101
|
|
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:bc2e66ae2124d056bbe13c91ab39f1f815d18e8ea9fb247d1e4d278f8c33f9b5
|
| 3 |
size 500897101
|
{checkpoint-38000/adapter_model → checkpoint-42000}/README.md
RENAMED
|
File without changes
|
{checkpoint-38000/adapter_model → checkpoint-42000}/adapter_config.json
RENAMED
|
File without changes
|
{checkpoint-38000/adapter_model → checkpoint-42000}/adapter_model.bin
RENAMED
|
@@ -1,3 +1,3 @@
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
-
oid sha256:
|
| 3 |
size 500897101
|
|
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:f723ba3028ed4671b1303b95a8e6ced28139e1673ad79569d94508cdbcf276e6
|
| 3 |
size 500897101
|
{checkpoint-38000 → checkpoint-42000}/optimizer.pt
RENAMED
|
@@ -1,3 +1,3 @@
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
-
oid sha256:
|
| 3 |
size 1001723453
|
|
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:89eb7c0106e859c3c23a264c2ccc5744239d7f1ca82d67c8c9de9180b96649da
|
| 3 |
size 1001723453
|
{checkpoint-38000 → checkpoint-42000}/rng_state.pth
RENAMED
|
@@ -1,3 +1,3 @@
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
-
oid sha256:
|
| 3 |
size 14575
|
|
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:6162a56e3fe60db7bec16127e13a48be509a4013bf0703ffc8b89024341614a9
|
| 3 |
size 14575
|
{checkpoint-38000 → checkpoint-42000}/scheduler.pt
RENAMED
|
@@ -1,3 +1,3 @@
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
-
oid sha256:
|
| 3 |
size 627
|
|
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:016ffa3122fe03449cce0003b2df603841d37baf6759e37edac391c19c3b0327
|
| 3 |
size 627
|
{checkpoint-38000 → checkpoint-42000}/trainer_state.json
RENAMED
|
@@ -1,8 +1,8 @@
|
|
| 1 |
{
|
| 2 |
-
"best_metric": 0.
|
| 3 |
-
"best_model_checkpoint": "./qlora-out/checkpoint-
|
| 4 |
-
"epoch": 1.
|
| 5 |
-
"global_step":
|
| 6 |
"is_hyper_param_search": false,
|
| 7 |
"is_local_process_zero": true,
|
| 8 |
"is_world_process_zero": true,
|
|
@@ -2590,11 +2590,283 @@
|
|
| 2590 |
"eval_samples_per_second": 0.423,
|
| 2591 |
"eval_steps_per_second": 0.423,
|
| 2592 |
"step": 38000
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 2593 |
}
|
| 2594 |
],
|
| 2595 |
"max_steps": 80463,
|
| 2596 |
"num_train_epochs": 3,
|
| 2597 |
-
"total_flos": 1.
|
| 2598 |
"trial_name": null,
|
| 2599 |
"trial_params": null
|
| 2600 |
}
|
|
|
|
| 1 |
{
|
| 2 |
+
"best_metric": 0.5362206697463989,
|
| 3 |
+
"best_model_checkpoint": "./qlora-out/checkpoint-42000",
|
| 4 |
+
"epoch": 1.565937138809142,
|
| 5 |
+
"global_step": 42000,
|
| 6 |
"is_hyper_param_search": false,
|
| 7 |
"is_local_process_zero": true,
|
| 8 |
"is_world_process_zero": true,
|
|
|
|
| 2590 |
"eval_samples_per_second": 0.423,
|
| 2591 |
"eval_steps_per_second": 0.423,
|
| 2592 |
"step": 38000
|
| 2593 |
+
},
|
| 2594 |
+
{
|
| 2595 |
+
"epoch": 1.42,
|
| 2596 |
+
"learning_rate": 0.00010833310040279531,
|
| 2597 |
+
"loss": 0.5161,
|
| 2598 |
+
"step": 38100
|
| 2599 |
+
},
|
| 2600 |
+
{
|
| 2601 |
+
"epoch": 1.42,
|
| 2602 |
+
"learning_rate": 0.00010794390806315602,
|
| 2603 |
+
"loss": 0.4929,
|
| 2604 |
+
"step": 38200
|
| 2605 |
+
},
|
| 2606 |
+
{
|
| 2607 |
+
"epoch": 1.43,
|
| 2608 |
+
"learning_rate": 0.0001075545945942978,
|
| 2609 |
+
"loss": 0.4819,
|
| 2610 |
+
"step": 38300
|
| 2611 |
+
},
|
| 2612 |
+
{
|
| 2613 |
+
"epoch": 1.43,
|
| 2614 |
+
"learning_rate": 0.00010716516593249742,
|
| 2615 |
+
"loss": 0.4594,
|
| 2616 |
+
"step": 38400
|
| 2617 |
+
},
|
| 2618 |
+
{
|
| 2619 |
+
"epoch": 1.44,
|
| 2620 |
+
"learning_rate": 0.00010677562801578798,
|
| 2621 |
+
"loss": 0.4802,
|
| 2622 |
+
"step": 38500
|
| 2623 |
+
},
|
| 2624 |
+
{
|
| 2625 |
+
"epoch": 1.44,
|
| 2626 |
+
"learning_rate": 0.00010638598678386864,
|
| 2627 |
+
"loss": 0.4793,
|
| 2628 |
+
"step": 38600
|
| 2629 |
+
},
|
| 2630 |
+
{
|
| 2631 |
+
"epoch": 1.44,
|
| 2632 |
+
"learning_rate": 0.00010599624817801383,
|
| 2633 |
+
"loss": 0.4912,
|
| 2634 |
+
"step": 38700
|
| 2635 |
+
},
|
| 2636 |
+
{
|
| 2637 |
+
"epoch": 1.45,
|
| 2638 |
+
"learning_rate": 0.0001056064181409828,
|
| 2639 |
+
"loss": 0.519,
|
| 2640 |
+
"step": 38800
|
| 2641 |
+
},
|
| 2642 |
+
{
|
| 2643 |
+
"epoch": 1.45,
|
| 2644 |
+
"learning_rate": 0.00010521650261692886,
|
| 2645 |
+
"loss": 0.4993,
|
| 2646 |
+
"step": 38900
|
| 2647 |
+
},
|
| 2648 |
+
{
|
| 2649 |
+
"epoch": 1.45,
|
| 2650 |
+
"learning_rate": 0.00010482650755130898,
|
| 2651 |
+
"loss": 0.4688,
|
| 2652 |
+
"step": 39000
|
| 2653 |
+
},
|
| 2654 |
+
{
|
| 2655 |
+
"epoch": 1.45,
|
| 2656 |
+
"eval_loss": 0.5486682057380676,
|
| 2657 |
+
"eval_runtime": 1265.4476,
|
| 2658 |
+
"eval_samples_per_second": 0.428,
|
| 2659 |
+
"eval_steps_per_second": 0.428,
|
| 2660 |
+
"step": 39000
|
| 2661 |
+
},
|
| 2662 |
+
{
|
| 2663 |
+
"epoch": 1.46,
|
| 2664 |
+
"learning_rate": 0.00010443643889079282,
|
| 2665 |
+
"loss": 0.4901,
|
| 2666 |
+
"step": 39100
|
| 2667 |
+
},
|
| 2668 |
+
{
|
| 2669 |
+
"epoch": 1.46,
|
| 2670 |
+
"learning_rate": 0.00010404630258317236,
|
| 2671 |
+
"loss": 0.4783,
|
| 2672 |
+
"step": 39200
|
| 2673 |
+
},
|
| 2674 |
+
{
|
| 2675 |
+
"epoch": 1.47,
|
| 2676 |
+
"learning_rate": 0.00010365610457727095,
|
| 2677 |
+
"loss": 0.5026,
|
| 2678 |
+
"step": 39300
|
| 2679 |
+
},
|
| 2680 |
+
{
|
| 2681 |
+
"epoch": 1.47,
|
| 2682 |
+
"learning_rate": 0.00010326585082285279,
|
| 2683 |
+
"loss": 0.4602,
|
| 2684 |
+
"step": 39400
|
| 2685 |
+
},
|
| 2686 |
+
{
|
| 2687 |
+
"epoch": 1.47,
|
| 2688 |
+
"learning_rate": 0.00010287554727053215,
|
| 2689 |
+
"loss": 0.5299,
|
| 2690 |
+
"step": 39500
|
| 2691 |
+
},
|
| 2692 |
+
{
|
| 2693 |
+
"epoch": 1.48,
|
| 2694 |
+
"learning_rate": 0.00010248519987168252,
|
| 2695 |
+
"loss": 0.5135,
|
| 2696 |
+
"step": 39600
|
| 2697 |
+
},
|
| 2698 |
+
{
|
| 2699 |
+
"epoch": 1.48,
|
| 2700 |
+
"learning_rate": 0.00010209481457834616,
|
| 2701 |
+
"loss": 0.4792,
|
| 2702 |
+
"step": 39700
|
| 2703 |
+
},
|
| 2704 |
+
{
|
| 2705 |
+
"epoch": 1.48,
|
| 2706 |
+
"learning_rate": 0.0001017043973431429,
|
| 2707 |
+
"loss": 0.5004,
|
| 2708 |
+
"step": 39800
|
| 2709 |
+
},
|
| 2710 |
+
{
|
| 2711 |
+
"epoch": 1.49,
|
| 2712 |
+
"learning_rate": 0.00010131395411917979,
|
| 2713 |
+
"loss": 0.5013,
|
| 2714 |
+
"step": 39900
|
| 2715 |
+
},
|
| 2716 |
+
{
|
| 2717 |
+
"epoch": 1.49,
|
| 2718 |
+
"learning_rate": 0.00010092349085996011,
|
| 2719 |
+
"loss": 0.501,
|
| 2720 |
+
"step": 40000
|
| 2721 |
+
},
|
| 2722 |
+
{
|
| 2723 |
+
"epoch": 1.49,
|
| 2724 |
+
"eval_loss": 0.5438262820243835,
|
| 2725 |
+
"eval_runtime": 1262.0322,
|
| 2726 |
+
"eval_samples_per_second": 0.429,
|
| 2727 |
+
"eval_steps_per_second": 0.429,
|
| 2728 |
+
"step": 40000
|
| 2729 |
+
},
|
| 2730 |
+
{
|
| 2731 |
+
"epoch": 1.5,
|
| 2732 |
+
"learning_rate": 0.0001005330135192927,
|
| 2733 |
+
"loss": 0.5043,
|
| 2734 |
+
"step": 40100
|
| 2735 |
+
},
|
| 2736 |
+
{
|
| 2737 |
+
"epoch": 1.5,
|
| 2738 |
+
"learning_rate": 0.000100142528051201,
|
| 2739 |
+
"loss": 0.4888,
|
| 2740 |
+
"step": 40200
|
| 2741 |
+
},
|
| 2742 |
+
{
|
| 2743 |
+
"epoch": 1.5,
|
| 2744 |
+
"learning_rate": 9.975204040983244e-05,
|
| 2745 |
+
"loss": 0.4866,
|
| 2746 |
+
"step": 40300
|
| 2747 |
+
},
|
| 2748 |
+
{
|
| 2749 |
+
"epoch": 1.51,
|
| 2750 |
+
"learning_rate": 9.936155654936761e-05,
|
| 2751 |
+
"loss": 0.4845,
|
| 2752 |
+
"step": 40400
|
| 2753 |
+
},
|
| 2754 |
+
{
|
| 2755 |
+
"epoch": 1.51,
|
| 2756 |
+
"learning_rate": 9.897108242392937e-05,
|
| 2757 |
+
"loss": 0.5027,
|
| 2758 |
+
"step": 40500
|
| 2759 |
+
},
|
| 2760 |
+
{
|
| 2761 |
+
"epoch": 1.51,
|
| 2762 |
+
"learning_rate": 9.858062398749225e-05,
|
| 2763 |
+
"loss": 0.5048,
|
| 2764 |
+
"step": 40600
|
| 2765 |
+
},
|
| 2766 |
+
{
|
| 2767 |
+
"epoch": 1.52,
|
| 2768 |
+
"learning_rate": 9.819018719379143e-05,
|
| 2769 |
+
"loss": 0.4985,
|
| 2770 |
+
"step": 40700
|
| 2771 |
+
},
|
| 2772 |
+
{
|
| 2773 |
+
"epoch": 1.52,
|
| 2774 |
+
"learning_rate": 9.77997779962322e-05,
|
| 2775 |
+
"loss": 0.4861,
|
| 2776 |
+
"step": 40800
|
| 2777 |
+
},
|
| 2778 |
+
{
|
| 2779 |
+
"epoch": 1.52,
|
| 2780 |
+
"learning_rate": 9.740940234779903e-05,
|
| 2781 |
+
"loss": 0.4743,
|
| 2782 |
+
"step": 40900
|
| 2783 |
+
},
|
| 2784 |
+
{
|
| 2785 |
+
"epoch": 1.53,
|
| 2786 |
+
"learning_rate": 9.701906620096474e-05,
|
| 2787 |
+
"loss": 0.5036,
|
| 2788 |
+
"step": 41000
|
| 2789 |
+
},
|
| 2790 |
+
{
|
| 2791 |
+
"epoch": 1.53,
|
| 2792 |
+
"eval_loss": 0.539782702922821,
|
| 2793 |
+
"eval_runtime": 1250.1524,
|
| 2794 |
+
"eval_samples_per_second": 0.434,
|
| 2795 |
+
"eval_steps_per_second": 0.434,
|
| 2796 |
+
"step": 41000
|
| 2797 |
+
},
|
| 2798 |
+
{
|
| 2799 |
+
"epoch": 1.53,
|
| 2800 |
+
"learning_rate": 9.662877550759995e-05,
|
| 2801 |
+
"loss": 0.4902,
|
| 2802 |
+
"step": 41100
|
| 2803 |
+
},
|
| 2804 |
+
{
|
| 2805 |
+
"epoch": 1.54,
|
| 2806 |
+
"learning_rate": 9.62385362188821e-05,
|
| 2807 |
+
"loss": 0.4568,
|
| 2808 |
+
"step": 41200
|
| 2809 |
+
},
|
| 2810 |
+
{
|
| 2811 |
+
"epoch": 1.54,
|
| 2812 |
+
"learning_rate": 9.584835428520491e-05,
|
| 2813 |
+
"loss": 0.491,
|
| 2814 |
+
"step": 41300
|
| 2815 |
+
},
|
| 2816 |
+
{
|
| 2817 |
+
"epoch": 1.54,
|
| 2818 |
+
"learning_rate": 9.545823565608745e-05,
|
| 2819 |
+
"loss": 0.5098,
|
| 2820 |
+
"step": 41400
|
| 2821 |
+
},
|
| 2822 |
+
{
|
| 2823 |
+
"epoch": 1.55,
|
| 2824 |
+
"learning_rate": 9.506818628008358e-05,
|
| 2825 |
+
"loss": 0.4964,
|
| 2826 |
+
"step": 41500
|
| 2827 |
+
},
|
| 2828 |
+
{
|
| 2829 |
+
"epoch": 1.55,
|
| 2830 |
+
"learning_rate": 9.467821210469116e-05,
|
| 2831 |
+
"loss": 0.4855,
|
| 2832 |
+
"step": 41600
|
| 2833 |
+
},
|
| 2834 |
+
{
|
| 2835 |
+
"epoch": 1.55,
|
| 2836 |
+
"learning_rate": 9.42883190762614e-05,
|
| 2837 |
+
"loss": 0.49,
|
| 2838 |
+
"step": 41700
|
| 2839 |
+
},
|
| 2840 |
+
{
|
| 2841 |
+
"epoch": 1.56,
|
| 2842 |
+
"learning_rate": 9.389851313990813e-05,
|
| 2843 |
+
"loss": 0.4981,
|
| 2844 |
+
"step": 41800
|
| 2845 |
+
},
|
| 2846 |
+
{
|
| 2847 |
+
"epoch": 1.56,
|
| 2848 |
+
"learning_rate": 9.350880023941727e-05,
|
| 2849 |
+
"loss": 0.4825,
|
| 2850 |
+
"step": 41900
|
| 2851 |
+
},
|
| 2852 |
+
{
|
| 2853 |
+
"epoch": 1.57,
|
| 2854 |
+
"learning_rate": 9.311918631715612e-05,
|
| 2855 |
+
"loss": 0.4751,
|
| 2856 |
+
"step": 42000
|
| 2857 |
+
},
|
| 2858 |
+
{
|
| 2859 |
+
"epoch": 1.57,
|
| 2860 |
+
"eval_loss": 0.5362206697463989,
|
| 2861 |
+
"eval_runtime": 1229.3972,
|
| 2862 |
+
"eval_samples_per_second": 0.441,
|
| 2863 |
+
"eval_steps_per_second": 0.441,
|
| 2864 |
+
"step": 42000
|
| 2865 |
}
|
| 2866 |
],
|
| 2867 |
"max_steps": 80463,
|
| 2868 |
"num_train_epochs": 3,
|
| 2869 |
+
"total_flos": 1.177938635424596e+19,
|
| 2870 |
"trial_name": null,
|
| 2871 |
"trial_params": null
|
| 2872 |
}
|
{checkpoint-38000 → checkpoint-42000}/training_args.bin
RENAMED
|
File without changes
|