Commit
·
4e08d51
1
Parent(s):
a774e2c
Training in progress, step 44000
Browse files- adapter_model.bin +1 -1
- {checkpoint-40000 → checkpoint-43000/adapter_model}/README.md +0 -0
- {checkpoint-40000 → checkpoint-43000/adapter_model}/adapter_config.json +0 -0
- {checkpoint-40000 → checkpoint-43000/adapter_model}/adapter_model.bin +1 -1
- {checkpoint-40000/adapter_model → checkpoint-44000}/README.md +0 -0
- {checkpoint-40000/adapter_model → checkpoint-44000}/adapter_config.json +0 -0
- {checkpoint-40000/adapter_model → checkpoint-44000}/adapter_model.bin +1 -1
- {checkpoint-40000 → checkpoint-44000}/optimizer.pt +1 -1
- {checkpoint-40000 → checkpoint-44000}/rng_state.pth +1 -1
- {checkpoint-40000 → checkpoint-44000}/scheduler.pt +1 -1
- {checkpoint-40000 → checkpoint-44000}/trainer_state.json +277 -5
- {checkpoint-40000 → checkpoint-44000}/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 sha256:514c6190fe2d77ecd2f858e923b2c7bec20b75928cee4654bcf237afb090151e
|
| 3 |
size 500897101
|
{checkpoint-40000 → checkpoint-43000/adapter_model}/README.md
RENAMED
|
File without changes
|
{checkpoint-40000 → checkpoint-43000/adapter_model}/adapter_config.json
RENAMED
|
File without changes
|
{checkpoint-40000 → checkpoint-43000/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:a98dbbf388b55a9ed29f37e7ec46fcecc3a3e1cd107ebdcea1335a06316bce14
|
| 3 |
size 500897101
|
{checkpoint-40000/adapter_model → checkpoint-44000}/README.md
RENAMED
|
File without changes
|
{checkpoint-40000/adapter_model → checkpoint-44000}/adapter_config.json
RENAMED
|
File without changes
|
{checkpoint-40000/adapter_model → checkpoint-44000}/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 sha256:514c6190fe2d77ecd2f858e923b2c7bec20b75928cee4654bcf237afb090151e
|
| 3 |
size 500897101
|
{checkpoint-40000 → checkpoint-44000}/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:c358d7d09699b43fe08d08ae6928e3d239f686bf90709539476817b4526e8156
|
| 3 |
size 1001723453
|
{checkpoint-40000 → checkpoint-44000}/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 sha256:0edbb7285779af4f2b9fba78491956e43d69071f56ba58ccc546a425b3c47636
|
| 3 |
size 14575
|
{checkpoint-40000 → checkpoint-44000}/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 sha256:4a53b5ec62bc72de547a30cfd6edc79f3190c27ad9d0a8e71792bbf9d33ce9b8
|
| 3 |
size 627
|
{checkpoint-40000 → checkpoint-44000}/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,
|
|
@@ -2726,11 +2726,283 @@
|
|
| 2726 |
"eval_samples_per_second": 0.429,
|
| 2727 |
"eval_steps_per_second": 0.429,
|
| 2728 |
"step": 40000
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 2729 |
}
|
| 2730 |
],
|
| 2731 |
"max_steps": 80463,
|
| 2732 |
"num_train_epochs": 3,
|
| 2733 |
-
"total_flos": 1.
|
| 2734 |
"trial_name": null,
|
| 2735 |
"trial_params": null
|
| 2736 |
}
|
|
|
|
| 1 |
{
|
| 2 |
+
"best_metric": 0.5285405516624451,
|
| 3 |
+
"best_model_checkpoint": "./qlora-out/checkpoint-44000",
|
| 4 |
+
"epoch": 1.6405055739905299,
|
| 5 |
+
"global_step": 44000,
|
| 6 |
"is_hyper_param_search": false,
|
| 7 |
"is_local_process_zero": true,
|
| 8 |
"is_world_process_zero": true,
|
|
|
|
| 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 |
+
"epoch": 1.57,
|
| 2868 |
+
"learning_rate": 9.272967731398264e-05,
|
| 2869 |
+
"loss": 0.4789,
|
| 2870 |
+
"step": 42100
|
| 2871 |
+
},
|
| 2872 |
+
{
|
| 2873 |
+
"epoch": 1.57,
|
| 2874 |
+
"learning_rate": 9.234027916915512e-05,
|
| 2875 |
+
"loss": 0.4644,
|
| 2876 |
+
"step": 42200
|
| 2877 |
+
},
|
| 2878 |
+
{
|
| 2879 |
+
"epoch": 1.58,
|
| 2880 |
+
"learning_rate": 9.195099782024136e-05,
|
| 2881 |
+
"loss": 0.4517,
|
| 2882 |
+
"step": 42300
|
| 2883 |
+
},
|
| 2884 |
+
{
|
| 2885 |
+
"epoch": 1.58,
|
| 2886 |
+
"learning_rate": 9.156183920302836e-05,
|
| 2887 |
+
"loss": 0.4768,
|
| 2888 |
+
"step": 42400
|
| 2889 |
+
},
|
| 2890 |
+
{
|
| 2891 |
+
"epoch": 1.58,
|
| 2892 |
+
"learning_rate": 9.117280925143156e-05,
|
| 2893 |
+
"loss": 0.4472,
|
| 2894 |
+
"step": 42500
|
| 2895 |
+
},
|
| 2896 |
+
{
|
| 2897 |
+
"epoch": 1.59,
|
| 2898 |
+
"learning_rate": 9.078391389740465e-05,
|
| 2899 |
+
"loss": 0.5002,
|
| 2900 |
+
"step": 42600
|
| 2901 |
+
},
|
| 2902 |
+
{
|
| 2903 |
+
"epoch": 1.59,
|
| 2904 |
+
"learning_rate": 9.039515907084884e-05,
|
| 2905 |
+
"loss": 0.4674,
|
| 2906 |
+
"step": 42700
|
| 2907 |
+
},
|
| 2908 |
+
{
|
| 2909 |
+
"epoch": 1.6,
|
| 2910 |
+
"learning_rate": 9.000655069952262e-05,
|
| 2911 |
+
"loss": 0.4704,
|
| 2912 |
+
"step": 42800
|
| 2913 |
+
},
|
| 2914 |
+
{
|
| 2915 |
+
"epoch": 1.6,
|
| 2916 |
+
"learning_rate": 8.961809470895141e-05,
|
| 2917 |
+
"loss": 0.4671,
|
| 2918 |
+
"step": 42900
|
| 2919 |
+
},
|
| 2920 |
+
{
|
| 2921 |
+
"epoch": 1.6,
|
| 2922 |
+
"learning_rate": 8.922979702233692e-05,
|
| 2923 |
+
"loss": 0.5007,
|
| 2924 |
+
"step": 43000
|
| 2925 |
+
},
|
| 2926 |
+
{
|
| 2927 |
+
"epoch": 1.6,
|
| 2928 |
+
"eval_loss": 0.5325181484222412,
|
| 2929 |
+
"eval_runtime": 1297.5192,
|
| 2930 |
+
"eval_samples_per_second": 0.418,
|
| 2931 |
+
"eval_steps_per_second": 0.418,
|
| 2932 |
+
"step": 43000
|
| 2933 |
+
},
|
| 2934 |
+
{
|
| 2935 |
+
"epoch": 1.61,
|
| 2936 |
+
"learning_rate": 8.884166356046725e-05,
|
| 2937 |
+
"loss": 0.4767,
|
| 2938 |
+
"step": 43100
|
| 2939 |
+
},
|
| 2940 |
+
{
|
| 2941 |
+
"epoch": 1.61,
|
| 2942 |
+
"learning_rate": 8.845370024162619e-05,
|
| 2943 |
+
"loss": 0.517,
|
| 2944 |
+
"step": 43200
|
| 2945 |
+
},
|
| 2946 |
+
{
|
| 2947 |
+
"epoch": 1.61,
|
| 2948 |
+
"learning_rate": 8.806591298150332e-05,
|
| 2949 |
+
"loss": 0.4742,
|
| 2950 |
+
"step": 43300
|
| 2951 |
+
},
|
| 2952 |
+
{
|
| 2953 |
+
"epoch": 1.62,
|
| 2954 |
+
"learning_rate": 8.767830769310362e-05,
|
| 2955 |
+
"loss": 0.4953,
|
| 2956 |
+
"step": 43400
|
| 2957 |
+
},
|
| 2958 |
+
{
|
| 2959 |
+
"epoch": 1.62,
|
| 2960 |
+
"learning_rate": 8.729089028665733e-05,
|
| 2961 |
+
"loss": 0.473,
|
| 2962 |
+
"step": 43500
|
| 2963 |
+
},
|
| 2964 |
+
{
|
| 2965 |
+
"epoch": 1.63,
|
| 2966 |
+
"learning_rate": 8.690366666952989e-05,
|
| 2967 |
+
"loss": 0.4695,
|
| 2968 |
+
"step": 43600
|
| 2969 |
+
},
|
| 2970 |
+
{
|
| 2971 |
+
"epoch": 1.63,
|
| 2972 |
+
"learning_rate": 8.651664274613183e-05,
|
| 2973 |
+
"loss": 0.4948,
|
| 2974 |
+
"step": 43700
|
| 2975 |
+
},
|
| 2976 |
+
{
|
| 2977 |
+
"epoch": 1.63,
|
| 2978 |
+
"learning_rate": 8.612982441782866e-05,
|
| 2979 |
+
"loss": 0.499,
|
| 2980 |
+
"step": 43800
|
| 2981 |
+
},
|
| 2982 |
+
{
|
| 2983 |
+
"epoch": 1.64,
|
| 2984 |
+
"learning_rate": 8.574321758285104e-05,
|
| 2985 |
+
"loss": 0.48,
|
| 2986 |
+
"step": 43900
|
| 2987 |
+
},
|
| 2988 |
+
{
|
| 2989 |
+
"epoch": 1.64,
|
| 2990 |
+
"learning_rate": 8.535682813620482e-05,
|
| 2991 |
+
"loss": 0.4672,
|
| 2992 |
+
"step": 44000
|
| 2993 |
+
},
|
| 2994 |
+
{
|
| 2995 |
+
"epoch": 1.64,
|
| 2996 |
+
"eval_loss": 0.5285405516624451,
|
| 2997 |
+
"eval_runtime": 1272.6033,
|
| 2998 |
+
"eval_samples_per_second": 0.426,
|
| 2999 |
+
"eval_steps_per_second": 0.426,
|
| 3000 |
+
"step": 44000
|
| 3001 |
}
|
| 3002 |
],
|
| 3003 |
"max_steps": 80463,
|
| 3004 |
"num_train_epochs": 3,
|
| 3005 |
+
"total_flos": 1.2339087922998313e+19,
|
| 3006 |
"trial_name": null,
|
| 3007 |
"trial_params": null
|
| 3008 |
}
|
{checkpoint-40000 → checkpoint-44000}/training_args.bin
RENAMED
|
File without changes
|