Uploaded checkpoint-2000
Browse files- model.safetensors +1 -1
- optimizer.pt +1 -1
- rng_state.pth +1 -1
- scheduler.pt +1 -1
- trainer_state.json +711 -3
model.safetensors
CHANGED
|
@@ -1,3 +1,3 @@
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
-
oid sha256:
|
| 3 |
size 2836579040
|
|
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:6f34a26801f7d37f64a99c14898ca0106179cdef8460793550938d8243950f81
|
| 3 |
size 2836579040
|
optimizer.pt
CHANGED
|
@@ -1,3 +1,3 @@
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
-
oid sha256:
|
| 3 |
size 5673376169
|
|
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:4add695eb3c80463223dce54d3c1dfd450589934a3767a55170cb5d7b58c1c37
|
| 3 |
size 5673376169
|
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 sha256:8c56c0d986fde7ce77bd50cb0ad4046e46bf3f78895b338e63f9217f7a569a8e
|
| 3 |
size 14244
|
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:b7dbe261a9604832f14328c92afa16e66095711f0d470c7aa4f09a20f6363131
|
| 3 |
size 1064
|
trainer_state.json
CHANGED
|
@@ -1,9 +1,9 @@
|
|
| 1 |
{
|
| 2 |
"best_metric": null,
|
| 3 |
"best_model_checkpoint": null,
|
| 4 |
-
"epoch": 0.
|
| 5 |
"eval_steps": 1000,
|
| 6 |
-
"global_step":
|
| 7 |
"is_hyper_param_search": false,
|
| 8 |
"is_local_process_zero": true,
|
| 9 |
"is_world_process_zero": true,
|
|
@@ -715,6 +715,714 @@
|
|
| 715 |
"eval_samples_per_second": 15.47,
|
| 716 |
"eval_steps_per_second": 15.47,
|
| 717 |
"step": 1000
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 718 |
}
|
| 719 |
],
|
| 720 |
"logging_steps": 10,
|
|
@@ -722,7 +1430,7 @@
|
|
| 722 |
"num_input_tokens_seen": 0,
|
| 723 |
"num_train_epochs": 1,
|
| 724 |
"save_steps": 1000,
|
| 725 |
-
"total_flos":
|
| 726 |
"train_batch_size": 1,
|
| 727 |
"trial_name": null,
|
| 728 |
"trial_params": null
|
|
|
|
| 1 |
{
|
| 2 |
"best_metric": null,
|
| 3 |
"best_model_checkpoint": null,
|
| 4 |
+
"epoch": 0.1,
|
| 5 |
"eval_steps": 1000,
|
| 6 |
+
"global_step": 2000,
|
| 7 |
"is_hyper_param_search": false,
|
| 8 |
"is_local_process_zero": true,
|
| 9 |
"is_world_process_zero": true,
|
|
|
|
| 715 |
"eval_samples_per_second": 15.47,
|
| 716 |
"eval_steps_per_second": 15.47,
|
| 717 |
"step": 1000
|
| 718 |
+
},
|
| 719 |
+
{
|
| 720 |
+
"epoch": 0.05,
|
| 721 |
+
"grad_norm": 71.5,
|
| 722 |
+
"learning_rate": 9.753333333333334e-07,
|
| 723 |
+
"loss": 1.5834,
|
| 724 |
+
"step": 1010
|
| 725 |
+
},
|
| 726 |
+
{
|
| 727 |
+
"epoch": 0.05,
|
| 728 |
+
"grad_norm": 71.0,
|
| 729 |
+
"learning_rate": 9.72888888888889e-07,
|
| 730 |
+
"loss": 1.4948,
|
| 731 |
+
"step": 1020
|
| 732 |
+
},
|
| 733 |
+
{
|
| 734 |
+
"epoch": 0.05,
|
| 735 |
+
"grad_norm": 71.5,
|
| 736 |
+
"learning_rate": 9.704444444444445e-07,
|
| 737 |
+
"loss": 1.4671,
|
| 738 |
+
"step": 1030
|
| 739 |
+
},
|
| 740 |
+
{
|
| 741 |
+
"epoch": 0.05,
|
| 742 |
+
"grad_norm": 70.0,
|
| 743 |
+
"learning_rate": 9.68e-07,
|
| 744 |
+
"loss": 1.341,
|
| 745 |
+
"step": 1040
|
| 746 |
+
},
|
| 747 |
+
{
|
| 748 |
+
"epoch": 0.05,
|
| 749 |
+
"grad_norm": 69.0,
|
| 750 |
+
"learning_rate": 9.655555555555555e-07,
|
| 751 |
+
"loss": 1.3023,
|
| 752 |
+
"step": 1050
|
| 753 |
+
},
|
| 754 |
+
{
|
| 755 |
+
"epoch": 0.05,
|
| 756 |
+
"grad_norm": 69.5,
|
| 757 |
+
"learning_rate": 9.631111111111111e-07,
|
| 758 |
+
"loss": 1.222,
|
| 759 |
+
"step": 1060
|
| 760 |
+
},
|
| 761 |
+
{
|
| 762 |
+
"epoch": 0.05,
|
| 763 |
+
"grad_norm": 64.5,
|
| 764 |
+
"learning_rate": 9.606666666666668e-07,
|
| 765 |
+
"loss": 1.1278,
|
| 766 |
+
"step": 1070
|
| 767 |
+
},
|
| 768 |
+
{
|
| 769 |
+
"epoch": 0.05,
|
| 770 |
+
"grad_norm": 62.5,
|
| 771 |
+
"learning_rate": 9.582222222222222e-07,
|
| 772 |
+
"loss": 1.0297,
|
| 773 |
+
"step": 1080
|
| 774 |
+
},
|
| 775 |
+
{
|
| 776 |
+
"epoch": 0.05,
|
| 777 |
+
"grad_norm": 60.5,
|
| 778 |
+
"learning_rate": 9.557777777777778e-07,
|
| 779 |
+
"loss": 0.9804,
|
| 780 |
+
"step": 1090
|
| 781 |
+
},
|
| 782 |
+
{
|
| 783 |
+
"epoch": 0.06,
|
| 784 |
+
"grad_norm": 56.0,
|
| 785 |
+
"learning_rate": 9.533333333333335e-07,
|
| 786 |
+
"loss": 0.9246,
|
| 787 |
+
"step": 1100
|
| 788 |
+
},
|
| 789 |
+
{
|
| 790 |
+
"epoch": 0.06,
|
| 791 |
+
"grad_norm": 54.5,
|
| 792 |
+
"learning_rate": 9.50888888888889e-07,
|
| 793 |
+
"loss": 0.8298,
|
| 794 |
+
"step": 1110
|
| 795 |
+
},
|
| 796 |
+
{
|
| 797 |
+
"epoch": 0.06,
|
| 798 |
+
"grad_norm": 50.0,
|
| 799 |
+
"learning_rate": 9.484444444444445e-07,
|
| 800 |
+
"loss": 0.7654,
|
| 801 |
+
"step": 1120
|
| 802 |
+
},
|
| 803 |
+
{
|
| 804 |
+
"epoch": 0.06,
|
| 805 |
+
"grad_norm": 54.0,
|
| 806 |
+
"learning_rate": 9.46e-07,
|
| 807 |
+
"loss": 0.7176,
|
| 808 |
+
"step": 1130
|
| 809 |
+
},
|
| 810 |
+
{
|
| 811 |
+
"epoch": 0.06,
|
| 812 |
+
"grad_norm": 47.25,
|
| 813 |
+
"learning_rate": 9.435555555555556e-07,
|
| 814 |
+
"loss": 0.6529,
|
| 815 |
+
"step": 1140
|
| 816 |
+
},
|
| 817 |
+
{
|
| 818 |
+
"epoch": 0.06,
|
| 819 |
+
"grad_norm": 45.75,
|
| 820 |
+
"learning_rate": 9.411111111111111e-07,
|
| 821 |
+
"loss": 0.6445,
|
| 822 |
+
"step": 1150
|
| 823 |
+
},
|
| 824 |
+
{
|
| 825 |
+
"epoch": 0.06,
|
| 826 |
+
"grad_norm": 47.75,
|
| 827 |
+
"learning_rate": 9.386666666666668e-07,
|
| 828 |
+
"loss": 0.6063,
|
| 829 |
+
"step": 1160
|
| 830 |
+
},
|
| 831 |
+
{
|
| 832 |
+
"epoch": 0.06,
|
| 833 |
+
"grad_norm": 43.25,
|
| 834 |
+
"learning_rate": 9.362222222222223e-07,
|
| 835 |
+
"loss": 0.5723,
|
| 836 |
+
"step": 1170
|
| 837 |
+
},
|
| 838 |
+
{
|
| 839 |
+
"epoch": 0.06,
|
| 840 |
+
"grad_norm": 41.0,
|
| 841 |
+
"learning_rate": 9.337777777777779e-07,
|
| 842 |
+
"loss": 0.5213,
|
| 843 |
+
"step": 1180
|
| 844 |
+
},
|
| 845 |
+
{
|
| 846 |
+
"epoch": 0.06,
|
| 847 |
+
"grad_norm": 35.75,
|
| 848 |
+
"learning_rate": 9.313333333333334e-07,
|
| 849 |
+
"loss": 0.5067,
|
| 850 |
+
"step": 1190
|
| 851 |
+
},
|
| 852 |
+
{
|
| 853 |
+
"epoch": 0.06,
|
| 854 |
+
"grad_norm": 35.0,
|
| 855 |
+
"learning_rate": 9.288888888888889e-07,
|
| 856 |
+
"loss": 0.4591,
|
| 857 |
+
"step": 1200
|
| 858 |
+
},
|
| 859 |
+
{
|
| 860 |
+
"epoch": 0.06,
|
| 861 |
+
"grad_norm": 38.5,
|
| 862 |
+
"learning_rate": 9.264444444444444e-07,
|
| 863 |
+
"loss": 0.4633,
|
| 864 |
+
"step": 1210
|
| 865 |
+
},
|
| 866 |
+
{
|
| 867 |
+
"epoch": 0.06,
|
| 868 |
+
"grad_norm": 35.0,
|
| 869 |
+
"learning_rate": 9.24e-07,
|
| 870 |
+
"loss": 0.4209,
|
| 871 |
+
"step": 1220
|
| 872 |
+
},
|
| 873 |
+
{
|
| 874 |
+
"epoch": 0.06,
|
| 875 |
+
"grad_norm": 33.0,
|
| 876 |
+
"learning_rate": 9.215555555555556e-07,
|
| 877 |
+
"loss": 0.4244,
|
| 878 |
+
"step": 1230
|
| 879 |
+
},
|
| 880 |
+
{
|
| 881 |
+
"epoch": 0.06,
|
| 882 |
+
"grad_norm": 29.5,
|
| 883 |
+
"learning_rate": 9.191111111111112e-07,
|
| 884 |
+
"loss": 0.3825,
|
| 885 |
+
"step": 1240
|
| 886 |
+
},
|
| 887 |
+
{
|
| 888 |
+
"epoch": 0.06,
|
| 889 |
+
"grad_norm": 28.75,
|
| 890 |
+
"learning_rate": 9.166666666666667e-07,
|
| 891 |
+
"loss": 0.3652,
|
| 892 |
+
"step": 1250
|
| 893 |
+
},
|
| 894 |
+
{
|
| 895 |
+
"epoch": 0.06,
|
| 896 |
+
"grad_norm": 26.5,
|
| 897 |
+
"learning_rate": 9.142222222222223e-07,
|
| 898 |
+
"loss": 0.3491,
|
| 899 |
+
"step": 1260
|
| 900 |
+
},
|
| 901 |
+
{
|
| 902 |
+
"epoch": 0.06,
|
| 903 |
+
"grad_norm": 30.5,
|
| 904 |
+
"learning_rate": 9.117777777777778e-07,
|
| 905 |
+
"loss": 0.3503,
|
| 906 |
+
"step": 1270
|
| 907 |
+
},
|
| 908 |
+
{
|
| 909 |
+
"epoch": 0.06,
|
| 910 |
+
"grad_norm": 26.0,
|
| 911 |
+
"learning_rate": 9.093333333333333e-07,
|
| 912 |
+
"loss": 0.3352,
|
| 913 |
+
"step": 1280
|
| 914 |
+
},
|
| 915 |
+
{
|
| 916 |
+
"epoch": 0.06,
|
| 917 |
+
"grad_norm": 24.0,
|
| 918 |
+
"learning_rate": 9.06888888888889e-07,
|
| 919 |
+
"loss": 0.2929,
|
| 920 |
+
"step": 1290
|
| 921 |
+
},
|
| 922 |
+
{
|
| 923 |
+
"epoch": 0.07,
|
| 924 |
+
"grad_norm": 21.0,
|
| 925 |
+
"learning_rate": 9.044444444444445e-07,
|
| 926 |
+
"loss": 0.2818,
|
| 927 |
+
"step": 1300
|
| 928 |
+
},
|
| 929 |
+
{
|
| 930 |
+
"epoch": 0.07,
|
| 931 |
+
"grad_norm": 26.25,
|
| 932 |
+
"learning_rate": 9.02e-07,
|
| 933 |
+
"loss": 0.2846,
|
| 934 |
+
"step": 1310
|
| 935 |
+
},
|
| 936 |
+
{
|
| 937 |
+
"epoch": 0.07,
|
| 938 |
+
"grad_norm": 22.625,
|
| 939 |
+
"learning_rate": 8.995555555555556e-07,
|
| 940 |
+
"loss": 0.2931,
|
| 941 |
+
"step": 1320
|
| 942 |
+
},
|
| 943 |
+
{
|
| 944 |
+
"epoch": 0.07,
|
| 945 |
+
"grad_norm": 23.5,
|
| 946 |
+
"learning_rate": 8.971111111111112e-07,
|
| 947 |
+
"loss": 0.2636,
|
| 948 |
+
"step": 1330
|
| 949 |
+
},
|
| 950 |
+
{
|
| 951 |
+
"epoch": 0.07,
|
| 952 |
+
"grad_norm": 21.75,
|
| 953 |
+
"learning_rate": 8.946666666666667e-07,
|
| 954 |
+
"loss": 0.2601,
|
| 955 |
+
"step": 1340
|
| 956 |
+
},
|
| 957 |
+
{
|
| 958 |
+
"epoch": 0.07,
|
| 959 |
+
"grad_norm": 21.875,
|
| 960 |
+
"learning_rate": 8.922222222222223e-07,
|
| 961 |
+
"loss": 0.2422,
|
| 962 |
+
"step": 1350
|
| 963 |
+
},
|
| 964 |
+
{
|
| 965 |
+
"epoch": 0.07,
|
| 966 |
+
"grad_norm": 19.875,
|
| 967 |
+
"learning_rate": 8.897777777777778e-07,
|
| 968 |
+
"loss": 0.2224,
|
| 969 |
+
"step": 1360
|
| 970 |
+
},
|
| 971 |
+
{
|
| 972 |
+
"epoch": 0.07,
|
| 973 |
+
"grad_norm": 19.375,
|
| 974 |
+
"learning_rate": 8.873333333333334e-07,
|
| 975 |
+
"loss": 0.2403,
|
| 976 |
+
"step": 1370
|
| 977 |
+
},
|
| 978 |
+
{
|
| 979 |
+
"epoch": 0.07,
|
| 980 |
+
"grad_norm": 18.25,
|
| 981 |
+
"learning_rate": 8.848888888888889e-07,
|
| 982 |
+
"loss": 0.2175,
|
| 983 |
+
"step": 1380
|
| 984 |
+
},
|
| 985 |
+
{
|
| 986 |
+
"epoch": 0.07,
|
| 987 |
+
"grad_norm": 17.375,
|
| 988 |
+
"learning_rate": 8.824444444444444e-07,
|
| 989 |
+
"loss": 0.2203,
|
| 990 |
+
"step": 1390
|
| 991 |
+
},
|
| 992 |
+
{
|
| 993 |
+
"epoch": 0.07,
|
| 994 |
+
"grad_norm": 18.625,
|
| 995 |
+
"learning_rate": 8.8e-07,
|
| 996 |
+
"loss": 0.1981,
|
| 997 |
+
"step": 1400
|
| 998 |
+
},
|
| 999 |
+
{
|
| 1000 |
+
"epoch": 0.07,
|
| 1001 |
+
"grad_norm": 17.25,
|
| 1002 |
+
"learning_rate": 8.775555555555557e-07,
|
| 1003 |
+
"loss": 0.1919,
|
| 1004 |
+
"step": 1410
|
| 1005 |
+
},
|
| 1006 |
+
{
|
| 1007 |
+
"epoch": 0.07,
|
| 1008 |
+
"grad_norm": 17.625,
|
| 1009 |
+
"learning_rate": 8.751111111111112e-07,
|
| 1010 |
+
"loss": 0.1952,
|
| 1011 |
+
"step": 1420
|
| 1012 |
+
},
|
| 1013 |
+
{
|
| 1014 |
+
"epoch": 0.07,
|
| 1015 |
+
"grad_norm": 16.125,
|
| 1016 |
+
"learning_rate": 8.726666666666667e-07,
|
| 1017 |
+
"loss": 0.1902,
|
| 1018 |
+
"step": 1430
|
| 1019 |
+
},
|
| 1020 |
+
{
|
| 1021 |
+
"epoch": 0.07,
|
| 1022 |
+
"grad_norm": 15.125,
|
| 1023 |
+
"learning_rate": 8.702222222222223e-07,
|
| 1024 |
+
"loss": 0.1776,
|
| 1025 |
+
"step": 1440
|
| 1026 |
+
},
|
| 1027 |
+
{
|
| 1028 |
+
"epoch": 0.07,
|
| 1029 |
+
"grad_norm": 15.4375,
|
| 1030 |
+
"learning_rate": 8.677777777777778e-07,
|
| 1031 |
+
"loss": 0.1818,
|
| 1032 |
+
"step": 1450
|
| 1033 |
+
},
|
| 1034 |
+
{
|
| 1035 |
+
"epoch": 0.07,
|
| 1036 |
+
"grad_norm": 18.0,
|
| 1037 |
+
"learning_rate": 8.653333333333333e-07,
|
| 1038 |
+
"loss": 0.1595,
|
| 1039 |
+
"step": 1460
|
| 1040 |
+
},
|
| 1041 |
+
{
|
| 1042 |
+
"epoch": 0.07,
|
| 1043 |
+
"grad_norm": 15.6875,
|
| 1044 |
+
"learning_rate": 8.62888888888889e-07,
|
| 1045 |
+
"loss": 0.1515,
|
| 1046 |
+
"step": 1470
|
| 1047 |
+
},
|
| 1048 |
+
{
|
| 1049 |
+
"epoch": 0.07,
|
| 1050 |
+
"grad_norm": 18.25,
|
| 1051 |
+
"learning_rate": 8.604444444444446e-07,
|
| 1052 |
+
"loss": 0.1645,
|
| 1053 |
+
"step": 1480
|
| 1054 |
+
},
|
| 1055 |
+
{
|
| 1056 |
+
"epoch": 0.07,
|
| 1057 |
+
"grad_norm": 16.0,
|
| 1058 |
+
"learning_rate": 8.580000000000001e-07,
|
| 1059 |
+
"loss": 0.1548,
|
| 1060 |
+
"step": 1490
|
| 1061 |
+
},
|
| 1062 |
+
{
|
| 1063 |
+
"epoch": 0.07,
|
| 1064 |
+
"grad_norm": 12.3125,
|
| 1065 |
+
"learning_rate": 8.555555555555556e-07,
|
| 1066 |
+
"loss": 0.1445,
|
| 1067 |
+
"step": 1500
|
| 1068 |
+
},
|
| 1069 |
+
{
|
| 1070 |
+
"epoch": 0.08,
|
| 1071 |
+
"grad_norm": 15.4375,
|
| 1072 |
+
"learning_rate": 8.531111111111111e-07,
|
| 1073 |
+
"loss": 0.1442,
|
| 1074 |
+
"step": 1510
|
| 1075 |
+
},
|
| 1076 |
+
{
|
| 1077 |
+
"epoch": 0.08,
|
| 1078 |
+
"grad_norm": 11.6875,
|
| 1079 |
+
"learning_rate": 8.506666666666667e-07,
|
| 1080 |
+
"loss": 0.1349,
|
| 1081 |
+
"step": 1520
|
| 1082 |
+
},
|
| 1083 |
+
{
|
| 1084 |
+
"epoch": 0.08,
|
| 1085 |
+
"grad_norm": 12.5625,
|
| 1086 |
+
"learning_rate": 8.482222222222222e-07,
|
| 1087 |
+
"loss": 0.1328,
|
| 1088 |
+
"step": 1530
|
| 1089 |
+
},
|
| 1090 |
+
{
|
| 1091 |
+
"epoch": 0.08,
|
| 1092 |
+
"grad_norm": 15.4375,
|
| 1093 |
+
"learning_rate": 8.457777777777778e-07,
|
| 1094 |
+
"loss": 0.1329,
|
| 1095 |
+
"step": 1540
|
| 1096 |
+
},
|
| 1097 |
+
{
|
| 1098 |
+
"epoch": 0.08,
|
| 1099 |
+
"grad_norm": 11.375,
|
| 1100 |
+
"learning_rate": 8.433333333333334e-07,
|
| 1101 |
+
"loss": 0.1273,
|
| 1102 |
+
"step": 1550
|
| 1103 |
+
},
|
| 1104 |
+
{
|
| 1105 |
+
"epoch": 0.08,
|
| 1106 |
+
"grad_norm": 11.6875,
|
| 1107 |
+
"learning_rate": 8.40888888888889e-07,
|
| 1108 |
+
"loss": 0.1383,
|
| 1109 |
+
"step": 1560
|
| 1110 |
+
},
|
| 1111 |
+
{
|
| 1112 |
+
"epoch": 0.08,
|
| 1113 |
+
"grad_norm": 13.875,
|
| 1114 |
+
"learning_rate": 8.384444444444445e-07,
|
| 1115 |
+
"loss": 0.1265,
|
| 1116 |
+
"step": 1570
|
| 1117 |
+
},
|
| 1118 |
+
{
|
| 1119 |
+
"epoch": 0.08,
|
| 1120 |
+
"grad_norm": 10.875,
|
| 1121 |
+
"learning_rate": 8.36e-07,
|
| 1122 |
+
"loss": 0.1264,
|
| 1123 |
+
"step": 1580
|
| 1124 |
+
},
|
| 1125 |
+
{
|
| 1126 |
+
"epoch": 0.08,
|
| 1127 |
+
"grad_norm": 11.25,
|
| 1128 |
+
"learning_rate": 8.335555555555555e-07,
|
| 1129 |
+
"loss": 0.1127,
|
| 1130 |
+
"step": 1590
|
| 1131 |
+
},
|
| 1132 |
+
{
|
| 1133 |
+
"epoch": 0.08,
|
| 1134 |
+
"grad_norm": 10.6875,
|
| 1135 |
+
"learning_rate": 8.311111111111112e-07,
|
| 1136 |
+
"loss": 0.112,
|
| 1137 |
+
"step": 1600
|
| 1138 |
+
},
|
| 1139 |
+
{
|
| 1140 |
+
"epoch": 0.08,
|
| 1141 |
+
"grad_norm": 9.5,
|
| 1142 |
+
"learning_rate": 8.286666666666667e-07,
|
| 1143 |
+
"loss": 0.1162,
|
| 1144 |
+
"step": 1610
|
| 1145 |
+
},
|
| 1146 |
+
{
|
| 1147 |
+
"epoch": 0.08,
|
| 1148 |
+
"grad_norm": 11.0625,
|
| 1149 |
+
"learning_rate": 8.262222222222222e-07,
|
| 1150 |
+
"loss": 0.0977,
|
| 1151 |
+
"step": 1620
|
| 1152 |
+
},
|
| 1153 |
+
{
|
| 1154 |
+
"epoch": 0.08,
|
| 1155 |
+
"grad_norm": 12.375,
|
| 1156 |
+
"learning_rate": 8.237777777777779e-07,
|
| 1157 |
+
"loss": 0.1047,
|
| 1158 |
+
"step": 1630
|
| 1159 |
+
},
|
| 1160 |
+
{
|
| 1161 |
+
"epoch": 0.08,
|
| 1162 |
+
"grad_norm": 12.375,
|
| 1163 |
+
"learning_rate": 8.213333333333334e-07,
|
| 1164 |
+
"loss": 0.1005,
|
| 1165 |
+
"step": 1640
|
| 1166 |
+
},
|
| 1167 |
+
{
|
| 1168 |
+
"epoch": 0.08,
|
| 1169 |
+
"grad_norm": 9.5625,
|
| 1170 |
+
"learning_rate": 8.188888888888889e-07,
|
| 1171 |
+
"loss": 0.1039,
|
| 1172 |
+
"step": 1650
|
| 1173 |
+
},
|
| 1174 |
+
{
|
| 1175 |
+
"epoch": 0.08,
|
| 1176 |
+
"grad_norm": 8.625,
|
| 1177 |
+
"learning_rate": 8.164444444444445e-07,
|
| 1178 |
+
"loss": 0.0944,
|
| 1179 |
+
"step": 1660
|
| 1180 |
+
},
|
| 1181 |
+
{
|
| 1182 |
+
"epoch": 0.08,
|
| 1183 |
+
"grad_norm": 10.1875,
|
| 1184 |
+
"learning_rate": 8.140000000000001e-07,
|
| 1185 |
+
"loss": 0.0916,
|
| 1186 |
+
"step": 1670
|
| 1187 |
+
},
|
| 1188 |
+
{
|
| 1189 |
+
"epoch": 0.08,
|
| 1190 |
+
"grad_norm": 7.84375,
|
| 1191 |
+
"learning_rate": 8.115555555555556e-07,
|
| 1192 |
+
"loss": 0.087,
|
| 1193 |
+
"step": 1680
|
| 1194 |
+
},
|
| 1195 |
+
{
|
| 1196 |
+
"epoch": 0.08,
|
| 1197 |
+
"grad_norm": 9.1875,
|
| 1198 |
+
"learning_rate": 8.091111111111111e-07,
|
| 1199 |
+
"loss": 0.0923,
|
| 1200 |
+
"step": 1690
|
| 1201 |
+
},
|
| 1202 |
+
{
|
| 1203 |
+
"epoch": 0.09,
|
| 1204 |
+
"grad_norm": 10.0,
|
| 1205 |
+
"learning_rate": 8.066666666666666e-07,
|
| 1206 |
+
"loss": 0.0992,
|
| 1207 |
+
"step": 1700
|
| 1208 |
+
},
|
| 1209 |
+
{
|
| 1210 |
+
"epoch": 0.09,
|
| 1211 |
+
"grad_norm": 9.0,
|
| 1212 |
+
"learning_rate": 8.042222222222223e-07,
|
| 1213 |
+
"loss": 0.0859,
|
| 1214 |
+
"step": 1710
|
| 1215 |
+
},
|
| 1216 |
+
{
|
| 1217 |
+
"epoch": 0.09,
|
| 1218 |
+
"grad_norm": 10.375,
|
| 1219 |
+
"learning_rate": 8.017777777777779e-07,
|
| 1220 |
+
"loss": 0.0862,
|
| 1221 |
+
"step": 1720
|
| 1222 |
+
},
|
| 1223 |
+
{
|
| 1224 |
+
"epoch": 0.09,
|
| 1225 |
+
"grad_norm": 8.125,
|
| 1226 |
+
"learning_rate": 7.993333333333334e-07,
|
| 1227 |
+
"loss": 0.0773,
|
| 1228 |
+
"step": 1730
|
| 1229 |
+
},
|
| 1230 |
+
{
|
| 1231 |
+
"epoch": 0.09,
|
| 1232 |
+
"grad_norm": 7.375,
|
| 1233 |
+
"learning_rate": 7.96888888888889e-07,
|
| 1234 |
+
"loss": 0.077,
|
| 1235 |
+
"step": 1740
|
| 1236 |
+
},
|
| 1237 |
+
{
|
| 1238 |
+
"epoch": 0.09,
|
| 1239 |
+
"grad_norm": 7.3125,
|
| 1240 |
+
"learning_rate": 7.944444444444445e-07,
|
| 1241 |
+
"loss": 0.0738,
|
| 1242 |
+
"step": 1750
|
| 1243 |
+
},
|
| 1244 |
+
{
|
| 1245 |
+
"epoch": 0.09,
|
| 1246 |
+
"grad_norm": 7.375,
|
| 1247 |
+
"learning_rate": 7.92e-07,
|
| 1248 |
+
"loss": 0.0703,
|
| 1249 |
+
"step": 1760
|
| 1250 |
+
},
|
| 1251 |
+
{
|
| 1252 |
+
"epoch": 0.09,
|
| 1253 |
+
"grad_norm": 8.125,
|
| 1254 |
+
"learning_rate": 7.895555555555555e-07,
|
| 1255 |
+
"loss": 0.0668,
|
| 1256 |
+
"step": 1770
|
| 1257 |
+
},
|
| 1258 |
+
{
|
| 1259 |
+
"epoch": 0.09,
|
| 1260 |
+
"grad_norm": 8.8125,
|
| 1261 |
+
"learning_rate": 7.87111111111111e-07,
|
| 1262 |
+
"loss": 0.0806,
|
| 1263 |
+
"step": 1780
|
| 1264 |
+
},
|
| 1265 |
+
{
|
| 1266 |
+
"epoch": 0.09,
|
| 1267 |
+
"grad_norm": 8.6875,
|
| 1268 |
+
"learning_rate": 7.846666666666668e-07,
|
| 1269 |
+
"loss": 0.0694,
|
| 1270 |
+
"step": 1790
|
| 1271 |
+
},
|
| 1272 |
+
{
|
| 1273 |
+
"epoch": 0.09,
|
| 1274 |
+
"grad_norm": 8.8125,
|
| 1275 |
+
"learning_rate": 7.822222222222223e-07,
|
| 1276 |
+
"loss": 0.0708,
|
| 1277 |
+
"step": 1800
|
| 1278 |
+
},
|
| 1279 |
+
{
|
| 1280 |
+
"epoch": 0.09,
|
| 1281 |
+
"grad_norm": 6.8125,
|
| 1282 |
+
"learning_rate": 7.797777777777778e-07,
|
| 1283 |
+
"loss": 0.0711,
|
| 1284 |
+
"step": 1810
|
| 1285 |
+
},
|
| 1286 |
+
{
|
| 1287 |
+
"epoch": 0.09,
|
| 1288 |
+
"grad_norm": 8.0625,
|
| 1289 |
+
"learning_rate": 7.773333333333334e-07,
|
| 1290 |
+
"loss": 0.0769,
|
| 1291 |
+
"step": 1820
|
| 1292 |
+
},
|
| 1293 |
+
{
|
| 1294 |
+
"epoch": 0.09,
|
| 1295 |
+
"grad_norm": 11.0625,
|
| 1296 |
+
"learning_rate": 7.748888888888889e-07,
|
| 1297 |
+
"loss": 0.0689,
|
| 1298 |
+
"step": 1830
|
| 1299 |
+
},
|
| 1300 |
+
{
|
| 1301 |
+
"epoch": 0.09,
|
| 1302 |
+
"grad_norm": 8.25,
|
| 1303 |
+
"learning_rate": 7.724444444444444e-07,
|
| 1304 |
+
"loss": 0.0651,
|
| 1305 |
+
"step": 1840
|
| 1306 |
+
},
|
| 1307 |
+
{
|
| 1308 |
+
"epoch": 0.09,
|
| 1309 |
+
"grad_norm": 7.9375,
|
| 1310 |
+
"learning_rate": 7.7e-07,
|
| 1311 |
+
"loss": 0.0731,
|
| 1312 |
+
"step": 1850
|
| 1313 |
+
},
|
| 1314 |
+
{
|
| 1315 |
+
"epoch": 0.09,
|
| 1316 |
+
"grad_norm": 6.65625,
|
| 1317 |
+
"learning_rate": 7.675555555555557e-07,
|
| 1318 |
+
"loss": 0.07,
|
| 1319 |
+
"step": 1860
|
| 1320 |
+
},
|
| 1321 |
+
{
|
| 1322 |
+
"epoch": 0.09,
|
| 1323 |
+
"grad_norm": 6.78125,
|
| 1324 |
+
"learning_rate": 7.651111111111112e-07,
|
| 1325 |
+
"loss": 0.0634,
|
| 1326 |
+
"step": 1870
|
| 1327 |
+
},
|
| 1328 |
+
{
|
| 1329 |
+
"epoch": 0.09,
|
| 1330 |
+
"grad_norm": 8.25,
|
| 1331 |
+
"learning_rate": 7.626666666666667e-07,
|
| 1332 |
+
"loss": 0.069,
|
| 1333 |
+
"step": 1880
|
| 1334 |
+
},
|
| 1335 |
+
{
|
| 1336 |
+
"epoch": 0.09,
|
| 1337 |
+
"grad_norm": 7.34375,
|
| 1338 |
+
"learning_rate": 7.602222222222222e-07,
|
| 1339 |
+
"loss": 0.0701,
|
| 1340 |
+
"step": 1890
|
| 1341 |
+
},
|
| 1342 |
+
{
|
| 1343 |
+
"epoch": 0.1,
|
| 1344 |
+
"grad_norm": 7.59375,
|
| 1345 |
+
"learning_rate": 7.577777777777778e-07,
|
| 1346 |
+
"loss": 0.0636,
|
| 1347 |
+
"step": 1900
|
| 1348 |
+
},
|
| 1349 |
+
{
|
| 1350 |
+
"epoch": 0.1,
|
| 1351 |
+
"grad_norm": 7.46875,
|
| 1352 |
+
"learning_rate": 7.553333333333334e-07,
|
| 1353 |
+
"loss": 0.059,
|
| 1354 |
+
"step": 1910
|
| 1355 |
+
},
|
| 1356 |
+
{
|
| 1357 |
+
"epoch": 0.1,
|
| 1358 |
+
"grad_norm": 8.625,
|
| 1359 |
+
"learning_rate": 7.528888888888889e-07,
|
| 1360 |
+
"loss": 0.063,
|
| 1361 |
+
"step": 1920
|
| 1362 |
+
},
|
| 1363 |
+
{
|
| 1364 |
+
"epoch": 0.1,
|
| 1365 |
+
"grad_norm": 7.8125,
|
| 1366 |
+
"learning_rate": 7.504444444444444e-07,
|
| 1367 |
+
"loss": 0.0612,
|
| 1368 |
+
"step": 1930
|
| 1369 |
+
},
|
| 1370 |
+
{
|
| 1371 |
+
"epoch": 0.1,
|
| 1372 |
+
"grad_norm": 8.125,
|
| 1373 |
+
"learning_rate": 7.480000000000001e-07,
|
| 1374 |
+
"loss": 0.0615,
|
| 1375 |
+
"step": 1940
|
| 1376 |
+
},
|
| 1377 |
+
{
|
| 1378 |
+
"epoch": 0.1,
|
| 1379 |
+
"grad_norm": 10.0625,
|
| 1380 |
+
"learning_rate": 7.455555555555556e-07,
|
| 1381 |
+
"loss": 0.059,
|
| 1382 |
+
"step": 1950
|
| 1383 |
+
},
|
| 1384 |
+
{
|
| 1385 |
+
"epoch": 0.1,
|
| 1386 |
+
"grad_norm": 6.09375,
|
| 1387 |
+
"learning_rate": 7.431111111111111e-07,
|
| 1388 |
+
"loss": 0.0644,
|
| 1389 |
+
"step": 1960
|
| 1390 |
+
},
|
| 1391 |
+
{
|
| 1392 |
+
"epoch": 0.1,
|
| 1393 |
+
"grad_norm": 7.0625,
|
| 1394 |
+
"learning_rate": 7.406666666666668e-07,
|
| 1395 |
+
"loss": 0.0658,
|
| 1396 |
+
"step": 1970
|
| 1397 |
+
},
|
| 1398 |
+
{
|
| 1399 |
+
"epoch": 0.1,
|
| 1400 |
+
"grad_norm": 5.875,
|
| 1401 |
+
"learning_rate": 7.382222222222223e-07,
|
| 1402 |
+
"loss": 0.059,
|
| 1403 |
+
"step": 1980
|
| 1404 |
+
},
|
| 1405 |
+
{
|
| 1406 |
+
"epoch": 0.1,
|
| 1407 |
+
"grad_norm": 6.84375,
|
| 1408 |
+
"learning_rate": 7.357777777777778e-07,
|
| 1409 |
+
"loss": 0.0617,
|
| 1410 |
+
"step": 1990
|
| 1411 |
+
},
|
| 1412 |
+
{
|
| 1413 |
+
"epoch": 0.1,
|
| 1414 |
+
"grad_norm": 6.3125,
|
| 1415 |
+
"learning_rate": 7.333333333333333e-07,
|
| 1416 |
+
"loss": 0.0613,
|
| 1417 |
+
"step": 2000
|
| 1418 |
+
},
|
| 1419 |
+
{
|
| 1420 |
+
"epoch": 0.1,
|
| 1421 |
+
"eval_loss": 0.06299971044063568,
|
| 1422 |
+
"eval_runtime": 64.6734,
|
| 1423 |
+
"eval_samples_per_second": 15.462,
|
| 1424 |
+
"eval_steps_per_second": 15.462,
|
| 1425 |
+
"step": 2000
|
| 1426 |
}
|
| 1427 |
],
|
| 1428 |
"logging_steps": 10,
|
|
|
|
| 1430 |
"num_input_tokens_seen": 0,
|
| 1431 |
"num_train_epochs": 1,
|
| 1432 |
"save_steps": 1000,
|
| 1433 |
+
"total_flos": 3.227844083712e+16,
|
| 1434 |
"train_batch_size": 1,
|
| 1435 |
"trial_name": null,
|
| 1436 |
"trial_params": null
|