Training in progress, step 200, checkpoint
Browse files- last-checkpoint/model.safetensors +1 -1
- last-checkpoint/optimizer.pt +1 -1
- last-checkpoint/rng_state_0.pth +1 -1
- last-checkpoint/rng_state_1.pth +1 -1
- last-checkpoint/rng_state_2.pth +1 -1
- last-checkpoint/rng_state_3.pth +1 -1
- last-checkpoint/scheduler.pt +1 -1
- last-checkpoint/trainer_state.json +710 -3
last-checkpoint/model.safetensors
CHANGED
|
@@ -1,3 +1,3 @@
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
-
oid sha256:
|
| 3 |
size 373077376
|
|
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:5fbd2e60526489e6f6f39cf288ada5ee34355bdd563beb7e5399d0ac622a5c3e
|
| 3 |
size 373077376
|
last-checkpoint/optimizer.pt
CHANGED
|
@@ -1,3 +1,3 @@
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
-
oid sha256:
|
| 3 |
size 422377867
|
|
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:3f6e8dff52c08eae75b95c752c0f6f01bfbb1ed09c5bce28a0e4593cda5e5c80
|
| 3 |
size 422377867
|
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 15365
|
|
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:ac0e770c61e3acef7bd144efee65a450ed487b244ed293a8dd801394ffcae775
|
| 3 |
size 15365
|
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 15365
|
|
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:21354c9980d6ba769d430eb9962ffba457eb04a66cb90e33b0934d3157fae7cf
|
| 3 |
size 15365
|
last-checkpoint/rng_state_2.pth
CHANGED
|
@@ -1,3 +1,3 @@
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
-
oid sha256:
|
| 3 |
size 15365
|
|
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:d574a8567b1636a399633d2b20273fdb3be4b888e37f0d331cd695b973f10463
|
| 3 |
size 15365
|
last-checkpoint/rng_state_3.pth
CHANGED
|
@@ -1,3 +1,3 @@
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
-
oid sha256:
|
| 3 |
size 15365
|
|
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:cca2b80df90b78b3ba9d9d2951c5b7aa80fc0a44b4512b495de6de40396fcb6e
|
| 3 |
size 15365
|
last-checkpoint/scheduler.pt
CHANGED
|
@@ -1,3 +1,3 @@
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
-
oid sha256:
|
| 3 |
size 1401
|
|
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:39a79e2827280868096ff650d0ee6e4723dddc824192c37a937a656d54903350
|
| 3 |
size 1401
|
last-checkpoint/trainer_state.json
CHANGED
|
@@ -2,9 +2,9 @@
|
|
| 2 |
"best_global_step": null,
|
| 3 |
"best_metric": null,
|
| 4 |
"best_model_checkpoint": null,
|
| 5 |
-
"epoch": 0.
|
| 6 |
"eval_steps": 100,
|
| 7 |
-
"global_step":
|
| 8 |
"is_hyper_param_search": false,
|
| 9 |
"is_local_process_zero": true,
|
| 10 |
"is_world_process_zero": true,
|
|
@@ -715,6 +715,713 @@
|
|
| 715 |
"eval_samples_per_second": 17.321,
|
| 716 |
"eval_steps_per_second": 1.083,
|
| 717 |
"step": 100
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 718 |
}
|
| 719 |
],
|
| 720 |
"logging_steps": 1,
|
|
@@ -734,7 +1441,7 @@
|
|
| 734 |
"attributes": {}
|
| 735 |
}
|
| 736 |
},
|
| 737 |
-
"total_flos":
|
| 738 |
"train_batch_size": 16,
|
| 739 |
"trial_name": null,
|
| 740 |
"trial_params": null
|
|
|
|
| 2 |
"best_global_step": null,
|
| 3 |
"best_metric": null,
|
| 4 |
"best_model_checkpoint": null,
|
| 5 |
+
"epoch": 0.8333333333333334,
|
| 6 |
"eval_steps": 100,
|
| 7 |
+
"global_step": 200,
|
| 8 |
"is_hyper_param_search": false,
|
| 9 |
"is_local_process_zero": true,
|
| 10 |
"is_world_process_zero": true,
|
|
|
|
| 715 |
"eval_samples_per_second": 17.321,
|
| 716 |
"eval_steps_per_second": 1.083,
|
| 717 |
"step": 100
|
| 718 |
+
},
|
| 719 |
+
{
|
| 720 |
+
"epoch": 0.42083333333333334,
|
| 721 |
+
"grad_norm": 0.466796875,
|
| 722 |
+
"learning_rate": 0.0006753187775963773,
|
| 723 |
+
"loss": 6.1547,
|
| 724 |
+
"step": 101
|
| 725 |
+
},
|
| 726 |
+
{
|
| 727 |
+
"epoch": 0.425,
|
| 728 |
+
"grad_norm": 0.443359375,
|
| 729 |
+
"learning_rate": 0.0006688502846768696,
|
| 730 |
+
"loss": 6.1169,
|
| 731 |
+
"step": 102
|
| 732 |
+
},
|
| 733 |
+
{
|
| 734 |
+
"epoch": 0.42916666666666664,
|
| 735 |
+
"grad_norm": 0.578125,
|
| 736 |
+
"learning_rate": 0.0006623497346023419,
|
| 737 |
+
"loss": 5.9032,
|
| 738 |
+
"step": 103
|
| 739 |
+
},
|
| 740 |
+
{
|
| 741 |
+
"epoch": 0.43333333333333335,
|
| 742 |
+
"grad_norm": 0.515625,
|
| 743 |
+
"learning_rate": 0.0006558183615379707,
|
| 744 |
+
"loss": 5.9968,
|
| 745 |
+
"step": 104
|
| 746 |
+
},
|
| 747 |
+
{
|
| 748 |
+
"epoch": 0.4375,
|
| 749 |
+
"grad_norm": 0.498046875,
|
| 750 |
+
"learning_rate": 0.0006492574055008473,
|
| 751 |
+
"loss": 6.0527,
|
| 752 |
+
"step": 105
|
| 753 |
+
},
|
| 754 |
+
{
|
| 755 |
+
"epoch": 0.44166666666666665,
|
| 756 |
+
"grad_norm": 0.45703125,
|
| 757 |
+
"learning_rate": 0.0006426681121245527,
|
| 758 |
+
"loss": 5.9248,
|
| 759 |
+
"step": 106
|
| 760 |
+
},
|
| 761 |
+
{
|
| 762 |
+
"epoch": 0.44583333333333336,
|
| 763 |
+
"grad_norm": 0.416015625,
|
| 764 |
+
"learning_rate": 0.0006360517324226675,
|
| 765 |
+
"loss": 5.8954,
|
| 766 |
+
"step": 107
|
| 767 |
+
},
|
| 768 |
+
{
|
| 769 |
+
"epoch": 0.45,
|
| 770 |
+
"grad_norm": 0.482421875,
|
| 771 |
+
"learning_rate": 0.0006294095225512603,
|
| 772 |
+
"loss": 5.7513,
|
| 773 |
+
"step": 108
|
| 774 |
+
},
|
| 775 |
+
{
|
| 776 |
+
"epoch": 0.45416666666666666,
|
| 777 |
+
"grad_norm": 0.388671875,
|
| 778 |
+
"learning_rate": 0.0006227427435703996,
|
| 779 |
+
"loss": 5.9773,
|
| 780 |
+
"step": 109
|
| 781 |
+
},
|
| 782 |
+
{
|
| 783 |
+
"epoch": 0.4583333333333333,
|
| 784 |
+
"grad_norm": 0.427734375,
|
| 785 |
+
"learning_rate": 0.0006160526612047339,
|
| 786 |
+
"loss": 5.8157,
|
| 787 |
+
"step": 110
|
| 788 |
+
},
|
| 789 |
+
{
|
| 790 |
+
"epoch": 0.4625,
|
| 791 |
+
"grad_norm": 0.484375,
|
| 792 |
+
"learning_rate": 0.0006093405456031879,
|
| 793 |
+
"loss": 5.9297,
|
| 794 |
+
"step": 111
|
| 795 |
+
},
|
| 796 |
+
{
|
| 797 |
+
"epoch": 0.4666666666666667,
|
| 798 |
+
"grad_norm": 0.40625,
|
| 799 |
+
"learning_rate": 0.0006026076710978171,
|
| 800 |
+
"loss": 5.9629,
|
| 801 |
+
"step": 112
|
| 802 |
+
},
|
| 803 |
+
{
|
| 804 |
+
"epoch": 0.4708333333333333,
|
| 805 |
+
"grad_norm": 0.40625,
|
| 806 |
+
"learning_rate": 0.0005958553159618693,
|
| 807 |
+
"loss": 5.9699,
|
| 808 |
+
"step": 113
|
| 809 |
+
},
|
| 810 |
+
{
|
| 811 |
+
"epoch": 0.475,
|
| 812 |
+
"grad_norm": 0.44140625,
|
| 813 |
+
"learning_rate": 0.0005890847621670966,
|
| 814 |
+
"loss": 5.9486,
|
| 815 |
+
"step": 114
|
| 816 |
+
},
|
| 817 |
+
{
|
| 818 |
+
"epoch": 0.4791666666666667,
|
| 819 |
+
"grad_norm": 0.365234375,
|
| 820 |
+
"learning_rate": 0.000582297295140367,
|
| 821 |
+
"loss": 5.9959,
|
| 822 |
+
"step": 115
|
| 823 |
+
},
|
| 824 |
+
{
|
| 825 |
+
"epoch": 0.48333333333333334,
|
| 826 |
+
"grad_norm": 0.3671875,
|
| 827 |
+
"learning_rate": 0.0005754942035196184,
|
| 828 |
+
"loss": 6.011,
|
| 829 |
+
"step": 116
|
| 830 |
+
},
|
| 831 |
+
{
|
| 832 |
+
"epoch": 0.4875,
|
| 833 |
+
"grad_norm": 0.3984375,
|
| 834 |
+
"learning_rate": 0.0005686767789092041,
|
| 835 |
+
"loss": 5.7683,
|
| 836 |
+
"step": 117
|
| 837 |
+
},
|
| 838 |
+
{
|
| 839 |
+
"epoch": 0.49166666666666664,
|
| 840 |
+
"grad_norm": 0.3125,
|
| 841 |
+
"learning_rate": 0.000561846315634674,
|
| 842 |
+
"loss": 5.9579,
|
| 843 |
+
"step": 118
|
| 844 |
+
},
|
| 845 |
+
{
|
| 846 |
+
"epoch": 0.49583333333333335,
|
| 847 |
+
"grad_norm": 0.376953125,
|
| 848 |
+
"learning_rate": 0.0005550041104970397,
|
| 849 |
+
"loss": 5.8793,
|
| 850 |
+
"step": 119
|
| 851 |
+
},
|
| 852 |
+
{
|
| 853 |
+
"epoch": 0.5,
|
| 854 |
+
"grad_norm": 0.44921875,
|
| 855 |
+
"learning_rate": 0.0005481514625265708,
|
| 856 |
+
"loss": 5.7256,
|
| 857 |
+
"step": 120
|
| 858 |
+
},
|
| 859 |
+
{
|
| 860 |
+
"epoch": 0.5041666666666667,
|
| 861 |
+
"grad_norm": 0.44921875,
|
| 862 |
+
"learning_rate": 0.0005412896727361663,
|
| 863 |
+
"loss": 5.8975,
|
| 864 |
+
"step": 121
|
| 865 |
+
},
|
| 866 |
+
{
|
| 867 |
+
"epoch": 0.5083333333333333,
|
| 868 |
+
"grad_norm": 0.46875,
|
| 869 |
+
"learning_rate": 0.0005344200438743488,
|
| 870 |
+
"loss": 5.959,
|
| 871 |
+
"step": 122
|
| 872 |
+
},
|
| 873 |
+
{
|
| 874 |
+
"epoch": 0.5125,
|
| 875 |
+
"grad_norm": 0.400390625,
|
| 876 |
+
"learning_rate": 0.0005275438801779327,
|
| 877 |
+
"loss": 5.8133,
|
| 878 |
+
"step": 123
|
| 879 |
+
},
|
| 880 |
+
{
|
| 881 |
+
"epoch": 0.5166666666666667,
|
| 882 |
+
"grad_norm": 0.37890625,
|
| 883 |
+
"learning_rate": 0.0005206624871244065,
|
| 884 |
+
"loss": 5.8534,
|
| 885 |
+
"step": 124
|
| 886 |
+
},
|
| 887 |
+
{
|
| 888 |
+
"epoch": 0.5208333333333334,
|
| 889 |
+
"grad_norm": 1.1796875,
|
| 890 |
+
"learning_rate": 0.0005137771711840811,
|
| 891 |
+
"loss": 5.8988,
|
| 892 |
+
"step": 125
|
| 893 |
+
},
|
| 894 |
+
{
|
| 895 |
+
"epoch": 0.525,
|
| 896 |
+
"grad_norm": 0.35546875,
|
| 897 |
+
"learning_rate": 0.0005068892395720482,
|
| 898 |
+
"loss": 6.0026,
|
| 899 |
+
"step": 126
|
| 900 |
+
},
|
| 901 |
+
{
|
| 902 |
+
"epoch": 0.5291666666666667,
|
| 903 |
+
"grad_norm": 0.443359375,
|
| 904 |
+
"learning_rate": 0.0005,
|
| 905 |
+
"loss": 5.8737,
|
| 906 |
+
"step": 127
|
| 907 |
+
},
|
| 908 |
+
{
|
| 909 |
+
"epoch": 0.5333333333333333,
|
| 910 |
+
"grad_norm": 0.5234375,
|
| 911 |
+
"learning_rate": 0.0004931107604279518,
|
| 912 |
+
"loss": 5.7772,
|
| 913 |
+
"step": 128
|
| 914 |
+
},
|
| 915 |
+
{
|
| 916 |
+
"epoch": 0.5375,
|
| 917 |
+
"grad_norm": 0.421875,
|
| 918 |
+
"learning_rate": 0.0004862228288159191,
|
| 919 |
+
"loss": 5.8457,
|
| 920 |
+
"step": 129
|
| 921 |
+
},
|
| 922 |
+
{
|
| 923 |
+
"epoch": 0.5416666666666666,
|
| 924 |
+
"grad_norm": 0.50390625,
|
| 925 |
+
"learning_rate": 0.0004793375128755934,
|
| 926 |
+
"loss": 5.7298,
|
| 927 |
+
"step": 130
|
| 928 |
+
},
|
| 929 |
+
{
|
| 930 |
+
"epoch": 0.5458333333333333,
|
| 931 |
+
"grad_norm": 0.4375,
|
| 932 |
+
"learning_rate": 0.0004724561198220672,
|
| 933 |
+
"loss": 6.0151,
|
| 934 |
+
"step": 131
|
| 935 |
+
},
|
| 936 |
+
{
|
| 937 |
+
"epoch": 0.55,
|
| 938 |
+
"grad_norm": 0.40234375,
|
| 939 |
+
"learning_rate": 0.0004655799561256514,
|
| 940 |
+
"loss": 5.7457,
|
| 941 |
+
"step": 132
|
| 942 |
+
},
|
| 943 |
+
{
|
| 944 |
+
"epoch": 0.5541666666666667,
|
| 945 |
+
"grad_norm": 0.470703125,
|
| 946 |
+
"learning_rate": 0.0004587103272638339,
|
| 947 |
+
"loss": 5.7534,
|
| 948 |
+
"step": 133
|
| 949 |
+
},
|
| 950 |
+
{
|
| 951 |
+
"epoch": 0.5583333333333333,
|
| 952 |
+
"grad_norm": 0.486328125,
|
| 953 |
+
"learning_rate": 0.0004518485374734292,
|
| 954 |
+
"loss": 5.7571,
|
| 955 |
+
"step": 134
|
| 956 |
+
},
|
| 957 |
+
{
|
| 958 |
+
"epoch": 0.5625,
|
| 959 |
+
"grad_norm": 0.40234375,
|
| 960 |
+
"learning_rate": 0.0004449958895029604,
|
| 961 |
+
"loss": 5.7288,
|
| 962 |
+
"step": 135
|
| 963 |
+
},
|
| 964 |
+
{
|
| 965 |
+
"epoch": 0.5666666666666667,
|
| 966 |
+
"grad_norm": 0.41015625,
|
| 967 |
+
"learning_rate": 0.00043815368436532617,
|
| 968 |
+
"loss": 5.8647,
|
| 969 |
+
"step": 136
|
| 970 |
+
},
|
| 971 |
+
{
|
| 972 |
+
"epoch": 0.5708333333333333,
|
| 973 |
+
"grad_norm": 0.37109375,
|
| 974 |
+
"learning_rate": 0.0004313232210907959,
|
| 975 |
+
"loss": 5.8594,
|
| 976 |
+
"step": 137
|
| 977 |
+
},
|
| 978 |
+
{
|
| 979 |
+
"epoch": 0.575,
|
| 980 |
+
"grad_norm": 0.38671875,
|
| 981 |
+
"learning_rate": 0.00042450579648038153,
|
| 982 |
+
"loss": 5.7164,
|
| 983 |
+
"step": 138
|
| 984 |
+
},
|
| 985 |
+
{
|
| 986 |
+
"epoch": 0.5791666666666667,
|
| 987 |
+
"grad_norm": 0.40234375,
|
| 988 |
+
"learning_rate": 0.00041770270485963295,
|
| 989 |
+
"loss": 5.7606,
|
| 990 |
+
"step": 139
|
| 991 |
+
},
|
| 992 |
+
{
|
| 993 |
+
"epoch": 0.5833333333333334,
|
| 994 |
+
"grad_norm": 0.37109375,
|
| 995 |
+
"learning_rate": 0.0004109152378329036,
|
| 996 |
+
"loss": 5.7807,
|
| 997 |
+
"step": 140
|
| 998 |
+
},
|
| 999 |
+
{
|
| 1000 |
+
"epoch": 0.5875,
|
| 1001 |
+
"grad_norm": 0.44921875,
|
| 1002 |
+
"learning_rate": 0.00040414468403813093,
|
| 1003 |
+
"loss": 5.7903,
|
| 1004 |
+
"step": 141
|
| 1005 |
+
},
|
| 1006 |
+
{
|
| 1007 |
+
"epoch": 0.5916666666666667,
|
| 1008 |
+
"grad_norm": 0.3046875,
|
| 1009 |
+
"learning_rate": 0.00039739232890218293,
|
| 1010 |
+
"loss": 5.6957,
|
| 1011 |
+
"step": 142
|
| 1012 |
+
},
|
| 1013 |
+
{
|
| 1014 |
+
"epoch": 0.5958333333333333,
|
| 1015 |
+
"grad_norm": 0.361328125,
|
| 1016 |
+
"learning_rate": 0.00039065945439681213,
|
| 1017 |
+
"loss": 5.6413,
|
| 1018 |
+
"step": 143
|
| 1019 |
+
},
|
| 1020 |
+
{
|
| 1021 |
+
"epoch": 0.6,
|
| 1022 |
+
"grad_norm": 0.384765625,
|
| 1023 |
+
"learning_rate": 0.0003839473387952662,
|
| 1024 |
+
"loss": 5.6618,
|
| 1025 |
+
"step": 144
|
| 1026 |
+
},
|
| 1027 |
+
{
|
| 1028 |
+
"epoch": 0.6041666666666666,
|
| 1029 |
+
"grad_norm": 0.392578125,
|
| 1030 |
+
"learning_rate": 0.00037725725642960046,
|
| 1031 |
+
"loss": 5.7087,
|
| 1032 |
+
"step": 145
|
| 1033 |
+
},
|
| 1034 |
+
{
|
| 1035 |
+
"epoch": 0.6083333333333333,
|
| 1036 |
+
"grad_norm": 0.46875,
|
| 1037 |
+
"learning_rate": 0.0003705904774487396,
|
| 1038 |
+
"loss": 5.5071,
|
| 1039 |
+
"step": 146
|
| 1040 |
+
},
|
| 1041 |
+
{
|
| 1042 |
+
"epoch": 0.6125,
|
| 1043 |
+
"grad_norm": 0.330078125,
|
| 1044 |
+
"learning_rate": 0.0003639482675773324,
|
| 1045 |
+
"loss": 5.6133,
|
| 1046 |
+
"step": 147
|
| 1047 |
+
},
|
| 1048 |
+
{
|
| 1049 |
+
"epoch": 0.6166666666666667,
|
| 1050 |
+
"grad_norm": 0.328125,
|
| 1051 |
+
"learning_rate": 0.00035733188787544744,
|
| 1052 |
+
"loss": 5.5889,
|
| 1053 |
+
"step": 148
|
| 1054 |
+
},
|
| 1055 |
+
{
|
| 1056 |
+
"epoch": 0.6208333333333333,
|
| 1057 |
+
"grad_norm": 0.341796875,
|
| 1058 |
+
"learning_rate": 0.00035074259449915284,
|
| 1059 |
+
"loss": 5.6028,
|
| 1060 |
+
"step": 149
|
| 1061 |
+
},
|
| 1062 |
+
{
|
| 1063 |
+
"epoch": 0.625,
|
| 1064 |
+
"grad_norm": 0.30078125,
|
| 1065 |
+
"learning_rate": 0.00034418163846202944,
|
| 1066 |
+
"loss": 5.8408,
|
| 1067 |
+
"step": 150
|
| 1068 |
+
},
|
| 1069 |
+
{
|
| 1070 |
+
"epoch": 0.6291666666666667,
|
| 1071 |
+
"grad_norm": 0.306640625,
|
| 1072 |
+
"learning_rate": 0.0003376502653976583,
|
| 1073 |
+
"loss": 5.8503,
|
| 1074 |
+
"step": 151
|
| 1075 |
+
},
|
| 1076 |
+
{
|
| 1077 |
+
"epoch": 0.6333333333333333,
|
| 1078 |
+
"grad_norm": 0.359375,
|
| 1079 |
+
"learning_rate": 0.0003311497153231305,
|
| 1080 |
+
"loss": 5.6345,
|
| 1081 |
+
"step": 152
|
| 1082 |
+
},
|
| 1083 |
+
{
|
| 1084 |
+
"epoch": 0.6375,
|
| 1085 |
+
"grad_norm": 0.365234375,
|
| 1086 |
+
"learning_rate": 0.00032468122240362287,
|
| 1087 |
+
"loss": 5.5636,
|
| 1088 |
+
"step": 153
|
| 1089 |
+
},
|
| 1090 |
+
{
|
| 1091 |
+
"epoch": 0.6416666666666667,
|
| 1092 |
+
"grad_norm": 0.30078125,
|
| 1093 |
+
"learning_rate": 0.000318246014718085,
|
| 1094 |
+
"loss": 5.7088,
|
| 1095 |
+
"step": 154
|
| 1096 |
+
},
|
| 1097 |
+
{
|
| 1098 |
+
"epoch": 0.6458333333333334,
|
| 1099 |
+
"grad_norm": 0.33984375,
|
| 1100 |
+
"learning_rate": 0.0003118453140260823,
|
| 1101 |
+
"loss": 5.7296,
|
| 1102 |
+
"step": 155
|
| 1103 |
+
},
|
| 1104 |
+
{
|
| 1105 |
+
"epoch": 0.65,
|
| 1106 |
+
"grad_norm": 0.275390625,
|
| 1107 |
+
"learning_rate": 0.000305480335535837,
|
| 1108 |
+
"loss": 5.7223,
|
| 1109 |
+
"step": 156
|
| 1110 |
+
},
|
| 1111 |
+
{
|
| 1112 |
+
"epoch": 0.6541666666666667,
|
| 1113 |
+
"grad_norm": 0.30859375,
|
| 1114 |
+
"learning_rate": 0.0002991522876735154,
|
| 1115 |
+
"loss": 5.5813,
|
| 1116 |
+
"step": 157
|
| 1117 |
+
},
|
| 1118 |
+
{
|
| 1119 |
+
"epoch": 0.6583333333333333,
|
| 1120 |
+
"grad_norm": 0.30859375,
|
| 1121 |
+
"learning_rate": 0.0002928623718538006,
|
| 1122 |
+
"loss": 5.6656,
|
| 1123 |
+
"step": 158
|
| 1124 |
+
},
|
| 1125 |
+
{
|
| 1126 |
+
"epoch": 0.6625,
|
| 1127 |
+
"grad_norm": 0.4140625,
|
| 1128 |
+
"learning_rate": 0.0002866117822517982,
|
| 1129 |
+
"loss": 5.5557,
|
| 1130 |
+
"step": 159
|
| 1131 |
+
},
|
| 1132 |
+
{
|
| 1133 |
+
"epoch": 0.6666666666666666,
|
| 1134 |
+
"grad_norm": 0.5234375,
|
| 1135 |
+
"learning_rate": 0.0002804017055763149,
|
| 1136 |
+
"loss": 5.6746,
|
| 1137 |
+
"step": 160
|
| 1138 |
+
},
|
| 1139 |
+
{
|
| 1140 |
+
"epoch": 0.6708333333333333,
|
| 1141 |
+
"grad_norm": 0.345703125,
|
| 1142 |
+
"learning_rate": 0.00027423332084455543,
|
| 1143 |
+
"loss": 5.4792,
|
| 1144 |
+
"step": 161
|
| 1145 |
+
},
|
| 1146 |
+
{
|
| 1147 |
+
"epoch": 0.675,
|
| 1148 |
+
"grad_norm": 0.271484375,
|
| 1149 |
+
"learning_rate": 0.0002681077991582797,
|
| 1150 |
+
"loss": 5.6576,
|
| 1151 |
+
"step": 162
|
| 1152 |
+
},
|
| 1153 |
+
{
|
| 1154 |
+
"epoch": 0.6791666666666667,
|
| 1155 |
+
"grad_norm": 0.33984375,
|
| 1156 |
+
"learning_rate": 0.0002620263034814632,
|
| 1157 |
+
"loss": 5.6155,
|
| 1158 |
+
"step": 163
|
| 1159 |
+
},
|
| 1160 |
+
{
|
| 1161 |
+
"epoch": 0.6833333333333333,
|
| 1162 |
+
"grad_norm": 0.255859375,
|
| 1163 |
+
"learning_rate": 0.00025598998841950106,
|
| 1164 |
+
"loss": 5.8009,
|
| 1165 |
+
"step": 164
|
| 1166 |
+
},
|
| 1167 |
+
{
|
| 1168 |
+
"epoch": 0.6875,
|
| 1169 |
+
"grad_norm": 0.33984375,
|
| 1170 |
+
"learning_rate": 0.0002500000000000001,
|
| 1171 |
+
"loss": 5.6332,
|
| 1172 |
+
"step": 165
|
| 1173 |
+
},
|
| 1174 |
+
{
|
| 1175 |
+
"epoch": 0.6916666666666667,
|
| 1176 |
+
"grad_norm": 0.322265625,
|
| 1177 |
+
"learning_rate": 0.00024405747545519962,
|
| 1178 |
+
"loss": 5.3542,
|
| 1179 |
+
"step": 166
|
| 1180 |
+
},
|
| 1181 |
+
{
|
| 1182 |
+
"epoch": 0.6958333333333333,
|
| 1183 |
+
"grad_norm": 0.259765625,
|
| 1184 |
+
"learning_rate": 0.00023816354300606107,
|
| 1185 |
+
"loss": 5.6408,
|
| 1186 |
+
"step": 167
|
| 1187 |
+
},
|
| 1188 |
+
{
|
| 1189 |
+
"epoch": 0.7,
|
| 1190 |
+
"grad_norm": 0.328125,
|
| 1191 |
+
"learning_rate": 0.0002323193216480698,
|
| 1192 |
+
"loss": 5.6203,
|
| 1193 |
+
"step": 168
|
| 1194 |
+
},
|
| 1195 |
+
{
|
| 1196 |
+
"epoch": 0.7041666666666667,
|
| 1197 |
+
"grad_norm": 0.25,
|
| 1198 |
+
"learning_rate": 0.00022652592093878665,
|
| 1199 |
+
"loss": 5.6601,
|
| 1200 |
+
"step": 169
|
| 1201 |
+
},
|
| 1202 |
+
{
|
| 1203 |
+
"epoch": 0.7083333333333334,
|
| 1204 |
+
"grad_norm": 0.251953125,
|
| 1205 |
+
"learning_rate": 0.0002207844407871929,
|
| 1206 |
+
"loss": 5.8144,
|
| 1207 |
+
"step": 170
|
| 1208 |
+
},
|
| 1209 |
+
{
|
| 1210 |
+
"epoch": 0.7125,
|
| 1211 |
+
"grad_norm": 0.29296875,
|
| 1212 |
+
"learning_rate": 0.0002150959712448669,
|
| 1213 |
+
"loss": 5.5789,
|
| 1214 |
+
"step": 171
|
| 1215 |
+
},
|
| 1216 |
+
{
|
| 1217 |
+
"epoch": 0.7166666666666667,
|
| 1218 |
+
"grad_norm": 0.2431640625,
|
| 1219 |
+
"learning_rate": 0.0002094615922990309,
|
| 1220 |
+
"loss": 5.6727,
|
| 1221 |
+
"step": 172
|
| 1222 |
+
},
|
| 1223 |
+
{
|
| 1224 |
+
"epoch": 0.7208333333333333,
|
| 1225 |
+
"grad_norm": 0.29296875,
|
| 1226 |
+
"learning_rate": 0.00020388237366751006,
|
| 1227 |
+
"loss": 5.4561,
|
| 1228 |
+
"step": 173
|
| 1229 |
+
},
|
| 1230 |
+
{
|
| 1231 |
+
"epoch": 0.725,
|
| 1232 |
+
"grad_norm": 0.2734375,
|
| 1233 |
+
"learning_rate": 0.00019835937459564064,
|
| 1234 |
+
"loss": 5.7376,
|
| 1235 |
+
"step": 174
|
| 1236 |
+
},
|
| 1237 |
+
{
|
| 1238 |
+
"epoch": 0.7291666666666666,
|
| 1239 |
+
"grad_norm": 0.27734375,
|
| 1240 |
+
"learning_rate": 0.00019289364365516608,
|
| 1241 |
+
"loss": 5.5645,
|
| 1242 |
+
"step": 175
|
| 1243 |
+
},
|
| 1244 |
+
{
|
| 1245 |
+
"epoch": 0.7333333333333333,
|
| 1246 |
+
"grad_norm": 0.322265625,
|
| 1247 |
+
"learning_rate": 0.00018748621854516078,
|
| 1248 |
+
"loss": 5.4465,
|
| 1249 |
+
"step": 176
|
| 1250 |
+
},
|
| 1251 |
+
{
|
| 1252 |
+
"epoch": 0.7375,
|
| 1253 |
+
"grad_norm": 0.275390625,
|
| 1254 |
+
"learning_rate": 0.0001821381258950161,
|
| 1255 |
+
"loss": 5.6323,
|
| 1256 |
+
"step": 177
|
| 1257 |
+
},
|
| 1258 |
+
{
|
| 1259 |
+
"epoch": 0.7416666666666667,
|
| 1260 |
+
"grad_norm": 0.279296875,
|
| 1261 |
+
"learning_rate": 0.0001768503810695295,
|
| 1262 |
+
"loss": 5.7042,
|
| 1263 |
+
"step": 178
|
| 1264 |
+
},
|
| 1265 |
+
{
|
| 1266 |
+
"epoch": 0.7458333333333333,
|
| 1267 |
+
"grad_norm": 0.3125,
|
| 1268 |
+
"learning_rate": 0.00017162398797613282,
|
| 1269 |
+
"loss": 5.549,
|
| 1270 |
+
"step": 179
|
| 1271 |
+
},
|
| 1272 |
+
{
|
| 1273 |
+
"epoch": 0.75,
|
| 1274 |
+
"grad_norm": 0.232421875,
|
| 1275 |
+
"learning_rate": 0.00016645993887429345,
|
| 1276 |
+
"loss": 5.7366,
|
| 1277 |
+
"step": 180
|
| 1278 |
+
},
|
| 1279 |
+
{
|
| 1280 |
+
"epoch": 0.7541666666666667,
|
| 1281 |
+
"grad_norm": 0.30078125,
|
| 1282 |
+
"learning_rate": 0.00016135921418712956,
|
| 1283 |
+
"loss": 5.7234,
|
| 1284 |
+
"step": 181
|
| 1285 |
+
},
|
| 1286 |
+
{
|
| 1287 |
+
"epoch": 0.7583333333333333,
|
| 1288 |
+
"grad_norm": 0.380859375,
|
| 1289 |
+
"learning_rate": 0.0001563227823152708,
|
| 1290 |
+
"loss": 5.4976,
|
| 1291 |
+
"step": 182
|
| 1292 |
+
},
|
| 1293 |
+
{
|
| 1294 |
+
"epoch": 0.7625,
|
| 1295 |
+
"grad_norm": 0.2890625,
|
| 1296 |
+
"learning_rate": 0.0001513515994530023,
|
| 1297 |
+
"loss": 5.4723,
|
| 1298 |
+
"step": 183
|
| 1299 |
+
},
|
| 1300 |
+
{
|
| 1301 |
+
"epoch": 0.7666666666666667,
|
| 1302 |
+
"grad_norm": 0.28125,
|
| 1303 |
+
"learning_rate": 0.00014644660940672628,
|
| 1304 |
+
"loss": 5.5005,
|
| 1305 |
+
"step": 184
|
| 1306 |
+
},
|
| 1307 |
+
{
|
| 1308 |
+
"epoch": 0.7708333333333334,
|
| 1309 |
+
"grad_norm": 0.291015625,
|
| 1310 |
+
"learning_rate": 0.00014160874341577446,
|
| 1311 |
+
"loss": 5.4176,
|
| 1312 |
+
"step": 185
|
| 1313 |
+
},
|
| 1314 |
+
{
|
| 1315 |
+
"epoch": 0.775,
|
| 1316 |
+
"grad_norm": 0.4296875,
|
| 1317 |
+
"learning_rate": 0.0001368389199756075,
|
| 1318 |
+
"loss": 5.5741,
|
| 1319 |
+
"step": 186
|
| 1320 |
+
},
|
| 1321 |
+
{
|
| 1322 |
+
"epoch": 0.7791666666666667,
|
| 1323 |
+
"grad_norm": 0.34375,
|
| 1324 |
+
"learning_rate": 0.0001321380446634342,
|
| 1325 |
+
"loss": 5.2794,
|
| 1326 |
+
"step": 187
|
| 1327 |
+
},
|
| 1328 |
+
{
|
| 1329 |
+
"epoch": 0.7833333333333333,
|
| 1330 |
+
"grad_norm": 0.3046875,
|
| 1331 |
+
"learning_rate": 0.0001275070099662815,
|
| 1332 |
+
"loss": 5.5068,
|
| 1333 |
+
"step": 188
|
| 1334 |
+
},
|
| 1335 |
+
{
|
| 1336 |
+
"epoch": 0.7875,
|
| 1337 |
+
"grad_norm": 0.294921875,
|
| 1338 |
+
"learning_rate": 0.00012294669511155192,
|
| 1339 |
+
"loss": 5.614,
|
| 1340 |
+
"step": 189
|
| 1341 |
+
},
|
| 1342 |
+
{
|
| 1343 |
+
"epoch": 0.7916666666666666,
|
| 1344 |
+
"grad_norm": 0.3125,
|
| 1345 |
+
"learning_rate": 0.00011845796590009682,
|
| 1346 |
+
"loss": 5.6004,
|
| 1347 |
+
"step": 190
|
| 1348 |
+
},
|
| 1349 |
+
{
|
| 1350 |
+
"epoch": 0.7958333333333333,
|
| 1351 |
+
"grad_norm": 0.26953125,
|
| 1352 |
+
"learning_rate": 0.00011404167454183955,
|
| 1353 |
+
"loss": 5.5943,
|
| 1354 |
+
"step": 191
|
| 1355 |
+
},
|
| 1356 |
+
{
|
| 1357 |
+
"epoch": 0.8,
|
| 1358 |
+
"grad_norm": 0.267578125,
|
| 1359 |
+
"learning_rate": 0.000109698659493979,
|
| 1360 |
+
"loss": 5.5778,
|
| 1361 |
+
"step": 192
|
| 1362 |
+
},
|
| 1363 |
+
{
|
| 1364 |
+
"epoch": 0.8041666666666667,
|
| 1365 |
+
"grad_norm": 0.294921875,
|
| 1366 |
+
"learning_rate": 0.00010542974530180327,
|
| 1367 |
+
"loss": 5.6804,
|
| 1368 |
+
"step": 193
|
| 1369 |
+
},
|
| 1370 |
+
{
|
| 1371 |
+
"epoch": 0.8083333333333333,
|
| 1372 |
+
"grad_norm": 0.25,
|
| 1373 |
+
"learning_rate": 0.00010123574244214551,
|
| 1374 |
+
"loss": 5.6639,
|
| 1375 |
+
"step": 194
|
| 1376 |
+
},
|
| 1377 |
+
{
|
| 1378 |
+
"epoch": 0.8125,
|
| 1379 |
+
"grad_norm": 0.3046875,
|
| 1380 |
+
"learning_rate": 9.711744716951093e-05,
|
| 1381 |
+
"loss": 5.4121,
|
| 1382 |
+
"step": 195
|
| 1383 |
+
},
|
| 1384 |
+
{
|
| 1385 |
+
"epoch": 0.8166666666666667,
|
| 1386 |
+
"grad_norm": 0.390625,
|
| 1387 |
+
"learning_rate": 9.307564136490254e-05,
|
| 1388 |
+
"loss": 5.4785,
|
| 1389 |
+
"step": 196
|
| 1390 |
+
},
|
| 1391 |
+
{
|
| 1392 |
+
"epoch": 0.8208333333333333,
|
| 1393 |
+
"grad_norm": 0.30859375,
|
| 1394 |
+
"learning_rate": 8.911109238737747e-05,
|
| 1395 |
+
"loss": 5.6047,
|
| 1396 |
+
"step": 197
|
| 1397 |
+
},
|
| 1398 |
+
{
|
| 1399 |
+
"epoch": 0.825,
|
| 1400 |
+
"grad_norm": 0.23046875,
|
| 1401 |
+
"learning_rate": 8.522455292835934e-05,
|
| 1402 |
+
"loss": 5.6457,
|
| 1403 |
+
"step": 198
|
| 1404 |
+
},
|
| 1405 |
+
{
|
| 1406 |
+
"epoch": 0.8291666666666667,
|
| 1407 |
+
"grad_norm": 0.232421875,
|
| 1408 |
+
"learning_rate": 8.141676086873573e-05,
|
| 1409 |
+
"loss": 5.6667,
|
| 1410 |
+
"step": 199
|
| 1411 |
+
},
|
| 1412 |
+
{
|
| 1413 |
+
"epoch": 0.8333333333333334,
|
| 1414 |
+
"grad_norm": 0.2109375,
|
| 1415 |
+
"learning_rate": 7.768843913876755e-05,
|
| 1416 |
+
"loss": 5.7239,
|
| 1417 |
+
"step": 200
|
| 1418 |
+
},
|
| 1419 |
+
{
|
| 1420 |
+
"epoch": 0.8333333333333334,
|
| 1421 |
+
"eval_runtime": 1.198,
|
| 1422 |
+
"eval_samples_per_second": 13.356,
|
| 1423 |
+
"eval_steps_per_second": 0.835,
|
| 1424 |
+
"step": 200
|
| 1425 |
}
|
| 1426 |
],
|
| 1427 |
"logging_steps": 1,
|
|
|
|
| 1441 |
"attributes": {}
|
| 1442 |
}
|
| 1443 |
},
|
| 1444 |
+
"total_flos": 1.018894554759168e+17,
|
| 1445 |
"train_batch_size": 16,
|
| 1446 |
"trial_name": null,
|
| 1447 |
"trial_params": null
|