Training in progress, step 1000, 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/rng_state_4.pth +1 -1
- last-checkpoint/rng_state_5.pth +1 -1
- last-checkpoint/rng_state_6.pth +1 -1
- last-checkpoint/rng_state_7.pth +1 -1
- last-checkpoint/scheduler.pt +1 -1
- last-checkpoint/trainer_state.json +711 -3
last-checkpoint/model.safetensors
CHANGED
|
@@ -1,3 +1,3 @@
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
-
oid sha256:
|
| 3 |
size 791869518
|
|
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:6415eeae002a06eaee3e6c8d4cb9f7505094d1f36ee4e29e435d17300ed0530d
|
| 3 |
size 791869518
|
last-checkpoint/optimizer.pt
CHANGED
|
@@ -1,3 +1,3 @@
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
-
oid sha256:
|
| 3 |
size 2375752250
|
|
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:a52ae8230dd79da1bc508a923dbccfd0826e740333dd2ce1ba9f9bed20ef632c
|
| 3 |
size 2375752250
|
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 15984
|
|
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:74386f26f36ed67f56395205881e5db2d0c28ffcbeed50dd95b28771d2dac588
|
| 3 |
size 15984
|
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 15984
|
|
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:41c88f9de084200454883a13c3717941ea3fd433e2f8735507fc30611f9c5501
|
| 3 |
size 15984
|
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 15984
|
|
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:965b00d4cb4710ebab57c8787b9925bb3f77b8eeba94a186ec4bc1c2f326ef3f
|
| 3 |
size 15984
|
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 15984
|
|
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:d5dc374b8b9a4c45c950f9d136feab85a767081fa59f0c7d68ed3a62060c4949
|
| 3 |
size 15984
|
last-checkpoint/rng_state_4.pth
CHANGED
|
@@ -1,3 +1,3 @@
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
-
oid sha256:
|
| 3 |
size 15984
|
|
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:5c7c212fb779217f1edac0baf44f67b608eefc1e0e4e3f5a9dd7eb557032c1bc
|
| 3 |
size 15984
|
last-checkpoint/rng_state_5.pth
CHANGED
|
@@ -1,3 +1,3 @@
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
-
oid sha256:
|
| 3 |
size 15984
|
|
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:86e1effd626ce1e95dd68a0c8089fe19218f2b24dfe9e45ed2cab1c0ebc10ba1
|
| 3 |
size 15984
|
last-checkpoint/rng_state_6.pth
CHANGED
|
@@ -1,3 +1,3 @@
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
-
oid sha256:
|
| 3 |
size 15984
|
|
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:799cc83f60dfc1c4243cfd6403592112414a2eec494e6832f10221c96ff62c20
|
| 3 |
size 15984
|
last-checkpoint/rng_state_7.pth
CHANGED
|
@@ -1,3 +1,3 @@
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
-
oid sha256:
|
| 3 |
size 15984
|
|
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:586777c398770c3255d3a1f48c7fef44ea9d89117c627c9ea490e16bfd9a49ba
|
| 3 |
size 15984
|
last-checkpoint/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:bf407b3cceb49ac6bb524b596a50ae77a5444821a3bb5dab22e8c85264a65a8c
|
| 3 |
size 1000
|
last-checkpoint/trainer_state.json
CHANGED
|
@@ -1,9 +1,9 @@
|
|
| 1 |
{
|
| 2 |
"best_metric": null,
|
| 3 |
"best_model_checkpoint": null,
|
| 4 |
-
"epoch": 0.
|
| 5 |
"eval_steps": 500,
|
| 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": 595.745,
|
| 716 |
"eval_steps_per_second": 37.241,
|
| 717 |
"step": 500
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 718 |
}
|
| 719 |
],
|
| 720 |
"logging_steps": 5,
|
|
@@ -734,7 +1442,7 @@
|
|
| 734 |
"attributes": {}
|
| 735 |
}
|
| 736 |
},
|
| 737 |
-
"total_flos":
|
| 738 |
"train_batch_size": 4,
|
| 739 |
"trial_name": null,
|
| 740 |
"trial_params": null
|
|
|
|
| 1 |
{
|
| 2 |
"best_metric": null,
|
| 3 |
"best_model_checkpoint": null,
|
| 4 |
+
"epoch": 0.72499886718927,
|
| 5 |
"eval_steps": 500,
|
| 6 |
+
"global_step": 1000,
|
| 7 |
"is_hyper_param_search": false,
|
| 8 |
"is_local_process_zero": true,
|
| 9 |
"is_world_process_zero": true,
|
|
|
|
| 715 |
"eval_samples_per_second": 595.745,
|
| 716 |
"eval_steps_per_second": 37.241,
|
| 717 |
"step": 500
|
| 718 |
+
},
|
| 719 |
+
{
|
| 720 |
+
"epoch": 0.36612442793058136,
|
| 721 |
+
"grad_norm": 100.625,
|
| 722 |
+
"learning_rate": 2.112812248186946e-06,
|
| 723 |
+
"loss": 90.367,
|
| 724 |
+
"step": 505
|
| 725 |
+
},
|
| 726 |
+
{
|
| 727 |
+
"epoch": 0.3697494222665277,
|
| 728 |
+
"grad_norm": 99.125,
|
| 729 |
+
"learning_rate": 2.1007252215954874e-06,
|
| 730 |
+
"loss": 89.764,
|
| 731 |
+
"step": 510
|
| 732 |
+
},
|
| 733 |
+
{
|
| 734 |
+
"epoch": 0.37337441660247406,
|
| 735 |
+
"grad_norm": 98.0625,
|
| 736 |
+
"learning_rate": 2.0886381950040293e-06,
|
| 737 |
+
"loss": 90.377,
|
| 738 |
+
"step": 515
|
| 739 |
+
},
|
| 740 |
+
{
|
| 741 |
+
"epoch": 0.37699941093842043,
|
| 742 |
+
"grad_norm": 97.125,
|
| 743 |
+
"learning_rate": 2.0765511684125707e-06,
|
| 744 |
+
"loss": 89.3218,
|
| 745 |
+
"step": 520
|
| 746 |
+
},
|
| 747 |
+
{
|
| 748 |
+
"epoch": 0.38062440527436675,
|
| 749 |
+
"grad_norm": 99.1875,
|
| 750 |
+
"learning_rate": 2.064464141821112e-06,
|
| 751 |
+
"loss": 90.3385,
|
| 752 |
+
"step": 525
|
| 753 |
+
},
|
| 754 |
+
{
|
| 755 |
+
"epoch": 0.3842493996103131,
|
| 756 |
+
"grad_norm": 101.125,
|
| 757 |
+
"learning_rate": 2.0523771152296536e-06,
|
| 758 |
+
"loss": 88.8969,
|
| 759 |
+
"step": 530
|
| 760 |
+
},
|
| 761 |
+
{
|
| 762 |
+
"epoch": 0.38787439394625944,
|
| 763 |
+
"grad_norm": 92.8125,
|
| 764 |
+
"learning_rate": 2.040290088638195e-06,
|
| 765 |
+
"loss": 90.3189,
|
| 766 |
+
"step": 535
|
| 767 |
+
},
|
| 768 |
+
{
|
| 769 |
+
"epoch": 0.3914993882822058,
|
| 770 |
+
"grad_norm": 93.0625,
|
| 771 |
+
"learning_rate": 2.0282030620467365e-06,
|
| 772 |
+
"loss": 88.3214,
|
| 773 |
+
"step": 540
|
| 774 |
+
},
|
| 775 |
+
{
|
| 776 |
+
"epoch": 0.39512438261815214,
|
| 777 |
+
"grad_norm": 96.0,
|
| 778 |
+
"learning_rate": 2.016116035455278e-06,
|
| 779 |
+
"loss": 90.5051,
|
| 780 |
+
"step": 545
|
| 781 |
+
},
|
| 782 |
+
{
|
| 783 |
+
"epoch": 0.3987493769540985,
|
| 784 |
+
"grad_norm": 97.75,
|
| 785 |
+
"learning_rate": 2.0040290088638194e-06,
|
| 786 |
+
"loss": 89.0853,
|
| 787 |
+
"step": 550
|
| 788 |
+
},
|
| 789 |
+
{
|
| 790 |
+
"epoch": 0.4023743712900449,
|
| 791 |
+
"grad_norm": 96.125,
|
| 792 |
+
"learning_rate": 1.991941982272361e-06,
|
| 793 |
+
"loss": 89.5784,
|
| 794 |
+
"step": 555
|
| 795 |
+
},
|
| 796 |
+
{
|
| 797 |
+
"epoch": 0.4059993656259912,
|
| 798 |
+
"grad_norm": 93.5625,
|
| 799 |
+
"learning_rate": 1.9798549556809023e-06,
|
| 800 |
+
"loss": 89.3408,
|
| 801 |
+
"step": 560
|
| 802 |
+
},
|
| 803 |
+
{
|
| 804 |
+
"epoch": 0.4096243599619376,
|
| 805 |
+
"grad_norm": 95.1875,
|
| 806 |
+
"learning_rate": 1.9677679290894438e-06,
|
| 807 |
+
"loss": 89.4404,
|
| 808 |
+
"step": 565
|
| 809 |
+
},
|
| 810 |
+
{
|
| 811 |
+
"epoch": 0.4132493542978839,
|
| 812 |
+
"grad_norm": 99.1875,
|
| 813 |
+
"learning_rate": 1.9556809024979856e-06,
|
| 814 |
+
"loss": 89.5664,
|
| 815 |
+
"step": 570
|
| 816 |
+
},
|
| 817 |
+
{
|
| 818 |
+
"epoch": 0.41687434863383027,
|
| 819 |
+
"grad_norm": 90.1875,
|
| 820 |
+
"learning_rate": 1.943593875906527e-06,
|
| 821 |
+
"loss": 89.5359,
|
| 822 |
+
"step": 575
|
| 823 |
+
},
|
| 824 |
+
{
|
| 825 |
+
"epoch": 0.4204993429697766,
|
| 826 |
+
"grad_norm": 100.625,
|
| 827 |
+
"learning_rate": 1.9315068493150685e-06,
|
| 828 |
+
"loss": 88.9519,
|
| 829 |
+
"step": 580
|
| 830 |
+
},
|
| 831 |
+
{
|
| 832 |
+
"epoch": 0.42412433730572296,
|
| 833 |
+
"grad_norm": 95.375,
|
| 834 |
+
"learning_rate": 1.91941982272361e-06,
|
| 835 |
+
"loss": 89.6489,
|
| 836 |
+
"step": 585
|
| 837 |
+
},
|
| 838 |
+
{
|
| 839 |
+
"epoch": 0.4277493316416693,
|
| 840 |
+
"grad_norm": 100.0,
|
| 841 |
+
"learning_rate": 1.9073327961321514e-06,
|
| 842 |
+
"loss": 88.3665,
|
| 843 |
+
"step": 590
|
| 844 |
+
},
|
| 845 |
+
{
|
| 846 |
+
"epoch": 0.43137432597761566,
|
| 847 |
+
"grad_norm": 92.5,
|
| 848 |
+
"learning_rate": 1.8952457695406929e-06,
|
| 849 |
+
"loss": 87.8933,
|
| 850 |
+
"step": 595
|
| 851 |
+
},
|
| 852 |
+
{
|
| 853 |
+
"epoch": 0.43499932031356203,
|
| 854 |
+
"grad_norm": 95.125,
|
| 855 |
+
"learning_rate": 1.8831587429492343e-06,
|
| 856 |
+
"loss": 89.0427,
|
| 857 |
+
"step": 600
|
| 858 |
+
},
|
| 859 |
+
{
|
| 860 |
+
"epoch": 0.43862431464950835,
|
| 861 |
+
"grad_norm": 93.5625,
|
| 862 |
+
"learning_rate": 1.8710717163577762e-06,
|
| 863 |
+
"loss": 88.9729,
|
| 864 |
+
"step": 605
|
| 865 |
+
},
|
| 866 |
+
{
|
| 867 |
+
"epoch": 0.4422493089854547,
|
| 868 |
+
"grad_norm": 94.125,
|
| 869 |
+
"learning_rate": 1.8589846897663176e-06,
|
| 870 |
+
"loss": 88.4542,
|
| 871 |
+
"step": 610
|
| 872 |
+
},
|
| 873 |
+
{
|
| 874 |
+
"epoch": 0.44587430332140104,
|
| 875 |
+
"grad_norm": 94.25,
|
| 876 |
+
"learning_rate": 1.846897663174859e-06,
|
| 877 |
+
"loss": 87.9236,
|
| 878 |
+
"step": 615
|
| 879 |
+
},
|
| 880 |
+
{
|
| 881 |
+
"epoch": 0.4494992976573474,
|
| 882 |
+
"grad_norm": 96.3125,
|
| 883 |
+
"learning_rate": 1.8348106365834005e-06,
|
| 884 |
+
"loss": 88.8232,
|
| 885 |
+
"step": 620
|
| 886 |
+
},
|
| 887 |
+
{
|
| 888 |
+
"epoch": 0.45312429199329374,
|
| 889 |
+
"grad_norm": 92.875,
|
| 890 |
+
"learning_rate": 1.822723609991942e-06,
|
| 891 |
+
"loss": 87.6141,
|
| 892 |
+
"step": 625
|
| 893 |
+
},
|
| 894 |
+
{
|
| 895 |
+
"epoch": 0.4567492863292401,
|
| 896 |
+
"grad_norm": 94.8125,
|
| 897 |
+
"learning_rate": 1.8106365834004834e-06,
|
| 898 |
+
"loss": 89.5391,
|
| 899 |
+
"step": 630
|
| 900 |
+
},
|
| 901 |
+
{
|
| 902 |
+
"epoch": 0.4603742806651865,
|
| 903 |
+
"grad_norm": 96.625,
|
| 904 |
+
"learning_rate": 1.7985495568090249e-06,
|
| 905 |
+
"loss": 87.6836,
|
| 906 |
+
"step": 635
|
| 907 |
+
},
|
| 908 |
+
{
|
| 909 |
+
"epoch": 0.4639992750011328,
|
| 910 |
+
"grad_norm": 98.5,
|
| 911 |
+
"learning_rate": 1.7864625302175668e-06,
|
| 912 |
+
"loss": 88.0843,
|
| 913 |
+
"step": 640
|
| 914 |
+
},
|
| 915 |
+
{
|
| 916 |
+
"epoch": 0.4676242693370792,
|
| 917 |
+
"grad_norm": 95.4375,
|
| 918 |
+
"learning_rate": 1.7743755036261082e-06,
|
| 919 |
+
"loss": 88.7523,
|
| 920 |
+
"step": 645
|
| 921 |
+
},
|
| 922 |
+
{
|
| 923 |
+
"epoch": 0.4712492636730255,
|
| 924 |
+
"grad_norm": 96.9375,
|
| 925 |
+
"learning_rate": 1.7622884770346497e-06,
|
| 926 |
+
"loss": 88.2255,
|
| 927 |
+
"step": 650
|
| 928 |
+
},
|
| 929 |
+
{
|
| 930 |
+
"epoch": 0.4748742580089719,
|
| 931 |
+
"grad_norm": 95.75,
|
| 932 |
+
"learning_rate": 1.750201450443191e-06,
|
| 933 |
+
"loss": 88.399,
|
| 934 |
+
"step": 655
|
| 935 |
+
},
|
| 936 |
+
{
|
| 937 |
+
"epoch": 0.4784992523449182,
|
| 938 |
+
"grad_norm": 99.4375,
|
| 939 |
+
"learning_rate": 1.7381144238517325e-06,
|
| 940 |
+
"loss": 86.5224,
|
| 941 |
+
"step": 660
|
| 942 |
+
},
|
| 943 |
+
{
|
| 944 |
+
"epoch": 0.48212424668086457,
|
| 945 |
+
"grad_norm": 93.125,
|
| 946 |
+
"learning_rate": 1.726027397260274e-06,
|
| 947 |
+
"loss": 86.3948,
|
| 948 |
+
"step": 665
|
| 949 |
+
},
|
| 950 |
+
{
|
| 951 |
+
"epoch": 0.4857492410168109,
|
| 952 |
+
"grad_norm": 96.3125,
|
| 953 |
+
"learning_rate": 1.7139403706688154e-06,
|
| 954 |
+
"loss": 87.6118,
|
| 955 |
+
"step": 670
|
| 956 |
+
},
|
| 957 |
+
{
|
| 958 |
+
"epoch": 0.48937423535275726,
|
| 959 |
+
"grad_norm": 95.8125,
|
| 960 |
+
"learning_rate": 1.701853344077357e-06,
|
| 961 |
+
"loss": 86.8485,
|
| 962 |
+
"step": 675
|
| 963 |
+
},
|
| 964 |
+
{
|
| 965 |
+
"epoch": 0.49299922968870363,
|
| 966 |
+
"grad_norm": 91.3125,
|
| 967 |
+
"learning_rate": 1.6897663174858985e-06,
|
| 968 |
+
"loss": 88.4679,
|
| 969 |
+
"step": 680
|
| 970 |
+
},
|
| 971 |
+
{
|
| 972 |
+
"epoch": 0.49662422402464995,
|
| 973 |
+
"grad_norm": 93.9375,
|
| 974 |
+
"learning_rate": 1.67767929089444e-06,
|
| 975 |
+
"loss": 87.9796,
|
| 976 |
+
"step": 685
|
| 977 |
+
},
|
| 978 |
+
{
|
| 979 |
+
"epoch": 0.5002492183605963,
|
| 980 |
+
"grad_norm": 97.3125,
|
| 981 |
+
"learning_rate": 1.6655922643029814e-06,
|
| 982 |
+
"loss": 87.4862,
|
| 983 |
+
"step": 690
|
| 984 |
+
},
|
| 985 |
+
{
|
| 986 |
+
"epoch": 0.5038742126965426,
|
| 987 |
+
"grad_norm": 91.125,
|
| 988 |
+
"learning_rate": 1.6535052377115229e-06,
|
| 989 |
+
"loss": 86.7279,
|
| 990 |
+
"step": 695
|
| 991 |
+
},
|
| 992 |
+
{
|
| 993 |
+
"epoch": 0.507499207032489,
|
| 994 |
+
"grad_norm": 94.5,
|
| 995 |
+
"learning_rate": 1.6414182111200643e-06,
|
| 996 |
+
"loss": 87.0621,
|
| 997 |
+
"step": 700
|
| 998 |
+
},
|
| 999 |
+
{
|
| 1000 |
+
"epoch": 0.5111242013684354,
|
| 1001 |
+
"grad_norm": 95.5,
|
| 1002 |
+
"learning_rate": 1.6293311845286058e-06,
|
| 1003 |
+
"loss": 87.104,
|
| 1004 |
+
"step": 705
|
| 1005 |
+
},
|
| 1006 |
+
{
|
| 1007 |
+
"epoch": 0.5147491957043817,
|
| 1008 |
+
"grad_norm": 94.9375,
|
| 1009 |
+
"learning_rate": 1.6172441579371477e-06,
|
| 1010 |
+
"loss": 86.7461,
|
| 1011 |
+
"step": 710
|
| 1012 |
+
},
|
| 1013 |
+
{
|
| 1014 |
+
"epoch": 0.518374190040328,
|
| 1015 |
+
"grad_norm": 93.125,
|
| 1016 |
+
"learning_rate": 1.605157131345689e-06,
|
| 1017 |
+
"loss": 87.189,
|
| 1018 |
+
"step": 715
|
| 1019 |
+
},
|
| 1020 |
+
{
|
| 1021 |
+
"epoch": 0.5219991843762745,
|
| 1022 |
+
"grad_norm": 96.5625,
|
| 1023 |
+
"learning_rate": 1.5930701047542306e-06,
|
| 1024 |
+
"loss": 86.5543,
|
| 1025 |
+
"step": 720
|
| 1026 |
+
},
|
| 1027 |
+
{
|
| 1028 |
+
"epoch": 0.5256241787122208,
|
| 1029 |
+
"grad_norm": 94.5,
|
| 1030 |
+
"learning_rate": 1.580983078162772e-06,
|
| 1031 |
+
"loss": 87.1241,
|
| 1032 |
+
"step": 725
|
| 1033 |
+
},
|
| 1034 |
+
{
|
| 1035 |
+
"epoch": 0.5292491730481671,
|
| 1036 |
+
"grad_norm": 93.375,
|
| 1037 |
+
"learning_rate": 1.5688960515713134e-06,
|
| 1038 |
+
"loss": 86.1505,
|
| 1039 |
+
"step": 730
|
| 1040 |
+
},
|
| 1041 |
+
{
|
| 1042 |
+
"epoch": 0.5328741673841134,
|
| 1043 |
+
"grad_norm": 94.625,
|
| 1044 |
+
"learning_rate": 1.556809024979855e-06,
|
| 1045 |
+
"loss": 87.0848,
|
| 1046 |
+
"step": 735
|
| 1047 |
+
},
|
| 1048 |
+
{
|
| 1049 |
+
"epoch": 0.5364991617200598,
|
| 1050 |
+
"grad_norm": 96.375,
|
| 1051 |
+
"learning_rate": 1.5447219983883963e-06,
|
| 1052 |
+
"loss": 86.5279,
|
| 1053 |
+
"step": 740
|
| 1054 |
+
},
|
| 1055 |
+
{
|
| 1056 |
+
"epoch": 0.5401241560560062,
|
| 1057 |
+
"grad_norm": 96.875,
|
| 1058 |
+
"learning_rate": 1.5326349717969382e-06,
|
| 1059 |
+
"loss": 87.5163,
|
| 1060 |
+
"step": 745
|
| 1061 |
+
},
|
| 1062 |
+
{
|
| 1063 |
+
"epoch": 0.5437491503919525,
|
| 1064 |
+
"grad_norm": 96.75,
|
| 1065 |
+
"learning_rate": 1.5205479452054797e-06,
|
| 1066 |
+
"loss": 84.3448,
|
| 1067 |
+
"step": 750
|
| 1068 |
+
},
|
| 1069 |
+
{
|
| 1070 |
+
"epoch": 0.5473741447278989,
|
| 1071 |
+
"grad_norm": 99.0,
|
| 1072 |
+
"learning_rate": 1.5084609186140211e-06,
|
| 1073 |
+
"loss": 86.8625,
|
| 1074 |
+
"step": 755
|
| 1075 |
+
},
|
| 1076 |
+
{
|
| 1077 |
+
"epoch": 0.5509991390638452,
|
| 1078 |
+
"grad_norm": 93.5,
|
| 1079 |
+
"learning_rate": 1.4963738920225626e-06,
|
| 1080 |
+
"loss": 85.2925,
|
| 1081 |
+
"step": 760
|
| 1082 |
+
},
|
| 1083 |
+
{
|
| 1084 |
+
"epoch": 0.5546241333997916,
|
| 1085 |
+
"grad_norm": 94.3125,
|
| 1086 |
+
"learning_rate": 1.484286865431104e-06,
|
| 1087 |
+
"loss": 87.5748,
|
| 1088 |
+
"step": 765
|
| 1089 |
+
},
|
| 1090 |
+
{
|
| 1091 |
+
"epoch": 0.5582491277357379,
|
| 1092 |
+
"grad_norm": 92.75,
|
| 1093 |
+
"learning_rate": 1.4721998388396455e-06,
|
| 1094 |
+
"loss": 86.6436,
|
| 1095 |
+
"step": 770
|
| 1096 |
+
},
|
| 1097 |
+
{
|
| 1098 |
+
"epoch": 0.5618741220716843,
|
| 1099 |
+
"grad_norm": 95.875,
|
| 1100 |
+
"learning_rate": 1.460112812248187e-06,
|
| 1101 |
+
"loss": 85.7076,
|
| 1102 |
+
"step": 775
|
| 1103 |
+
},
|
| 1104 |
+
{
|
| 1105 |
+
"epoch": 0.5654991164076306,
|
| 1106 |
+
"grad_norm": 96.625,
|
| 1107 |
+
"learning_rate": 1.4480257856567283e-06,
|
| 1108 |
+
"loss": 86.4671,
|
| 1109 |
+
"step": 780
|
| 1110 |
+
},
|
| 1111 |
+
{
|
| 1112 |
+
"epoch": 0.5691241107435769,
|
| 1113 |
+
"grad_norm": 95.375,
|
| 1114 |
+
"learning_rate": 1.43593875906527e-06,
|
| 1115 |
+
"loss": 85.973,
|
| 1116 |
+
"step": 785
|
| 1117 |
+
},
|
| 1118 |
+
{
|
| 1119 |
+
"epoch": 0.5727491050795234,
|
| 1120 |
+
"grad_norm": 95.75,
|
| 1121 |
+
"learning_rate": 1.4238517324738115e-06,
|
| 1122 |
+
"loss": 84.9478,
|
| 1123 |
+
"step": 790
|
| 1124 |
+
},
|
| 1125 |
+
{
|
| 1126 |
+
"epoch": 0.5763740994154697,
|
| 1127 |
+
"grad_norm": 94.875,
|
| 1128 |
+
"learning_rate": 1.411764705882353e-06,
|
| 1129 |
+
"loss": 85.6088,
|
| 1130 |
+
"step": 795
|
| 1131 |
+
},
|
| 1132 |
+
{
|
| 1133 |
+
"epoch": 0.579999093751416,
|
| 1134 |
+
"grad_norm": 93.75,
|
| 1135 |
+
"learning_rate": 1.3996776792908943e-06,
|
| 1136 |
+
"loss": 86.5241,
|
| 1137 |
+
"step": 800
|
| 1138 |
+
},
|
| 1139 |
+
{
|
| 1140 |
+
"epoch": 0.5836240880873623,
|
| 1141 |
+
"grad_norm": 93.0625,
|
| 1142 |
+
"learning_rate": 1.387590652699436e-06,
|
| 1143 |
+
"loss": 85.3578,
|
| 1144 |
+
"step": 805
|
| 1145 |
+
},
|
| 1146 |
+
{
|
| 1147 |
+
"epoch": 0.5872490824233088,
|
| 1148 |
+
"grad_norm": 94.25,
|
| 1149 |
+
"learning_rate": 1.3755036261079775e-06,
|
| 1150 |
+
"loss": 86.0338,
|
| 1151 |
+
"step": 810
|
| 1152 |
+
},
|
| 1153 |
+
{
|
| 1154 |
+
"epoch": 0.5908740767592551,
|
| 1155 |
+
"grad_norm": 92.6875,
|
| 1156 |
+
"learning_rate": 1.363416599516519e-06,
|
| 1157 |
+
"loss": 84.5091,
|
| 1158 |
+
"step": 815
|
| 1159 |
+
},
|
| 1160 |
+
{
|
| 1161 |
+
"epoch": 0.5944990710952014,
|
| 1162 |
+
"grad_norm": 92.8125,
|
| 1163 |
+
"learning_rate": 1.3513295729250606e-06,
|
| 1164 |
+
"loss": 85.7377,
|
| 1165 |
+
"step": 820
|
| 1166 |
+
},
|
| 1167 |
+
{
|
| 1168 |
+
"epoch": 0.5981240654311477,
|
| 1169 |
+
"grad_norm": 91.125,
|
| 1170 |
+
"learning_rate": 1.339242546333602e-06,
|
| 1171 |
+
"loss": 84.7625,
|
| 1172 |
+
"step": 825
|
| 1173 |
+
},
|
| 1174 |
+
{
|
| 1175 |
+
"epoch": 0.6017490597670941,
|
| 1176 |
+
"grad_norm": 93.1875,
|
| 1177 |
+
"learning_rate": 1.3271555197421435e-06,
|
| 1178 |
+
"loss": 84.8545,
|
| 1179 |
+
"step": 830
|
| 1180 |
+
},
|
| 1181 |
+
{
|
| 1182 |
+
"epoch": 0.6053740541030405,
|
| 1183 |
+
"grad_norm": 94.8125,
|
| 1184 |
+
"learning_rate": 1.315068493150685e-06,
|
| 1185 |
+
"loss": 85.5188,
|
| 1186 |
+
"step": 835
|
| 1187 |
+
},
|
| 1188 |
+
{
|
| 1189 |
+
"epoch": 0.6089990484389868,
|
| 1190 |
+
"grad_norm": 91.75,
|
| 1191 |
+
"learning_rate": 1.3029814665592266e-06,
|
| 1192 |
+
"loss": 86.0601,
|
| 1193 |
+
"step": 840
|
| 1194 |
+
},
|
| 1195 |
+
{
|
| 1196 |
+
"epoch": 0.6126240427749332,
|
| 1197 |
+
"grad_norm": 94.375,
|
| 1198 |
+
"learning_rate": 1.290894439967768e-06,
|
| 1199 |
+
"loss": 85.4887,
|
| 1200 |
+
"step": 845
|
| 1201 |
+
},
|
| 1202 |
+
{
|
| 1203 |
+
"epoch": 0.6162490371108795,
|
| 1204 |
+
"grad_norm": 95.375,
|
| 1205 |
+
"learning_rate": 1.2788074133763095e-06,
|
| 1206 |
+
"loss": 85.1952,
|
| 1207 |
+
"step": 850
|
| 1208 |
+
},
|
| 1209 |
+
{
|
| 1210 |
+
"epoch": 0.6198740314468258,
|
| 1211 |
+
"grad_norm": 95.1875,
|
| 1212 |
+
"learning_rate": 1.2667203867848511e-06,
|
| 1213 |
+
"loss": 84.1906,
|
| 1214 |
+
"step": 855
|
| 1215 |
+
},
|
| 1216 |
+
{
|
| 1217 |
+
"epoch": 0.6234990257827722,
|
| 1218 |
+
"grad_norm": 92.8125,
|
| 1219 |
+
"learning_rate": 1.2546333601933926e-06,
|
| 1220 |
+
"loss": 84.6792,
|
| 1221 |
+
"step": 860
|
| 1222 |
+
},
|
| 1223 |
+
{
|
| 1224 |
+
"epoch": 0.6271240201187186,
|
| 1225 |
+
"grad_norm": 92.625,
|
| 1226 |
+
"learning_rate": 1.242546333601934e-06,
|
| 1227 |
+
"loss": 85.1767,
|
| 1228 |
+
"step": 865
|
| 1229 |
+
},
|
| 1230 |
+
{
|
| 1231 |
+
"epoch": 0.6307490144546649,
|
| 1232 |
+
"grad_norm": 91.3125,
|
| 1233 |
+
"learning_rate": 1.2304593070104755e-06,
|
| 1234 |
+
"loss": 84.0814,
|
| 1235 |
+
"step": 870
|
| 1236 |
+
},
|
| 1237 |
+
{
|
| 1238 |
+
"epoch": 0.6343740087906112,
|
| 1239 |
+
"grad_norm": 96.1875,
|
| 1240 |
+
"learning_rate": 1.218372280419017e-06,
|
| 1241 |
+
"loss": 84.6096,
|
| 1242 |
+
"step": 875
|
| 1243 |
+
},
|
| 1244 |
+
{
|
| 1245 |
+
"epoch": 0.6379990031265577,
|
| 1246 |
+
"grad_norm": 90.5625,
|
| 1247 |
+
"learning_rate": 1.2062852538275584e-06,
|
| 1248 |
+
"loss": 84.8486,
|
| 1249 |
+
"step": 880
|
| 1250 |
+
},
|
| 1251 |
+
{
|
| 1252 |
+
"epoch": 0.641623997462504,
|
| 1253 |
+
"grad_norm": 92.6875,
|
| 1254 |
+
"learning_rate": 1.1941982272360998e-06,
|
| 1255 |
+
"loss": 84.4379,
|
| 1256 |
+
"step": 885
|
| 1257 |
+
},
|
| 1258 |
+
{
|
| 1259 |
+
"epoch": 0.6452489917984503,
|
| 1260 |
+
"grad_norm": 96.5,
|
| 1261 |
+
"learning_rate": 1.1821112006446415e-06,
|
| 1262 |
+
"loss": 85.6003,
|
| 1263 |
+
"step": 890
|
| 1264 |
+
},
|
| 1265 |
+
{
|
| 1266 |
+
"epoch": 0.6488739861343966,
|
| 1267 |
+
"grad_norm": 96.25,
|
| 1268 |
+
"learning_rate": 1.170024174053183e-06,
|
| 1269 |
+
"loss": 83.5816,
|
| 1270 |
+
"step": 895
|
| 1271 |
+
},
|
| 1272 |
+
{
|
| 1273 |
+
"epoch": 0.652498980470343,
|
| 1274 |
+
"grad_norm": 90.25,
|
| 1275 |
+
"learning_rate": 1.1579371474617244e-06,
|
| 1276 |
+
"loss": 84.2162,
|
| 1277 |
+
"step": 900
|
| 1278 |
+
},
|
| 1279 |
+
{
|
| 1280 |
+
"epoch": 0.6561239748062894,
|
| 1281 |
+
"grad_norm": 98.5,
|
| 1282 |
+
"learning_rate": 1.1458501208702658e-06,
|
| 1283 |
+
"loss": 83.1881,
|
| 1284 |
+
"step": 905
|
| 1285 |
+
},
|
| 1286 |
+
{
|
| 1287 |
+
"epoch": 0.6597489691422357,
|
| 1288 |
+
"grad_norm": 92.0,
|
| 1289 |
+
"learning_rate": 1.1337630942788075e-06,
|
| 1290 |
+
"loss": 84.7499,
|
| 1291 |
+
"step": 910
|
| 1292 |
+
},
|
| 1293 |
+
{
|
| 1294 |
+
"epoch": 0.6633739634781821,
|
| 1295 |
+
"grad_norm": 93.3125,
|
| 1296 |
+
"learning_rate": 1.121676067687349e-06,
|
| 1297 |
+
"loss": 83.7881,
|
| 1298 |
+
"step": 915
|
| 1299 |
+
},
|
| 1300 |
+
{
|
| 1301 |
+
"epoch": 0.6669989578141284,
|
| 1302 |
+
"grad_norm": 94.75,
|
| 1303 |
+
"learning_rate": 1.1095890410958904e-06,
|
| 1304 |
+
"loss": 83.8075,
|
| 1305 |
+
"step": 920
|
| 1306 |
+
},
|
| 1307 |
+
{
|
| 1308 |
+
"epoch": 0.6706239521500748,
|
| 1309 |
+
"grad_norm": 93.4375,
|
| 1310 |
+
"learning_rate": 1.097502014504432e-06,
|
| 1311 |
+
"loss": 83.9208,
|
| 1312 |
+
"step": 925
|
| 1313 |
+
},
|
| 1314 |
+
{
|
| 1315 |
+
"epoch": 0.6742489464860211,
|
| 1316 |
+
"grad_norm": 96.0,
|
| 1317 |
+
"learning_rate": 1.0854149879129735e-06,
|
| 1318 |
+
"loss": 83.8316,
|
| 1319 |
+
"step": 930
|
| 1320 |
+
},
|
| 1321 |
+
{
|
| 1322 |
+
"epoch": 0.6778739408219675,
|
| 1323 |
+
"grad_norm": 97.375,
|
| 1324 |
+
"learning_rate": 1.073327961321515e-06,
|
| 1325 |
+
"loss": 83.665,
|
| 1326 |
+
"step": 935
|
| 1327 |
+
},
|
| 1328 |
+
{
|
| 1329 |
+
"epoch": 0.6814989351579138,
|
| 1330 |
+
"grad_norm": 93.875,
|
| 1331 |
+
"learning_rate": 1.0612409347300566e-06,
|
| 1332 |
+
"loss": 84.09,
|
| 1333 |
+
"step": 940
|
| 1334 |
+
},
|
| 1335 |
+
{
|
| 1336 |
+
"epoch": 0.6851239294938601,
|
| 1337 |
+
"grad_norm": 95.125,
|
| 1338 |
+
"learning_rate": 1.049153908138598e-06,
|
| 1339 |
+
"loss": 83.6363,
|
| 1340 |
+
"step": 945
|
| 1341 |
+
},
|
| 1342 |
+
{
|
| 1343 |
+
"epoch": 0.6887489238298066,
|
| 1344 |
+
"grad_norm": 93.5,
|
| 1345 |
+
"learning_rate": 1.0370668815471395e-06,
|
| 1346 |
+
"loss": 82.6956,
|
| 1347 |
+
"step": 950
|
| 1348 |
+
},
|
| 1349 |
+
{
|
| 1350 |
+
"epoch": 0.6923739181657529,
|
| 1351 |
+
"grad_norm": 91.5,
|
| 1352 |
+
"learning_rate": 1.024979854955681e-06,
|
| 1353 |
+
"loss": 83.8282,
|
| 1354 |
+
"step": 955
|
| 1355 |
+
},
|
| 1356 |
+
{
|
| 1357 |
+
"epoch": 0.6959989125016992,
|
| 1358 |
+
"grad_norm": 101.25,
|
| 1359 |
+
"learning_rate": 1.0128928283642226e-06,
|
| 1360 |
+
"loss": 83.393,
|
| 1361 |
+
"step": 960
|
| 1362 |
+
},
|
| 1363 |
+
{
|
| 1364 |
+
"epoch": 0.6996239068376455,
|
| 1365 |
+
"grad_norm": 93.6875,
|
| 1366 |
+
"learning_rate": 1.000805801772764e-06,
|
| 1367 |
+
"loss": 82.5698,
|
| 1368 |
+
"step": 965
|
| 1369 |
+
},
|
| 1370 |
+
{
|
| 1371 |
+
"epoch": 0.703248901173592,
|
| 1372 |
+
"grad_norm": 94.3125,
|
| 1373 |
+
"learning_rate": 9.887187751813055e-07,
|
| 1374 |
+
"loss": 82.9595,
|
| 1375 |
+
"step": 970
|
| 1376 |
+
},
|
| 1377 |
+
{
|
| 1378 |
+
"epoch": 0.7068738955095383,
|
| 1379 |
+
"grad_norm": 93.5,
|
| 1380 |
+
"learning_rate": 9.76631748589847e-07,
|
| 1381 |
+
"loss": 82.2969,
|
| 1382 |
+
"step": 975
|
| 1383 |
+
},
|
| 1384 |
+
{
|
| 1385 |
+
"epoch": 0.7104988898454846,
|
| 1386 |
+
"grad_norm": 91.125,
|
| 1387 |
+
"learning_rate": 9.645447219983884e-07,
|
| 1388 |
+
"loss": 84.9823,
|
| 1389 |
+
"step": 980
|
| 1390 |
+
},
|
| 1391 |
+
{
|
| 1392 |
+
"epoch": 0.7141238841814309,
|
| 1393 |
+
"grad_norm": 93.0,
|
| 1394 |
+
"learning_rate": 9.524576954069299e-07,
|
| 1395 |
+
"loss": 84.2716,
|
| 1396 |
+
"step": 985
|
| 1397 |
+
},
|
| 1398 |
+
{
|
| 1399 |
+
"epoch": 0.7177488785173773,
|
| 1400 |
+
"grad_norm": 90.5,
|
| 1401 |
+
"learning_rate": 9.403706688154714e-07,
|
| 1402 |
+
"loss": 84.1091,
|
| 1403 |
+
"step": 990
|
| 1404 |
+
},
|
| 1405 |
+
{
|
| 1406 |
+
"epoch": 0.7213738728533237,
|
| 1407 |
+
"grad_norm": 94.5625,
|
| 1408 |
+
"learning_rate": 9.282836422240129e-07,
|
| 1409 |
+
"loss": 82.6404,
|
| 1410 |
+
"step": 995
|
| 1411 |
+
},
|
| 1412 |
+
{
|
| 1413 |
+
"epoch": 0.72499886718927,
|
| 1414 |
+
"grad_norm": 94.5625,
|
| 1415 |
+
"learning_rate": 9.161966156325544e-07,
|
| 1416 |
+
"loss": 82.809,
|
| 1417 |
+
"step": 1000
|
| 1418 |
+
},
|
| 1419 |
+
{
|
| 1420 |
+
"epoch": 0.72499886718927,
|
| 1421 |
+
"eval_loss": NaN,
|
| 1422 |
+
"eval_runtime": 124.4615,
|
| 1423 |
+
"eval_samples_per_second": 597.277,
|
| 1424 |
+
"eval_steps_per_second": 37.337,
|
| 1425 |
+
"step": 1000
|
| 1426 |
}
|
| 1427 |
],
|
| 1428 |
"logging_steps": 5,
|
|
|
|
| 1442 |
"attributes": {}
|
| 1443 |
}
|
| 1444 |
},
|
| 1445 |
+
"total_flos": 4.332357992788787e+18,
|
| 1446 |
"train_batch_size": 4,
|
| 1447 |
"trial_name": null,
|
| 1448 |
"trial_params": null
|