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:dd78f70bc690f6c96ef8249e16cd3eaed70120ea55548486454d9b19a469e2f9
|
| 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:bdfdcc66bbfd0f7601e8207cf92cc9b52dd079dc47d6e19e50c732b619c60dd8
|
| 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:fb96e4663ec39ec9f3920ec91963e2a7dc93e87ff87c51ea8d11abe793c9c1d1
|
| 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": 595.254,
|
| 716 |
"eval_steps_per_second": 37.204,
|
| 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": 0.8904223941232122,
|
| 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": 595.254,
|
| 716 |
"eval_steps_per_second": 37.204,
|
| 717 |
"step": 500
|
| 718 |
+
},
|
| 719 |
+
{
|
| 720 |
+
"epoch": 0.4496633090322222,
|
| 721 |
+
"grad_norm": 153.0,
|
| 722 |
+
"learning_rate": 1.8356435643564356e-06,
|
| 723 |
+
"loss": 97.0697,
|
| 724 |
+
"step": 505
|
| 725 |
+
},
|
| 726 |
+
{
|
| 727 |
+
"epoch": 0.45411542100283825,
|
| 728 |
+
"grad_norm": 152.375,
|
| 729 |
+
"learning_rate": 1.8207920792079207e-06,
|
| 730 |
+
"loss": 96.6341,
|
| 731 |
+
"step": 510
|
| 732 |
+
},
|
| 733 |
+
{
|
| 734 |
+
"epoch": 0.45856753297345426,
|
| 735 |
+
"grad_norm": 148.625,
|
| 736 |
+
"learning_rate": 1.8059405940594058e-06,
|
| 737 |
+
"loss": 97.238,
|
| 738 |
+
"step": 515
|
| 739 |
+
},
|
| 740 |
+
{
|
| 741 |
+
"epoch": 0.46301964494407033,
|
| 742 |
+
"grad_norm": 156.5,
|
| 743 |
+
"learning_rate": 1.7910891089108912e-06,
|
| 744 |
+
"loss": 97.0582,
|
| 745 |
+
"step": 520
|
| 746 |
+
},
|
| 747 |
+
{
|
| 748 |
+
"epoch": 0.4674717569146864,
|
| 749 |
+
"grad_norm": 156.875,
|
| 750 |
+
"learning_rate": 1.7762376237623763e-06,
|
| 751 |
+
"loss": 99.9485,
|
| 752 |
+
"step": 525
|
| 753 |
+
},
|
| 754 |
+
{
|
| 755 |
+
"epoch": 0.47192386888530247,
|
| 756 |
+
"grad_norm": 150.5,
|
| 757 |
+
"learning_rate": 1.7613861386138614e-06,
|
| 758 |
+
"loss": 98.0501,
|
| 759 |
+
"step": 530
|
| 760 |
+
},
|
| 761 |
+
{
|
| 762 |
+
"epoch": 0.47637598085591853,
|
| 763 |
+
"grad_norm": 149.0,
|
| 764 |
+
"learning_rate": 1.7465346534653465e-06,
|
| 765 |
+
"loss": 97.4984,
|
| 766 |
+
"step": 535
|
| 767 |
+
},
|
| 768 |
+
{
|
| 769 |
+
"epoch": 0.4808280928265346,
|
| 770 |
+
"grad_norm": 159.0,
|
| 771 |
+
"learning_rate": 1.7316831683168316e-06,
|
| 772 |
+
"loss": 94.833,
|
| 773 |
+
"step": 540
|
| 774 |
+
},
|
| 775 |
+
{
|
| 776 |
+
"epoch": 0.48528020479715067,
|
| 777 |
+
"grad_norm": 154.0,
|
| 778 |
+
"learning_rate": 1.7168316831683167e-06,
|
| 779 |
+
"loss": 96.7634,
|
| 780 |
+
"step": 545
|
| 781 |
+
},
|
| 782 |
+
{
|
| 783 |
+
"epoch": 0.4897323167677667,
|
| 784 |
+
"grad_norm": 145.75,
|
| 785 |
+
"learning_rate": 1.7019801980198019e-06,
|
| 786 |
+
"loss": 96.3564,
|
| 787 |
+
"step": 550
|
| 788 |
+
},
|
| 789 |
+
{
|
| 790 |
+
"epoch": 0.49418442873838275,
|
| 791 |
+
"grad_norm": 150.75,
|
| 792 |
+
"learning_rate": 1.6871287128712874e-06,
|
| 793 |
+
"loss": 96.0712,
|
| 794 |
+
"step": 555
|
| 795 |
+
},
|
| 796 |
+
{
|
| 797 |
+
"epoch": 0.4986365407089988,
|
| 798 |
+
"grad_norm": 149.625,
|
| 799 |
+
"learning_rate": 1.6722772277227725e-06,
|
| 800 |
+
"loss": 97.4836,
|
| 801 |
+
"step": 560
|
| 802 |
+
},
|
| 803 |
+
{
|
| 804 |
+
"epoch": 0.5030886526796149,
|
| 805 |
+
"grad_norm": 145.375,
|
| 806 |
+
"learning_rate": 1.6574257425742576e-06,
|
| 807 |
+
"loss": 98.0151,
|
| 808 |
+
"step": 565
|
| 809 |
+
},
|
| 810 |
+
{
|
| 811 |
+
"epoch": 0.5075407646502309,
|
| 812 |
+
"grad_norm": 156.875,
|
| 813 |
+
"learning_rate": 1.6425742574257427e-06,
|
| 814 |
+
"loss": 97.3034,
|
| 815 |
+
"step": 570
|
| 816 |
+
},
|
| 817 |
+
{
|
| 818 |
+
"epoch": 0.511992876620847,
|
| 819 |
+
"grad_norm": 149.125,
|
| 820 |
+
"learning_rate": 1.6277227722772279e-06,
|
| 821 |
+
"loss": 97.3657,
|
| 822 |
+
"step": 575
|
| 823 |
+
},
|
| 824 |
+
{
|
| 825 |
+
"epoch": 0.516444988591463,
|
| 826 |
+
"grad_norm": 153.0,
|
| 827 |
+
"learning_rate": 1.612871287128713e-06,
|
| 828 |
+
"loss": 96.8924,
|
| 829 |
+
"step": 580
|
| 830 |
+
},
|
| 831 |
+
{
|
| 832 |
+
"epoch": 0.5208971005620792,
|
| 833 |
+
"grad_norm": 151.0,
|
| 834 |
+
"learning_rate": 1.598019801980198e-06,
|
| 835 |
+
"loss": 97.888,
|
| 836 |
+
"step": 585
|
| 837 |
+
},
|
| 838 |
+
{
|
| 839 |
+
"epoch": 0.5253492125326952,
|
| 840 |
+
"grad_norm": 157.5,
|
| 841 |
+
"learning_rate": 1.5831683168316832e-06,
|
| 842 |
+
"loss": 95.7355,
|
| 843 |
+
"step": 590
|
| 844 |
+
},
|
| 845 |
+
{
|
| 846 |
+
"epoch": 0.5298013245033113,
|
| 847 |
+
"grad_norm": 150.5,
|
| 848 |
+
"learning_rate": 1.5683168316831683e-06,
|
| 849 |
+
"loss": 94.4425,
|
| 850 |
+
"step": 595
|
| 851 |
+
},
|
| 852 |
+
{
|
| 853 |
+
"epoch": 0.5342534364739273,
|
| 854 |
+
"grad_norm": 153.0,
|
| 855 |
+
"learning_rate": 1.5534653465346534e-06,
|
| 856 |
+
"loss": 96.4229,
|
| 857 |
+
"step": 600
|
| 858 |
+
},
|
| 859 |
+
{
|
| 860 |
+
"epoch": 0.5387055484445433,
|
| 861 |
+
"grad_norm": 152.875,
|
| 862 |
+
"learning_rate": 1.5386138613861388e-06,
|
| 863 |
+
"loss": 96.716,
|
| 864 |
+
"step": 605
|
| 865 |
+
},
|
| 866 |
+
{
|
| 867 |
+
"epoch": 0.5431576604151594,
|
| 868 |
+
"grad_norm": 151.75,
|
| 869 |
+
"learning_rate": 1.5237623762376239e-06,
|
| 870 |
+
"loss": 96.9523,
|
| 871 |
+
"step": 610
|
| 872 |
+
},
|
| 873 |
+
{
|
| 874 |
+
"epoch": 0.5476097723857755,
|
| 875 |
+
"grad_norm": 149.0,
|
| 876 |
+
"learning_rate": 1.508910891089109e-06,
|
| 877 |
+
"loss": 96.1208,
|
| 878 |
+
"step": 615
|
| 879 |
+
},
|
| 880 |
+
{
|
| 881 |
+
"epoch": 0.5520618843563916,
|
| 882 |
+
"grad_norm": 151.0,
|
| 883 |
+
"learning_rate": 1.4940594059405941e-06,
|
| 884 |
+
"loss": 96.7012,
|
| 885 |
+
"step": 620
|
| 886 |
+
},
|
| 887 |
+
{
|
| 888 |
+
"epoch": 0.5565139963270076,
|
| 889 |
+
"grad_norm": 148.75,
|
| 890 |
+
"learning_rate": 1.4792079207920792e-06,
|
| 891 |
+
"loss": 95.9622,
|
| 892 |
+
"step": 625
|
| 893 |
+
},
|
| 894 |
+
{
|
| 895 |
+
"epoch": 0.5609661082976237,
|
| 896 |
+
"grad_norm": 148.0,
|
| 897 |
+
"learning_rate": 1.4643564356435644e-06,
|
| 898 |
+
"loss": 96.7668,
|
| 899 |
+
"step": 630
|
| 900 |
+
},
|
| 901 |
+
{
|
| 902 |
+
"epoch": 0.5654182202682397,
|
| 903 |
+
"grad_norm": 149.0,
|
| 904 |
+
"learning_rate": 1.4495049504950495e-06,
|
| 905 |
+
"loss": 95.4153,
|
| 906 |
+
"step": 635
|
| 907 |
+
},
|
| 908 |
+
{
|
| 909 |
+
"epoch": 0.5698703322388559,
|
| 910 |
+
"grad_norm": 149.375,
|
| 911 |
+
"learning_rate": 1.4346534653465346e-06,
|
| 912 |
+
"loss": 95.8311,
|
| 913 |
+
"step": 640
|
| 914 |
+
},
|
| 915 |
+
{
|
| 916 |
+
"epoch": 0.5743224442094719,
|
| 917 |
+
"grad_norm": 147.25,
|
| 918 |
+
"learning_rate": 1.4198019801980197e-06,
|
| 919 |
+
"loss": 97.1089,
|
| 920 |
+
"step": 645
|
| 921 |
+
},
|
| 922 |
+
{
|
| 923 |
+
"epoch": 0.5787745561800879,
|
| 924 |
+
"grad_norm": 149.75,
|
| 925 |
+
"learning_rate": 1.404950495049505e-06,
|
| 926 |
+
"loss": 96.5063,
|
| 927 |
+
"step": 650
|
| 928 |
+
},
|
| 929 |
+
{
|
| 930 |
+
"epoch": 0.583226668150704,
|
| 931 |
+
"grad_norm": 151.0,
|
| 932 |
+
"learning_rate": 1.3900990099009902e-06,
|
| 933 |
+
"loss": 97.1067,
|
| 934 |
+
"step": 655
|
| 935 |
+
},
|
| 936 |
+
{
|
| 937 |
+
"epoch": 0.58767878012132,
|
| 938 |
+
"grad_norm": 157.625,
|
| 939 |
+
"learning_rate": 1.3752475247524753e-06,
|
| 940 |
+
"loss": 96.2562,
|
| 941 |
+
"step": 660
|
| 942 |
+
},
|
| 943 |
+
{
|
| 944 |
+
"epoch": 0.5921308920919361,
|
| 945 |
+
"grad_norm": 152.375,
|
| 946 |
+
"learning_rate": 1.3603960396039604e-06,
|
| 947 |
+
"loss": 95.2903,
|
| 948 |
+
"step": 665
|
| 949 |
+
},
|
| 950 |
+
{
|
| 951 |
+
"epoch": 0.5965830040625522,
|
| 952 |
+
"grad_norm": 151.625,
|
| 953 |
+
"learning_rate": 1.3455445544554455e-06,
|
| 954 |
+
"loss": 95.4937,
|
| 955 |
+
"step": 670
|
| 956 |
+
},
|
| 957 |
+
{
|
| 958 |
+
"epoch": 0.6010351160331683,
|
| 959 |
+
"grad_norm": 151.625,
|
| 960 |
+
"learning_rate": 1.3306930693069308e-06,
|
| 961 |
+
"loss": 95.8772,
|
| 962 |
+
"step": 675
|
| 963 |
+
},
|
| 964 |
+
{
|
| 965 |
+
"epoch": 0.6054872280037843,
|
| 966 |
+
"grad_norm": 154.625,
|
| 967 |
+
"learning_rate": 1.315841584158416e-06,
|
| 968 |
+
"loss": 96.8501,
|
| 969 |
+
"step": 680
|
| 970 |
+
},
|
| 971 |
+
{
|
| 972 |
+
"epoch": 0.6099393399744003,
|
| 973 |
+
"grad_norm": 151.5,
|
| 974 |
+
"learning_rate": 1.300990099009901e-06,
|
| 975 |
+
"loss": 96.9895,
|
| 976 |
+
"step": 685
|
| 977 |
+
},
|
| 978 |
+
{
|
| 979 |
+
"epoch": 0.6143914519450164,
|
| 980 |
+
"grad_norm": 148.875,
|
| 981 |
+
"learning_rate": 1.2861386138613862e-06,
|
| 982 |
+
"loss": 94.545,
|
| 983 |
+
"step": 690
|
| 984 |
+
},
|
| 985 |
+
{
|
| 986 |
+
"epoch": 0.6188435639156324,
|
| 987 |
+
"grad_norm": 152.75,
|
| 988 |
+
"learning_rate": 1.2712871287128713e-06,
|
| 989 |
+
"loss": 95.3633,
|
| 990 |
+
"step": 695
|
| 991 |
+
},
|
| 992 |
+
{
|
| 993 |
+
"epoch": 0.6232956758862486,
|
| 994 |
+
"grad_norm": 153.875,
|
| 995 |
+
"learning_rate": 1.2564356435643564e-06,
|
| 996 |
+
"loss": 94.8764,
|
| 997 |
+
"step": 700
|
| 998 |
+
},
|
| 999 |
+
{
|
| 1000 |
+
"epoch": 0.6277477878568646,
|
| 1001 |
+
"grad_norm": 151.375,
|
| 1002 |
+
"learning_rate": 1.2415841584158415e-06,
|
| 1003 |
+
"loss": 94.5789,
|
| 1004 |
+
"step": 705
|
| 1005 |
+
},
|
| 1006 |
+
{
|
| 1007 |
+
"epoch": 0.6321998998274807,
|
| 1008 |
+
"grad_norm": 145.75,
|
| 1009 |
+
"learning_rate": 1.2267326732673267e-06,
|
| 1010 |
+
"loss": 94.5253,
|
| 1011 |
+
"step": 710
|
| 1012 |
+
},
|
| 1013 |
+
{
|
| 1014 |
+
"epoch": 0.6366520117980967,
|
| 1015 |
+
"grad_norm": 151.25,
|
| 1016 |
+
"learning_rate": 1.211881188118812e-06,
|
| 1017 |
+
"loss": 95.8074,
|
| 1018 |
+
"step": 715
|
| 1019 |
+
},
|
| 1020 |
+
{
|
| 1021 |
+
"epoch": 0.6411041237687127,
|
| 1022 |
+
"grad_norm": 152.875,
|
| 1023 |
+
"learning_rate": 1.197029702970297e-06,
|
| 1024 |
+
"loss": 94.3003,
|
| 1025 |
+
"step": 720
|
| 1026 |
+
},
|
| 1027 |
+
{
|
| 1028 |
+
"epoch": 0.6455562357393289,
|
| 1029 |
+
"grad_norm": 149.75,
|
| 1030 |
+
"learning_rate": 1.1821782178217822e-06,
|
| 1031 |
+
"loss": 93.6894,
|
| 1032 |
+
"step": 725
|
| 1033 |
+
},
|
| 1034 |
+
{
|
| 1035 |
+
"epoch": 0.6500083477099449,
|
| 1036 |
+
"grad_norm": 160.5,
|
| 1037 |
+
"learning_rate": 1.1673267326732673e-06,
|
| 1038 |
+
"loss": 93.642,
|
| 1039 |
+
"step": 730
|
| 1040 |
+
},
|
| 1041 |
+
{
|
| 1042 |
+
"epoch": 0.654460459680561,
|
| 1043 |
+
"grad_norm": 149.125,
|
| 1044 |
+
"learning_rate": 1.1524752475247524e-06,
|
| 1045 |
+
"loss": 95.5531,
|
| 1046 |
+
"step": 735
|
| 1047 |
+
},
|
| 1048 |
+
{
|
| 1049 |
+
"epoch": 0.658912571651177,
|
| 1050 |
+
"grad_norm": 155.625,
|
| 1051 |
+
"learning_rate": 1.1376237623762376e-06,
|
| 1052 |
+
"loss": 95.4458,
|
| 1053 |
+
"step": 740
|
| 1054 |
+
},
|
| 1055 |
+
{
|
| 1056 |
+
"epoch": 0.6633646836217931,
|
| 1057 |
+
"grad_norm": 146.5,
|
| 1058 |
+
"learning_rate": 1.1227722772277229e-06,
|
| 1059 |
+
"loss": 95.5642,
|
| 1060 |
+
"step": 745
|
| 1061 |
+
},
|
| 1062 |
+
{
|
| 1063 |
+
"epoch": 0.6678167955924091,
|
| 1064 |
+
"grad_norm": 150.875,
|
| 1065 |
+
"learning_rate": 1.107920792079208e-06,
|
| 1066 |
+
"loss": 91.5858,
|
| 1067 |
+
"step": 750
|
| 1068 |
+
},
|
| 1069 |
+
{
|
| 1070 |
+
"epoch": 0.6722689075630253,
|
| 1071 |
+
"grad_norm": 159.875,
|
| 1072 |
+
"learning_rate": 1.0930693069306931e-06,
|
| 1073 |
+
"loss": 95.2966,
|
| 1074 |
+
"step": 755
|
| 1075 |
+
},
|
| 1076 |
+
{
|
| 1077 |
+
"epoch": 0.6767210195336413,
|
| 1078 |
+
"grad_norm": 147.5,
|
| 1079 |
+
"learning_rate": 1.0782178217821782e-06,
|
| 1080 |
+
"loss": 93.9319,
|
| 1081 |
+
"step": 760
|
| 1082 |
+
},
|
| 1083 |
+
{
|
| 1084 |
+
"epoch": 0.6811731315042573,
|
| 1085 |
+
"grad_norm": 151.25,
|
| 1086 |
+
"learning_rate": 1.0633663366336634e-06,
|
| 1087 |
+
"loss": 96.4605,
|
| 1088 |
+
"step": 765
|
| 1089 |
+
},
|
| 1090 |
+
{
|
| 1091 |
+
"epoch": 0.6856252434748734,
|
| 1092 |
+
"grad_norm": 150.75,
|
| 1093 |
+
"learning_rate": 1.0485148514851485e-06,
|
| 1094 |
+
"loss": 95.4815,
|
| 1095 |
+
"step": 770
|
| 1096 |
+
},
|
| 1097 |
+
{
|
| 1098 |
+
"epoch": 0.6900773554454894,
|
| 1099 |
+
"grad_norm": 152.625,
|
| 1100 |
+
"learning_rate": 1.0336633663366336e-06,
|
| 1101 |
+
"loss": 95.7014,
|
| 1102 |
+
"step": 775
|
| 1103 |
+
},
|
| 1104 |
+
{
|
| 1105 |
+
"epoch": 0.6945294674161056,
|
| 1106 |
+
"grad_norm": 150.0,
|
| 1107 |
+
"learning_rate": 1.018811881188119e-06,
|
| 1108 |
+
"loss": 95.7446,
|
| 1109 |
+
"step": 780
|
| 1110 |
+
},
|
| 1111 |
+
{
|
| 1112 |
+
"epoch": 0.6989815793867216,
|
| 1113 |
+
"grad_norm": 149.75,
|
| 1114 |
+
"learning_rate": 1.003960396039604e-06,
|
| 1115 |
+
"loss": 91.6439,
|
| 1116 |
+
"step": 785
|
| 1117 |
+
},
|
| 1118 |
+
{
|
| 1119 |
+
"epoch": 0.7034336913573377,
|
| 1120 |
+
"grad_norm": 151.0,
|
| 1121 |
+
"learning_rate": 9.891089108910892e-07,
|
| 1122 |
+
"loss": 94.8783,
|
| 1123 |
+
"step": 790
|
| 1124 |
+
},
|
| 1125 |
+
{
|
| 1126 |
+
"epoch": 0.7078858033279537,
|
| 1127 |
+
"grad_norm": 147.875,
|
| 1128 |
+
"learning_rate": 9.742574257425743e-07,
|
| 1129 |
+
"loss": 94.2932,
|
| 1130 |
+
"step": 795
|
| 1131 |
+
},
|
| 1132 |
+
{
|
| 1133 |
+
"epoch": 0.7123379152985697,
|
| 1134 |
+
"grad_norm": 152.5,
|
| 1135 |
+
"learning_rate": 9.594059405940594e-07,
|
| 1136 |
+
"loss": 94.3391,
|
| 1137 |
+
"step": 800
|
| 1138 |
+
},
|
| 1139 |
+
{
|
| 1140 |
+
"epoch": 0.7167900272691858,
|
| 1141 |
+
"grad_norm": 151.625,
|
| 1142 |
+
"learning_rate": 9.445544554455446e-07,
|
| 1143 |
+
"loss": 93.1912,
|
| 1144 |
+
"step": 805
|
| 1145 |
+
},
|
| 1146 |
+
{
|
| 1147 |
+
"epoch": 0.7212421392398018,
|
| 1148 |
+
"grad_norm": 150.5,
|
| 1149 |
+
"learning_rate": 9.297029702970297e-07,
|
| 1150 |
+
"loss": 94.2237,
|
| 1151 |
+
"step": 810
|
| 1152 |
+
},
|
| 1153 |
+
{
|
| 1154 |
+
"epoch": 0.725694251210418,
|
| 1155 |
+
"grad_norm": 146.25,
|
| 1156 |
+
"learning_rate": 9.148514851485148e-07,
|
| 1157 |
+
"loss": 93.2389,
|
| 1158 |
+
"step": 815
|
| 1159 |
+
},
|
| 1160 |
+
{
|
| 1161 |
+
"epoch": 0.730146363181034,
|
| 1162 |
+
"grad_norm": 149.875,
|
| 1163 |
+
"learning_rate": 9e-07,
|
| 1164 |
+
"loss": 94.0122,
|
| 1165 |
+
"step": 820
|
| 1166 |
+
},
|
| 1167 |
+
{
|
| 1168 |
+
"epoch": 0.7345984751516501,
|
| 1169 |
+
"grad_norm": 148.375,
|
| 1170 |
+
"learning_rate": 8.851485148514851e-07,
|
| 1171 |
+
"loss": 94.1428,
|
| 1172 |
+
"step": 825
|
| 1173 |
+
},
|
| 1174 |
+
{
|
| 1175 |
+
"epoch": 0.7390505871222661,
|
| 1176 |
+
"grad_norm": 146.5,
|
| 1177 |
+
"learning_rate": 8.702970297029703e-07,
|
| 1178 |
+
"loss": 92.5999,
|
| 1179 |
+
"step": 830
|
| 1180 |
+
},
|
| 1181 |
+
{
|
| 1182 |
+
"epoch": 0.7435026990928821,
|
| 1183 |
+
"grad_norm": 158.375,
|
| 1184 |
+
"learning_rate": 8.554455445544554e-07,
|
| 1185 |
+
"loss": 93.4235,
|
| 1186 |
+
"step": 835
|
| 1187 |
+
},
|
| 1188 |
+
{
|
| 1189 |
+
"epoch": 0.7479548110634983,
|
| 1190 |
+
"grad_norm": 143.25,
|
| 1191 |
+
"learning_rate": 8.405940594059407e-07,
|
| 1192 |
+
"loss": 94.7491,
|
| 1193 |
+
"step": 840
|
| 1194 |
+
},
|
| 1195 |
+
{
|
| 1196 |
+
"epoch": 0.7524069230341143,
|
| 1197 |
+
"grad_norm": 149.375,
|
| 1198 |
+
"learning_rate": 8.257425742574259e-07,
|
| 1199 |
+
"loss": 93.3154,
|
| 1200 |
+
"step": 845
|
| 1201 |
+
},
|
| 1202 |
+
{
|
| 1203 |
+
"epoch": 0.7568590350047304,
|
| 1204 |
+
"grad_norm": 153.25,
|
| 1205 |
+
"learning_rate": 8.10891089108911e-07,
|
| 1206 |
+
"loss": 91.7949,
|
| 1207 |
+
"step": 850
|
| 1208 |
+
},
|
| 1209 |
+
{
|
| 1210 |
+
"epoch": 0.7613111469753464,
|
| 1211 |
+
"grad_norm": 151.375,
|
| 1212 |
+
"learning_rate": 7.960396039603961e-07,
|
| 1213 |
+
"loss": 91.7284,
|
| 1214 |
+
"step": 855
|
| 1215 |
+
},
|
| 1216 |
+
{
|
| 1217 |
+
"epoch": 0.7657632589459625,
|
| 1218 |
+
"grad_norm": 152.125,
|
| 1219 |
+
"learning_rate": 7.811881188118812e-07,
|
| 1220 |
+
"loss": 93.6502,
|
| 1221 |
+
"step": 860
|
| 1222 |
+
},
|
| 1223 |
+
{
|
| 1224 |
+
"epoch": 0.7702153709165785,
|
| 1225 |
+
"grad_norm": 151.125,
|
| 1226 |
+
"learning_rate": 7.663366336633663e-07,
|
| 1227 |
+
"loss": 94.1399,
|
| 1228 |
+
"step": 865
|
| 1229 |
+
},
|
| 1230 |
+
{
|
| 1231 |
+
"epoch": 0.7746674828871946,
|
| 1232 |
+
"grad_norm": 151.0,
|
| 1233 |
+
"learning_rate": 7.514851485148515e-07,
|
| 1234 |
+
"loss": 92.5499,
|
| 1235 |
+
"step": 870
|
| 1236 |
+
},
|
| 1237 |
+
{
|
| 1238 |
+
"epoch": 0.7791195948578107,
|
| 1239 |
+
"grad_norm": 150.875,
|
| 1240 |
+
"learning_rate": 7.366336633663367e-07,
|
| 1241 |
+
"loss": 93.4188,
|
| 1242 |
+
"step": 875
|
| 1243 |
+
},
|
| 1244 |
+
{
|
| 1245 |
+
"epoch": 0.7835717068284267,
|
| 1246 |
+
"grad_norm": 159.5,
|
| 1247 |
+
"learning_rate": 7.217821782178218e-07,
|
| 1248 |
+
"loss": 90.1312,
|
| 1249 |
+
"step": 880
|
| 1250 |
+
},
|
| 1251 |
+
{
|
| 1252 |
+
"epoch": 0.7880238187990428,
|
| 1253 |
+
"grad_norm": 144.25,
|
| 1254 |
+
"learning_rate": 7.069306930693069e-07,
|
| 1255 |
+
"loss": 91.4897,
|
| 1256 |
+
"step": 885
|
| 1257 |
+
},
|
| 1258 |
+
{
|
| 1259 |
+
"epoch": 0.7924759307696588,
|
| 1260 |
+
"grad_norm": 147.875,
|
| 1261 |
+
"learning_rate": 6.920792079207921e-07,
|
| 1262 |
+
"loss": 93.3765,
|
| 1263 |
+
"step": 890
|
| 1264 |
+
},
|
| 1265 |
+
{
|
| 1266 |
+
"epoch": 0.796928042740275,
|
| 1267 |
+
"grad_norm": 155.0,
|
| 1268 |
+
"learning_rate": 6.772277227722772e-07,
|
| 1269 |
+
"loss": 93.9429,
|
| 1270 |
+
"step": 895
|
| 1271 |
+
},
|
| 1272 |
+
{
|
| 1273 |
+
"epoch": 0.801380154710891,
|
| 1274 |
+
"grad_norm": 145.5,
|
| 1275 |
+
"learning_rate": 6.623762376237624e-07,
|
| 1276 |
+
"loss": 91.7795,
|
| 1277 |
+
"step": 900
|
| 1278 |
+
},
|
| 1279 |
+
{
|
| 1280 |
+
"epoch": 0.8058322666815071,
|
| 1281 |
+
"grad_norm": 148.375,
|
| 1282 |
+
"learning_rate": 6.475247524752476e-07,
|
| 1283 |
+
"loss": 92.5831,
|
| 1284 |
+
"step": 905
|
| 1285 |
+
},
|
| 1286 |
+
{
|
| 1287 |
+
"epoch": 0.8102843786521231,
|
| 1288 |
+
"grad_norm": 152.5,
|
| 1289 |
+
"learning_rate": 6.326732673267327e-07,
|
| 1290 |
+
"loss": 92.3641,
|
| 1291 |
+
"step": 910
|
| 1292 |
+
},
|
| 1293 |
+
{
|
| 1294 |
+
"epoch": 0.8147364906227391,
|
| 1295 |
+
"grad_norm": 148.75,
|
| 1296 |
+
"learning_rate": 6.178217821782178e-07,
|
| 1297 |
+
"loss": 91.3071,
|
| 1298 |
+
"step": 915
|
| 1299 |
+
},
|
| 1300 |
+
{
|
| 1301 |
+
"epoch": 0.8191886025933552,
|
| 1302 |
+
"grad_norm": 152.375,
|
| 1303 |
+
"learning_rate": 6.02970297029703e-07,
|
| 1304 |
+
"loss": 90.2066,
|
| 1305 |
+
"step": 920
|
| 1306 |
+
},
|
| 1307 |
+
{
|
| 1308 |
+
"epoch": 0.8236407145639713,
|
| 1309 |
+
"grad_norm": 148.75,
|
| 1310 |
+
"learning_rate": 5.881188118811882e-07,
|
| 1311 |
+
"loss": 91.9865,
|
| 1312 |
+
"step": 925
|
| 1313 |
+
},
|
| 1314 |
+
{
|
| 1315 |
+
"epoch": 0.8280928265345874,
|
| 1316 |
+
"grad_norm": 145.375,
|
| 1317 |
+
"learning_rate": 5.732673267326733e-07,
|
| 1318 |
+
"loss": 90.7357,
|
| 1319 |
+
"step": 930
|
| 1320 |
+
},
|
| 1321 |
+
{
|
| 1322 |
+
"epoch": 0.8325449385052034,
|
| 1323 |
+
"grad_norm": 153.5,
|
| 1324 |
+
"learning_rate": 5.584158415841584e-07,
|
| 1325 |
+
"loss": 94.0521,
|
| 1326 |
+
"step": 935
|
| 1327 |
+
},
|
| 1328 |
+
{
|
| 1329 |
+
"epoch": 0.8369970504758195,
|
| 1330 |
+
"grad_norm": 147.625,
|
| 1331 |
+
"learning_rate": 5.435643564356436e-07,
|
| 1332 |
+
"loss": 92.177,
|
| 1333 |
+
"step": 940
|
| 1334 |
+
},
|
| 1335 |
+
{
|
| 1336 |
+
"epoch": 0.8414491624464355,
|
| 1337 |
+
"grad_norm": 144.75,
|
| 1338 |
+
"learning_rate": 5.287128712871287e-07,
|
| 1339 |
+
"loss": 93.7151,
|
| 1340 |
+
"step": 945
|
| 1341 |
+
},
|
| 1342 |
+
{
|
| 1343 |
+
"epoch": 0.8459012744170515,
|
| 1344 |
+
"grad_norm": 145.375,
|
| 1345 |
+
"learning_rate": 5.138613861386139e-07,
|
| 1346 |
+
"loss": 90.6827,
|
| 1347 |
+
"step": 950
|
| 1348 |
+
},
|
| 1349 |
+
{
|
| 1350 |
+
"epoch": 0.8503533863876677,
|
| 1351 |
+
"grad_norm": 149.625,
|
| 1352 |
+
"learning_rate": 4.990099009900991e-07,
|
| 1353 |
+
"loss": 90.1044,
|
| 1354 |
+
"step": 955
|
| 1355 |
+
},
|
| 1356 |
+
{
|
| 1357 |
+
"epoch": 0.8548054983582837,
|
| 1358 |
+
"grad_norm": 147.5,
|
| 1359 |
+
"learning_rate": 4.841584158415842e-07,
|
| 1360 |
+
"loss": 90.5797,
|
| 1361 |
+
"step": 960
|
| 1362 |
+
},
|
| 1363 |
+
{
|
| 1364 |
+
"epoch": 0.8592576103288998,
|
| 1365 |
+
"grad_norm": 150.75,
|
| 1366 |
+
"learning_rate": 4.693069306930693e-07,
|
| 1367 |
+
"loss": 90.0814,
|
| 1368 |
+
"step": 965
|
| 1369 |
+
},
|
| 1370 |
+
{
|
| 1371 |
+
"epoch": 0.8637097222995158,
|
| 1372 |
+
"grad_norm": 152.5,
|
| 1373 |
+
"learning_rate": 4.5445544554455447e-07,
|
| 1374 |
+
"loss": 91.7237,
|
| 1375 |
+
"step": 970
|
| 1376 |
+
},
|
| 1377 |
+
{
|
| 1378 |
+
"epoch": 0.8681618342701319,
|
| 1379 |
+
"grad_norm": 152.625,
|
| 1380 |
+
"learning_rate": 4.396039603960396e-07,
|
| 1381 |
+
"loss": 91.0753,
|
| 1382 |
+
"step": 975
|
| 1383 |
+
},
|
| 1384 |
+
{
|
| 1385 |
+
"epoch": 0.872613946240748,
|
| 1386 |
+
"grad_norm": 152.0,
|
| 1387 |
+
"learning_rate": 4.247524752475247e-07,
|
| 1388 |
+
"loss": 91.9936,
|
| 1389 |
+
"step": 980
|
| 1390 |
+
},
|
| 1391 |
+
{
|
| 1392 |
+
"epoch": 0.877066058211364,
|
| 1393 |
+
"grad_norm": 143.5,
|
| 1394 |
+
"learning_rate": 4.099009900990099e-07,
|
| 1395 |
+
"loss": 91.7998,
|
| 1396 |
+
"step": 985
|
| 1397 |
+
},
|
| 1398 |
+
{
|
| 1399 |
+
"epoch": 0.8815181701819801,
|
| 1400 |
+
"grad_norm": 146.25,
|
| 1401 |
+
"learning_rate": 3.950495049504951e-07,
|
| 1402 |
+
"loss": 91.8213,
|
| 1403 |
+
"step": 990
|
| 1404 |
+
},
|
| 1405 |
+
{
|
| 1406 |
+
"epoch": 0.8859702821525961,
|
| 1407 |
+
"grad_norm": 153.375,
|
| 1408 |
+
"learning_rate": 3.801980198019802e-07,
|
| 1409 |
+
"loss": 90.3063,
|
| 1410 |
+
"step": 995
|
| 1411 |
+
},
|
| 1412 |
+
{
|
| 1413 |
+
"epoch": 0.8904223941232122,
|
| 1414 |
+
"grad_norm": 145.25,
|
| 1415 |
+
"learning_rate": 3.653465346534653e-07,
|
| 1416 |
+
"loss": 91.2036,
|
| 1417 |
+
"step": 1000
|
| 1418 |
+
},
|
| 1419 |
+
{
|
| 1420 |
+
"epoch": 0.8904223941232122,
|
| 1421 |
+
"eval_loss": NaN,
|
| 1422 |
+
"eval_runtime": 101.482,
|
| 1423 |
+
"eval_samples_per_second": 596.431,
|
| 1424 |
+
"eval_steps_per_second": 37.278,
|
| 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
|