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 sha256:617ff76d496774425760760c864d36a829690e93c00f300e5fd6a772bec23af2
|
| 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:bdd691842c9b3eecd0100adcd2271f6b8b162e162a40b960523591fe8491a784
|
| 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 sha256:079541ad505a3dc0e80239afe57cfe11e44b1de1d72b78fa14ede3d018356e24
|
| 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": 606.217,
|
| 716 |
"eval_steps_per_second": 37.889,
|
| 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": 0.2311754549460711,
|
| 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": 606.217,
|
| 716 |
"eval_steps_per_second": 37.889,
|
| 717 |
"step": 500
|
| 718 |
+
},
|
| 719 |
+
{
|
| 720 |
+
"epoch": 0.1167436047477659,
|
| 721 |
+
"grad_norm": 178.25,
|
| 722 |
+
"learning_rate": 9.815005138746147e-06,
|
| 723 |
+
"loss": 73.0971,
|
| 724 |
+
"step": 505
|
| 725 |
+
},
|
| 726 |
+
{
|
| 727 |
+
"epoch": 0.11789948202249625,
|
| 728 |
+
"grad_norm": 176.125,
|
| 729 |
+
"learning_rate": 9.802158273381296e-06,
|
| 730 |
+
"loss": 73.0252,
|
| 731 |
+
"step": 510
|
| 732 |
+
},
|
| 733 |
+
{
|
| 734 |
+
"epoch": 0.11905535929722662,
|
| 735 |
+
"grad_norm": 204.625,
|
| 736 |
+
"learning_rate": 9.789311408016445e-06,
|
| 737 |
+
"loss": 74.122,
|
| 738 |
+
"step": 515
|
| 739 |
+
},
|
| 740 |
+
{
|
| 741 |
+
"epoch": 0.12021123657195697,
|
| 742 |
+
"grad_norm": 175.25,
|
| 743 |
+
"learning_rate": 9.776464542651593e-06,
|
| 744 |
+
"loss": 75.304,
|
| 745 |
+
"step": 520
|
| 746 |
+
},
|
| 747 |
+
{
|
| 748 |
+
"epoch": 0.12136711384668733,
|
| 749 |
+
"grad_norm": 228.125,
|
| 750 |
+
"learning_rate": 9.763617677286742e-06,
|
| 751 |
+
"loss": 80.1305,
|
| 752 |
+
"step": 525
|
| 753 |
+
},
|
| 754 |
+
{
|
| 755 |
+
"epoch": 0.12252299112141768,
|
| 756 |
+
"grad_norm": 351.25,
|
| 757 |
+
"learning_rate": 9.750770811921891e-06,
|
| 758 |
+
"loss": 98.8667,
|
| 759 |
+
"step": 530
|
| 760 |
+
},
|
| 761 |
+
{
|
| 762 |
+
"epoch": 0.12367886839614804,
|
| 763 |
+
"grad_norm": 253.75,
|
| 764 |
+
"learning_rate": 9.73792394655704e-06,
|
| 765 |
+
"loss": 90.5137,
|
| 766 |
+
"step": 535
|
| 767 |
+
},
|
| 768 |
+
{
|
| 769 |
+
"epoch": 0.12483474567087839,
|
| 770 |
+
"grad_norm": 204.0,
|
| 771 |
+
"learning_rate": 9.72507708119219e-06,
|
| 772 |
+
"loss": 85.6656,
|
| 773 |
+
"step": 540
|
| 774 |
+
},
|
| 775 |
+
{
|
| 776 |
+
"epoch": 0.12599062294560875,
|
| 777 |
+
"grad_norm": 164.875,
|
| 778 |
+
"learning_rate": 9.712230215827338e-06,
|
| 779 |
+
"loss": 81.4311,
|
| 780 |
+
"step": 545
|
| 781 |
+
},
|
| 782 |
+
{
|
| 783 |
+
"epoch": 0.1271465002203391,
|
| 784 |
+
"grad_norm": 171.75,
|
| 785 |
+
"learning_rate": 9.699383350462488e-06,
|
| 786 |
+
"loss": 78.962,
|
| 787 |
+
"step": 550
|
| 788 |
+
},
|
| 789 |
+
{
|
| 790 |
+
"epoch": 0.12830237749506945,
|
| 791 |
+
"grad_norm": 181.375,
|
| 792 |
+
"learning_rate": 9.686536485097637e-06,
|
| 793 |
+
"loss": 78.2494,
|
| 794 |
+
"step": 555
|
| 795 |
+
},
|
| 796 |
+
{
|
| 797 |
+
"epoch": 0.12945825476979983,
|
| 798 |
+
"grad_norm": 167.125,
|
| 799 |
+
"learning_rate": 9.673689619732786e-06,
|
| 800 |
+
"loss": 76.5218,
|
| 801 |
+
"step": 560
|
| 802 |
+
},
|
| 803 |
+
{
|
| 804 |
+
"epoch": 0.13061413204453018,
|
| 805 |
+
"grad_norm": 174.625,
|
| 806 |
+
"learning_rate": 9.660842754367935e-06,
|
| 807 |
+
"loss": 76.3271,
|
| 808 |
+
"step": 565
|
| 809 |
+
},
|
| 810 |
+
{
|
| 811 |
+
"epoch": 0.13177000931926053,
|
| 812 |
+
"grad_norm": 163.125,
|
| 813 |
+
"learning_rate": 9.647995889003084e-06,
|
| 814 |
+
"loss": 77.2531,
|
| 815 |
+
"step": 570
|
| 816 |
+
},
|
| 817 |
+
{
|
| 818 |
+
"epoch": 0.13292588659399088,
|
| 819 |
+
"grad_norm": 185.875,
|
| 820 |
+
"learning_rate": 9.635149023638232e-06,
|
| 821 |
+
"loss": 76.3901,
|
| 822 |
+
"step": 575
|
| 823 |
+
},
|
| 824 |
+
{
|
| 825 |
+
"epoch": 0.13408176386872123,
|
| 826 |
+
"grad_norm": 160.5,
|
| 827 |
+
"learning_rate": 9.622302158273383e-06,
|
| 828 |
+
"loss": 76.1426,
|
| 829 |
+
"step": 580
|
| 830 |
+
},
|
| 831 |
+
{
|
| 832 |
+
"epoch": 0.1352376411434516,
|
| 833 |
+
"grad_norm": 170.75,
|
| 834 |
+
"learning_rate": 9.609455292908532e-06,
|
| 835 |
+
"loss": 75.3614,
|
| 836 |
+
"step": 585
|
| 837 |
+
},
|
| 838 |
+
{
|
| 839 |
+
"epoch": 0.13639351841818195,
|
| 840 |
+
"grad_norm": 186.125,
|
| 841 |
+
"learning_rate": 9.59660842754368e-06,
|
| 842 |
+
"loss": 75.3318,
|
| 843 |
+
"step": 590
|
| 844 |
+
},
|
| 845 |
+
{
|
| 846 |
+
"epoch": 0.1375493956929123,
|
| 847 |
+
"grad_norm": 176.5,
|
| 848 |
+
"learning_rate": 9.58376156217883e-06,
|
| 849 |
+
"loss": 73.9041,
|
| 850 |
+
"step": 595
|
| 851 |
+
},
|
| 852 |
+
{
|
| 853 |
+
"epoch": 0.13870527296764265,
|
| 854 |
+
"grad_norm": 170.25,
|
| 855 |
+
"learning_rate": 9.570914696813978e-06,
|
| 856 |
+
"loss": 74.9994,
|
| 857 |
+
"step": 600
|
| 858 |
+
},
|
| 859 |
+
{
|
| 860 |
+
"epoch": 0.13986115024237303,
|
| 861 |
+
"grad_norm": 222.625,
|
| 862 |
+
"learning_rate": 9.558067831449127e-06,
|
| 863 |
+
"loss": 76.4041,
|
| 864 |
+
"step": 605
|
| 865 |
+
},
|
| 866 |
+
{
|
| 867 |
+
"epoch": 0.14101702751710338,
|
| 868 |
+
"grad_norm": 174.625,
|
| 869 |
+
"learning_rate": 9.545220966084276e-06,
|
| 870 |
+
"loss": 76.0331,
|
| 871 |
+
"step": 610
|
| 872 |
+
},
|
| 873 |
+
{
|
| 874 |
+
"epoch": 0.14217290479183373,
|
| 875 |
+
"grad_norm": 169.875,
|
| 876 |
+
"learning_rate": 9.532374100719425e-06,
|
| 877 |
+
"loss": 74.6435,
|
| 878 |
+
"step": 615
|
| 879 |
+
},
|
| 880 |
+
{
|
| 881 |
+
"epoch": 0.14332878206656408,
|
| 882 |
+
"grad_norm": 188.75,
|
| 883 |
+
"learning_rate": 9.519527235354574e-06,
|
| 884 |
+
"loss": 76.2372,
|
| 885 |
+
"step": 620
|
| 886 |
+
},
|
| 887 |
+
{
|
| 888 |
+
"epoch": 0.14448465934129442,
|
| 889 |
+
"grad_norm": 164.25,
|
| 890 |
+
"learning_rate": 9.506680369989724e-06,
|
| 891 |
+
"loss": 75.1062,
|
| 892 |
+
"step": 625
|
| 893 |
+
},
|
| 894 |
+
{
|
| 895 |
+
"epoch": 0.1456405366160248,
|
| 896 |
+
"grad_norm": 165.125,
|
| 897 |
+
"learning_rate": 9.493833504624871e-06,
|
| 898 |
+
"loss": 75.9294,
|
| 899 |
+
"step": 630
|
| 900 |
+
},
|
| 901 |
+
{
|
| 902 |
+
"epoch": 0.14679641389075515,
|
| 903 |
+
"grad_norm": 189.75,
|
| 904 |
+
"learning_rate": 9.480986639260022e-06,
|
| 905 |
+
"loss": 75.5001,
|
| 906 |
+
"step": 635
|
| 907 |
+
},
|
| 908 |
+
{
|
| 909 |
+
"epoch": 0.1479522911654855,
|
| 910 |
+
"grad_norm": 228.5,
|
| 911 |
+
"learning_rate": 9.468139773895171e-06,
|
| 912 |
+
"loss": 74.1863,
|
| 913 |
+
"step": 640
|
| 914 |
+
},
|
| 915 |
+
{
|
| 916 |
+
"epoch": 0.14910816844021585,
|
| 917 |
+
"grad_norm": 222.0,
|
| 918 |
+
"learning_rate": 9.45529290853032e-06,
|
| 919 |
+
"loss": 76.5168,
|
| 920 |
+
"step": 645
|
| 921 |
+
},
|
| 922 |
+
{
|
| 923 |
+
"epoch": 0.15026404571494623,
|
| 924 |
+
"grad_norm": 187.875,
|
| 925 |
+
"learning_rate": 9.442446043165469e-06,
|
| 926 |
+
"loss": 74.2555,
|
| 927 |
+
"step": 650
|
| 928 |
+
},
|
| 929 |
+
{
|
| 930 |
+
"epoch": 0.15141992298967658,
|
| 931 |
+
"grad_norm": 159.0,
|
| 932 |
+
"learning_rate": 9.429599177800617e-06,
|
| 933 |
+
"loss": 75.0447,
|
| 934 |
+
"step": 655
|
| 935 |
+
},
|
| 936 |
+
{
|
| 937 |
+
"epoch": 0.15257580026440692,
|
| 938 |
+
"grad_norm": 158.75,
|
| 939 |
+
"learning_rate": 9.416752312435766e-06,
|
| 940 |
+
"loss": 73.3359,
|
| 941 |
+
"step": 660
|
| 942 |
+
},
|
| 943 |
+
{
|
| 944 |
+
"epoch": 0.15373167753913727,
|
| 945 |
+
"grad_norm": 170.625,
|
| 946 |
+
"learning_rate": 9.403905447070915e-06,
|
| 947 |
+
"loss": 73.3929,
|
| 948 |
+
"step": 665
|
| 949 |
+
},
|
| 950 |
+
{
|
| 951 |
+
"epoch": 0.15488755481386765,
|
| 952 |
+
"grad_norm": 202.5,
|
| 953 |
+
"learning_rate": 9.391058581706064e-06,
|
| 954 |
+
"loss": 75.0084,
|
| 955 |
+
"step": 670
|
| 956 |
+
},
|
| 957 |
+
{
|
| 958 |
+
"epoch": 0.156043432088598,
|
| 959 |
+
"grad_norm": 194.625,
|
| 960 |
+
"learning_rate": 9.378211716341213e-06,
|
| 961 |
+
"loss": 74.5239,
|
| 962 |
+
"step": 675
|
| 963 |
+
},
|
| 964 |
+
{
|
| 965 |
+
"epoch": 0.15719930936332835,
|
| 966 |
+
"grad_norm": 178.375,
|
| 967 |
+
"learning_rate": 9.365364850976363e-06,
|
| 968 |
+
"loss": 75.5832,
|
| 969 |
+
"step": 680
|
| 970 |
+
},
|
| 971 |
+
{
|
| 972 |
+
"epoch": 0.1583551866380587,
|
| 973 |
+
"grad_norm": 162.75,
|
| 974 |
+
"learning_rate": 9.35251798561151e-06,
|
| 975 |
+
"loss": 74.5449,
|
| 976 |
+
"step": 685
|
| 977 |
+
},
|
| 978 |
+
{
|
| 979 |
+
"epoch": 0.15951106391278905,
|
| 980 |
+
"grad_norm": 184.75,
|
| 981 |
+
"learning_rate": 9.339671120246661e-06,
|
| 982 |
+
"loss": 74.4716,
|
| 983 |
+
"step": 690
|
| 984 |
+
},
|
| 985 |
+
{
|
| 986 |
+
"epoch": 0.16066694118751942,
|
| 987 |
+
"grad_norm": 175.0,
|
| 988 |
+
"learning_rate": 9.32682425488181e-06,
|
| 989 |
+
"loss": 75.2684,
|
| 990 |
+
"step": 695
|
| 991 |
+
},
|
| 992 |
+
{
|
| 993 |
+
"epoch": 0.16182281846224977,
|
| 994 |
+
"grad_norm": 173.75,
|
| 995 |
+
"learning_rate": 9.313977389516959e-06,
|
| 996 |
+
"loss": 73.7116,
|
| 997 |
+
"step": 700
|
| 998 |
+
},
|
| 999 |
+
{
|
| 1000 |
+
"epoch": 0.16297869573698012,
|
| 1001 |
+
"grad_norm": 165.375,
|
| 1002 |
+
"learning_rate": 9.301130524152108e-06,
|
| 1003 |
+
"loss": 73.4956,
|
| 1004 |
+
"step": 705
|
| 1005 |
+
},
|
| 1006 |
+
{
|
| 1007 |
+
"epoch": 0.16413457301171047,
|
| 1008 |
+
"grad_norm": 175.875,
|
| 1009 |
+
"learning_rate": 9.288283658787256e-06,
|
| 1010 |
+
"loss": 73.5028,
|
| 1011 |
+
"step": 710
|
| 1012 |
+
},
|
| 1013 |
+
{
|
| 1014 |
+
"epoch": 0.16529045028644085,
|
| 1015 |
+
"grad_norm": 171.375,
|
| 1016 |
+
"learning_rate": 9.275436793422405e-06,
|
| 1017 |
+
"loss": 73.6011,
|
| 1018 |
+
"step": 715
|
| 1019 |
+
},
|
| 1020 |
+
{
|
| 1021 |
+
"epoch": 0.1664463275611712,
|
| 1022 |
+
"grad_norm": 175.75,
|
| 1023 |
+
"learning_rate": 9.262589928057554e-06,
|
| 1024 |
+
"loss": 74.965,
|
| 1025 |
+
"step": 720
|
| 1026 |
+
},
|
| 1027 |
+
{
|
| 1028 |
+
"epoch": 0.16760220483590155,
|
| 1029 |
+
"grad_norm": 160.625,
|
| 1030 |
+
"learning_rate": 9.249743062692705e-06,
|
| 1031 |
+
"loss": 74.7802,
|
| 1032 |
+
"step": 725
|
| 1033 |
+
},
|
| 1034 |
+
{
|
| 1035 |
+
"epoch": 0.1687580821106319,
|
| 1036 |
+
"grad_norm": 163.875,
|
| 1037 |
+
"learning_rate": 9.236896197327852e-06,
|
| 1038 |
+
"loss": 74.4619,
|
| 1039 |
+
"step": 730
|
| 1040 |
+
},
|
| 1041 |
+
{
|
| 1042 |
+
"epoch": 0.16991395938536225,
|
| 1043 |
+
"grad_norm": 173.0,
|
| 1044 |
+
"learning_rate": 9.224049331963002e-06,
|
| 1045 |
+
"loss": 74.9448,
|
| 1046 |
+
"step": 735
|
| 1047 |
+
},
|
| 1048 |
+
{
|
| 1049 |
+
"epoch": 0.17106983666009262,
|
| 1050 |
+
"grad_norm": 193.375,
|
| 1051 |
+
"learning_rate": 9.21120246659815e-06,
|
| 1052 |
+
"loss": 74.6345,
|
| 1053 |
+
"step": 740
|
| 1054 |
+
},
|
| 1055 |
+
{
|
| 1056 |
+
"epoch": 0.17222571393482297,
|
| 1057 |
+
"grad_norm": 168.25,
|
| 1058 |
+
"learning_rate": 9.1983556012333e-06,
|
| 1059 |
+
"loss": 74.5997,
|
| 1060 |
+
"step": 745
|
| 1061 |
+
},
|
| 1062 |
+
{
|
| 1063 |
+
"epoch": 0.17338159120955332,
|
| 1064 |
+
"grad_norm": 181.375,
|
| 1065 |
+
"learning_rate": 9.185508735868449e-06,
|
| 1066 |
+
"loss": 72.9167,
|
| 1067 |
+
"step": 750
|
| 1068 |
+
},
|
| 1069 |
+
{
|
| 1070 |
+
"epoch": 0.17453746848428367,
|
| 1071 |
+
"grad_norm": 209.875,
|
| 1072 |
+
"learning_rate": 9.172661870503598e-06,
|
| 1073 |
+
"loss": 75.2782,
|
| 1074 |
+
"step": 755
|
| 1075 |
+
},
|
| 1076 |
+
{
|
| 1077 |
+
"epoch": 0.17569334575901405,
|
| 1078 |
+
"grad_norm": 165.25,
|
| 1079 |
+
"learning_rate": 9.159815005138747e-06,
|
| 1080 |
+
"loss": 73.2717,
|
| 1081 |
+
"step": 760
|
| 1082 |
+
},
|
| 1083 |
+
{
|
| 1084 |
+
"epoch": 0.1768492230337444,
|
| 1085 |
+
"grad_norm": 168.75,
|
| 1086 |
+
"learning_rate": 9.146968139773897e-06,
|
| 1087 |
+
"loss": 75.1429,
|
| 1088 |
+
"step": 765
|
| 1089 |
+
},
|
| 1090 |
+
{
|
| 1091 |
+
"epoch": 0.17800510030847475,
|
| 1092 |
+
"grad_norm": 161.0,
|
| 1093 |
+
"learning_rate": 9.134121274409044e-06,
|
| 1094 |
+
"loss": 74.8651,
|
| 1095 |
+
"step": 770
|
| 1096 |
+
},
|
| 1097 |
+
{
|
| 1098 |
+
"epoch": 0.1791609775832051,
|
| 1099 |
+
"grad_norm": 186.375,
|
| 1100 |
+
"learning_rate": 9.121274409044195e-06,
|
| 1101 |
+
"loss": 75.679,
|
| 1102 |
+
"step": 775
|
| 1103 |
+
},
|
| 1104 |
+
{
|
| 1105 |
+
"epoch": 0.18031685485793547,
|
| 1106 |
+
"grad_norm": 183.875,
|
| 1107 |
+
"learning_rate": 9.108427543679344e-06,
|
| 1108 |
+
"loss": 74.1174,
|
| 1109 |
+
"step": 780
|
| 1110 |
+
},
|
| 1111 |
+
{
|
| 1112 |
+
"epoch": 0.18147273213266582,
|
| 1113 |
+
"grad_norm": 160.0,
|
| 1114 |
+
"learning_rate": 9.095580678314493e-06,
|
| 1115 |
+
"loss": 74.8214,
|
| 1116 |
+
"step": 785
|
| 1117 |
+
},
|
| 1118 |
+
{
|
| 1119 |
+
"epoch": 0.18262860940739617,
|
| 1120 |
+
"grad_norm": 173.625,
|
| 1121 |
+
"learning_rate": 9.082733812949641e-06,
|
| 1122 |
+
"loss": 72.8824,
|
| 1123 |
+
"step": 790
|
| 1124 |
+
},
|
| 1125 |
+
{
|
| 1126 |
+
"epoch": 0.18378448668212652,
|
| 1127 |
+
"grad_norm": 184.75,
|
| 1128 |
+
"learning_rate": 9.06988694758479e-06,
|
| 1129 |
+
"loss": 75.2577,
|
| 1130 |
+
"step": 795
|
| 1131 |
+
},
|
| 1132 |
+
{
|
| 1133 |
+
"epoch": 0.18494036395685687,
|
| 1134 |
+
"grad_norm": 171.75,
|
| 1135 |
+
"learning_rate": 9.057040082219939e-06,
|
| 1136 |
+
"loss": 74.0234,
|
| 1137 |
+
"step": 800
|
| 1138 |
+
},
|
| 1139 |
+
{
|
| 1140 |
+
"epoch": 0.18609624123158724,
|
| 1141 |
+
"grad_norm": 166.125,
|
| 1142 |
+
"learning_rate": 9.044193216855088e-06,
|
| 1143 |
+
"loss": 74.5028,
|
| 1144 |
+
"step": 805
|
| 1145 |
+
},
|
| 1146 |
+
{
|
| 1147 |
+
"epoch": 0.1872521185063176,
|
| 1148 |
+
"grad_norm": 181.25,
|
| 1149 |
+
"learning_rate": 9.031346351490237e-06,
|
| 1150 |
+
"loss": 74.7906,
|
| 1151 |
+
"step": 810
|
| 1152 |
+
},
|
| 1153 |
+
{
|
| 1154 |
+
"epoch": 0.18840799578104794,
|
| 1155 |
+
"grad_norm": 168.25,
|
| 1156 |
+
"learning_rate": 9.018499486125386e-06,
|
| 1157 |
+
"loss": 73.6598,
|
| 1158 |
+
"step": 815
|
| 1159 |
+
},
|
| 1160 |
+
{
|
| 1161 |
+
"epoch": 0.1895638730557783,
|
| 1162 |
+
"grad_norm": 213.625,
|
| 1163 |
+
"learning_rate": 9.005652620760536e-06,
|
| 1164 |
+
"loss": 74.4191,
|
| 1165 |
+
"step": 820
|
| 1166 |
+
},
|
| 1167 |
+
{
|
| 1168 |
+
"epoch": 0.19071975033050867,
|
| 1169 |
+
"grad_norm": 184.5,
|
| 1170 |
+
"learning_rate": 8.992805755395683e-06,
|
| 1171 |
+
"loss": 74.0181,
|
| 1172 |
+
"step": 825
|
| 1173 |
+
},
|
| 1174 |
+
{
|
| 1175 |
+
"epoch": 0.19187562760523902,
|
| 1176 |
+
"grad_norm": 168.75,
|
| 1177 |
+
"learning_rate": 8.979958890030834e-06,
|
| 1178 |
+
"loss": 72.935,
|
| 1179 |
+
"step": 830
|
| 1180 |
+
},
|
| 1181 |
+
{
|
| 1182 |
+
"epoch": 0.19303150487996937,
|
| 1183 |
+
"grad_norm": 176.625,
|
| 1184 |
+
"learning_rate": 8.967112024665983e-06,
|
| 1185 |
+
"loss": 74.5831,
|
| 1186 |
+
"step": 835
|
| 1187 |
+
},
|
| 1188 |
+
{
|
| 1189 |
+
"epoch": 0.19418738215469972,
|
| 1190 |
+
"grad_norm": 168.625,
|
| 1191 |
+
"learning_rate": 8.954265159301132e-06,
|
| 1192 |
+
"loss": 75.029,
|
| 1193 |
+
"step": 840
|
| 1194 |
+
},
|
| 1195 |
+
{
|
| 1196 |
+
"epoch": 0.1953432594294301,
|
| 1197 |
+
"grad_norm": 188.0,
|
| 1198 |
+
"learning_rate": 8.94141829393628e-06,
|
| 1199 |
+
"loss": 75.1383,
|
| 1200 |
+
"step": 845
|
| 1201 |
+
},
|
| 1202 |
+
{
|
| 1203 |
+
"epoch": 0.19649913670416044,
|
| 1204 |
+
"grad_norm": 163.75,
|
| 1205 |
+
"learning_rate": 8.92857142857143e-06,
|
| 1206 |
+
"loss": 73.7028,
|
| 1207 |
+
"step": 850
|
| 1208 |
+
},
|
| 1209 |
+
{
|
| 1210 |
+
"epoch": 0.1976550139788908,
|
| 1211 |
+
"grad_norm": 168.75,
|
| 1212 |
+
"learning_rate": 8.915724563206578e-06,
|
| 1213 |
+
"loss": 73.8515,
|
| 1214 |
+
"step": 855
|
| 1215 |
+
},
|
| 1216 |
+
{
|
| 1217 |
+
"epoch": 0.19881089125362114,
|
| 1218 |
+
"grad_norm": 199.875,
|
| 1219 |
+
"learning_rate": 8.902877697841727e-06,
|
| 1220 |
+
"loss": 74.1701,
|
| 1221 |
+
"step": 860
|
| 1222 |
+
},
|
| 1223 |
+
{
|
| 1224 |
+
"epoch": 0.1999667685283515,
|
| 1225 |
+
"grad_norm": 157.5,
|
| 1226 |
+
"learning_rate": 8.890030832476876e-06,
|
| 1227 |
+
"loss": 75.171,
|
| 1228 |
+
"step": 865
|
| 1229 |
+
},
|
| 1230 |
+
{
|
| 1231 |
+
"epoch": 0.20112264580308187,
|
| 1232 |
+
"grad_norm": 189.875,
|
| 1233 |
+
"learning_rate": 8.877183967112025e-06,
|
| 1234 |
+
"loss": 73.9979,
|
| 1235 |
+
"step": 870
|
| 1236 |
+
},
|
| 1237 |
+
{
|
| 1238 |
+
"epoch": 0.20227852307781222,
|
| 1239 |
+
"grad_norm": 160.75,
|
| 1240 |
+
"learning_rate": 8.864337101747175e-06,
|
| 1241 |
+
"loss": 73.9971,
|
| 1242 |
+
"step": 875
|
| 1243 |
+
},
|
| 1244 |
+
{
|
| 1245 |
+
"epoch": 0.20343440035254257,
|
| 1246 |
+
"grad_norm": 170.75,
|
| 1247 |
+
"learning_rate": 8.851490236382322e-06,
|
| 1248 |
+
"loss": 73.9474,
|
| 1249 |
+
"step": 880
|
| 1250 |
+
},
|
| 1251 |
+
{
|
| 1252 |
+
"epoch": 0.20459027762727292,
|
| 1253 |
+
"grad_norm": 161.875,
|
| 1254 |
+
"learning_rate": 8.838643371017473e-06,
|
| 1255 |
+
"loss": 73.6239,
|
| 1256 |
+
"step": 885
|
| 1257 |
+
},
|
| 1258 |
+
{
|
| 1259 |
+
"epoch": 0.2057461549020033,
|
| 1260 |
+
"grad_norm": 208.25,
|
| 1261 |
+
"learning_rate": 8.825796505652622e-06,
|
| 1262 |
+
"loss": 74.3579,
|
| 1263 |
+
"step": 890
|
| 1264 |
+
},
|
| 1265 |
+
{
|
| 1266 |
+
"epoch": 0.20690203217673364,
|
| 1267 |
+
"grad_norm": 182.375,
|
| 1268 |
+
"learning_rate": 8.81294964028777e-06,
|
| 1269 |
+
"loss": 73.6577,
|
| 1270 |
+
"step": 895
|
| 1271 |
+
},
|
| 1272 |
+
{
|
| 1273 |
+
"epoch": 0.208057909451464,
|
| 1274 |
+
"grad_norm": 167.375,
|
| 1275 |
+
"learning_rate": 8.80010277492292e-06,
|
| 1276 |
+
"loss": 73.492,
|
| 1277 |
+
"step": 900
|
| 1278 |
+
},
|
| 1279 |
+
{
|
| 1280 |
+
"epoch": 0.20921378672619434,
|
| 1281 |
+
"grad_norm": 202.625,
|
| 1282 |
+
"learning_rate": 8.787255909558068e-06,
|
| 1283 |
+
"loss": 73.0436,
|
| 1284 |
+
"step": 905
|
| 1285 |
+
},
|
| 1286 |
+
{
|
| 1287 |
+
"epoch": 0.2103696640009247,
|
| 1288 |
+
"grad_norm": 156.5,
|
| 1289 |
+
"learning_rate": 8.774409044193217e-06,
|
| 1290 |
+
"loss": 74.3569,
|
| 1291 |
+
"step": 910
|
| 1292 |
+
},
|
| 1293 |
+
{
|
| 1294 |
+
"epoch": 0.21152554127565507,
|
| 1295 |
+
"grad_norm": 167.75,
|
| 1296 |
+
"learning_rate": 8.761562178828366e-06,
|
| 1297 |
+
"loss": 73.5566,
|
| 1298 |
+
"step": 915
|
| 1299 |
+
},
|
| 1300 |
+
{
|
| 1301 |
+
"epoch": 0.21268141855038541,
|
| 1302 |
+
"grad_norm": 159.0,
|
| 1303 |
+
"learning_rate": 8.748715313463516e-06,
|
| 1304 |
+
"loss": 73.4377,
|
| 1305 |
+
"step": 920
|
| 1306 |
+
},
|
| 1307 |
+
{
|
| 1308 |
+
"epoch": 0.21383729582511576,
|
| 1309 |
+
"grad_norm": 197.5,
|
| 1310 |
+
"learning_rate": 8.735868448098664e-06,
|
| 1311 |
+
"loss": 73.5647,
|
| 1312 |
+
"step": 925
|
| 1313 |
+
},
|
| 1314 |
+
{
|
| 1315 |
+
"epoch": 0.2149931730998461,
|
| 1316 |
+
"grad_norm": 157.75,
|
| 1317 |
+
"learning_rate": 8.723021582733814e-06,
|
| 1318 |
+
"loss": 74.3935,
|
| 1319 |
+
"step": 930
|
| 1320 |
+
},
|
| 1321 |
+
{
|
| 1322 |
+
"epoch": 0.2161490503745765,
|
| 1323 |
+
"grad_norm": 160.0,
|
| 1324 |
+
"learning_rate": 8.710174717368961e-06,
|
| 1325 |
+
"loss": 73.5235,
|
| 1326 |
+
"step": 935
|
| 1327 |
+
},
|
| 1328 |
+
{
|
| 1329 |
+
"epoch": 0.21730492764930684,
|
| 1330 |
+
"grad_norm": 193.5,
|
| 1331 |
+
"learning_rate": 8.697327852004112e-06,
|
| 1332 |
+
"loss": 74.8152,
|
| 1333 |
+
"step": 940
|
| 1334 |
+
},
|
| 1335 |
+
{
|
| 1336 |
+
"epoch": 0.2184608049240372,
|
| 1337 |
+
"grad_norm": 166.125,
|
| 1338 |
+
"learning_rate": 8.68448098663926e-06,
|
| 1339 |
+
"loss": 74.962,
|
| 1340 |
+
"step": 945
|
| 1341 |
+
},
|
| 1342 |
+
{
|
| 1343 |
+
"epoch": 0.21961668219876754,
|
| 1344 |
+
"grad_norm": 187.75,
|
| 1345 |
+
"learning_rate": 8.67163412127441e-06,
|
| 1346 |
+
"loss": 73.3269,
|
| 1347 |
+
"step": 950
|
| 1348 |
+
},
|
| 1349 |
+
{
|
| 1350 |
+
"epoch": 0.22077255947349791,
|
| 1351 |
+
"grad_norm": 177.625,
|
| 1352 |
+
"learning_rate": 8.658787255909558e-06,
|
| 1353 |
+
"loss": 72.6171,
|
| 1354 |
+
"step": 955
|
| 1355 |
+
},
|
| 1356 |
+
{
|
| 1357 |
+
"epoch": 0.22192843674822826,
|
| 1358 |
+
"grad_norm": 207.875,
|
| 1359 |
+
"learning_rate": 8.645940390544709e-06,
|
| 1360 |
+
"loss": 73.4562,
|
| 1361 |
+
"step": 960
|
| 1362 |
+
},
|
| 1363 |
+
{
|
| 1364 |
+
"epoch": 0.2230843140229586,
|
| 1365 |
+
"grad_norm": 197.0,
|
| 1366 |
+
"learning_rate": 8.633093525179856e-06,
|
| 1367 |
+
"loss": 72.9479,
|
| 1368 |
+
"step": 965
|
| 1369 |
+
},
|
| 1370 |
+
{
|
| 1371 |
+
"epoch": 0.22424019129768896,
|
| 1372 |
+
"grad_norm": 178.75,
|
| 1373 |
+
"learning_rate": 8.620246659815007e-06,
|
| 1374 |
+
"loss": 73.2069,
|
| 1375 |
+
"step": 970
|
| 1376 |
+
},
|
| 1377 |
+
{
|
| 1378 |
+
"epoch": 0.2253960685724193,
|
| 1379 |
+
"grad_norm": 184.375,
|
| 1380 |
+
"learning_rate": 8.607399794450156e-06,
|
| 1381 |
+
"loss": 72.4369,
|
| 1382 |
+
"step": 975
|
| 1383 |
+
},
|
| 1384 |
+
{
|
| 1385 |
+
"epoch": 0.2265519458471497,
|
| 1386 |
+
"grad_norm": 193.625,
|
| 1387 |
+
"learning_rate": 8.594552929085304e-06,
|
| 1388 |
+
"loss": 75.2301,
|
| 1389 |
+
"step": 980
|
| 1390 |
+
},
|
| 1391 |
+
{
|
| 1392 |
+
"epoch": 0.22770782312188004,
|
| 1393 |
+
"grad_norm": 160.75,
|
| 1394 |
+
"learning_rate": 8.581706063720453e-06,
|
| 1395 |
+
"loss": 74.4445,
|
| 1396 |
+
"step": 985
|
| 1397 |
+
},
|
| 1398 |
+
{
|
| 1399 |
+
"epoch": 0.2288637003966104,
|
| 1400 |
+
"grad_norm": 188.0,
|
| 1401 |
+
"learning_rate": 8.568859198355602e-06,
|
| 1402 |
+
"loss": 73.1566,
|
| 1403 |
+
"step": 990
|
| 1404 |
+
},
|
| 1405 |
+
{
|
| 1406 |
+
"epoch": 0.23001957767134074,
|
| 1407 |
+
"grad_norm": 166.0,
|
| 1408 |
+
"learning_rate": 8.556012332990751e-06,
|
| 1409 |
+
"loss": 72.9151,
|
| 1410 |
+
"step": 995
|
| 1411 |
+
},
|
| 1412 |
+
{
|
| 1413 |
+
"epoch": 0.2311754549460711,
|
| 1414 |
+
"grad_norm": 165.5,
|
| 1415 |
+
"learning_rate": 8.5431654676259e-06,
|
| 1416 |
+
"loss": 72.4977,
|
| 1417 |
+
"step": 1000
|
| 1418 |
+
},
|
| 1419 |
+
{
|
| 1420 |
+
"epoch": 0.2311754549460711,
|
| 1421 |
+
"eval_loss": NaN,
|
| 1422 |
+
"eval_runtime": 382.2843,
|
| 1423 |
+
"eval_samples_per_second": 609.845,
|
| 1424 |
+
"eval_steps_per_second": 38.116,
|
| 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
|