Training in progress, epoch 2, checkpoint
Browse files- last-checkpoint/model-00001-of-00003.safetensors +1 -1
- last-checkpoint/model-00002-of-00003.safetensors +1 -1
- last-checkpoint/model-00003-of-00003.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 +691 -3
last-checkpoint/model-00001-of-00003.safetensors
CHANGED
|
@@ -1,3 +1,3 @@
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
-
oid sha256:
|
| 3 |
size 4949453792
|
|
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:35e7bd1801d0d4245d02fcc221540d160d9ca4436762abd40ab696436db37997
|
| 3 |
size 4949453792
|
last-checkpoint/model-00002-of-00003.safetensors
CHANGED
|
@@ -1,3 +1,3 @@
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
-
oid sha256:
|
| 3 |
size 4999819336
|
|
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:0fac833f220a4de94d80f1c6128e85dbd0534b37e1778d74f64d089406b9a9cd
|
| 3 |
size 4999819336
|
last-checkpoint/model-00003-of-00003.safetensors
CHANGED
|
@@ -1,3 +1,3 @@
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
-
oid sha256:
|
| 3 |
size 4546807800
|
|
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:a5d6ebf26b088bbc6f219afaf49ff803ae69e0485761a48a6794d7aaaccc4eb7
|
| 3 |
size 4546807800
|
last-checkpoint/optimizer.pt
CHANGED
|
@@ -1,3 +1,3 @@
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
-
oid sha256:
|
| 3 |
size 28992348490
|
|
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:84bf6c04da13948bebd4d30d4b13cd682886c3f4762b4f85e343d6b5fe8ad40e
|
| 3 |
size 28992348490
|
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 sha256:06fea830cf5ad73ec00d500ea6fb952740ac936f18e93fa2d32abde1ea3ead92
|
| 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:be561d1df19be227394d8ea607c54262a06c9bf880af0aa5e04a52596a2a6cb0
|
| 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 sha256:03f3e24417a59435f5a8450a4aeb0f09cc92734b5c3b45a0701b2c043c415c05
|
| 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 sha256:2bea02744c29f30024590ab1629a0e7b7dabbf1e8476456c2e7c5ce46dc35c28
|
| 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 sha256:041be966454b60c86af576fc1eb7f34189114689abff8f9622b947110f7334c8
|
| 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:b85766f6596d15a810177d77dd259d9b50588cf100ec5f8ebff5fed881d57957
|
| 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 sha256:8be75d04b1ebe614241b88fd010a5dda1b7bf703c00c6ebe310ca07975830fe7
|
| 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 sha256:4699833a7ab4cb692996ef7567f934c0bac79d6a067963a873f89a38e412bd48
|
| 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 1064
|
|
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:a23384b1a4df8f5cde36ecc73a9742d7fae5940c0f154d6cc580286c571d0dba
|
| 3 |
size 1064
|
last-checkpoint/trainer_state.json
CHANGED
|
@@ -1,9 +1,9 @@
|
|
| 1 |
{
|
| 2 |
"best_metric": null,
|
| 3 |
"best_model_checkpoint": null,
|
| 4 |
-
"epoch":
|
| 5 |
"eval_steps": 300,
|
| 6 |
-
"global_step":
|
| 7 |
"is_hyper_param_search": false,
|
| 8 |
"is_local_process_zero": true,
|
| 9 |
"is_world_process_zero": true,
|
|
@@ -694,6 +694,694 @@
|
|
| 694 |
"learning_rate": 1.5043989923528937e-06,
|
| 695 |
"loss": 1.2183,
|
| 696 |
"step": 480
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 697 |
}
|
| 698 |
],
|
| 699 |
"logging_steps": 5,
|
|
@@ -713,7 +1401,7 @@
|
|
| 713 |
"attributes": {}
|
| 714 |
}
|
| 715 |
},
|
| 716 |
-
"total_flos":
|
| 717 |
"train_batch_size": 8,
|
| 718 |
"trial_name": null,
|
| 719 |
"trial_params": null
|
|
|
|
| 1 |
{
|
| 2 |
"best_metric": null,
|
| 3 |
"best_model_checkpoint": null,
|
| 4 |
+
"epoch": 2.0,
|
| 5 |
"eval_steps": 300,
|
| 6 |
+
"global_step": 962,
|
| 7 |
"is_hyper_param_search": false,
|
| 8 |
"is_local_process_zero": true,
|
| 9 |
"is_world_process_zero": true,
|
|
|
|
| 694 |
"learning_rate": 1.5043989923528937e-06,
|
| 695 |
"loss": 1.2183,
|
| 696 |
"step": 480
|
| 697 |
+
},
|
| 698 |
+
{
|
| 699 |
+
"epoch": 1.0083160083160083,
|
| 700 |
+
"grad_norm": 3.453125,
|
| 701 |
+
"learning_rate": 1.4949567433771448e-06,
|
| 702 |
+
"loss": 1.1826,
|
| 703 |
+
"step": 485
|
| 704 |
+
},
|
| 705 |
+
{
|
| 706 |
+
"epoch": 1.0187110187110187,
|
| 707 |
+
"grad_norm": 3.609375,
|
| 708 |
+
"learning_rate": 1.4854556812111887e-06,
|
| 709 |
+
"loss": 1.1925,
|
| 710 |
+
"step": 490
|
| 711 |
+
},
|
| 712 |
+
{
|
| 713 |
+
"epoch": 1.0291060291060292,
|
| 714 |
+
"grad_norm": 3.5,
|
| 715 |
+
"learning_rate": 1.4758969348178766e-06,
|
| 716 |
+
"loss": 1.19,
|
| 717 |
+
"step": 495
|
| 718 |
+
},
|
| 719 |
+
{
|
| 720 |
+
"epoch": 1.0395010395010396,
|
| 721 |
+
"grad_norm": 3.53125,
|
| 722 |
+
"learning_rate": 1.4662816400143836e-06,
|
| 723 |
+
"loss": 1.1963,
|
| 724 |
+
"step": 500
|
| 725 |
+
},
|
| 726 |
+
{
|
| 727 |
+
"epoch": 1.04989604989605,
|
| 728 |
+
"grad_norm": 3.640625,
|
| 729 |
+
"learning_rate": 1.4566109393372433e-06,
|
| 730 |
+
"loss": 1.1872,
|
| 731 |
+
"step": 505
|
| 732 |
+
},
|
| 733 |
+
{
|
| 734 |
+
"epoch": 1.0602910602910602,
|
| 735 |
+
"grad_norm": 3.578125,
|
| 736 |
+
"learning_rate": 1.4468859819065882e-06,
|
| 737 |
+
"loss": 1.1833,
|
| 738 |
+
"step": 510
|
| 739 |
+
},
|
| 740 |
+
{
|
| 741 |
+
"epoch": 1.0706860706860706,
|
| 742 |
+
"grad_norm": 3.5625,
|
| 743 |
+
"learning_rate": 1.4371079232896044e-06,
|
| 744 |
+
"loss": 1.1815,
|
| 745 |
+
"step": 515
|
| 746 |
+
},
|
| 747 |
+
{
|
| 748 |
+
"epoch": 1.0810810810810811,
|
| 749 |
+
"grad_norm": 3.53125,
|
| 750 |
+
"learning_rate": 1.4272779253632212e-06,
|
| 751 |
+
"loss": 1.1855,
|
| 752 |
+
"step": 520
|
| 753 |
+
},
|
| 754 |
+
{
|
| 755 |
+
"epoch": 1.0914760914760915,
|
| 756 |
+
"grad_norm": 3.59375,
|
| 757 |
+
"learning_rate": 1.4173971561760518e-06,
|
| 758 |
+
"loss": 1.188,
|
| 759 |
+
"step": 525
|
| 760 |
+
},
|
| 761 |
+
{
|
| 762 |
+
"epoch": 1.1018711018711018,
|
| 763 |
+
"grad_norm": 3.609375,
|
| 764 |
+
"learning_rate": 1.4074667898096009e-06,
|
| 765 |
+
"loss": 1.1873,
|
| 766 |
+
"step": 530
|
| 767 |
+
},
|
| 768 |
+
{
|
| 769 |
+
"epoch": 1.1122661122661124,
|
| 770 |
+
"grad_norm": 3.53125,
|
| 771 |
+
"learning_rate": 1.397488006238752e-06,
|
| 772 |
+
"loss": 1.1945,
|
| 773 |
+
"step": 535
|
| 774 |
+
},
|
| 775 |
+
{
|
| 776 |
+
"epoch": 1.1226611226611227,
|
| 777 |
+
"grad_norm": 3.5,
|
| 778 |
+
"learning_rate": 1.387461991191559e-06,
|
| 779 |
+
"loss": 1.1856,
|
| 780 |
+
"step": 540
|
| 781 |
+
},
|
| 782 |
+
{
|
| 783 |
+
"epoch": 1.133056133056133,
|
| 784 |
+
"grad_norm": 3.609375,
|
| 785 |
+
"learning_rate": 1.3773899360083524e-06,
|
| 786 |
+
"loss": 1.1868,
|
| 787 |
+
"step": 545
|
| 788 |
+
},
|
| 789 |
+
{
|
| 790 |
+
"epoch": 1.1434511434511434,
|
| 791 |
+
"grad_norm": 3.609375,
|
| 792 |
+
"learning_rate": 1.3672730375001773e-06,
|
| 793 |
+
"loss": 1.1791,
|
| 794 |
+
"step": 550
|
| 795 |
+
},
|
| 796 |
+
{
|
| 797 |
+
"epoch": 1.1538461538461537,
|
| 798 |
+
"grad_norm": 3.65625,
|
| 799 |
+
"learning_rate": 1.357112497806582e-06,
|
| 800 |
+
"loss": 1.1969,
|
| 801 |
+
"step": 555
|
| 802 |
+
},
|
| 803 |
+
{
|
| 804 |
+
"epoch": 1.1642411642411643,
|
| 805 |
+
"grad_norm": 3.609375,
|
| 806 |
+
"learning_rate": 1.3469095242527764e-06,
|
| 807 |
+
"loss": 1.1828,
|
| 808 |
+
"step": 560
|
| 809 |
+
},
|
| 810 |
+
{
|
| 811 |
+
"epoch": 1.1746361746361746,
|
| 812 |
+
"grad_norm": 3.515625,
|
| 813 |
+
"learning_rate": 1.3366653292061682e-06,
|
| 814 |
+
"loss": 1.1803,
|
| 815 |
+
"step": 565
|
| 816 |
+
},
|
| 817 |
+
{
|
| 818 |
+
"epoch": 1.185031185031185,
|
| 819 |
+
"grad_norm": 3.53125,
|
| 820 |
+
"learning_rate": 1.3263811299323063e-06,
|
| 821 |
+
"loss": 1.1803,
|
| 822 |
+
"step": 570
|
| 823 |
+
},
|
| 824 |
+
{
|
| 825 |
+
"epoch": 1.1954261954261955,
|
| 826 |
+
"grad_norm": 3.6875,
|
| 827 |
+
"learning_rate": 1.3160581484502382e-06,
|
| 828 |
+
"loss": 1.1823,
|
| 829 |
+
"step": 575
|
| 830 |
+
},
|
| 831 |
+
{
|
| 832 |
+
"epoch": 1.2058212058212059,
|
| 833 |
+
"grad_norm": 3.71875,
|
| 834 |
+
"learning_rate": 1.3056976113873037e-06,
|
| 835 |
+
"loss": 1.1832,
|
| 836 |
+
"step": 580
|
| 837 |
+
},
|
| 838 |
+
{
|
| 839 |
+
"epoch": 1.2162162162162162,
|
| 840 |
+
"grad_norm": 3.625,
|
| 841 |
+
"learning_rate": 1.2953007498333807e-06,
|
| 842 |
+
"loss": 1.1841,
|
| 843 |
+
"step": 585
|
| 844 |
+
},
|
| 845 |
+
{
|
| 846 |
+
"epoch": 1.2266112266112266,
|
| 847 |
+
"grad_norm": 3.59375,
|
| 848 |
+
"learning_rate": 1.284868799194602e-06,
|
| 849 |
+
"loss": 1.1804,
|
| 850 |
+
"step": 590
|
| 851 |
+
},
|
| 852 |
+
{
|
| 853 |
+
"epoch": 1.237006237006237,
|
| 854 |
+
"grad_norm": 3.5,
|
| 855 |
+
"learning_rate": 1.2744029990465574e-06,
|
| 856 |
+
"loss": 1.1811,
|
| 857 |
+
"step": 595
|
| 858 |
+
},
|
| 859 |
+
{
|
| 860 |
+
"epoch": 1.2474012474012475,
|
| 861 |
+
"grad_norm": 3.5625,
|
| 862 |
+
"learning_rate": 1.2639045929870018e-06,
|
| 863 |
+
"loss": 1.1794,
|
| 864 |
+
"step": 600
|
| 865 |
+
},
|
| 866 |
+
{
|
| 867 |
+
"epoch": 1.2474012474012475,
|
| 868 |
+
"eval_loss": 1.2036519050598145,
|
| 869 |
+
"eval_runtime": 10.8624,
|
| 870 |
+
"eval_samples_per_second": 85.616,
|
| 871 |
+
"eval_steps_per_second": 2.762,
|
| 872 |
+
"step": 600
|
| 873 |
+
},
|
| 874 |
+
{
|
| 875 |
+
"epoch": 1.2577962577962578,
|
| 876 |
+
"grad_norm": 3.609375,
|
| 877 |
+
"learning_rate": 1.2533748284880842e-06,
|
| 878 |
+
"loss": 1.1905,
|
| 879 |
+
"step": 605
|
| 880 |
+
},
|
| 881 |
+
{
|
| 882 |
+
"epoch": 1.2681912681912682,
|
| 883 |
+
"grad_norm": 3.5625,
|
| 884 |
+
"learning_rate": 1.2428149567481184e-06,
|
| 885 |
+
"loss": 1.1836,
|
| 886 |
+
"step": 610
|
| 887 |
+
},
|
| 888 |
+
{
|
| 889 |
+
"epoch": 1.2785862785862787,
|
| 890 |
+
"grad_norm": 3.5625,
|
| 891 |
+
"learning_rate": 1.2322262325429063e-06,
|
| 892 |
+
"loss": 1.1823,
|
| 893 |
+
"step": 615
|
| 894 |
+
},
|
| 895 |
+
{
|
| 896 |
+
"epoch": 1.288981288981289,
|
| 897 |
+
"grad_norm": 3.5625,
|
| 898 |
+
"learning_rate": 1.2216099140766436e-06,
|
| 899 |
+
"loss": 1.1836,
|
| 900 |
+
"step": 620
|
| 901 |
+
},
|
| 902 |
+
{
|
| 903 |
+
"epoch": 1.2993762993762994,
|
| 904 |
+
"grad_norm": 3.71875,
|
| 905 |
+
"learning_rate": 1.2109672628324104e-06,
|
| 906 |
+
"loss": 1.1837,
|
| 907 |
+
"step": 625
|
| 908 |
+
},
|
| 909 |
+
{
|
| 910 |
+
"epoch": 1.3097713097713097,
|
| 911 |
+
"grad_norm": 3.578125,
|
| 912 |
+
"learning_rate": 1.2002995434222767e-06,
|
| 913 |
+
"loss": 1.1827,
|
| 914 |
+
"step": 630
|
| 915 |
+
},
|
| 916 |
+
{
|
| 917 |
+
"epoch": 1.32016632016632,
|
| 918 |
+
"grad_norm": 3.6875,
|
| 919 |
+
"learning_rate": 1.1896080234370355e-06,
|
| 920 |
+
"loss": 1.1803,
|
| 921 |
+
"step": 635
|
| 922 |
+
},
|
| 923 |
+
{
|
| 924 |
+
"epoch": 1.3305613305613306,
|
| 925 |
+
"grad_norm": 3.515625,
|
| 926 |
+
"learning_rate": 1.178893973295581e-06,
|
| 927 |
+
"loss": 1.1788,
|
| 928 |
+
"step": 640
|
| 929 |
+
},
|
| 930 |
+
{
|
| 931 |
+
"epoch": 1.340956340956341,
|
| 932 |
+
"grad_norm": 3.640625,
|
| 933 |
+
"learning_rate": 1.1681586660939504e-06,
|
| 934 |
+
"loss": 1.1918,
|
| 935 |
+
"step": 645
|
| 936 |
+
},
|
| 937 |
+
{
|
| 938 |
+
"epoch": 1.3513513513513513,
|
| 939 |
+
"grad_norm": 3.578125,
|
| 940 |
+
"learning_rate": 1.1574033774540505e-06,
|
| 941 |
+
"loss": 1.1796,
|
| 942 |
+
"step": 650
|
| 943 |
+
},
|
| 944 |
+
{
|
| 945 |
+
"epoch": 1.3617463617463619,
|
| 946 |
+
"grad_norm": 3.65625,
|
| 947 |
+
"learning_rate": 1.1466293853720795e-06,
|
| 948 |
+
"loss": 1.1837,
|
| 949 |
+
"step": 655
|
| 950 |
+
},
|
| 951 |
+
{
|
| 952 |
+
"epoch": 1.3721413721413722,
|
| 953 |
+
"grad_norm": 3.609375,
|
| 954 |
+
"learning_rate": 1.1358379700666703e-06,
|
| 955 |
+
"loss": 1.1776,
|
| 956 |
+
"step": 660
|
| 957 |
+
},
|
| 958 |
+
{
|
| 959 |
+
"epoch": 1.3825363825363826,
|
| 960 |
+
"grad_norm": 3.546875,
|
| 961 |
+
"learning_rate": 1.1250304138267701e-06,
|
| 962 |
+
"loss": 1.181,
|
| 963 |
+
"step": 665
|
| 964 |
+
},
|
| 965 |
+
{
|
| 966 |
+
"epoch": 1.392931392931393,
|
| 967 |
+
"grad_norm": 3.6875,
|
| 968 |
+
"learning_rate": 1.11420800085927e-06,
|
| 969 |
+
"loss": 1.1764,
|
| 970 |
+
"step": 670
|
| 971 |
+
},
|
| 972 |
+
{
|
| 973 |
+
"epoch": 1.4033264033264032,
|
| 974 |
+
"grad_norm": 3.65625,
|
| 975 |
+
"learning_rate": 1.1033720171364108e-06,
|
| 976 |
+
"loss": 1.1792,
|
| 977 |
+
"step": 675
|
| 978 |
+
},
|
| 979 |
+
{
|
| 980 |
+
"epoch": 1.4137214137214138,
|
| 981 |
+
"grad_norm": 3.546875,
|
| 982 |
+
"learning_rate": 1.092523750242977e-06,
|
| 983 |
+
"loss": 1.1784,
|
| 984 |
+
"step": 680
|
| 985 |
+
},
|
| 986 |
+
{
|
| 987 |
+
"epoch": 1.4241164241164241,
|
| 988 |
+
"grad_norm": 3.65625,
|
| 989 |
+
"learning_rate": 1.0816644892232997e-06,
|
| 990 |
+
"loss": 1.1855,
|
| 991 |
+
"step": 685
|
| 992 |
+
},
|
| 993 |
+
{
|
| 994 |
+
"epoch": 1.4345114345114345,
|
| 995 |
+
"grad_norm": 3.640625,
|
| 996 |
+
"learning_rate": 1.070795524428086e-06,
|
| 997 |
+
"loss": 1.1782,
|
| 998 |
+
"step": 690
|
| 999 |
+
},
|
| 1000 |
+
{
|
| 1001 |
+
"epoch": 1.444906444906445,
|
| 1002 |
+
"grad_norm": 3.5625,
|
| 1003 |
+
"learning_rate": 1.0599181473610938e-06,
|
| 1004 |
+
"loss": 1.1837,
|
| 1005 |
+
"step": 695
|
| 1006 |
+
},
|
| 1007 |
+
{
|
| 1008 |
+
"epoch": 1.4553014553014554,
|
| 1009 |
+
"grad_norm": 3.59375,
|
| 1010 |
+
"learning_rate": 1.049033650525668e-06,
|
| 1011 |
+
"loss": 1.1786,
|
| 1012 |
+
"step": 700
|
| 1013 |
+
},
|
| 1014 |
+
{
|
| 1015 |
+
"epoch": 1.4656964656964657,
|
| 1016 |
+
"grad_norm": 3.5625,
|
| 1017 |
+
"learning_rate": 1.0381433272711585e-06,
|
| 1018 |
+
"loss": 1.1747,
|
| 1019 |
+
"step": 705
|
| 1020 |
+
},
|
| 1021 |
+
{
|
| 1022 |
+
"epoch": 1.476091476091476,
|
| 1023 |
+
"grad_norm": 3.71875,
|
| 1024 |
+
"learning_rate": 1.0272484716392408e-06,
|
| 1025 |
+
"loss": 1.1854,
|
| 1026 |
+
"step": 710
|
| 1027 |
+
},
|
| 1028 |
+
{
|
| 1029 |
+
"epoch": 1.4864864864864864,
|
| 1030 |
+
"grad_norm": 3.59375,
|
| 1031 |
+
"learning_rate": 1.0163503782101484e-06,
|
| 1032 |
+
"loss": 1.1755,
|
| 1033 |
+
"step": 715
|
| 1034 |
+
},
|
| 1035 |
+
{
|
| 1036 |
+
"epoch": 1.496881496881497,
|
| 1037 |
+
"grad_norm": 3.578125,
|
| 1038 |
+
"learning_rate": 1.0054503419488454e-06,
|
| 1039 |
+
"loss": 1.1795,
|
| 1040 |
+
"step": 720
|
| 1041 |
+
},
|
| 1042 |
+
{
|
| 1043 |
+
"epoch": 1.5072765072765073,
|
| 1044 |
+
"grad_norm": 3.578125,
|
| 1045 |
+
"learning_rate": 9.945496580511543e-07,
|
| 1046 |
+
"loss": 1.1846,
|
| 1047 |
+
"step": 725
|
| 1048 |
+
},
|
| 1049 |
+
{
|
| 1050 |
+
"epoch": 1.5176715176715176,
|
| 1051 |
+
"grad_norm": 3.6875,
|
| 1052 |
+
"learning_rate": 9.836496217898518e-07,
|
| 1053 |
+
"loss": 1.1806,
|
| 1054 |
+
"step": 730
|
| 1055 |
+
},
|
| 1056 |
+
{
|
| 1057 |
+
"epoch": 1.5280665280665282,
|
| 1058 |
+
"grad_norm": 3.609375,
|
| 1059 |
+
"learning_rate": 9.72751528360759e-07,
|
| 1060 |
+
"loss": 1.1667,
|
| 1061 |
+
"step": 735
|
| 1062 |
+
},
|
| 1063 |
+
{
|
| 1064 |
+
"epoch": 1.5384615384615383,
|
| 1065 |
+
"grad_norm": 3.65625,
|
| 1066 |
+
"learning_rate": 9.618566727288414e-07,
|
| 1067 |
+
"loss": 1.185,
|
| 1068 |
+
"step": 740
|
| 1069 |
+
},
|
| 1070 |
+
{
|
| 1071 |
+
"epoch": 1.5488565488565489,
|
| 1072 |
+
"grad_norm": 3.703125,
|
| 1073 |
+
"learning_rate": 9.509663494743321e-07,
|
| 1074 |
+
"loss": 1.191,
|
| 1075 |
+
"step": 745
|
| 1076 |
+
},
|
| 1077 |
+
{
|
| 1078 |
+
"epoch": 1.5592515592515592,
|
| 1079 |
+
"grad_norm": 4.9375,
|
| 1080 |
+
"learning_rate": 9.400818526389062e-07,
|
| 1081 |
+
"loss": 1.1769,
|
| 1082 |
+
"step": 750
|
| 1083 |
+
},
|
| 1084 |
+
{
|
| 1085 |
+
"epoch": 1.5696465696465696,
|
| 1086 |
+
"grad_norm": 3.546875,
|
| 1087 |
+
"learning_rate": 9.292044755719138e-07,
|
| 1088 |
+
"loss": 1.1741,
|
| 1089 |
+
"step": 755
|
| 1090 |
+
},
|
| 1091 |
+
{
|
| 1092 |
+
"epoch": 1.5800415800415801,
|
| 1093 |
+
"grad_norm": 3.546875,
|
| 1094 |
+
"learning_rate": 9.183355107767003e-07,
|
| 1095 |
+
"loss": 1.1771,
|
| 1096 |
+
"step": 760
|
| 1097 |
+
},
|
| 1098 |
+
{
|
| 1099 |
+
"epoch": 1.5904365904365905,
|
| 1100 |
+
"grad_norm": 3.59375,
|
| 1101 |
+
"learning_rate": 9.07476249757023e-07,
|
| 1102 |
+
"loss": 1.1725,
|
| 1103 |
+
"step": 765
|
| 1104 |
+
},
|
| 1105 |
+
{
|
| 1106 |
+
"epoch": 1.6008316008316008,
|
| 1107 |
+
"grad_norm": 3.59375,
|
| 1108 |
+
"learning_rate": 8.966279828635894e-07,
|
| 1109 |
+
"loss": 1.1801,
|
| 1110 |
+
"step": 770
|
| 1111 |
+
},
|
| 1112 |
+
{
|
| 1113 |
+
"epoch": 1.6112266112266114,
|
| 1114 |
+
"grad_norm": 3.609375,
|
| 1115 |
+
"learning_rate": 8.8579199914073e-07,
|
| 1116 |
+
"loss": 1.1697,
|
| 1117 |
+
"step": 775
|
| 1118 |
+
},
|
| 1119 |
+
{
|
| 1120 |
+
"epoch": 1.6216216216216215,
|
| 1121 |
+
"grad_norm": 3.546875,
|
| 1122 |
+
"learning_rate": 8.749695861732299e-07,
|
| 1123 |
+
"loss": 1.175,
|
| 1124 |
+
"step": 780
|
| 1125 |
+
},
|
| 1126 |
+
{
|
| 1127 |
+
"epoch": 1.632016632016632,
|
| 1128 |
+
"grad_norm": 3.578125,
|
| 1129 |
+
"learning_rate": 8.641620299333295e-07,
|
| 1130 |
+
"loss": 1.1818,
|
| 1131 |
+
"step": 785
|
| 1132 |
+
},
|
| 1133 |
+
{
|
| 1134 |
+
"epoch": 1.6424116424116424,
|
| 1135 |
+
"grad_norm": 3.65625,
|
| 1136 |
+
"learning_rate": 8.533706146279207e-07,
|
| 1137 |
+
"loss": 1.1697,
|
| 1138 |
+
"step": 790
|
| 1139 |
+
},
|
| 1140 |
+
{
|
| 1141 |
+
"epoch": 1.6528066528066527,
|
| 1142 |
+
"grad_norm": 3.59375,
|
| 1143 |
+
"learning_rate": 8.425966225459493e-07,
|
| 1144 |
+
"loss": 1.1702,
|
| 1145 |
+
"step": 795
|
| 1146 |
+
},
|
| 1147 |
+
{
|
| 1148 |
+
"epoch": 1.6632016632016633,
|
| 1149 |
+
"grad_norm": 3.65625,
|
| 1150 |
+
"learning_rate": 8.318413339060495e-07,
|
| 1151 |
+
"loss": 1.1793,
|
| 1152 |
+
"step": 800
|
| 1153 |
+
},
|
| 1154 |
+
{
|
| 1155 |
+
"epoch": 1.6735966735966736,
|
| 1156 |
+
"grad_norm": 3.65625,
|
| 1157 |
+
"learning_rate": 8.21106026704419e-07,
|
| 1158 |
+
"loss": 1.1801,
|
| 1159 |
+
"step": 805
|
| 1160 |
+
},
|
| 1161 |
+
{
|
| 1162 |
+
"epoch": 1.683991683991684,
|
| 1163 |
+
"grad_norm": 3.59375,
|
| 1164 |
+
"learning_rate": 8.103919765629645e-07,
|
| 1165 |
+
"loss": 1.1797,
|
| 1166 |
+
"step": 810
|
| 1167 |
+
},
|
| 1168 |
+
{
|
| 1169 |
+
"epoch": 1.6943866943866945,
|
| 1170 |
+
"grad_norm": 3.5625,
|
| 1171 |
+
"learning_rate": 7.997004565777233e-07,
|
| 1172 |
+
"loss": 1.1659,
|
| 1173 |
+
"step": 815
|
| 1174 |
+
},
|
| 1175 |
+
{
|
| 1176 |
+
"epoch": 1.7047817047817047,
|
| 1177 |
+
"grad_norm": 3.671875,
|
| 1178 |
+
"learning_rate": 7.890327371675895e-07,
|
| 1179 |
+
"loss": 1.1743,
|
| 1180 |
+
"step": 820
|
| 1181 |
+
},
|
| 1182 |
+
{
|
| 1183 |
+
"epoch": 1.7151767151767152,
|
| 1184 |
+
"grad_norm": 3.671875,
|
| 1185 |
+
"learning_rate": 7.783900859233562e-07,
|
| 1186 |
+
"loss": 1.1849,
|
| 1187 |
+
"step": 825
|
| 1188 |
+
},
|
| 1189 |
+
{
|
| 1190 |
+
"epoch": 1.7255717255717256,
|
| 1191 |
+
"grad_norm": 3.625,
|
| 1192 |
+
"learning_rate": 7.677737674570936e-07,
|
| 1193 |
+
"loss": 1.1791,
|
| 1194 |
+
"step": 830
|
| 1195 |
+
},
|
| 1196 |
+
{
|
| 1197 |
+
"epoch": 1.735966735966736,
|
| 1198 |
+
"grad_norm": 3.578125,
|
| 1199 |
+
"learning_rate": 7.571850432518819e-07,
|
| 1200 |
+
"loss": 1.1789,
|
| 1201 |
+
"step": 835
|
| 1202 |
+
},
|
| 1203 |
+
{
|
| 1204 |
+
"epoch": 1.7463617463617465,
|
| 1205 |
+
"grad_norm": 3.5625,
|
| 1206 |
+
"learning_rate": 7.466251715119156e-07,
|
| 1207 |
+
"loss": 1.1773,
|
| 1208 |
+
"step": 840
|
| 1209 |
+
},
|
| 1210 |
+
{
|
| 1211 |
+
"epoch": 1.7567567567567568,
|
| 1212 |
+
"grad_norm": 3.609375,
|
| 1213 |
+
"learning_rate": 7.360954070129981e-07,
|
| 1214 |
+
"loss": 1.175,
|
| 1215 |
+
"step": 845
|
| 1216 |
+
},
|
| 1217 |
+
{
|
| 1218 |
+
"epoch": 1.7671517671517671,
|
| 1219 |
+
"grad_norm": 3.609375,
|
| 1220 |
+
"learning_rate": 7.255970009534425e-07,
|
| 1221 |
+
"loss": 1.1816,
|
| 1222 |
+
"step": 850
|
| 1223 |
+
},
|
| 1224 |
+
{
|
| 1225 |
+
"epoch": 1.7775467775467777,
|
| 1226 |
+
"grad_norm": 3.734375,
|
| 1227 |
+
"learning_rate": 7.151312008053979e-07,
|
| 1228 |
+
"loss": 1.1848,
|
| 1229 |
+
"step": 855
|
| 1230 |
+
},
|
| 1231 |
+
{
|
| 1232 |
+
"epoch": 1.7879417879417878,
|
| 1233 |
+
"grad_norm": 3.5625,
|
| 1234 |
+
"learning_rate": 7.046992501666195e-07,
|
| 1235 |
+
"loss": 1.175,
|
| 1236 |
+
"step": 860
|
| 1237 |
+
},
|
| 1238 |
+
{
|
| 1239 |
+
"epoch": 1.7983367983367984,
|
| 1240 |
+
"grad_norm": 3.625,
|
| 1241 |
+
"learning_rate": 6.943023886126965e-07,
|
| 1242 |
+
"loss": 1.1753,
|
| 1243 |
+
"step": 865
|
| 1244 |
+
},
|
| 1245 |
+
{
|
| 1246 |
+
"epoch": 1.8087318087318087,
|
| 1247 |
+
"grad_norm": 3.5625,
|
| 1248 |
+
"learning_rate": 6.839418515497618e-07,
|
| 1249 |
+
"loss": 1.1718,
|
| 1250 |
+
"step": 870
|
| 1251 |
+
},
|
| 1252 |
+
{
|
| 1253 |
+
"epoch": 1.819126819126819,
|
| 1254 |
+
"grad_norm": 3.59375,
|
| 1255 |
+
"learning_rate": 6.736188700676935e-07,
|
| 1256 |
+
"loss": 1.164,
|
| 1257 |
+
"step": 875
|
| 1258 |
+
},
|
| 1259 |
+
{
|
| 1260 |
+
"epoch": 1.8295218295218296,
|
| 1261 |
+
"grad_norm": 3.59375,
|
| 1262 |
+
"learning_rate": 6.633346707938319e-07,
|
| 1263 |
+
"loss": 1.1761,
|
| 1264 |
+
"step": 880
|
| 1265 |
+
},
|
| 1266 |
+
{
|
| 1267 |
+
"epoch": 1.83991683991684,
|
| 1268 |
+
"grad_norm": 3.671875,
|
| 1269 |
+
"learning_rate": 6.530904757472236e-07,
|
| 1270 |
+
"loss": 1.1869,
|
| 1271 |
+
"step": 885
|
| 1272 |
+
},
|
| 1273 |
+
{
|
| 1274 |
+
"epoch": 1.8503118503118503,
|
| 1275 |
+
"grad_norm": 3.59375,
|
| 1276 |
+
"learning_rate": 6.42887502193418e-07,
|
| 1277 |
+
"loss": 1.1836,
|
| 1278 |
+
"step": 890
|
| 1279 |
+
},
|
| 1280 |
+
{
|
| 1281 |
+
"epoch": 1.8607068607068609,
|
| 1282 |
+
"grad_norm": 3.578125,
|
| 1283 |
+
"learning_rate": 6.327269624998227e-07,
|
| 1284 |
+
"loss": 1.1699,
|
| 1285 |
+
"step": 895
|
| 1286 |
+
},
|
| 1287 |
+
{
|
| 1288 |
+
"epoch": 1.871101871101871,
|
| 1289 |
+
"grad_norm": 3.65625,
|
| 1290 |
+
"learning_rate": 6.226100639916474e-07,
|
| 1291 |
+
"loss": 1.1743,
|
| 1292 |
+
"step": 900
|
| 1293 |
+
},
|
| 1294 |
+
{
|
| 1295 |
+
"epoch": 1.871101871101871,
|
| 1296 |
+
"eval_loss": 1.1942965984344482,
|
| 1297 |
+
"eval_runtime": 10.786,
|
| 1298 |
+
"eval_samples_per_second": 86.223,
|
| 1299 |
+
"eval_steps_per_second": 2.781,
|
| 1300 |
+
"step": 900
|
| 1301 |
+
},
|
| 1302 |
+
{
|
| 1303 |
+
"epoch": 1.8814968814968815,
|
| 1304 |
+
"grad_norm": 3.53125,
|
| 1305 |
+
"learning_rate": 6.125380088084408e-07,
|
| 1306 |
+
"loss": 1.1797,
|
| 1307 |
+
"step": 905
|
| 1308 |
+
},
|
| 1309 |
+
{
|
| 1310 |
+
"epoch": 1.8918918918918919,
|
| 1311 |
+
"grad_norm": 3.71875,
|
| 1312 |
+
"learning_rate": 6.025119937612481e-07,
|
| 1313 |
+
"loss": 1.1758,
|
| 1314 |
+
"step": 910
|
| 1315 |
+
},
|
| 1316 |
+
{
|
| 1317 |
+
"epoch": 1.9022869022869022,
|
| 1318 |
+
"grad_norm": 3.65625,
|
| 1319 |
+
"learning_rate": 5.925332101903994e-07,
|
| 1320 |
+
"loss": 1.1783,
|
| 1321 |
+
"step": 915
|
| 1322 |
+
},
|
| 1323 |
+
{
|
| 1324 |
+
"epoch": 1.9126819126819128,
|
| 1325 |
+
"grad_norm": 3.71875,
|
| 1326 |
+
"learning_rate": 5.826028438239479e-07,
|
| 1327 |
+
"loss": 1.1763,
|
| 1328 |
+
"step": 920
|
| 1329 |
+
},
|
| 1330 |
+
{
|
| 1331 |
+
"epoch": 1.9230769230769231,
|
| 1332 |
+
"grad_norm": 3.546875,
|
| 1333 |
+
"learning_rate": 5.727220746367791e-07,
|
| 1334 |
+
"loss": 1.18,
|
| 1335 |
+
"step": 925
|
| 1336 |
+
},
|
| 1337 |
+
{
|
| 1338 |
+
"epoch": 1.9334719334719335,
|
| 1339 |
+
"grad_norm": 3.546875,
|
| 1340 |
+
"learning_rate": 5.628920767103957e-07,
|
| 1341 |
+
"loss": 1.1782,
|
| 1342 |
+
"step": 930
|
| 1343 |
+
},
|
| 1344 |
+
{
|
| 1345 |
+
"epoch": 1.943866943866944,
|
| 1346 |
+
"grad_norm": 3.65625,
|
| 1347 |
+
"learning_rate": 5.531140180934119e-07,
|
| 1348 |
+
"loss": 1.1772,
|
| 1349 |
+
"step": 935
|
| 1350 |
+
},
|
| 1351 |
+
{
|
| 1352 |
+
"epoch": 1.9542619542619541,
|
| 1353 |
+
"grad_norm": 3.546875,
|
| 1354 |
+
"learning_rate": 5.433890606627568e-07,
|
| 1355 |
+
"loss": 1.175,
|
| 1356 |
+
"step": 940
|
| 1357 |
+
},
|
| 1358 |
+
{
|
| 1359 |
+
"epoch": 1.9646569646569647,
|
| 1360 |
+
"grad_norm": 3.59375,
|
| 1361 |
+
"learning_rate": 5.337183599856164e-07,
|
| 1362 |
+
"loss": 1.1745,
|
| 1363 |
+
"step": 945
|
| 1364 |
+
},
|
| 1365 |
+
{
|
| 1366 |
+
"epoch": 1.975051975051975,
|
| 1367 |
+
"grad_norm": 3.5625,
|
| 1368 |
+
"learning_rate": 5.241030651821231e-07,
|
| 1369 |
+
"loss": 1.1662,
|
| 1370 |
+
"step": 950
|
| 1371 |
+
},
|
| 1372 |
+
{
|
| 1373 |
+
"epoch": 1.9854469854469854,
|
| 1374 |
+
"grad_norm": 3.609375,
|
| 1375 |
+
"learning_rate": 5.145443187888114e-07,
|
| 1376 |
+
"loss": 1.171,
|
| 1377 |
+
"step": 955
|
| 1378 |
+
},
|
| 1379 |
+
{
|
| 1380 |
+
"epoch": 1.995841995841996,
|
| 1381 |
+
"grad_norm": 3.640625,
|
| 1382 |
+
"learning_rate": 5.050432566228552e-07,
|
| 1383 |
+
"loss": 1.1831,
|
| 1384 |
+
"step": 960
|
| 1385 |
}
|
| 1386 |
],
|
| 1387 |
"logging_steps": 5,
|
|
|
|
| 1401 |
"attributes": {}
|
| 1402 |
}
|
| 1403 |
},
|
| 1404 |
+
"total_flos": 5.381932813203276e+18,
|
| 1405 |
"train_batch_size": 8,
|
| 1406 |
"trial_name": null,
|
| 1407 |
"trial_params": null
|