Training in progress, step 2020, checkpoint
Browse files
last-checkpoint/optimizer.pt
CHANGED
|
@@ -1,3 +1,3 @@
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
-
oid sha256:
|
| 3 |
size 1657
|
|
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:f81cf03e1b17591f60580ba7c8226945d1300b6f988a995687a59ba1751fe885
|
| 3 |
size 1657
|
last-checkpoint/rng_state.pth
CHANGED
|
@@ -1,3 +1,3 @@
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
-
oid sha256:
|
| 3 |
size 14709
|
|
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:a8c4a03519a7c965ae85f1d83247241546f756653ea5de5190de7d08d7468c35
|
| 3 |
size 14709
|
last-checkpoint/scheduler.pt
CHANGED
|
@@ -1,3 +1,3 @@
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
-
oid sha256:
|
| 3 |
size 1465
|
|
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:857e7c9fa4409f57728415ac7ee974353cb8a65f06e52a28d53c52c5235d936a
|
| 3 |
size 1465
|
last-checkpoint/trainer_state.json
CHANGED
|
@@ -2,9 +2,9 @@
|
|
| 2 |
"best_global_step": null,
|
| 3 |
"best_metric": null,
|
| 4 |
"best_model_checkpoint": null,
|
| 5 |
-
"epoch": 15.
|
| 6 |
"eval_steps": 500,
|
| 7 |
-
"global_step":
|
| 8 |
"is_hyper_param_search": false,
|
| 9 |
"is_local_process_zero": true,
|
| 10 |
"is_world_process_zero": true,
|
|
@@ -3626,6 +3626,24 @@
|
|
| 3626 |
"mean_token_accuracy": 0.6472333878278732,
|
| 3627 |
"num_tokens": 7328009.0,
|
| 3628 |
"step": 2010
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 3629 |
}
|
| 3630 |
],
|
| 3631 |
"logging_steps": 5,
|
|
@@ -3645,7 +3663,7 @@
|
|
| 3645 |
"attributes": {}
|
| 3646 |
}
|
| 3647 |
},
|
| 3648 |
-
"total_flos": 3.
|
| 3649 |
"train_batch_size": 4,
|
| 3650 |
"trial_name": null,
|
| 3651 |
"trial_params": null
|
|
|
|
| 2 |
"best_global_step": null,
|
| 3 |
"best_metric": null,
|
| 4 |
"best_model_checkpoint": null,
|
| 5 |
+
"epoch": 15.538461538461538,
|
| 6 |
"eval_steps": 500,
|
| 7 |
+
"global_step": 2020,
|
| 8 |
"is_hyper_param_search": false,
|
| 9 |
"is_local_process_zero": true,
|
| 10 |
"is_world_process_zero": true,
|
|
|
|
| 3626 |
"mean_token_accuracy": 0.6472333878278732,
|
| 3627 |
"num_tokens": 7328009.0,
|
| 3628 |
"step": 2010
|
| 3629 |
+
},
|
| 3630 |
+
{
|
| 3631 |
+
"epoch": 15.5,
|
| 3632 |
+
"grad_norm": 0.0,
|
| 3633 |
+
"learning_rate": 3.993e-05,
|
| 3634 |
+
"loss": 1.713,
|
| 3635 |
+
"mean_token_accuracy": 0.6539407223463058,
|
| 3636 |
+
"num_tokens": 7346790.0,
|
| 3637 |
+
"step": 2015
|
| 3638 |
+
},
|
| 3639 |
+
{
|
| 3640 |
+
"epoch": 15.538461538461538,
|
| 3641 |
+
"grad_norm": 0.0,
|
| 3642 |
+
"learning_rate": 3.9905000000000005e-05,
|
| 3643 |
+
"loss": 1.7869,
|
| 3644 |
+
"mean_token_accuracy": 0.6394775450229645,
|
| 3645 |
+
"num_tokens": 7364128.0,
|
| 3646 |
+
"step": 2020
|
| 3647 |
}
|
| 3648 |
],
|
| 3649 |
"logging_steps": 5,
|
|
|
|
| 3663 |
"attributes": {}
|
| 3664 |
}
|
| 3665 |
},
|
| 3666 |
+
"total_flos": 3.7298158248738816e+17,
|
| 3667 |
"train_batch_size": 4,
|
| 3668 |
"trial_name": null,
|
| 3669 |
"trial_params": null
|