Upload trainer_state.json with huggingface_hub
Browse files- trainer_state.json +2103 -3
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,
|
|
@@ -1407,6 +1407,2106 @@
|
|
| 1407 |
"learning_rate": 4e-05,
|
| 1408 |
"loss": 0.0251,
|
| 1409 |
"step": 2000
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1410 |
}
|
| 1411 |
],
|
| 1412 |
"logging_steps": 10,
|
|
@@ -1426,7 +3526,7 @@
|
|
| 1426 |
"attributes": {}
|
| 1427 |
}
|
| 1428 |
},
|
| 1429 |
-
"total_flos":
|
| 1430 |
"train_batch_size": 1,
|
| 1431 |
"trial_name": null,
|
| 1432 |
"trial_params": null
|
|
|
|
| 1 |
{
|
| 2 |
"best_metric": null,
|
| 3 |
"best_model_checkpoint": null,
|
| 4 |
+
"epoch": 0.5,
|
| 5 |
"eval_steps": 500,
|
| 6 |
+
"global_step": 5000,
|
| 7 |
"is_hyper_param_search": false,
|
| 8 |
"is_local_process_zero": true,
|
| 9 |
"is_world_process_zero": true,
|
|
|
|
| 1407 |
"learning_rate": 4e-05,
|
| 1408 |
"loss": 0.0251,
|
| 1409 |
"step": 2000
|
| 1410 |
+
},
|
| 1411 |
+
{
|
| 1412 |
+
"epoch": 0.201,
|
| 1413 |
+
"grad_norm": 0.17456738650798798,
|
| 1414 |
+
"learning_rate": 3.995e-05,
|
| 1415 |
+
"loss": 0.0315,
|
| 1416 |
+
"step": 2010
|
| 1417 |
+
},
|
| 1418 |
+
{
|
| 1419 |
+
"epoch": 0.202,
|
| 1420 |
+
"grad_norm": 0.15893588960170746,
|
| 1421 |
+
"learning_rate": 3.99e-05,
|
| 1422 |
+
"loss": 0.0243,
|
| 1423 |
+
"step": 2020
|
| 1424 |
+
},
|
| 1425 |
+
{
|
| 1426 |
+
"epoch": 0.203,
|
| 1427 |
+
"grad_norm": 0.14638105034828186,
|
| 1428 |
+
"learning_rate": 3.9850000000000006e-05,
|
| 1429 |
+
"loss": 0.0247,
|
| 1430 |
+
"step": 2030
|
| 1431 |
+
},
|
| 1432 |
+
{
|
| 1433 |
+
"epoch": 0.204,
|
| 1434 |
+
"grad_norm": 0.1714017242193222,
|
| 1435 |
+
"learning_rate": 3.9800000000000005e-05,
|
| 1436 |
+
"loss": 0.0252,
|
| 1437 |
+
"step": 2040
|
| 1438 |
+
},
|
| 1439 |
+
{
|
| 1440 |
+
"epoch": 0.205,
|
| 1441 |
+
"grad_norm": 0.18679572641849518,
|
| 1442 |
+
"learning_rate": 3.9750000000000004e-05,
|
| 1443 |
+
"loss": 0.0234,
|
| 1444 |
+
"step": 2050
|
| 1445 |
+
},
|
| 1446 |
+
{
|
| 1447 |
+
"epoch": 0.206,
|
| 1448 |
+
"grad_norm": 0.10623681545257568,
|
| 1449 |
+
"learning_rate": 3.97e-05,
|
| 1450 |
+
"loss": 0.0256,
|
| 1451 |
+
"step": 2060
|
| 1452 |
+
},
|
| 1453 |
+
{
|
| 1454 |
+
"epoch": 0.207,
|
| 1455 |
+
"grad_norm": 0.18566076457500458,
|
| 1456 |
+
"learning_rate": 3.965e-05,
|
| 1457 |
+
"loss": 0.0238,
|
| 1458 |
+
"step": 2070
|
| 1459 |
+
},
|
| 1460 |
+
{
|
| 1461 |
+
"epoch": 0.208,
|
| 1462 |
+
"grad_norm": 0.12487553805112839,
|
| 1463 |
+
"learning_rate": 3.960000000000001e-05,
|
| 1464 |
+
"loss": 0.0226,
|
| 1465 |
+
"step": 2080
|
| 1466 |
+
},
|
| 1467 |
+
{
|
| 1468 |
+
"epoch": 0.209,
|
| 1469 |
+
"grad_norm": 0.13191473484039307,
|
| 1470 |
+
"learning_rate": 3.9550000000000006e-05,
|
| 1471 |
+
"loss": 0.0232,
|
| 1472 |
+
"step": 2090
|
| 1473 |
+
},
|
| 1474 |
+
{
|
| 1475 |
+
"epoch": 0.21,
|
| 1476 |
+
"grad_norm": 0.225613072514534,
|
| 1477 |
+
"learning_rate": 3.9500000000000005e-05,
|
| 1478 |
+
"loss": 0.0226,
|
| 1479 |
+
"step": 2100
|
| 1480 |
+
},
|
| 1481 |
+
{
|
| 1482 |
+
"epoch": 0.211,
|
| 1483 |
+
"grad_norm": 0.10896781831979752,
|
| 1484 |
+
"learning_rate": 3.9450000000000003e-05,
|
| 1485 |
+
"loss": 0.0206,
|
| 1486 |
+
"step": 2110
|
| 1487 |
+
},
|
| 1488 |
+
{
|
| 1489 |
+
"epoch": 0.212,
|
| 1490 |
+
"grad_norm": 0.16153796017169952,
|
| 1491 |
+
"learning_rate": 3.94e-05,
|
| 1492 |
+
"loss": 0.0195,
|
| 1493 |
+
"step": 2120
|
| 1494 |
+
},
|
| 1495 |
+
{
|
| 1496 |
+
"epoch": 0.213,
|
| 1497 |
+
"grad_norm": 0.19171251356601715,
|
| 1498 |
+
"learning_rate": 3.935e-05,
|
| 1499 |
+
"loss": 0.0203,
|
| 1500 |
+
"step": 2130
|
| 1501 |
+
},
|
| 1502 |
+
{
|
| 1503 |
+
"epoch": 0.214,
|
| 1504 |
+
"grad_norm": 0.13199982047080994,
|
| 1505 |
+
"learning_rate": 3.9300000000000007e-05,
|
| 1506 |
+
"loss": 0.0194,
|
| 1507 |
+
"step": 2140
|
| 1508 |
+
},
|
| 1509 |
+
{
|
| 1510 |
+
"epoch": 0.215,
|
| 1511 |
+
"grad_norm": 0.12839478254318237,
|
| 1512 |
+
"learning_rate": 3.9250000000000005e-05,
|
| 1513 |
+
"loss": 0.0227,
|
| 1514 |
+
"step": 2150
|
| 1515 |
+
},
|
| 1516 |
+
{
|
| 1517 |
+
"epoch": 0.216,
|
| 1518 |
+
"grad_norm": 0.10787441581487656,
|
| 1519 |
+
"learning_rate": 3.9200000000000004e-05,
|
| 1520 |
+
"loss": 0.0195,
|
| 1521 |
+
"step": 2160
|
| 1522 |
+
},
|
| 1523 |
+
{
|
| 1524 |
+
"epoch": 0.217,
|
| 1525 |
+
"grad_norm": 0.1551046371459961,
|
| 1526 |
+
"learning_rate": 3.915e-05,
|
| 1527 |
+
"loss": 0.019,
|
| 1528 |
+
"step": 2170
|
| 1529 |
+
},
|
| 1530 |
+
{
|
| 1531 |
+
"epoch": 0.218,
|
| 1532 |
+
"grad_norm": 0.18844197690486908,
|
| 1533 |
+
"learning_rate": 3.91e-05,
|
| 1534 |
+
"loss": 0.02,
|
| 1535 |
+
"step": 2180
|
| 1536 |
+
},
|
| 1537 |
+
{
|
| 1538 |
+
"epoch": 0.219,
|
| 1539 |
+
"grad_norm": 0.21247665584087372,
|
| 1540 |
+
"learning_rate": 3.905e-05,
|
| 1541 |
+
"loss": 0.0206,
|
| 1542 |
+
"step": 2190
|
| 1543 |
+
},
|
| 1544 |
+
{
|
| 1545 |
+
"epoch": 0.22,
|
| 1546 |
+
"grad_norm": 0.11881183087825775,
|
| 1547 |
+
"learning_rate": 3.9000000000000006e-05,
|
| 1548 |
+
"loss": 0.0176,
|
| 1549 |
+
"step": 2200
|
| 1550 |
+
},
|
| 1551 |
+
{
|
| 1552 |
+
"epoch": 0.221,
|
| 1553 |
+
"grad_norm": 0.16291823983192444,
|
| 1554 |
+
"learning_rate": 3.8950000000000005e-05,
|
| 1555 |
+
"loss": 0.0178,
|
| 1556 |
+
"step": 2210
|
| 1557 |
+
},
|
| 1558 |
+
{
|
| 1559 |
+
"epoch": 0.222,
|
| 1560 |
+
"grad_norm": 0.14063787460327148,
|
| 1561 |
+
"learning_rate": 3.8900000000000004e-05,
|
| 1562 |
+
"loss": 0.0177,
|
| 1563 |
+
"step": 2220
|
| 1564 |
+
},
|
| 1565 |
+
{
|
| 1566 |
+
"epoch": 0.223,
|
| 1567 |
+
"grad_norm": 0.15583930909633636,
|
| 1568 |
+
"learning_rate": 3.885e-05,
|
| 1569 |
+
"loss": 0.0185,
|
| 1570 |
+
"step": 2230
|
| 1571 |
+
},
|
| 1572 |
+
{
|
| 1573 |
+
"epoch": 0.224,
|
| 1574 |
+
"grad_norm": 0.16128882765769958,
|
| 1575 |
+
"learning_rate": 3.88e-05,
|
| 1576 |
+
"loss": 0.0168,
|
| 1577 |
+
"step": 2240
|
| 1578 |
+
},
|
| 1579 |
+
{
|
| 1580 |
+
"epoch": 0.225,
|
| 1581 |
+
"grad_norm": 0.1588473916053772,
|
| 1582 |
+
"learning_rate": 3.875e-05,
|
| 1583 |
+
"loss": 0.0167,
|
| 1584 |
+
"step": 2250
|
| 1585 |
+
},
|
| 1586 |
+
{
|
| 1587 |
+
"epoch": 0.226,
|
| 1588 |
+
"grad_norm": 0.10487533360719681,
|
| 1589 |
+
"learning_rate": 3.8700000000000006e-05,
|
| 1590 |
+
"loss": 0.0154,
|
| 1591 |
+
"step": 2260
|
| 1592 |
+
},
|
| 1593 |
+
{
|
| 1594 |
+
"epoch": 0.227,
|
| 1595 |
+
"grad_norm": 0.2638506591320038,
|
| 1596 |
+
"learning_rate": 3.8650000000000004e-05,
|
| 1597 |
+
"loss": 0.0179,
|
| 1598 |
+
"step": 2270
|
| 1599 |
+
},
|
| 1600 |
+
{
|
| 1601 |
+
"epoch": 0.228,
|
| 1602 |
+
"grad_norm": 0.12504911422729492,
|
| 1603 |
+
"learning_rate": 3.86e-05,
|
| 1604 |
+
"loss": 0.016,
|
| 1605 |
+
"step": 2280
|
| 1606 |
+
},
|
| 1607 |
+
{
|
| 1608 |
+
"epoch": 0.229,
|
| 1609 |
+
"grad_norm": 0.11655262857675552,
|
| 1610 |
+
"learning_rate": 3.855e-05,
|
| 1611 |
+
"loss": 0.0164,
|
| 1612 |
+
"step": 2290
|
| 1613 |
+
},
|
| 1614 |
+
{
|
| 1615 |
+
"epoch": 0.23,
|
| 1616 |
+
"grad_norm": 0.10052930563688278,
|
| 1617 |
+
"learning_rate": 3.85e-05,
|
| 1618 |
+
"loss": 0.0143,
|
| 1619 |
+
"step": 2300
|
| 1620 |
+
},
|
| 1621 |
+
{
|
| 1622 |
+
"epoch": 0.231,
|
| 1623 |
+
"grad_norm": 0.07682032138109207,
|
| 1624 |
+
"learning_rate": 3.845e-05,
|
| 1625 |
+
"loss": 0.0165,
|
| 1626 |
+
"step": 2310
|
| 1627 |
+
},
|
| 1628 |
+
{
|
| 1629 |
+
"epoch": 0.232,
|
| 1630 |
+
"grad_norm": 0.12146533280611038,
|
| 1631 |
+
"learning_rate": 3.8400000000000005e-05,
|
| 1632 |
+
"loss": 0.0147,
|
| 1633 |
+
"step": 2320
|
| 1634 |
+
},
|
| 1635 |
+
{
|
| 1636 |
+
"epoch": 0.233,
|
| 1637 |
+
"grad_norm": 0.16349685192108154,
|
| 1638 |
+
"learning_rate": 3.8350000000000004e-05,
|
| 1639 |
+
"loss": 0.0166,
|
| 1640 |
+
"step": 2330
|
| 1641 |
+
},
|
| 1642 |
+
{
|
| 1643 |
+
"epoch": 0.234,
|
| 1644 |
+
"grad_norm": 0.10822432488203049,
|
| 1645 |
+
"learning_rate": 3.83e-05,
|
| 1646 |
+
"loss": 0.0146,
|
| 1647 |
+
"step": 2340
|
| 1648 |
+
},
|
| 1649 |
+
{
|
| 1650 |
+
"epoch": 0.235,
|
| 1651 |
+
"grad_norm": 0.0805143415927887,
|
| 1652 |
+
"learning_rate": 3.825e-05,
|
| 1653 |
+
"loss": 0.0131,
|
| 1654 |
+
"step": 2350
|
| 1655 |
+
},
|
| 1656 |
+
{
|
| 1657 |
+
"epoch": 0.236,
|
| 1658 |
+
"grad_norm": 0.08285068720579147,
|
| 1659 |
+
"learning_rate": 3.82e-05,
|
| 1660 |
+
"loss": 0.0134,
|
| 1661 |
+
"step": 2360
|
| 1662 |
+
},
|
| 1663 |
+
{
|
| 1664 |
+
"epoch": 0.237,
|
| 1665 |
+
"grad_norm": 0.0886882022023201,
|
| 1666 |
+
"learning_rate": 3.8150000000000006e-05,
|
| 1667 |
+
"loss": 0.0133,
|
| 1668 |
+
"step": 2370
|
| 1669 |
+
},
|
| 1670 |
+
{
|
| 1671 |
+
"epoch": 0.238,
|
| 1672 |
+
"grad_norm": 0.08793161064386368,
|
| 1673 |
+
"learning_rate": 3.8100000000000005e-05,
|
| 1674 |
+
"loss": 0.0142,
|
| 1675 |
+
"step": 2380
|
| 1676 |
+
},
|
| 1677 |
+
{
|
| 1678 |
+
"epoch": 0.239,
|
| 1679 |
+
"grad_norm": 0.08325997740030289,
|
| 1680 |
+
"learning_rate": 3.805e-05,
|
| 1681 |
+
"loss": 0.012,
|
| 1682 |
+
"step": 2390
|
| 1683 |
+
},
|
| 1684 |
+
{
|
| 1685 |
+
"epoch": 0.24,
|
| 1686 |
+
"grad_norm": 0.10990972816944122,
|
| 1687 |
+
"learning_rate": 3.8e-05,
|
| 1688 |
+
"loss": 0.0134,
|
| 1689 |
+
"step": 2400
|
| 1690 |
+
},
|
| 1691 |
+
{
|
| 1692 |
+
"epoch": 0.241,
|
| 1693 |
+
"grad_norm": 0.06695697456598282,
|
| 1694 |
+
"learning_rate": 3.795e-05,
|
| 1695 |
+
"loss": 0.0119,
|
| 1696 |
+
"step": 2410
|
| 1697 |
+
},
|
| 1698 |
+
{
|
| 1699 |
+
"epoch": 0.242,
|
| 1700 |
+
"grad_norm": 0.08304648846387863,
|
| 1701 |
+
"learning_rate": 3.79e-05,
|
| 1702 |
+
"loss": 0.0128,
|
| 1703 |
+
"step": 2420
|
| 1704 |
+
},
|
| 1705 |
+
{
|
| 1706 |
+
"epoch": 0.243,
|
| 1707 |
+
"grad_norm": 0.09513472020626068,
|
| 1708 |
+
"learning_rate": 3.7850000000000005e-05,
|
| 1709 |
+
"loss": 0.0146,
|
| 1710 |
+
"step": 2430
|
| 1711 |
+
},
|
| 1712 |
+
{
|
| 1713 |
+
"epoch": 0.244,
|
| 1714 |
+
"grad_norm": 0.07892587035894394,
|
| 1715 |
+
"learning_rate": 3.7800000000000004e-05,
|
| 1716 |
+
"loss": 0.0116,
|
| 1717 |
+
"step": 2440
|
| 1718 |
+
},
|
| 1719 |
+
{
|
| 1720 |
+
"epoch": 0.245,
|
| 1721 |
+
"grad_norm": 0.12630197405815125,
|
| 1722 |
+
"learning_rate": 3.775e-05,
|
| 1723 |
+
"loss": 0.0132,
|
| 1724 |
+
"step": 2450
|
| 1725 |
+
},
|
| 1726 |
+
{
|
| 1727 |
+
"epoch": 0.246,
|
| 1728 |
+
"grad_norm": 0.08250122517347336,
|
| 1729 |
+
"learning_rate": 3.77e-05,
|
| 1730 |
+
"loss": 0.013,
|
| 1731 |
+
"step": 2460
|
| 1732 |
+
},
|
| 1733 |
+
{
|
| 1734 |
+
"epoch": 0.247,
|
| 1735 |
+
"grad_norm": 0.09903154522180557,
|
| 1736 |
+
"learning_rate": 3.765e-05,
|
| 1737 |
+
"loss": 0.0117,
|
| 1738 |
+
"step": 2470
|
| 1739 |
+
},
|
| 1740 |
+
{
|
| 1741 |
+
"epoch": 0.248,
|
| 1742 |
+
"grad_norm": 0.09059987217187881,
|
| 1743 |
+
"learning_rate": 3.76e-05,
|
| 1744 |
+
"loss": 0.0111,
|
| 1745 |
+
"step": 2480
|
| 1746 |
+
},
|
| 1747 |
+
{
|
| 1748 |
+
"epoch": 0.249,
|
| 1749 |
+
"grad_norm": 0.09777077287435532,
|
| 1750 |
+
"learning_rate": 3.7550000000000005e-05,
|
| 1751 |
+
"loss": 0.0142,
|
| 1752 |
+
"step": 2490
|
| 1753 |
+
},
|
| 1754 |
+
{
|
| 1755 |
+
"epoch": 0.25,
|
| 1756 |
+
"grad_norm": 0.1801980435848236,
|
| 1757 |
+
"learning_rate": 3.7500000000000003e-05,
|
| 1758 |
+
"loss": 0.0121,
|
| 1759 |
+
"step": 2500
|
| 1760 |
+
},
|
| 1761 |
+
{
|
| 1762 |
+
"epoch": 0.251,
|
| 1763 |
+
"grad_norm": 0.08936703950166702,
|
| 1764 |
+
"learning_rate": 3.745e-05,
|
| 1765 |
+
"loss": 0.0112,
|
| 1766 |
+
"step": 2510
|
| 1767 |
+
},
|
| 1768 |
+
{
|
| 1769 |
+
"epoch": 0.252,
|
| 1770 |
+
"grad_norm": 0.09601296484470367,
|
| 1771 |
+
"learning_rate": 3.74e-05,
|
| 1772 |
+
"loss": 0.0116,
|
| 1773 |
+
"step": 2520
|
| 1774 |
+
},
|
| 1775 |
+
{
|
| 1776 |
+
"epoch": 0.253,
|
| 1777 |
+
"grad_norm": 0.08924739062786102,
|
| 1778 |
+
"learning_rate": 3.735e-05,
|
| 1779 |
+
"loss": 0.0119,
|
| 1780 |
+
"step": 2530
|
| 1781 |
+
},
|
| 1782 |
+
{
|
| 1783 |
+
"epoch": 0.254,
|
| 1784 |
+
"grad_norm": 0.07558383047580719,
|
| 1785 |
+
"learning_rate": 3.73e-05,
|
| 1786 |
+
"loss": 0.0108,
|
| 1787 |
+
"step": 2540
|
| 1788 |
+
},
|
| 1789 |
+
{
|
| 1790 |
+
"epoch": 0.255,
|
| 1791 |
+
"grad_norm": 0.05701779946684837,
|
| 1792 |
+
"learning_rate": 3.7250000000000004e-05,
|
| 1793 |
+
"loss": 0.011,
|
| 1794 |
+
"step": 2550
|
| 1795 |
+
},
|
| 1796 |
+
{
|
| 1797 |
+
"epoch": 0.256,
|
| 1798 |
+
"grad_norm": 0.0955379456281662,
|
| 1799 |
+
"learning_rate": 3.72e-05,
|
| 1800 |
+
"loss": 0.0112,
|
| 1801 |
+
"step": 2560
|
| 1802 |
+
},
|
| 1803 |
+
{
|
| 1804 |
+
"epoch": 0.257,
|
| 1805 |
+
"grad_norm": 0.07837249338626862,
|
| 1806 |
+
"learning_rate": 3.715e-05,
|
| 1807 |
+
"loss": 0.0111,
|
| 1808 |
+
"step": 2570
|
| 1809 |
+
},
|
| 1810 |
+
{
|
| 1811 |
+
"epoch": 0.258,
|
| 1812 |
+
"grad_norm": 0.09438953548669815,
|
| 1813 |
+
"learning_rate": 3.71e-05,
|
| 1814 |
+
"loss": 0.0121,
|
| 1815 |
+
"step": 2580
|
| 1816 |
+
},
|
| 1817 |
+
{
|
| 1818 |
+
"epoch": 0.259,
|
| 1819 |
+
"grad_norm": 0.08802532404661179,
|
| 1820 |
+
"learning_rate": 3.705e-05,
|
| 1821 |
+
"loss": 0.0098,
|
| 1822 |
+
"step": 2590
|
| 1823 |
+
},
|
| 1824 |
+
{
|
| 1825 |
+
"epoch": 0.26,
|
| 1826 |
+
"grad_norm": 0.0785641148686409,
|
| 1827 |
+
"learning_rate": 3.7e-05,
|
| 1828 |
+
"loss": 0.0106,
|
| 1829 |
+
"step": 2600
|
| 1830 |
+
},
|
| 1831 |
+
{
|
| 1832 |
+
"epoch": 0.261,
|
| 1833 |
+
"grad_norm": 0.10036404430866241,
|
| 1834 |
+
"learning_rate": 3.6950000000000004e-05,
|
| 1835 |
+
"loss": 0.0107,
|
| 1836 |
+
"step": 2610
|
| 1837 |
+
},
|
| 1838 |
+
{
|
| 1839 |
+
"epoch": 0.262,
|
| 1840 |
+
"grad_norm": 0.0663432776927948,
|
| 1841 |
+
"learning_rate": 3.69e-05,
|
| 1842 |
+
"loss": 0.0098,
|
| 1843 |
+
"step": 2620
|
| 1844 |
+
},
|
| 1845 |
+
{
|
| 1846 |
+
"epoch": 0.263,
|
| 1847 |
+
"grad_norm": 0.06886564195156097,
|
| 1848 |
+
"learning_rate": 3.685e-05,
|
| 1849 |
+
"loss": 0.0094,
|
| 1850 |
+
"step": 2630
|
| 1851 |
+
},
|
| 1852 |
+
{
|
| 1853 |
+
"epoch": 0.264,
|
| 1854 |
+
"grad_norm": 0.06641737371683121,
|
| 1855 |
+
"learning_rate": 3.68e-05,
|
| 1856 |
+
"loss": 0.0112,
|
| 1857 |
+
"step": 2640
|
| 1858 |
+
},
|
| 1859 |
+
{
|
| 1860 |
+
"epoch": 0.265,
|
| 1861 |
+
"grad_norm": 0.1470363438129425,
|
| 1862 |
+
"learning_rate": 3.675e-05,
|
| 1863 |
+
"loss": 0.0118,
|
| 1864 |
+
"step": 2650
|
| 1865 |
+
},
|
| 1866 |
+
{
|
| 1867 |
+
"epoch": 0.266,
|
| 1868 |
+
"grad_norm": 0.08694775402545929,
|
| 1869 |
+
"learning_rate": 3.6700000000000004e-05,
|
| 1870 |
+
"loss": 0.0105,
|
| 1871 |
+
"step": 2660
|
| 1872 |
+
},
|
| 1873 |
+
{
|
| 1874 |
+
"epoch": 0.267,
|
| 1875 |
+
"grad_norm": 0.08168693631887436,
|
| 1876 |
+
"learning_rate": 3.665e-05,
|
| 1877 |
+
"loss": 0.0101,
|
| 1878 |
+
"step": 2670
|
| 1879 |
+
},
|
| 1880 |
+
{
|
| 1881 |
+
"epoch": 0.268,
|
| 1882 |
+
"grad_norm": 0.06114206463098526,
|
| 1883 |
+
"learning_rate": 3.66e-05,
|
| 1884 |
+
"loss": 0.0097,
|
| 1885 |
+
"step": 2680
|
| 1886 |
+
},
|
| 1887 |
+
{
|
| 1888 |
+
"epoch": 0.269,
|
| 1889 |
+
"grad_norm": 0.09011895209550858,
|
| 1890 |
+
"learning_rate": 3.655e-05,
|
| 1891 |
+
"loss": 0.0101,
|
| 1892 |
+
"step": 2690
|
| 1893 |
+
},
|
| 1894 |
+
{
|
| 1895 |
+
"epoch": 0.27,
|
| 1896 |
+
"grad_norm": 0.06499819457530975,
|
| 1897 |
+
"learning_rate": 3.65e-05,
|
| 1898 |
+
"loss": 0.0089,
|
| 1899 |
+
"step": 2700
|
| 1900 |
+
},
|
| 1901 |
+
{
|
| 1902 |
+
"epoch": 0.271,
|
| 1903 |
+
"grad_norm": 0.08157055824995041,
|
| 1904 |
+
"learning_rate": 3.645e-05,
|
| 1905 |
+
"loss": 0.0099,
|
| 1906 |
+
"step": 2710
|
| 1907 |
+
},
|
| 1908 |
+
{
|
| 1909 |
+
"epoch": 0.272,
|
| 1910 |
+
"grad_norm": 0.06255003809928894,
|
| 1911 |
+
"learning_rate": 3.6400000000000004e-05,
|
| 1912 |
+
"loss": 0.0091,
|
| 1913 |
+
"step": 2720
|
| 1914 |
+
},
|
| 1915 |
+
{
|
| 1916 |
+
"epoch": 0.273,
|
| 1917 |
+
"grad_norm": 0.13641099631786346,
|
| 1918 |
+
"learning_rate": 3.635e-05,
|
| 1919 |
+
"loss": 0.01,
|
| 1920 |
+
"step": 2730
|
| 1921 |
+
},
|
| 1922 |
+
{
|
| 1923 |
+
"epoch": 0.274,
|
| 1924 |
+
"grad_norm": 0.06449054926633835,
|
| 1925 |
+
"learning_rate": 3.63e-05,
|
| 1926 |
+
"loss": 0.0087,
|
| 1927 |
+
"step": 2740
|
| 1928 |
+
},
|
| 1929 |
+
{
|
| 1930 |
+
"epoch": 0.275,
|
| 1931 |
+
"grad_norm": 0.09242594987154007,
|
| 1932 |
+
"learning_rate": 3.625e-05,
|
| 1933 |
+
"loss": 0.0084,
|
| 1934 |
+
"step": 2750
|
| 1935 |
+
},
|
| 1936 |
+
{
|
| 1937 |
+
"epoch": 0.276,
|
| 1938 |
+
"grad_norm": 0.14216932654380798,
|
| 1939 |
+
"learning_rate": 3.62e-05,
|
| 1940 |
+
"loss": 0.01,
|
| 1941 |
+
"step": 2760
|
| 1942 |
+
},
|
| 1943 |
+
{
|
| 1944 |
+
"epoch": 0.277,
|
| 1945 |
+
"grad_norm": 0.11992328613996506,
|
| 1946 |
+
"learning_rate": 3.615e-05,
|
| 1947 |
+
"loss": 0.0107,
|
| 1948 |
+
"step": 2770
|
| 1949 |
+
},
|
| 1950 |
+
{
|
| 1951 |
+
"epoch": 0.278,
|
| 1952 |
+
"grad_norm": 0.10537979751825333,
|
| 1953 |
+
"learning_rate": 3.61e-05,
|
| 1954 |
+
"loss": 0.0101,
|
| 1955 |
+
"step": 2780
|
| 1956 |
+
},
|
| 1957 |
+
{
|
| 1958 |
+
"epoch": 0.279,
|
| 1959 |
+
"grad_norm": 0.06420467048883438,
|
| 1960 |
+
"learning_rate": 3.605e-05,
|
| 1961 |
+
"loss": 0.0088,
|
| 1962 |
+
"step": 2790
|
| 1963 |
+
},
|
| 1964 |
+
{
|
| 1965 |
+
"epoch": 0.28,
|
| 1966 |
+
"grad_norm": 0.10813489556312561,
|
| 1967 |
+
"learning_rate": 3.6e-05,
|
| 1968 |
+
"loss": 0.0093,
|
| 1969 |
+
"step": 2800
|
| 1970 |
+
},
|
| 1971 |
+
{
|
| 1972 |
+
"epoch": 0.281,
|
| 1973 |
+
"grad_norm": 0.05735234543681145,
|
| 1974 |
+
"learning_rate": 3.595e-05,
|
| 1975 |
+
"loss": 0.0087,
|
| 1976 |
+
"step": 2810
|
| 1977 |
+
},
|
| 1978 |
+
{
|
| 1979 |
+
"epoch": 0.282,
|
| 1980 |
+
"grad_norm": 0.05712522938847542,
|
| 1981 |
+
"learning_rate": 3.59e-05,
|
| 1982 |
+
"loss": 0.0082,
|
| 1983 |
+
"step": 2820
|
| 1984 |
+
},
|
| 1985 |
+
{
|
| 1986 |
+
"epoch": 0.283,
|
| 1987 |
+
"grad_norm": 0.07710873335599899,
|
| 1988 |
+
"learning_rate": 3.585e-05,
|
| 1989 |
+
"loss": 0.0088,
|
| 1990 |
+
"step": 2830
|
| 1991 |
+
},
|
| 1992 |
+
{
|
| 1993 |
+
"epoch": 0.284,
|
| 1994 |
+
"grad_norm": 0.11007268726825714,
|
| 1995 |
+
"learning_rate": 3.58e-05,
|
| 1996 |
+
"loss": 0.0075,
|
| 1997 |
+
"step": 2840
|
| 1998 |
+
},
|
| 1999 |
+
{
|
| 2000 |
+
"epoch": 0.285,
|
| 2001 |
+
"grad_norm": 0.07825978100299835,
|
| 2002 |
+
"learning_rate": 3.575e-05,
|
| 2003 |
+
"loss": 0.0089,
|
| 2004 |
+
"step": 2850
|
| 2005 |
+
},
|
| 2006 |
+
{
|
| 2007 |
+
"epoch": 0.286,
|
| 2008 |
+
"grad_norm": 0.06950812041759491,
|
| 2009 |
+
"learning_rate": 3.57e-05,
|
| 2010 |
+
"loss": 0.0077,
|
| 2011 |
+
"step": 2860
|
| 2012 |
+
},
|
| 2013 |
+
{
|
| 2014 |
+
"epoch": 0.287,
|
| 2015 |
+
"grad_norm": 0.052572544664144516,
|
| 2016 |
+
"learning_rate": 3.565e-05,
|
| 2017 |
+
"loss": 0.0076,
|
| 2018 |
+
"step": 2870
|
| 2019 |
+
},
|
| 2020 |
+
{
|
| 2021 |
+
"epoch": 0.288,
|
| 2022 |
+
"grad_norm": 0.06588669121265411,
|
| 2023 |
+
"learning_rate": 3.56e-05,
|
| 2024 |
+
"loss": 0.0073,
|
| 2025 |
+
"step": 2880
|
| 2026 |
+
},
|
| 2027 |
+
{
|
| 2028 |
+
"epoch": 0.289,
|
| 2029 |
+
"grad_norm": 0.052969031035900116,
|
| 2030 |
+
"learning_rate": 3.555e-05,
|
| 2031 |
+
"loss": 0.0066,
|
| 2032 |
+
"step": 2890
|
| 2033 |
+
},
|
| 2034 |
+
{
|
| 2035 |
+
"epoch": 0.29,
|
| 2036 |
+
"grad_norm": 0.04204658418893814,
|
| 2037 |
+
"learning_rate": 3.55e-05,
|
| 2038 |
+
"loss": 0.0064,
|
| 2039 |
+
"step": 2900
|
| 2040 |
+
},
|
| 2041 |
+
{
|
| 2042 |
+
"epoch": 0.291,
|
| 2043 |
+
"grad_norm": 0.04765693470835686,
|
| 2044 |
+
"learning_rate": 3.545e-05,
|
| 2045 |
+
"loss": 0.0067,
|
| 2046 |
+
"step": 2910
|
| 2047 |
+
},
|
| 2048 |
+
{
|
| 2049 |
+
"epoch": 0.292,
|
| 2050 |
+
"grad_norm": 0.06796044856309891,
|
| 2051 |
+
"learning_rate": 3.54e-05,
|
| 2052 |
+
"loss": 0.0069,
|
| 2053 |
+
"step": 2920
|
| 2054 |
+
},
|
| 2055 |
+
{
|
| 2056 |
+
"epoch": 0.293,
|
| 2057 |
+
"grad_norm": 0.12173280119895935,
|
| 2058 |
+
"learning_rate": 3.535e-05,
|
| 2059 |
+
"loss": 0.0126,
|
| 2060 |
+
"step": 2930
|
| 2061 |
+
},
|
| 2062 |
+
{
|
| 2063 |
+
"epoch": 0.294,
|
| 2064 |
+
"grad_norm": 0.09393921494483948,
|
| 2065 |
+
"learning_rate": 3.53e-05,
|
| 2066 |
+
"loss": 0.0096,
|
| 2067 |
+
"step": 2940
|
| 2068 |
+
},
|
| 2069 |
+
{
|
| 2070 |
+
"epoch": 0.295,
|
| 2071 |
+
"grad_norm": 0.08246493339538574,
|
| 2072 |
+
"learning_rate": 3.525e-05,
|
| 2073 |
+
"loss": 0.0084,
|
| 2074 |
+
"step": 2950
|
| 2075 |
+
},
|
| 2076 |
+
{
|
| 2077 |
+
"epoch": 0.296,
|
| 2078 |
+
"grad_norm": 0.04482726752758026,
|
| 2079 |
+
"learning_rate": 3.52e-05,
|
| 2080 |
+
"loss": 0.0073,
|
| 2081 |
+
"step": 2960
|
| 2082 |
+
},
|
| 2083 |
+
{
|
| 2084 |
+
"epoch": 0.297,
|
| 2085 |
+
"grad_norm": 0.1147686317563057,
|
| 2086 |
+
"learning_rate": 3.515e-05,
|
| 2087 |
+
"loss": 0.0094,
|
| 2088 |
+
"step": 2970
|
| 2089 |
+
},
|
| 2090 |
+
{
|
| 2091 |
+
"epoch": 0.298,
|
| 2092 |
+
"grad_norm": 0.09143181890249252,
|
| 2093 |
+
"learning_rate": 3.51e-05,
|
| 2094 |
+
"loss": 0.0075,
|
| 2095 |
+
"step": 2980
|
| 2096 |
+
},
|
| 2097 |
+
{
|
| 2098 |
+
"epoch": 0.299,
|
| 2099 |
+
"grad_norm": 0.05911434814333916,
|
| 2100 |
+
"learning_rate": 3.505e-05,
|
| 2101 |
+
"loss": 0.0075,
|
| 2102 |
+
"step": 2990
|
| 2103 |
+
},
|
| 2104 |
+
{
|
| 2105 |
+
"epoch": 0.3,
|
| 2106 |
+
"grad_norm": 0.04372965916991234,
|
| 2107 |
+
"learning_rate": 3.5e-05,
|
| 2108 |
+
"loss": 0.0072,
|
| 2109 |
+
"step": 3000
|
| 2110 |
+
},
|
| 2111 |
+
{
|
| 2112 |
+
"epoch": 0.301,
|
| 2113 |
+
"grad_norm": 0.05518479272723198,
|
| 2114 |
+
"learning_rate": 3.495e-05,
|
| 2115 |
+
"loss": 0.0068,
|
| 2116 |
+
"step": 3010
|
| 2117 |
+
},
|
| 2118 |
+
{
|
| 2119 |
+
"epoch": 0.302,
|
| 2120 |
+
"grad_norm": 0.04555105045437813,
|
| 2121 |
+
"learning_rate": 3.49e-05,
|
| 2122 |
+
"loss": 0.0064,
|
| 2123 |
+
"step": 3020
|
| 2124 |
+
},
|
| 2125 |
+
{
|
| 2126 |
+
"epoch": 0.303,
|
| 2127 |
+
"grad_norm": 0.03831150382757187,
|
| 2128 |
+
"learning_rate": 3.485e-05,
|
| 2129 |
+
"loss": 0.007,
|
| 2130 |
+
"step": 3030
|
| 2131 |
+
},
|
| 2132 |
+
{
|
| 2133 |
+
"epoch": 0.304,
|
| 2134 |
+
"grad_norm": 0.04596749320626259,
|
| 2135 |
+
"learning_rate": 3.48e-05,
|
| 2136 |
+
"loss": 0.0065,
|
| 2137 |
+
"step": 3040
|
| 2138 |
+
},
|
| 2139 |
+
{
|
| 2140 |
+
"epoch": 0.305,
|
| 2141 |
+
"grad_norm": 0.07694078236818314,
|
| 2142 |
+
"learning_rate": 3.475e-05,
|
| 2143 |
+
"loss": 0.0059,
|
| 2144 |
+
"step": 3050
|
| 2145 |
+
},
|
| 2146 |
+
{
|
| 2147 |
+
"epoch": 0.306,
|
| 2148 |
+
"grad_norm": 0.12307348102331161,
|
| 2149 |
+
"learning_rate": 3.4699999999999996e-05,
|
| 2150 |
+
"loss": 0.0099,
|
| 2151 |
+
"step": 3060
|
| 2152 |
+
},
|
| 2153 |
+
{
|
| 2154 |
+
"epoch": 0.307,
|
| 2155 |
+
"grad_norm": 0.059611763805150986,
|
| 2156 |
+
"learning_rate": 3.465e-05,
|
| 2157 |
+
"loss": 0.0067,
|
| 2158 |
+
"step": 3070
|
| 2159 |
+
},
|
| 2160 |
+
{
|
| 2161 |
+
"epoch": 0.308,
|
| 2162 |
+
"grad_norm": 0.07357806712388992,
|
| 2163 |
+
"learning_rate": 3.46e-05,
|
| 2164 |
+
"loss": 0.007,
|
| 2165 |
+
"step": 3080
|
| 2166 |
+
},
|
| 2167 |
+
{
|
| 2168 |
+
"epoch": 0.309,
|
| 2169 |
+
"grad_norm": 0.060446444898843765,
|
| 2170 |
+
"learning_rate": 3.455e-05,
|
| 2171 |
+
"loss": 0.0063,
|
| 2172 |
+
"step": 3090
|
| 2173 |
+
},
|
| 2174 |
+
{
|
| 2175 |
+
"epoch": 0.31,
|
| 2176 |
+
"grad_norm": 0.05178246274590492,
|
| 2177 |
+
"learning_rate": 3.45e-05,
|
| 2178 |
+
"loss": 0.0064,
|
| 2179 |
+
"step": 3100
|
| 2180 |
+
},
|
| 2181 |
+
{
|
| 2182 |
+
"epoch": 0.311,
|
| 2183 |
+
"grad_norm": 0.4560135006904602,
|
| 2184 |
+
"learning_rate": 3.445e-05,
|
| 2185 |
+
"loss": 0.0076,
|
| 2186 |
+
"step": 3110
|
| 2187 |
+
},
|
| 2188 |
+
{
|
| 2189 |
+
"epoch": 0.312,
|
| 2190 |
+
"grad_norm": 0.10910088568925858,
|
| 2191 |
+
"learning_rate": 3.4399999999999996e-05,
|
| 2192 |
+
"loss": 0.0082,
|
| 2193 |
+
"step": 3120
|
| 2194 |
+
},
|
| 2195 |
+
{
|
| 2196 |
+
"epoch": 0.313,
|
| 2197 |
+
"grad_norm": 0.05087321624159813,
|
| 2198 |
+
"learning_rate": 3.435e-05,
|
| 2199 |
+
"loss": 0.0061,
|
| 2200 |
+
"step": 3130
|
| 2201 |
+
},
|
| 2202 |
+
{
|
| 2203 |
+
"epoch": 0.314,
|
| 2204 |
+
"grad_norm": 0.055152568966150284,
|
| 2205 |
+
"learning_rate": 3.430000000000001e-05,
|
| 2206 |
+
"loss": 0.0065,
|
| 2207 |
+
"step": 3140
|
| 2208 |
+
},
|
| 2209 |
+
{
|
| 2210 |
+
"epoch": 0.315,
|
| 2211 |
+
"grad_norm": 0.48375555872917175,
|
| 2212 |
+
"learning_rate": 3.4250000000000006e-05,
|
| 2213 |
+
"loss": 0.0081,
|
| 2214 |
+
"step": 3150
|
| 2215 |
+
},
|
| 2216 |
+
{
|
| 2217 |
+
"epoch": 0.316,
|
| 2218 |
+
"grad_norm": 0.12223263829946518,
|
| 2219 |
+
"learning_rate": 3.4200000000000005e-05,
|
| 2220 |
+
"loss": 0.0098,
|
| 2221 |
+
"step": 3160
|
| 2222 |
+
},
|
| 2223 |
+
{
|
| 2224 |
+
"epoch": 0.317,
|
| 2225 |
+
"grad_norm": 0.15452681481838226,
|
| 2226 |
+
"learning_rate": 3.415e-05,
|
| 2227 |
+
"loss": 0.0087,
|
| 2228 |
+
"step": 3170
|
| 2229 |
+
},
|
| 2230 |
+
{
|
| 2231 |
+
"epoch": 0.318,
|
| 2232 |
+
"grad_norm": 0.06133843585848808,
|
| 2233 |
+
"learning_rate": 3.41e-05,
|
| 2234 |
+
"loss": 0.0087,
|
| 2235 |
+
"step": 3180
|
| 2236 |
+
},
|
| 2237 |
+
{
|
| 2238 |
+
"epoch": 0.319,
|
| 2239 |
+
"grad_norm": 0.04037950187921524,
|
| 2240 |
+
"learning_rate": 3.405e-05,
|
| 2241 |
+
"loss": 0.0062,
|
| 2242 |
+
"step": 3190
|
| 2243 |
+
},
|
| 2244 |
+
{
|
| 2245 |
+
"epoch": 0.32,
|
| 2246 |
+
"grad_norm": 0.05707933381199837,
|
| 2247 |
+
"learning_rate": 3.4000000000000007e-05,
|
| 2248 |
+
"loss": 0.0061,
|
| 2249 |
+
"step": 3200
|
| 2250 |
+
},
|
| 2251 |
+
{
|
| 2252 |
+
"epoch": 0.321,
|
| 2253 |
+
"grad_norm": 0.04124099016189575,
|
| 2254 |
+
"learning_rate": 3.3950000000000005e-05,
|
| 2255 |
+
"loss": 0.0058,
|
| 2256 |
+
"step": 3210
|
| 2257 |
+
},
|
| 2258 |
+
{
|
| 2259 |
+
"epoch": 0.322,
|
| 2260 |
+
"grad_norm": 0.12988638877868652,
|
| 2261 |
+
"learning_rate": 3.3900000000000004e-05,
|
| 2262 |
+
"loss": 0.0062,
|
| 2263 |
+
"step": 3220
|
| 2264 |
+
},
|
| 2265 |
+
{
|
| 2266 |
+
"epoch": 0.323,
|
| 2267 |
+
"grad_norm": 0.04961306229233742,
|
| 2268 |
+
"learning_rate": 3.385e-05,
|
| 2269 |
+
"loss": 0.0057,
|
| 2270 |
+
"step": 3230
|
| 2271 |
+
},
|
| 2272 |
+
{
|
| 2273 |
+
"epoch": 0.324,
|
| 2274 |
+
"grad_norm": 0.05354069173336029,
|
| 2275 |
+
"learning_rate": 3.38e-05,
|
| 2276 |
+
"loss": 0.0062,
|
| 2277 |
+
"step": 3240
|
| 2278 |
+
},
|
| 2279 |
+
{
|
| 2280 |
+
"epoch": 0.325,
|
| 2281 |
+
"grad_norm": 0.04944461211562157,
|
| 2282 |
+
"learning_rate": 3.375000000000001e-05,
|
| 2283 |
+
"loss": 0.0061,
|
| 2284 |
+
"step": 3250
|
| 2285 |
+
},
|
| 2286 |
+
{
|
| 2287 |
+
"epoch": 0.326,
|
| 2288 |
+
"grad_norm": 0.03429180383682251,
|
| 2289 |
+
"learning_rate": 3.3700000000000006e-05,
|
| 2290 |
+
"loss": 0.0055,
|
| 2291 |
+
"step": 3260
|
| 2292 |
+
},
|
| 2293 |
+
{
|
| 2294 |
+
"epoch": 0.327,
|
| 2295 |
+
"grad_norm": 0.05271946266293526,
|
| 2296 |
+
"learning_rate": 3.3650000000000005e-05,
|
| 2297 |
+
"loss": 0.0054,
|
| 2298 |
+
"step": 3270
|
| 2299 |
+
},
|
| 2300 |
+
{
|
| 2301 |
+
"epoch": 0.328,
|
| 2302 |
+
"grad_norm": 0.03602539002895355,
|
| 2303 |
+
"learning_rate": 3.3600000000000004e-05,
|
| 2304 |
+
"loss": 0.0049,
|
| 2305 |
+
"step": 3280
|
| 2306 |
+
},
|
| 2307 |
+
{
|
| 2308 |
+
"epoch": 0.329,
|
| 2309 |
+
"grad_norm": 0.03325178474187851,
|
| 2310 |
+
"learning_rate": 3.355e-05,
|
| 2311 |
+
"loss": 0.0052,
|
| 2312 |
+
"step": 3290
|
| 2313 |
+
},
|
| 2314 |
+
{
|
| 2315 |
+
"epoch": 0.33,
|
| 2316 |
+
"grad_norm": 0.03728079795837402,
|
| 2317 |
+
"learning_rate": 3.35e-05,
|
| 2318 |
+
"loss": 0.0057,
|
| 2319 |
+
"step": 3300
|
| 2320 |
+
},
|
| 2321 |
+
{
|
| 2322 |
+
"epoch": 0.331,
|
| 2323 |
+
"grad_norm": 0.053768668323755264,
|
| 2324 |
+
"learning_rate": 3.345000000000001e-05,
|
| 2325 |
+
"loss": 0.0055,
|
| 2326 |
+
"step": 3310
|
| 2327 |
+
},
|
| 2328 |
+
{
|
| 2329 |
+
"epoch": 0.332,
|
| 2330 |
+
"grad_norm": 0.054501548409461975,
|
| 2331 |
+
"learning_rate": 3.3400000000000005e-05,
|
| 2332 |
+
"loss": 0.0053,
|
| 2333 |
+
"step": 3320
|
| 2334 |
+
},
|
| 2335 |
+
{
|
| 2336 |
+
"epoch": 0.333,
|
| 2337 |
+
"grad_norm": 0.05519956722855568,
|
| 2338 |
+
"learning_rate": 3.3350000000000004e-05,
|
| 2339 |
+
"loss": 0.0053,
|
| 2340 |
+
"step": 3330
|
| 2341 |
+
},
|
| 2342 |
+
{
|
| 2343 |
+
"epoch": 0.334,
|
| 2344 |
+
"grad_norm": 0.05373954027891159,
|
| 2345 |
+
"learning_rate": 3.33e-05,
|
| 2346 |
+
"loss": 0.006,
|
| 2347 |
+
"step": 3340
|
| 2348 |
+
},
|
| 2349 |
+
{
|
| 2350 |
+
"epoch": 0.335,
|
| 2351 |
+
"grad_norm": 0.04272560030221939,
|
| 2352 |
+
"learning_rate": 3.325e-05,
|
| 2353 |
+
"loss": 0.0057,
|
| 2354 |
+
"step": 3350
|
| 2355 |
+
},
|
| 2356 |
+
{
|
| 2357 |
+
"epoch": 0.336,
|
| 2358 |
+
"grad_norm": 0.047061894088983536,
|
| 2359 |
+
"learning_rate": 3.32e-05,
|
| 2360 |
+
"loss": 0.0048,
|
| 2361 |
+
"step": 3360
|
| 2362 |
+
},
|
| 2363 |
+
{
|
| 2364 |
+
"epoch": 0.337,
|
| 2365 |
+
"grad_norm": 0.032794494181871414,
|
| 2366 |
+
"learning_rate": 3.3150000000000006e-05,
|
| 2367 |
+
"loss": 0.0046,
|
| 2368 |
+
"step": 3370
|
| 2369 |
+
},
|
| 2370 |
+
{
|
| 2371 |
+
"epoch": 0.338,
|
| 2372 |
+
"grad_norm": 0.027148200199007988,
|
| 2373 |
+
"learning_rate": 3.3100000000000005e-05,
|
| 2374 |
+
"loss": 0.0046,
|
| 2375 |
+
"step": 3380
|
| 2376 |
+
},
|
| 2377 |
+
{
|
| 2378 |
+
"epoch": 0.339,
|
| 2379 |
+
"grad_norm": 0.035516317933797836,
|
| 2380 |
+
"learning_rate": 3.3050000000000004e-05,
|
| 2381 |
+
"loss": 0.0048,
|
| 2382 |
+
"step": 3390
|
| 2383 |
+
},
|
| 2384 |
+
{
|
| 2385 |
+
"epoch": 0.34,
|
| 2386 |
+
"grad_norm": 0.046294230967760086,
|
| 2387 |
+
"learning_rate": 3.3e-05,
|
| 2388 |
+
"loss": 0.0063,
|
| 2389 |
+
"step": 3400
|
| 2390 |
+
},
|
| 2391 |
+
{
|
| 2392 |
+
"epoch": 0.341,
|
| 2393 |
+
"grad_norm": 0.07840899378061295,
|
| 2394 |
+
"learning_rate": 3.295e-05,
|
| 2395 |
+
"loss": 0.0047,
|
| 2396 |
+
"step": 3410
|
| 2397 |
+
},
|
| 2398 |
+
{
|
| 2399 |
+
"epoch": 0.342,
|
| 2400 |
+
"grad_norm": 0.04392802715301514,
|
| 2401 |
+
"learning_rate": 3.29e-05,
|
| 2402 |
+
"loss": 0.0048,
|
| 2403 |
+
"step": 3420
|
| 2404 |
+
},
|
| 2405 |
+
{
|
| 2406 |
+
"epoch": 0.343,
|
| 2407 |
+
"grad_norm": 0.04237942770123482,
|
| 2408 |
+
"learning_rate": 3.2850000000000006e-05,
|
| 2409 |
+
"loss": 0.0047,
|
| 2410 |
+
"step": 3430
|
| 2411 |
+
},
|
| 2412 |
+
{
|
| 2413 |
+
"epoch": 0.344,
|
| 2414 |
+
"grad_norm": 0.03379204496741295,
|
| 2415 |
+
"learning_rate": 3.2800000000000004e-05,
|
| 2416 |
+
"loss": 0.005,
|
| 2417 |
+
"step": 3440
|
| 2418 |
+
},
|
| 2419 |
+
{
|
| 2420 |
+
"epoch": 0.345,
|
| 2421 |
+
"grad_norm": 0.3130718469619751,
|
| 2422 |
+
"learning_rate": 3.275e-05,
|
| 2423 |
+
"loss": 0.0108,
|
| 2424 |
+
"step": 3450
|
| 2425 |
+
},
|
| 2426 |
+
{
|
| 2427 |
+
"epoch": 0.346,
|
| 2428 |
+
"grad_norm": 0.13712112605571747,
|
| 2429 |
+
"learning_rate": 3.27e-05,
|
| 2430 |
+
"loss": 0.0095,
|
| 2431 |
+
"step": 3460
|
| 2432 |
+
},
|
| 2433 |
+
{
|
| 2434 |
+
"epoch": 0.347,
|
| 2435 |
+
"grad_norm": 0.12321494519710541,
|
| 2436 |
+
"learning_rate": 3.265e-05,
|
| 2437 |
+
"loss": 0.0075,
|
| 2438 |
+
"step": 3470
|
| 2439 |
+
},
|
| 2440 |
+
{
|
| 2441 |
+
"epoch": 0.348,
|
| 2442 |
+
"grad_norm": 0.06602940708398819,
|
| 2443 |
+
"learning_rate": 3.26e-05,
|
| 2444 |
+
"loss": 0.0062,
|
| 2445 |
+
"step": 3480
|
| 2446 |
+
},
|
| 2447 |
+
{
|
| 2448 |
+
"epoch": 0.349,
|
| 2449 |
+
"grad_norm": 0.08250287175178528,
|
| 2450 |
+
"learning_rate": 3.2550000000000005e-05,
|
| 2451 |
+
"loss": 0.0053,
|
| 2452 |
+
"step": 3490
|
| 2453 |
+
},
|
| 2454 |
+
{
|
| 2455 |
+
"epoch": 0.35,
|
| 2456 |
+
"grad_norm": 0.04638442397117615,
|
| 2457 |
+
"learning_rate": 3.2500000000000004e-05,
|
| 2458 |
+
"loss": 0.0052,
|
| 2459 |
+
"step": 3500
|
| 2460 |
+
},
|
| 2461 |
+
{
|
| 2462 |
+
"epoch": 0.351,
|
| 2463 |
+
"grad_norm": 0.043373119086027145,
|
| 2464 |
+
"learning_rate": 3.245e-05,
|
| 2465 |
+
"loss": 0.0053,
|
| 2466 |
+
"step": 3510
|
| 2467 |
+
},
|
| 2468 |
+
{
|
| 2469 |
+
"epoch": 0.352,
|
| 2470 |
+
"grad_norm": 0.04895636439323425,
|
| 2471 |
+
"learning_rate": 3.24e-05,
|
| 2472 |
+
"loss": 0.0043,
|
| 2473 |
+
"step": 3520
|
| 2474 |
+
},
|
| 2475 |
+
{
|
| 2476 |
+
"epoch": 0.353,
|
| 2477 |
+
"grad_norm": 0.04256746545433998,
|
| 2478 |
+
"learning_rate": 3.235e-05,
|
| 2479 |
+
"loss": 0.0044,
|
| 2480 |
+
"step": 3530
|
| 2481 |
+
},
|
| 2482 |
+
{
|
| 2483 |
+
"epoch": 0.354,
|
| 2484 |
+
"grad_norm": 0.02967280149459839,
|
| 2485 |
+
"learning_rate": 3.2300000000000006e-05,
|
| 2486 |
+
"loss": 0.0046,
|
| 2487 |
+
"step": 3540
|
| 2488 |
+
},
|
| 2489 |
+
{
|
| 2490 |
+
"epoch": 0.355,
|
| 2491 |
+
"grad_norm": 0.02590947411954403,
|
| 2492 |
+
"learning_rate": 3.2250000000000005e-05,
|
| 2493 |
+
"loss": 0.0044,
|
| 2494 |
+
"step": 3550
|
| 2495 |
+
},
|
| 2496 |
+
{
|
| 2497 |
+
"epoch": 0.356,
|
| 2498 |
+
"grad_norm": 0.026240160688757896,
|
| 2499 |
+
"learning_rate": 3.2200000000000003e-05,
|
| 2500 |
+
"loss": 0.0041,
|
| 2501 |
+
"step": 3560
|
| 2502 |
+
},
|
| 2503 |
+
{
|
| 2504 |
+
"epoch": 0.357,
|
| 2505 |
+
"grad_norm": 0.048163361847400665,
|
| 2506 |
+
"learning_rate": 3.215e-05,
|
| 2507 |
+
"loss": 0.0048,
|
| 2508 |
+
"step": 3570
|
| 2509 |
+
},
|
| 2510 |
+
{
|
| 2511 |
+
"epoch": 0.358,
|
| 2512 |
+
"grad_norm": 0.04310280829668045,
|
| 2513 |
+
"learning_rate": 3.21e-05,
|
| 2514 |
+
"loss": 0.0044,
|
| 2515 |
+
"step": 3580
|
| 2516 |
+
},
|
| 2517 |
+
{
|
| 2518 |
+
"epoch": 0.359,
|
| 2519 |
+
"grad_norm": 0.027477843686938286,
|
| 2520 |
+
"learning_rate": 3.205e-05,
|
| 2521 |
+
"loss": 0.004,
|
| 2522 |
+
"step": 3590
|
| 2523 |
+
},
|
| 2524 |
+
{
|
| 2525 |
+
"epoch": 0.36,
|
| 2526 |
+
"grad_norm": 0.02773194946348667,
|
| 2527 |
+
"learning_rate": 3.2000000000000005e-05,
|
| 2528 |
+
"loss": 0.0046,
|
| 2529 |
+
"step": 3600
|
| 2530 |
+
},
|
| 2531 |
+
{
|
| 2532 |
+
"epoch": 0.361,
|
| 2533 |
+
"grad_norm": 0.027110638096928596,
|
| 2534 |
+
"learning_rate": 3.1950000000000004e-05,
|
| 2535 |
+
"loss": 0.004,
|
| 2536 |
+
"step": 3610
|
| 2537 |
+
},
|
| 2538 |
+
{
|
| 2539 |
+
"epoch": 0.362,
|
| 2540 |
+
"grad_norm": 0.04346521571278572,
|
| 2541 |
+
"learning_rate": 3.19e-05,
|
| 2542 |
+
"loss": 0.0039,
|
| 2543 |
+
"step": 3620
|
| 2544 |
+
},
|
| 2545 |
+
{
|
| 2546 |
+
"epoch": 0.363,
|
| 2547 |
+
"grad_norm": 0.024588119238615036,
|
| 2548 |
+
"learning_rate": 3.185e-05,
|
| 2549 |
+
"loss": 0.0041,
|
| 2550 |
+
"step": 3630
|
| 2551 |
+
},
|
| 2552 |
+
{
|
| 2553 |
+
"epoch": 0.364,
|
| 2554 |
+
"grad_norm": 0.03631160408258438,
|
| 2555 |
+
"learning_rate": 3.18e-05,
|
| 2556 |
+
"loss": 0.0039,
|
| 2557 |
+
"step": 3640
|
| 2558 |
+
},
|
| 2559 |
+
{
|
| 2560 |
+
"epoch": 0.365,
|
| 2561 |
+
"grad_norm": 0.028497323393821716,
|
| 2562 |
+
"learning_rate": 3.175e-05,
|
| 2563 |
+
"loss": 0.0045,
|
| 2564 |
+
"step": 3650
|
| 2565 |
+
},
|
| 2566 |
+
{
|
| 2567 |
+
"epoch": 0.366,
|
| 2568 |
+
"grad_norm": 0.06324070692062378,
|
| 2569 |
+
"learning_rate": 3.1700000000000005e-05,
|
| 2570 |
+
"loss": 0.004,
|
| 2571 |
+
"step": 3660
|
| 2572 |
+
},
|
| 2573 |
+
{
|
| 2574 |
+
"epoch": 0.367,
|
| 2575 |
+
"grad_norm": 0.22121182084083557,
|
| 2576 |
+
"learning_rate": 3.1650000000000004e-05,
|
| 2577 |
+
"loss": 0.0078,
|
| 2578 |
+
"step": 3670
|
| 2579 |
+
},
|
| 2580 |
+
{
|
| 2581 |
+
"epoch": 0.368,
|
| 2582 |
+
"grad_norm": 0.130402073264122,
|
| 2583 |
+
"learning_rate": 3.16e-05,
|
| 2584 |
+
"loss": 0.0068,
|
| 2585 |
+
"step": 3680
|
| 2586 |
+
},
|
| 2587 |
+
{
|
| 2588 |
+
"epoch": 0.369,
|
| 2589 |
+
"grad_norm": 0.11193361133337021,
|
| 2590 |
+
"learning_rate": 3.155e-05,
|
| 2591 |
+
"loss": 0.0052,
|
| 2592 |
+
"step": 3690
|
| 2593 |
+
},
|
| 2594 |
+
{
|
| 2595 |
+
"epoch": 0.37,
|
| 2596 |
+
"grad_norm": 0.043261051177978516,
|
| 2597 |
+
"learning_rate": 3.15e-05,
|
| 2598 |
+
"loss": 0.0046,
|
| 2599 |
+
"step": 3700
|
| 2600 |
+
},
|
| 2601 |
+
{
|
| 2602 |
+
"epoch": 0.371,
|
| 2603 |
+
"grad_norm": 0.05132100731134415,
|
| 2604 |
+
"learning_rate": 3.145e-05,
|
| 2605 |
+
"loss": 0.0041,
|
| 2606 |
+
"step": 3710
|
| 2607 |
+
},
|
| 2608 |
+
{
|
| 2609 |
+
"epoch": 0.372,
|
| 2610 |
+
"grad_norm": 0.03498228266835213,
|
| 2611 |
+
"learning_rate": 3.1400000000000004e-05,
|
| 2612 |
+
"loss": 0.0041,
|
| 2613 |
+
"step": 3720
|
| 2614 |
+
},
|
| 2615 |
+
{
|
| 2616 |
+
"epoch": 0.373,
|
| 2617 |
+
"grad_norm": 0.03594733029603958,
|
| 2618 |
+
"learning_rate": 3.135e-05,
|
| 2619 |
+
"loss": 0.0037,
|
| 2620 |
+
"step": 3730
|
| 2621 |
+
},
|
| 2622 |
+
{
|
| 2623 |
+
"epoch": 0.374,
|
| 2624 |
+
"grad_norm": 0.05542735382914543,
|
| 2625 |
+
"learning_rate": 3.13e-05,
|
| 2626 |
+
"loss": 0.0042,
|
| 2627 |
+
"step": 3740
|
| 2628 |
+
},
|
| 2629 |
+
{
|
| 2630 |
+
"epoch": 0.375,
|
| 2631 |
+
"grad_norm": 0.03302931785583496,
|
| 2632 |
+
"learning_rate": 3.125e-05,
|
| 2633 |
+
"loss": 0.0048,
|
| 2634 |
+
"step": 3750
|
| 2635 |
+
},
|
| 2636 |
+
{
|
| 2637 |
+
"epoch": 0.376,
|
| 2638 |
+
"grad_norm": 0.028392167761921883,
|
| 2639 |
+
"learning_rate": 3.12e-05,
|
| 2640 |
+
"loss": 0.0042,
|
| 2641 |
+
"step": 3760
|
| 2642 |
+
},
|
| 2643 |
+
{
|
| 2644 |
+
"epoch": 0.377,
|
| 2645 |
+
"grad_norm": 0.05274713411927223,
|
| 2646 |
+
"learning_rate": 3.115e-05,
|
| 2647 |
+
"loss": 0.0045,
|
| 2648 |
+
"step": 3770
|
| 2649 |
+
},
|
| 2650 |
+
{
|
| 2651 |
+
"epoch": 0.378,
|
| 2652 |
+
"grad_norm": 0.024890929460525513,
|
| 2653 |
+
"learning_rate": 3.1100000000000004e-05,
|
| 2654 |
+
"loss": 0.0038,
|
| 2655 |
+
"step": 3780
|
| 2656 |
+
},
|
| 2657 |
+
{
|
| 2658 |
+
"epoch": 0.379,
|
| 2659 |
+
"grad_norm": 0.02797631174325943,
|
| 2660 |
+
"learning_rate": 3.105e-05,
|
| 2661 |
+
"loss": 0.0036,
|
| 2662 |
+
"step": 3790
|
| 2663 |
+
},
|
| 2664 |
+
{
|
| 2665 |
+
"epoch": 0.38,
|
| 2666 |
+
"grad_norm": 0.033390454947948456,
|
| 2667 |
+
"learning_rate": 3.1e-05,
|
| 2668 |
+
"loss": 0.0039,
|
| 2669 |
+
"step": 3800
|
| 2670 |
+
},
|
| 2671 |
+
{
|
| 2672 |
+
"epoch": 0.381,
|
| 2673 |
+
"grad_norm": 0.024741416797041893,
|
| 2674 |
+
"learning_rate": 3.095e-05,
|
| 2675 |
+
"loss": 0.0042,
|
| 2676 |
+
"step": 3810
|
| 2677 |
+
},
|
| 2678 |
+
{
|
| 2679 |
+
"epoch": 0.382,
|
| 2680 |
+
"grad_norm": 0.05398337543010712,
|
| 2681 |
+
"learning_rate": 3.09e-05,
|
| 2682 |
+
"loss": 0.0049,
|
| 2683 |
+
"step": 3820
|
| 2684 |
+
},
|
| 2685 |
+
{
|
| 2686 |
+
"epoch": 0.383,
|
| 2687 |
+
"grad_norm": 0.027936646714806557,
|
| 2688 |
+
"learning_rate": 3.0850000000000004e-05,
|
| 2689 |
+
"loss": 0.0034,
|
| 2690 |
+
"step": 3830
|
| 2691 |
+
},
|
| 2692 |
+
{
|
| 2693 |
+
"epoch": 0.384,
|
| 2694 |
+
"grad_norm": 0.02413495071232319,
|
| 2695 |
+
"learning_rate": 3.08e-05,
|
| 2696 |
+
"loss": 0.0036,
|
| 2697 |
+
"step": 3840
|
| 2698 |
+
},
|
| 2699 |
+
{
|
| 2700 |
+
"epoch": 0.385,
|
| 2701 |
+
"grad_norm": 0.037689995020627975,
|
| 2702 |
+
"learning_rate": 3.075e-05,
|
| 2703 |
+
"loss": 0.0033,
|
| 2704 |
+
"step": 3850
|
| 2705 |
+
},
|
| 2706 |
+
{
|
| 2707 |
+
"epoch": 0.386,
|
| 2708 |
+
"grad_norm": 0.028174949809908867,
|
| 2709 |
+
"learning_rate": 3.07e-05,
|
| 2710 |
+
"loss": 0.0036,
|
| 2711 |
+
"step": 3860
|
| 2712 |
+
},
|
| 2713 |
+
{
|
| 2714 |
+
"epoch": 0.387,
|
| 2715 |
+
"grad_norm": 0.064354807138443,
|
| 2716 |
+
"learning_rate": 3.065e-05,
|
| 2717 |
+
"loss": 0.0037,
|
| 2718 |
+
"step": 3870
|
| 2719 |
+
},
|
| 2720 |
+
{
|
| 2721 |
+
"epoch": 0.388,
|
| 2722 |
+
"grad_norm": 0.028341595083475113,
|
| 2723 |
+
"learning_rate": 3.06e-05,
|
| 2724 |
+
"loss": 0.0034,
|
| 2725 |
+
"step": 3880
|
| 2726 |
+
},
|
| 2727 |
+
{
|
| 2728 |
+
"epoch": 0.389,
|
| 2729 |
+
"grad_norm": 0.06142325699329376,
|
| 2730 |
+
"learning_rate": 3.0550000000000004e-05,
|
| 2731 |
+
"loss": 0.004,
|
| 2732 |
+
"step": 3890
|
| 2733 |
+
},
|
| 2734 |
+
{
|
| 2735 |
+
"epoch": 0.39,
|
| 2736 |
+
"grad_norm": 0.03553822636604309,
|
| 2737 |
+
"learning_rate": 3.05e-05,
|
| 2738 |
+
"loss": 0.004,
|
| 2739 |
+
"step": 3900
|
| 2740 |
+
},
|
| 2741 |
+
{
|
| 2742 |
+
"epoch": 0.391,
|
| 2743 |
+
"grad_norm": 0.025645367801189423,
|
| 2744 |
+
"learning_rate": 3.045e-05,
|
| 2745 |
+
"loss": 0.0032,
|
| 2746 |
+
"step": 3910
|
| 2747 |
+
},
|
| 2748 |
+
{
|
| 2749 |
+
"epoch": 0.392,
|
| 2750 |
+
"grad_norm": 0.053947921842336655,
|
| 2751 |
+
"learning_rate": 3.04e-05,
|
| 2752 |
+
"loss": 0.0042,
|
| 2753 |
+
"step": 3920
|
| 2754 |
+
},
|
| 2755 |
+
{
|
| 2756 |
+
"epoch": 0.393,
|
| 2757 |
+
"grad_norm": 0.040126167237758636,
|
| 2758 |
+
"learning_rate": 3.035e-05,
|
| 2759 |
+
"loss": 0.0038,
|
| 2760 |
+
"step": 3930
|
| 2761 |
+
},
|
| 2762 |
+
{
|
| 2763 |
+
"epoch": 0.394,
|
| 2764 |
+
"grad_norm": 0.02956206165254116,
|
| 2765 |
+
"learning_rate": 3.03e-05,
|
| 2766 |
+
"loss": 0.0035,
|
| 2767 |
+
"step": 3940
|
| 2768 |
+
},
|
| 2769 |
+
{
|
| 2770 |
+
"epoch": 0.395,
|
| 2771 |
+
"grad_norm": 0.11952024698257446,
|
| 2772 |
+
"learning_rate": 3.025e-05,
|
| 2773 |
+
"loss": 0.0057,
|
| 2774 |
+
"step": 3950
|
| 2775 |
+
},
|
| 2776 |
+
{
|
| 2777 |
+
"epoch": 0.396,
|
| 2778 |
+
"grad_norm": 0.04155363142490387,
|
| 2779 |
+
"learning_rate": 3.02e-05,
|
| 2780 |
+
"loss": 0.0034,
|
| 2781 |
+
"step": 3960
|
| 2782 |
+
},
|
| 2783 |
+
{
|
| 2784 |
+
"epoch": 0.397,
|
| 2785 |
+
"grad_norm": 0.03884551301598549,
|
| 2786 |
+
"learning_rate": 3.015e-05,
|
| 2787 |
+
"loss": 0.0037,
|
| 2788 |
+
"step": 3970
|
| 2789 |
+
},
|
| 2790 |
+
{
|
| 2791 |
+
"epoch": 0.398,
|
| 2792 |
+
"grad_norm": 0.033869728446006775,
|
| 2793 |
+
"learning_rate": 3.01e-05,
|
| 2794 |
+
"loss": 0.0033,
|
| 2795 |
+
"step": 3980
|
| 2796 |
+
},
|
| 2797 |
+
{
|
| 2798 |
+
"epoch": 0.399,
|
| 2799 |
+
"grad_norm": 0.027508044615387917,
|
| 2800 |
+
"learning_rate": 3.0050000000000002e-05,
|
| 2801 |
+
"loss": 0.0038,
|
| 2802 |
+
"step": 3990
|
| 2803 |
+
},
|
| 2804 |
+
{
|
| 2805 |
+
"epoch": 0.4,
|
| 2806 |
+
"grad_norm": 0.019838711246848106,
|
| 2807 |
+
"learning_rate": 3e-05,
|
| 2808 |
+
"loss": 0.0034,
|
| 2809 |
+
"step": 4000
|
| 2810 |
+
},
|
| 2811 |
+
{
|
| 2812 |
+
"epoch": 0.401,
|
| 2813 |
+
"grad_norm": 0.042124535888433456,
|
| 2814 |
+
"learning_rate": 2.995e-05,
|
| 2815 |
+
"loss": 0.003,
|
| 2816 |
+
"step": 4010
|
| 2817 |
+
},
|
| 2818 |
+
{
|
| 2819 |
+
"epoch": 0.402,
|
| 2820 |
+
"grad_norm": 0.03583139553666115,
|
| 2821 |
+
"learning_rate": 2.9900000000000002e-05,
|
| 2822 |
+
"loss": 0.0029,
|
| 2823 |
+
"step": 4020
|
| 2824 |
+
},
|
| 2825 |
+
{
|
| 2826 |
+
"epoch": 0.403,
|
| 2827 |
+
"grad_norm": 0.024187223985791206,
|
| 2828 |
+
"learning_rate": 2.985e-05,
|
| 2829 |
+
"loss": 0.0031,
|
| 2830 |
+
"step": 4030
|
| 2831 |
+
},
|
| 2832 |
+
{
|
| 2833 |
+
"epoch": 0.404,
|
| 2834 |
+
"grad_norm": 0.02509123831987381,
|
| 2835 |
+
"learning_rate": 2.98e-05,
|
| 2836 |
+
"loss": 0.003,
|
| 2837 |
+
"step": 4040
|
| 2838 |
+
},
|
| 2839 |
+
{
|
| 2840 |
+
"epoch": 0.405,
|
| 2841 |
+
"grad_norm": 0.015798581764101982,
|
| 2842 |
+
"learning_rate": 2.975e-05,
|
| 2843 |
+
"loss": 0.003,
|
| 2844 |
+
"step": 4050
|
| 2845 |
+
},
|
| 2846 |
+
{
|
| 2847 |
+
"epoch": 0.406,
|
| 2848 |
+
"grad_norm": 0.01964486949145794,
|
| 2849 |
+
"learning_rate": 2.97e-05,
|
| 2850 |
+
"loss": 0.0032,
|
| 2851 |
+
"step": 4060
|
| 2852 |
+
},
|
| 2853 |
+
{
|
| 2854 |
+
"epoch": 0.407,
|
| 2855 |
+
"grad_norm": 0.025820232927799225,
|
| 2856 |
+
"learning_rate": 2.965e-05,
|
| 2857 |
+
"loss": 0.0032,
|
| 2858 |
+
"step": 4070
|
| 2859 |
+
},
|
| 2860 |
+
{
|
| 2861 |
+
"epoch": 0.408,
|
| 2862 |
+
"grad_norm": 0.03453589975833893,
|
| 2863 |
+
"learning_rate": 2.96e-05,
|
| 2864 |
+
"loss": 0.003,
|
| 2865 |
+
"step": 4080
|
| 2866 |
+
},
|
| 2867 |
+
{
|
| 2868 |
+
"epoch": 0.409,
|
| 2869 |
+
"grad_norm": 0.022311529144644737,
|
| 2870 |
+
"learning_rate": 2.955e-05,
|
| 2871 |
+
"loss": 0.0025,
|
| 2872 |
+
"step": 4090
|
| 2873 |
+
},
|
| 2874 |
+
{
|
| 2875 |
+
"epoch": 0.41,
|
| 2876 |
+
"grad_norm": 0.02296466939151287,
|
| 2877 |
+
"learning_rate": 2.95e-05,
|
| 2878 |
+
"loss": 0.003,
|
| 2879 |
+
"step": 4100
|
| 2880 |
+
},
|
| 2881 |
+
{
|
| 2882 |
+
"epoch": 0.411,
|
| 2883 |
+
"grad_norm": 0.022816313430666924,
|
| 2884 |
+
"learning_rate": 2.945e-05,
|
| 2885 |
+
"loss": 0.0032,
|
| 2886 |
+
"step": 4110
|
| 2887 |
+
},
|
| 2888 |
+
{
|
| 2889 |
+
"epoch": 0.412,
|
| 2890 |
+
"grad_norm": 0.021030904725193977,
|
| 2891 |
+
"learning_rate": 2.94e-05,
|
| 2892 |
+
"loss": 0.003,
|
| 2893 |
+
"step": 4120
|
| 2894 |
+
},
|
| 2895 |
+
{
|
| 2896 |
+
"epoch": 0.413,
|
| 2897 |
+
"grad_norm": 0.02336346171796322,
|
| 2898 |
+
"learning_rate": 2.935e-05,
|
| 2899 |
+
"loss": 0.0028,
|
| 2900 |
+
"step": 4130
|
| 2901 |
+
},
|
| 2902 |
+
{
|
| 2903 |
+
"epoch": 0.414,
|
| 2904 |
+
"grad_norm": 0.019582638517022133,
|
| 2905 |
+
"learning_rate": 2.93e-05,
|
| 2906 |
+
"loss": 0.0027,
|
| 2907 |
+
"step": 4140
|
| 2908 |
+
},
|
| 2909 |
+
{
|
| 2910 |
+
"epoch": 0.415,
|
| 2911 |
+
"grad_norm": 0.031429585069417953,
|
| 2912 |
+
"learning_rate": 2.925e-05,
|
| 2913 |
+
"loss": 0.0027,
|
| 2914 |
+
"step": 4150
|
| 2915 |
+
},
|
| 2916 |
+
{
|
| 2917 |
+
"epoch": 0.416,
|
| 2918 |
+
"grad_norm": 0.027804825454950333,
|
| 2919 |
+
"learning_rate": 2.9199999999999998e-05,
|
| 2920 |
+
"loss": 0.0027,
|
| 2921 |
+
"step": 4160
|
| 2922 |
+
},
|
| 2923 |
+
{
|
| 2924 |
+
"epoch": 0.417,
|
| 2925 |
+
"grad_norm": 0.022006656974554062,
|
| 2926 |
+
"learning_rate": 2.915e-05,
|
| 2927 |
+
"loss": 0.003,
|
| 2928 |
+
"step": 4170
|
| 2929 |
+
},
|
| 2930 |
+
{
|
| 2931 |
+
"epoch": 0.418,
|
| 2932 |
+
"grad_norm": 0.052478939294815063,
|
| 2933 |
+
"learning_rate": 2.91e-05,
|
| 2934 |
+
"loss": 0.0044,
|
| 2935 |
+
"step": 4180
|
| 2936 |
+
},
|
| 2937 |
+
{
|
| 2938 |
+
"epoch": 0.419,
|
| 2939 |
+
"grad_norm": 0.03854925185441971,
|
| 2940 |
+
"learning_rate": 2.9049999999999998e-05,
|
| 2941 |
+
"loss": 0.0036,
|
| 2942 |
+
"step": 4190
|
| 2943 |
+
},
|
| 2944 |
+
{
|
| 2945 |
+
"epoch": 0.42,
|
| 2946 |
+
"grad_norm": 0.02749469131231308,
|
| 2947 |
+
"learning_rate": 2.9e-05,
|
| 2948 |
+
"loss": 0.0033,
|
| 2949 |
+
"step": 4200
|
| 2950 |
+
},
|
| 2951 |
+
{
|
| 2952 |
+
"epoch": 0.421,
|
| 2953 |
+
"grad_norm": 0.01697971485555172,
|
| 2954 |
+
"learning_rate": 2.895e-05,
|
| 2955 |
+
"loss": 0.0032,
|
| 2956 |
+
"step": 4210
|
| 2957 |
+
},
|
| 2958 |
+
{
|
| 2959 |
+
"epoch": 0.422,
|
| 2960 |
+
"grad_norm": 0.05183997377753258,
|
| 2961 |
+
"learning_rate": 2.8899999999999998e-05,
|
| 2962 |
+
"loss": 0.0029,
|
| 2963 |
+
"step": 4220
|
| 2964 |
+
},
|
| 2965 |
+
{
|
| 2966 |
+
"epoch": 0.423,
|
| 2967 |
+
"grad_norm": 0.030102815479040146,
|
| 2968 |
+
"learning_rate": 2.885e-05,
|
| 2969 |
+
"loss": 0.0028,
|
| 2970 |
+
"step": 4230
|
| 2971 |
+
},
|
| 2972 |
+
{
|
| 2973 |
+
"epoch": 0.424,
|
| 2974 |
+
"grad_norm": 0.027241216972470284,
|
| 2975 |
+
"learning_rate": 2.88e-05,
|
| 2976 |
+
"loss": 0.0033,
|
| 2977 |
+
"step": 4240
|
| 2978 |
+
},
|
| 2979 |
+
{
|
| 2980 |
+
"epoch": 0.425,
|
| 2981 |
+
"grad_norm": 0.01855759136378765,
|
| 2982 |
+
"learning_rate": 2.8749999999999997e-05,
|
| 2983 |
+
"loss": 0.0024,
|
| 2984 |
+
"step": 4250
|
| 2985 |
+
},
|
| 2986 |
+
{
|
| 2987 |
+
"epoch": 0.426,
|
| 2988 |
+
"grad_norm": 0.019300928339362144,
|
| 2989 |
+
"learning_rate": 2.87e-05,
|
| 2990 |
+
"loss": 0.0024,
|
| 2991 |
+
"step": 4260
|
| 2992 |
+
},
|
| 2993 |
+
{
|
| 2994 |
+
"epoch": 0.427,
|
| 2995 |
+
"grad_norm": 0.01639522798359394,
|
| 2996 |
+
"learning_rate": 2.865e-05,
|
| 2997 |
+
"loss": 0.0026,
|
| 2998 |
+
"step": 4270
|
| 2999 |
+
},
|
| 3000 |
+
{
|
| 3001 |
+
"epoch": 0.428,
|
| 3002 |
+
"grad_norm": 0.027084793895483017,
|
| 3003 |
+
"learning_rate": 2.86e-05,
|
| 3004 |
+
"loss": 0.0027,
|
| 3005 |
+
"step": 4280
|
| 3006 |
+
},
|
| 3007 |
+
{
|
| 3008 |
+
"epoch": 0.429,
|
| 3009 |
+
"grad_norm": 0.021206015720963478,
|
| 3010 |
+
"learning_rate": 2.855e-05,
|
| 3011 |
+
"loss": 0.0025,
|
| 3012 |
+
"step": 4290
|
| 3013 |
+
},
|
| 3014 |
+
{
|
| 3015 |
+
"epoch": 0.43,
|
| 3016 |
+
"grad_norm": 0.0655827671289444,
|
| 3017 |
+
"learning_rate": 2.8499999999999998e-05,
|
| 3018 |
+
"loss": 0.0027,
|
| 3019 |
+
"step": 4300
|
| 3020 |
+
},
|
| 3021 |
+
{
|
| 3022 |
+
"epoch": 0.431,
|
| 3023 |
+
"grad_norm": 0.03779730945825577,
|
| 3024 |
+
"learning_rate": 2.845e-05,
|
| 3025 |
+
"loss": 0.0024,
|
| 3026 |
+
"step": 4310
|
| 3027 |
+
},
|
| 3028 |
+
{
|
| 3029 |
+
"epoch": 0.432,
|
| 3030 |
+
"grad_norm": 0.045267749577760696,
|
| 3031 |
+
"learning_rate": 2.84e-05,
|
| 3032 |
+
"loss": 0.0031,
|
| 3033 |
+
"step": 4320
|
| 3034 |
+
},
|
| 3035 |
+
{
|
| 3036 |
+
"epoch": 0.433,
|
| 3037 |
+
"grad_norm": 0.017473919317126274,
|
| 3038 |
+
"learning_rate": 2.8349999999999998e-05,
|
| 3039 |
+
"loss": 0.0027,
|
| 3040 |
+
"step": 4330
|
| 3041 |
+
},
|
| 3042 |
+
{
|
| 3043 |
+
"epoch": 0.434,
|
| 3044 |
+
"grad_norm": 0.019513197243213654,
|
| 3045 |
+
"learning_rate": 2.83e-05,
|
| 3046 |
+
"loss": 0.0024,
|
| 3047 |
+
"step": 4340
|
| 3048 |
+
},
|
| 3049 |
+
{
|
| 3050 |
+
"epoch": 0.435,
|
| 3051 |
+
"grad_norm": 0.01616765186190605,
|
| 3052 |
+
"learning_rate": 2.825e-05,
|
| 3053 |
+
"loss": 0.0025,
|
| 3054 |
+
"step": 4350
|
| 3055 |
+
},
|
| 3056 |
+
{
|
| 3057 |
+
"epoch": 0.436,
|
| 3058 |
+
"grad_norm": 0.02270474284887314,
|
| 3059 |
+
"learning_rate": 2.8199999999999998e-05,
|
| 3060 |
+
"loss": 0.0024,
|
| 3061 |
+
"step": 4360
|
| 3062 |
+
},
|
| 3063 |
+
{
|
| 3064 |
+
"epoch": 0.437,
|
| 3065 |
+
"grad_norm": 0.02363002672791481,
|
| 3066 |
+
"learning_rate": 2.815e-05,
|
| 3067 |
+
"loss": 0.0025,
|
| 3068 |
+
"step": 4370
|
| 3069 |
+
},
|
| 3070 |
+
{
|
| 3071 |
+
"epoch": 0.438,
|
| 3072 |
+
"grad_norm": 0.023898936808109283,
|
| 3073 |
+
"learning_rate": 2.8100000000000005e-05,
|
| 3074 |
+
"loss": 0.0023,
|
| 3075 |
+
"step": 4380
|
| 3076 |
+
},
|
| 3077 |
+
{
|
| 3078 |
+
"epoch": 0.439,
|
| 3079 |
+
"grad_norm": 0.01270793005824089,
|
| 3080 |
+
"learning_rate": 2.8050000000000004e-05,
|
| 3081 |
+
"loss": 0.0022,
|
| 3082 |
+
"step": 4390
|
| 3083 |
+
},
|
| 3084 |
+
{
|
| 3085 |
+
"epoch": 0.44,
|
| 3086 |
+
"grad_norm": 0.03917006403207779,
|
| 3087 |
+
"learning_rate": 2.8000000000000003e-05,
|
| 3088 |
+
"loss": 0.0029,
|
| 3089 |
+
"step": 4400
|
| 3090 |
+
},
|
| 3091 |
+
{
|
| 3092 |
+
"epoch": 0.441,
|
| 3093 |
+
"grad_norm": 0.028284449130296707,
|
| 3094 |
+
"learning_rate": 2.7950000000000005e-05,
|
| 3095 |
+
"loss": 0.0023,
|
| 3096 |
+
"step": 4410
|
| 3097 |
+
},
|
| 3098 |
+
{
|
| 3099 |
+
"epoch": 0.442,
|
| 3100 |
+
"grad_norm": 0.017493903636932373,
|
| 3101 |
+
"learning_rate": 2.7900000000000004e-05,
|
| 3102 |
+
"loss": 0.0022,
|
| 3103 |
+
"step": 4420
|
| 3104 |
+
},
|
| 3105 |
+
{
|
| 3106 |
+
"epoch": 0.443,
|
| 3107 |
+
"grad_norm": 0.03160572797060013,
|
| 3108 |
+
"learning_rate": 2.7850000000000003e-05,
|
| 3109 |
+
"loss": 0.0025,
|
| 3110 |
+
"step": 4430
|
| 3111 |
+
},
|
| 3112 |
+
{
|
| 3113 |
+
"epoch": 0.444,
|
| 3114 |
+
"grad_norm": 0.022926049306988716,
|
| 3115 |
+
"learning_rate": 2.7800000000000005e-05,
|
| 3116 |
+
"loss": 0.0024,
|
| 3117 |
+
"step": 4440
|
| 3118 |
+
},
|
| 3119 |
+
{
|
| 3120 |
+
"epoch": 0.445,
|
| 3121 |
+
"grad_norm": 0.032902974635362625,
|
| 3122 |
+
"learning_rate": 2.7750000000000004e-05,
|
| 3123 |
+
"loss": 0.0025,
|
| 3124 |
+
"step": 4450
|
| 3125 |
+
},
|
| 3126 |
+
{
|
| 3127 |
+
"epoch": 0.446,
|
| 3128 |
+
"grad_norm": 0.017781972885131836,
|
| 3129 |
+
"learning_rate": 2.7700000000000002e-05,
|
| 3130 |
+
"loss": 0.0022,
|
| 3131 |
+
"step": 4460
|
| 3132 |
+
},
|
| 3133 |
+
{
|
| 3134 |
+
"epoch": 0.447,
|
| 3135 |
+
"grad_norm": 0.1209416389465332,
|
| 3136 |
+
"learning_rate": 2.7650000000000005e-05,
|
| 3137 |
+
"loss": 0.0028,
|
| 3138 |
+
"step": 4470
|
| 3139 |
+
},
|
| 3140 |
+
{
|
| 3141 |
+
"epoch": 0.448,
|
| 3142 |
+
"grad_norm": 0.03747338801622391,
|
| 3143 |
+
"learning_rate": 2.7600000000000003e-05,
|
| 3144 |
+
"loss": 0.0024,
|
| 3145 |
+
"step": 4480
|
| 3146 |
+
},
|
| 3147 |
+
{
|
| 3148 |
+
"epoch": 0.449,
|
| 3149 |
+
"grad_norm": 0.04210735112428665,
|
| 3150 |
+
"learning_rate": 2.7550000000000002e-05,
|
| 3151 |
+
"loss": 0.0024,
|
| 3152 |
+
"step": 4490
|
| 3153 |
+
},
|
| 3154 |
+
{
|
| 3155 |
+
"epoch": 0.45,
|
| 3156 |
+
"grad_norm": 0.02306324429810047,
|
| 3157 |
+
"learning_rate": 2.7500000000000004e-05,
|
| 3158 |
+
"loss": 0.0022,
|
| 3159 |
+
"step": 4500
|
| 3160 |
+
},
|
| 3161 |
+
{
|
| 3162 |
+
"epoch": 0.451,
|
| 3163 |
+
"grad_norm": 0.027622856199741364,
|
| 3164 |
+
"learning_rate": 2.7450000000000003e-05,
|
| 3165 |
+
"loss": 0.0026,
|
| 3166 |
+
"step": 4510
|
| 3167 |
+
},
|
| 3168 |
+
{
|
| 3169 |
+
"epoch": 0.452,
|
| 3170 |
+
"grad_norm": 0.014202162623405457,
|
| 3171 |
+
"learning_rate": 2.7400000000000002e-05,
|
| 3172 |
+
"loss": 0.0022,
|
| 3173 |
+
"step": 4520
|
| 3174 |
+
},
|
| 3175 |
+
{
|
| 3176 |
+
"epoch": 0.453,
|
| 3177 |
+
"grad_norm": 0.051466915756464005,
|
| 3178 |
+
"learning_rate": 2.7350000000000004e-05,
|
| 3179 |
+
"loss": 0.0031,
|
| 3180 |
+
"step": 4530
|
| 3181 |
+
},
|
| 3182 |
+
{
|
| 3183 |
+
"epoch": 0.454,
|
| 3184 |
+
"grad_norm": 0.052768610417842865,
|
| 3185 |
+
"learning_rate": 2.7300000000000003e-05,
|
| 3186 |
+
"loss": 0.0028,
|
| 3187 |
+
"step": 4540
|
| 3188 |
+
},
|
| 3189 |
+
{
|
| 3190 |
+
"epoch": 0.455,
|
| 3191 |
+
"grad_norm": 0.02291076071560383,
|
| 3192 |
+
"learning_rate": 2.725e-05,
|
| 3193 |
+
"loss": 0.0023,
|
| 3194 |
+
"step": 4550
|
| 3195 |
+
},
|
| 3196 |
+
{
|
| 3197 |
+
"epoch": 0.456,
|
| 3198 |
+
"grad_norm": 0.027942989021539688,
|
| 3199 |
+
"learning_rate": 2.7200000000000004e-05,
|
| 3200 |
+
"loss": 0.0023,
|
| 3201 |
+
"step": 4560
|
| 3202 |
+
},
|
| 3203 |
+
{
|
| 3204 |
+
"epoch": 0.457,
|
| 3205 |
+
"grad_norm": 0.01529670413583517,
|
| 3206 |
+
"learning_rate": 2.7150000000000003e-05,
|
| 3207 |
+
"loss": 0.0024,
|
| 3208 |
+
"step": 4570
|
| 3209 |
+
},
|
| 3210 |
+
{
|
| 3211 |
+
"epoch": 0.458,
|
| 3212 |
+
"grad_norm": 0.02945224940776825,
|
| 3213 |
+
"learning_rate": 2.7100000000000005e-05,
|
| 3214 |
+
"loss": 0.0024,
|
| 3215 |
+
"step": 4580
|
| 3216 |
+
},
|
| 3217 |
+
{
|
| 3218 |
+
"epoch": 0.459,
|
| 3219 |
+
"grad_norm": 0.027197351679205894,
|
| 3220 |
+
"learning_rate": 2.7050000000000004e-05,
|
| 3221 |
+
"loss": 0.0024,
|
| 3222 |
+
"step": 4590
|
| 3223 |
+
},
|
| 3224 |
+
{
|
| 3225 |
+
"epoch": 0.46,
|
| 3226 |
+
"grad_norm": 0.022022951394319534,
|
| 3227 |
+
"learning_rate": 2.7000000000000002e-05,
|
| 3228 |
+
"loss": 0.0023,
|
| 3229 |
+
"step": 4600
|
| 3230 |
+
},
|
| 3231 |
+
{
|
| 3232 |
+
"epoch": 0.461,
|
| 3233 |
+
"grad_norm": 0.019739823415875435,
|
| 3234 |
+
"learning_rate": 2.6950000000000005e-05,
|
| 3235 |
+
"loss": 0.0022,
|
| 3236 |
+
"step": 4610
|
| 3237 |
+
},
|
| 3238 |
+
{
|
| 3239 |
+
"epoch": 0.462,
|
| 3240 |
+
"grad_norm": 0.06794995814561844,
|
| 3241 |
+
"learning_rate": 2.6900000000000003e-05,
|
| 3242 |
+
"loss": 0.0027,
|
| 3243 |
+
"step": 4620
|
| 3244 |
+
},
|
| 3245 |
+
{
|
| 3246 |
+
"epoch": 0.463,
|
| 3247 |
+
"grad_norm": 0.049228962510824203,
|
| 3248 |
+
"learning_rate": 2.6850000000000002e-05,
|
| 3249 |
+
"loss": 0.0026,
|
| 3250 |
+
"step": 4630
|
| 3251 |
+
},
|
| 3252 |
+
{
|
| 3253 |
+
"epoch": 0.464,
|
| 3254 |
+
"grad_norm": 0.0241558700799942,
|
| 3255 |
+
"learning_rate": 2.6800000000000004e-05,
|
| 3256 |
+
"loss": 0.0021,
|
| 3257 |
+
"step": 4640
|
| 3258 |
+
},
|
| 3259 |
+
{
|
| 3260 |
+
"epoch": 0.465,
|
| 3261 |
+
"grad_norm": 0.024576248601078987,
|
| 3262 |
+
"learning_rate": 2.6750000000000003e-05,
|
| 3263 |
+
"loss": 0.0022,
|
| 3264 |
+
"step": 4650
|
| 3265 |
+
},
|
| 3266 |
+
{
|
| 3267 |
+
"epoch": 0.466,
|
| 3268 |
+
"grad_norm": 0.030337205156683922,
|
| 3269 |
+
"learning_rate": 2.6700000000000002e-05,
|
| 3270 |
+
"loss": 0.0024,
|
| 3271 |
+
"step": 4660
|
| 3272 |
+
},
|
| 3273 |
+
{
|
| 3274 |
+
"epoch": 0.467,
|
| 3275 |
+
"grad_norm": 0.015081087127327919,
|
| 3276 |
+
"learning_rate": 2.6650000000000004e-05,
|
| 3277 |
+
"loss": 0.002,
|
| 3278 |
+
"step": 4670
|
| 3279 |
+
},
|
| 3280 |
+
{
|
| 3281 |
+
"epoch": 0.468,
|
| 3282 |
+
"grad_norm": 0.026368912309408188,
|
| 3283 |
+
"learning_rate": 2.6600000000000003e-05,
|
| 3284 |
+
"loss": 0.0022,
|
| 3285 |
+
"step": 4680
|
| 3286 |
+
},
|
| 3287 |
+
{
|
| 3288 |
+
"epoch": 0.469,
|
| 3289 |
+
"grad_norm": 0.018447600305080414,
|
| 3290 |
+
"learning_rate": 2.655e-05,
|
| 3291 |
+
"loss": 0.0022,
|
| 3292 |
+
"step": 4690
|
| 3293 |
+
},
|
| 3294 |
+
{
|
| 3295 |
+
"epoch": 0.47,
|
| 3296 |
+
"grad_norm": 0.018314722925424576,
|
| 3297 |
+
"learning_rate": 2.6500000000000004e-05,
|
| 3298 |
+
"loss": 0.0019,
|
| 3299 |
+
"step": 4700
|
| 3300 |
+
},
|
| 3301 |
+
{
|
| 3302 |
+
"epoch": 0.471,
|
| 3303 |
+
"grad_norm": 0.02361704409122467,
|
| 3304 |
+
"learning_rate": 2.6450000000000003e-05,
|
| 3305 |
+
"loss": 0.0023,
|
| 3306 |
+
"step": 4710
|
| 3307 |
+
},
|
| 3308 |
+
{
|
| 3309 |
+
"epoch": 0.472,
|
| 3310 |
+
"grad_norm": 0.02032247930765152,
|
| 3311 |
+
"learning_rate": 2.64e-05,
|
| 3312 |
+
"loss": 0.002,
|
| 3313 |
+
"step": 4720
|
| 3314 |
+
},
|
| 3315 |
+
{
|
| 3316 |
+
"epoch": 0.473,
|
| 3317 |
+
"grad_norm": 0.017889728769659996,
|
| 3318 |
+
"learning_rate": 2.6350000000000004e-05,
|
| 3319 |
+
"loss": 0.0019,
|
| 3320 |
+
"step": 4730
|
| 3321 |
+
},
|
| 3322 |
+
{
|
| 3323 |
+
"epoch": 0.474,
|
| 3324 |
+
"grad_norm": 0.01962173730134964,
|
| 3325 |
+
"learning_rate": 2.6300000000000002e-05,
|
| 3326 |
+
"loss": 0.0019,
|
| 3327 |
+
"step": 4740
|
| 3328 |
+
},
|
| 3329 |
+
{
|
| 3330 |
+
"epoch": 0.475,
|
| 3331 |
+
"grad_norm": 0.015778113156557083,
|
| 3332 |
+
"learning_rate": 2.625e-05,
|
| 3333 |
+
"loss": 0.0021,
|
| 3334 |
+
"step": 4750
|
| 3335 |
+
},
|
| 3336 |
+
{
|
| 3337 |
+
"epoch": 0.476,
|
| 3338 |
+
"grad_norm": 0.01894952729344368,
|
| 3339 |
+
"learning_rate": 2.6200000000000003e-05,
|
| 3340 |
+
"loss": 0.0023,
|
| 3341 |
+
"step": 4760
|
| 3342 |
+
},
|
| 3343 |
+
{
|
| 3344 |
+
"epoch": 0.477,
|
| 3345 |
+
"grad_norm": 0.0239462498575449,
|
| 3346 |
+
"learning_rate": 2.6150000000000002e-05,
|
| 3347 |
+
"loss": 0.002,
|
| 3348 |
+
"step": 4770
|
| 3349 |
+
},
|
| 3350 |
+
{
|
| 3351 |
+
"epoch": 0.478,
|
| 3352 |
+
"grad_norm": 0.025406278669834137,
|
| 3353 |
+
"learning_rate": 2.61e-05,
|
| 3354 |
+
"loss": 0.0021,
|
| 3355 |
+
"step": 4780
|
| 3356 |
+
},
|
| 3357 |
+
{
|
| 3358 |
+
"epoch": 0.479,
|
| 3359 |
+
"grad_norm": 0.01813661865890026,
|
| 3360 |
+
"learning_rate": 2.6050000000000003e-05,
|
| 3361 |
+
"loss": 0.0023,
|
| 3362 |
+
"step": 4790
|
| 3363 |
+
},
|
| 3364 |
+
{
|
| 3365 |
+
"epoch": 0.48,
|
| 3366 |
+
"grad_norm": 0.01979188807308674,
|
| 3367 |
+
"learning_rate": 2.6000000000000002e-05,
|
| 3368 |
+
"loss": 0.0019,
|
| 3369 |
+
"step": 4800
|
| 3370 |
+
},
|
| 3371 |
+
{
|
| 3372 |
+
"epoch": 0.481,
|
| 3373 |
+
"grad_norm": 0.02219184674322605,
|
| 3374 |
+
"learning_rate": 2.595e-05,
|
| 3375 |
+
"loss": 0.002,
|
| 3376 |
+
"step": 4810
|
| 3377 |
+
},
|
| 3378 |
+
{
|
| 3379 |
+
"epoch": 0.482,
|
| 3380 |
+
"grad_norm": 0.012867514044046402,
|
| 3381 |
+
"learning_rate": 2.5900000000000003e-05,
|
| 3382 |
+
"loss": 0.0017,
|
| 3383 |
+
"step": 4820
|
| 3384 |
+
},
|
| 3385 |
+
{
|
| 3386 |
+
"epoch": 0.483,
|
| 3387 |
+
"grad_norm": 0.014178570359945297,
|
| 3388 |
+
"learning_rate": 2.585e-05,
|
| 3389 |
+
"loss": 0.0023,
|
| 3390 |
+
"step": 4830
|
| 3391 |
+
},
|
| 3392 |
+
{
|
| 3393 |
+
"epoch": 0.484,
|
| 3394 |
+
"grad_norm": 0.013582895509898663,
|
| 3395 |
+
"learning_rate": 2.58e-05,
|
| 3396 |
+
"loss": 0.002,
|
| 3397 |
+
"step": 4840
|
| 3398 |
+
},
|
| 3399 |
+
{
|
| 3400 |
+
"epoch": 0.485,
|
| 3401 |
+
"grad_norm": 0.02718137763440609,
|
| 3402 |
+
"learning_rate": 2.5750000000000002e-05,
|
| 3403 |
+
"loss": 0.0019,
|
| 3404 |
+
"step": 4850
|
| 3405 |
+
},
|
| 3406 |
+
{
|
| 3407 |
+
"epoch": 0.486,
|
| 3408 |
+
"grad_norm": 0.016559738665819168,
|
| 3409 |
+
"learning_rate": 2.57e-05,
|
| 3410 |
+
"loss": 0.0019,
|
| 3411 |
+
"step": 4860
|
| 3412 |
+
},
|
| 3413 |
+
{
|
| 3414 |
+
"epoch": 0.487,
|
| 3415 |
+
"grad_norm": 0.01447515282779932,
|
| 3416 |
+
"learning_rate": 2.5650000000000003e-05,
|
| 3417 |
+
"loss": 0.0017,
|
| 3418 |
+
"step": 4870
|
| 3419 |
+
},
|
| 3420 |
+
{
|
| 3421 |
+
"epoch": 0.488,
|
| 3422 |
+
"grad_norm": 0.010251802392303944,
|
| 3423 |
+
"learning_rate": 2.5600000000000002e-05,
|
| 3424 |
+
"loss": 0.0017,
|
| 3425 |
+
"step": 4880
|
| 3426 |
+
},
|
| 3427 |
+
{
|
| 3428 |
+
"epoch": 0.489,
|
| 3429 |
+
"grad_norm": 0.021138856187462807,
|
| 3430 |
+
"learning_rate": 2.555e-05,
|
| 3431 |
+
"loss": 0.0023,
|
| 3432 |
+
"step": 4890
|
| 3433 |
+
},
|
| 3434 |
+
{
|
| 3435 |
+
"epoch": 0.49,
|
| 3436 |
+
"grad_norm": 0.026664163917303085,
|
| 3437 |
+
"learning_rate": 2.5500000000000003e-05,
|
| 3438 |
+
"loss": 0.0021,
|
| 3439 |
+
"step": 4900
|
| 3440 |
+
},
|
| 3441 |
+
{
|
| 3442 |
+
"epoch": 0.491,
|
| 3443 |
+
"grad_norm": 0.012794408947229385,
|
| 3444 |
+
"learning_rate": 2.5450000000000002e-05,
|
| 3445 |
+
"loss": 0.0017,
|
| 3446 |
+
"step": 4910
|
| 3447 |
+
},
|
| 3448 |
+
{
|
| 3449 |
+
"epoch": 0.492,
|
| 3450 |
+
"grad_norm": 0.013725240714848042,
|
| 3451 |
+
"learning_rate": 2.54e-05,
|
| 3452 |
+
"loss": 0.0018,
|
| 3453 |
+
"step": 4920
|
| 3454 |
+
},
|
| 3455 |
+
{
|
| 3456 |
+
"epoch": 0.493,
|
| 3457 |
+
"grad_norm": 0.01432815007865429,
|
| 3458 |
+
"learning_rate": 2.5350000000000003e-05,
|
| 3459 |
+
"loss": 0.0017,
|
| 3460 |
+
"step": 4930
|
| 3461 |
+
},
|
| 3462 |
+
{
|
| 3463 |
+
"epoch": 0.494,
|
| 3464 |
+
"grad_norm": 0.014761424623429775,
|
| 3465 |
+
"learning_rate": 2.5300000000000002e-05,
|
| 3466 |
+
"loss": 0.0019,
|
| 3467 |
+
"step": 4940
|
| 3468 |
+
},
|
| 3469 |
+
{
|
| 3470 |
+
"epoch": 0.495,
|
| 3471 |
+
"grad_norm": 0.06742983311414719,
|
| 3472 |
+
"learning_rate": 2.525e-05,
|
| 3473 |
+
"loss": 0.0025,
|
| 3474 |
+
"step": 4950
|
| 3475 |
+
},
|
| 3476 |
+
{
|
| 3477 |
+
"epoch": 0.496,
|
| 3478 |
+
"grad_norm": 0.024261673912405968,
|
| 3479 |
+
"learning_rate": 2.5200000000000003e-05,
|
| 3480 |
+
"loss": 0.0021,
|
| 3481 |
+
"step": 4960
|
| 3482 |
+
},
|
| 3483 |
+
{
|
| 3484 |
+
"epoch": 0.497,
|
| 3485 |
+
"grad_norm": 0.02116272784769535,
|
| 3486 |
+
"learning_rate": 2.515e-05,
|
| 3487 |
+
"loss": 0.002,
|
| 3488 |
+
"step": 4970
|
| 3489 |
+
},
|
| 3490 |
+
{
|
| 3491 |
+
"epoch": 0.498,
|
| 3492 |
+
"grad_norm": 0.014996570535004139,
|
| 3493 |
+
"learning_rate": 2.51e-05,
|
| 3494 |
+
"loss": 0.0017,
|
| 3495 |
+
"step": 4980
|
| 3496 |
+
},
|
| 3497 |
+
{
|
| 3498 |
+
"epoch": 0.499,
|
| 3499 |
+
"grad_norm": 0.014970551244914532,
|
| 3500 |
+
"learning_rate": 2.5050000000000002e-05,
|
| 3501 |
+
"loss": 0.002,
|
| 3502 |
+
"step": 4990
|
| 3503 |
+
},
|
| 3504 |
+
{
|
| 3505 |
+
"epoch": 0.5,
|
| 3506 |
+
"grad_norm": 0.01756688393652439,
|
| 3507 |
+
"learning_rate": 2.5e-05,
|
| 3508 |
+
"loss": 0.002,
|
| 3509 |
+
"step": 5000
|
| 3510 |
}
|
| 3511 |
],
|
| 3512 |
"logging_steps": 10,
|
|
|
|
| 3526 |
"attributes": {}
|
| 3527 |
}
|
| 3528 |
},
|
| 3529 |
+
"total_flos": 1.8481101668352e+17,
|
| 3530 |
"train_batch_size": 1,
|
| 3531 |
"trial_name": null,
|
| 3532 |
"trial_params": null
|