Upload multi_student_2/checkpoint-1800/trainer_state.json with huggingface_hub
Browse files
multi_student_2/checkpoint-1800/trainer_state.json
ADDED
|
@@ -0,0 +1,1294 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"best_global_step": null,
|
| 3 |
+
"best_metric": null,
|
| 4 |
+
"best_model_checkpoint": null,
|
| 5 |
+
"epoch": 1.4276080920269734,
|
| 6 |
+
"eval_steps": 500,
|
| 7 |
+
"global_step": 1800,
|
| 8 |
+
"is_hyper_param_search": false,
|
| 9 |
+
"is_local_process_zero": true,
|
| 10 |
+
"is_world_process_zero": true,
|
| 11 |
+
"log_history": [
|
| 12 |
+
{
|
| 13 |
+
"epoch": 0.007933359777865925,
|
| 14 |
+
"grad_norm": 19.14600299430099,
|
| 15 |
+
"learning_rate": 8.9197224975223e-07,
|
| 16 |
+
"loss": 8.2489,
|
| 17 |
+
"step": 10
|
| 18 |
+
},
|
| 19 |
+
{
|
| 20 |
+
"epoch": 0.01586671955573185,
|
| 21 |
+
"grad_norm": 18.01915458135243,
|
| 22 |
+
"learning_rate": 1.8830525272547078e-06,
|
| 23 |
+
"loss": 8.1177,
|
| 24 |
+
"step": 20
|
| 25 |
+
},
|
| 26 |
+
{
|
| 27 |
+
"epoch": 0.02380007933359778,
|
| 28 |
+
"grad_norm": 9.33822311099433,
|
| 29 |
+
"learning_rate": 2.8741328047571853e-06,
|
| 30 |
+
"loss": 7.4735,
|
| 31 |
+
"step": 30
|
| 32 |
+
},
|
| 33 |
+
{
|
| 34 |
+
"epoch": 0.0317334391114637,
|
| 35 |
+
"grad_norm": 4.906820841029322,
|
| 36 |
+
"learning_rate": 3.8652130822596635e-06,
|
| 37 |
+
"loss": 6.7427,
|
| 38 |
+
"step": 40
|
| 39 |
+
},
|
| 40 |
+
{
|
| 41 |
+
"epoch": 0.039666798889329634,
|
| 42 |
+
"grad_norm": 3.0516622673060785,
|
| 43 |
+
"learning_rate": 4.8562933597621405e-06,
|
| 44 |
+
"loss": 6.2707,
|
| 45 |
+
"step": 50
|
| 46 |
+
},
|
| 47 |
+
{
|
| 48 |
+
"epoch": 0.04760015866719556,
|
| 49 |
+
"grad_norm": 3.0497788345357493,
|
| 50 |
+
"learning_rate": 5.847373637264618e-06,
|
| 51 |
+
"loss": 5.9319,
|
| 52 |
+
"step": 60
|
| 53 |
+
},
|
| 54 |
+
{
|
| 55 |
+
"epoch": 0.055533518445061485,
|
| 56 |
+
"grad_norm": 3.678614138586657,
|
| 57 |
+
"learning_rate": 6.838453914767097e-06,
|
| 58 |
+
"loss": 5.5292,
|
| 59 |
+
"step": 70
|
| 60 |
+
},
|
| 61 |
+
{
|
| 62 |
+
"epoch": 0.0634668782229274,
|
| 63 |
+
"grad_norm": 4.261516409919679,
|
| 64 |
+
"learning_rate": 7.829534192269576e-06,
|
| 65 |
+
"loss": 4.9099,
|
| 66 |
+
"step": 80
|
| 67 |
+
},
|
| 68 |
+
{
|
| 69 |
+
"epoch": 0.07140023800079334,
|
| 70 |
+
"grad_norm": 4.6638020698067875,
|
| 71 |
+
"learning_rate": 8.820614469772052e-06,
|
| 72 |
+
"loss": 4.2313,
|
| 73 |
+
"step": 90
|
| 74 |
+
},
|
| 75 |
+
{
|
| 76 |
+
"epoch": 0.07933359777865927,
|
| 77 |
+
"grad_norm": 4.5951745742169,
|
| 78 |
+
"learning_rate": 9.81169474727453e-06,
|
| 79 |
+
"loss": 3.663,
|
| 80 |
+
"step": 100
|
| 81 |
+
},
|
| 82 |
+
{
|
| 83 |
+
"epoch": 0.0872669575565252,
|
| 84 |
+
"grad_norm": 3.9132985800446947,
|
| 85 |
+
"learning_rate": 1.0802775024777007e-05,
|
| 86 |
+
"loss": 3.2806,
|
| 87 |
+
"step": 110
|
| 88 |
+
},
|
| 89 |
+
{
|
| 90 |
+
"epoch": 0.09520031733439112,
|
| 91 |
+
"grad_norm": 3.864223439251188,
|
| 92 |
+
"learning_rate": 1.1793855302279485e-05,
|
| 93 |
+
"loss": 3.0464,
|
| 94 |
+
"step": 120
|
| 95 |
+
},
|
| 96 |
+
{
|
| 97 |
+
"epoch": 0.10313367711225704,
|
| 98 |
+
"grad_norm": 3.5022825994548232,
|
| 99 |
+
"learning_rate": 1.2784935579781963e-05,
|
| 100 |
+
"loss": 2.8886,
|
| 101 |
+
"step": 130
|
| 102 |
+
},
|
| 103 |
+
{
|
| 104 |
+
"epoch": 0.11106703689012297,
|
| 105 |
+
"grad_norm": 5.1510905388003,
|
| 106 |
+
"learning_rate": 1.377601585728444e-05,
|
| 107 |
+
"loss": 2.7668,
|
| 108 |
+
"step": 140
|
| 109 |
+
},
|
| 110 |
+
{
|
| 111 |
+
"epoch": 0.1190003966679889,
|
| 112 |
+
"grad_norm": 7.21771904799698,
|
| 113 |
+
"learning_rate": 1.4767096134786917e-05,
|
| 114 |
+
"loss": 2.6692,
|
| 115 |
+
"step": 150
|
| 116 |
+
},
|
| 117 |
+
{
|
| 118 |
+
"epoch": 0.1269337564458548,
|
| 119 |
+
"grad_norm": 9.497867401163038,
|
| 120 |
+
"learning_rate": 1.5758176412289396e-05,
|
| 121 |
+
"loss": 2.5934,
|
| 122 |
+
"step": 160
|
| 123 |
+
},
|
| 124 |
+
{
|
| 125 |
+
"epoch": 0.13486711622372075,
|
| 126 |
+
"grad_norm": 8.479404348386907,
|
| 127 |
+
"learning_rate": 1.6749256689791874e-05,
|
| 128 |
+
"loss": 2.5471,
|
| 129 |
+
"step": 170
|
| 130 |
+
},
|
| 131 |
+
{
|
| 132 |
+
"epoch": 0.14280047600158668,
|
| 133 |
+
"grad_norm": 13.33424723235721,
|
| 134 |
+
"learning_rate": 1.7740336967294352e-05,
|
| 135 |
+
"loss": 2.4827,
|
| 136 |
+
"step": 180
|
| 137 |
+
},
|
| 138 |
+
{
|
| 139 |
+
"epoch": 0.1507338357794526,
|
| 140 |
+
"grad_norm": 9.736118445504315,
|
| 141 |
+
"learning_rate": 1.873141724479683e-05,
|
| 142 |
+
"loss": 2.4223,
|
| 143 |
+
"step": 190
|
| 144 |
+
},
|
| 145 |
+
{
|
| 146 |
+
"epoch": 0.15866719555731854,
|
| 147 |
+
"grad_norm": 12.548385792238333,
|
| 148 |
+
"learning_rate": 1.9722497522299307e-05,
|
| 149 |
+
"loss": 2.3752,
|
| 150 |
+
"step": 200
|
| 151 |
+
},
|
| 152 |
+
{
|
| 153 |
+
"epoch": 0.16660055533518445,
|
| 154 |
+
"grad_norm": 15.910150113440148,
|
| 155 |
+
"learning_rate": 2.0713577799801785e-05,
|
| 156 |
+
"loss": 2.3339,
|
| 157 |
+
"step": 210
|
| 158 |
+
},
|
| 159 |
+
{
|
| 160 |
+
"epoch": 0.1745339151130504,
|
| 161 |
+
"grad_norm": 12.024121798784853,
|
| 162 |
+
"learning_rate": 2.1704658077304263e-05,
|
| 163 |
+
"loss": 2.2732,
|
| 164 |
+
"step": 220
|
| 165 |
+
},
|
| 166 |
+
{
|
| 167 |
+
"epoch": 0.1824672748909163,
|
| 168 |
+
"grad_norm": 18.143793557621525,
|
| 169 |
+
"learning_rate": 2.269573835480674e-05,
|
| 170 |
+
"loss": 2.2154,
|
| 171 |
+
"step": 230
|
| 172 |
+
},
|
| 173 |
+
{
|
| 174 |
+
"epoch": 0.19040063466878224,
|
| 175 |
+
"grad_norm": 11.930035885996817,
|
| 176 |
+
"learning_rate": 2.368681863230922e-05,
|
| 177 |
+
"loss": 2.1602,
|
| 178 |
+
"step": 240
|
| 179 |
+
},
|
| 180 |
+
{
|
| 181 |
+
"epoch": 0.19833399444664815,
|
| 182 |
+
"grad_norm": 12.424586861990159,
|
| 183 |
+
"learning_rate": 2.4677898909811696e-05,
|
| 184 |
+
"loss": 2.0887,
|
| 185 |
+
"step": 250
|
| 186 |
+
},
|
| 187 |
+
{
|
| 188 |
+
"epoch": 0.2062673542245141,
|
| 189 |
+
"grad_norm": 11.063682700121975,
|
| 190 |
+
"learning_rate": 2.5668979187314174e-05,
|
| 191 |
+
"loss": 1.9843,
|
| 192 |
+
"step": 260
|
| 193 |
+
},
|
| 194 |
+
{
|
| 195 |
+
"epoch": 0.21420071400238,
|
| 196 |
+
"grad_norm": 9.80005982816964,
|
| 197 |
+
"learning_rate": 2.6660059464816652e-05,
|
| 198 |
+
"loss": 1.871,
|
| 199 |
+
"step": 270
|
| 200 |
+
},
|
| 201 |
+
{
|
| 202 |
+
"epoch": 0.22213407378024594,
|
| 203 |
+
"grad_norm": 6.8486046511969985,
|
| 204 |
+
"learning_rate": 2.7651139742319126e-05,
|
| 205 |
+
"loss": 1.7654,
|
| 206 |
+
"step": 280
|
| 207 |
+
},
|
| 208 |
+
{
|
| 209 |
+
"epoch": 0.23006743355811185,
|
| 210 |
+
"grad_norm": 4.650133260417432,
|
| 211 |
+
"learning_rate": 2.8642220019821607e-05,
|
| 212 |
+
"loss": 1.6631,
|
| 213 |
+
"step": 290
|
| 214 |
+
},
|
| 215 |
+
{
|
| 216 |
+
"epoch": 0.2380007933359778,
|
| 217 |
+
"grad_norm": 8.484902376448611,
|
| 218 |
+
"learning_rate": 2.9633300297324085e-05,
|
| 219 |
+
"loss": 1.587,
|
| 220 |
+
"step": 300
|
| 221 |
+
},
|
| 222 |
+
{
|
| 223 |
+
"epoch": 0.2459341531138437,
|
| 224 |
+
"grad_norm": 8.266296207037957,
|
| 225 |
+
"learning_rate": 3.062438057482656e-05,
|
| 226 |
+
"loss": 1.5209,
|
| 227 |
+
"step": 310
|
| 228 |
+
},
|
| 229 |
+
{
|
| 230 |
+
"epoch": 0.2538675128917096,
|
| 231 |
+
"grad_norm": 6.035258103391185,
|
| 232 |
+
"learning_rate": 3.161546085232904e-05,
|
| 233 |
+
"loss": 1.4691,
|
| 234 |
+
"step": 320
|
| 235 |
+
},
|
| 236 |
+
{
|
| 237 |
+
"epoch": 0.2618008726695756,
|
| 238 |
+
"grad_norm": 6.090335574638085,
|
| 239 |
+
"learning_rate": 3.260654112983152e-05,
|
| 240 |
+
"loss": 1.4213,
|
| 241 |
+
"step": 330
|
| 242 |
+
},
|
| 243 |
+
{
|
| 244 |
+
"epoch": 0.2697342324474415,
|
| 245 |
+
"grad_norm": 7.504625709152101,
|
| 246 |
+
"learning_rate": 3.359762140733399e-05,
|
| 247 |
+
"loss": 1.3809,
|
| 248 |
+
"step": 340
|
| 249 |
+
},
|
| 250 |
+
{
|
| 251 |
+
"epoch": 0.2776675922253074,
|
| 252 |
+
"grad_norm": 5.216278847793629,
|
| 253 |
+
"learning_rate": 3.458870168483647e-05,
|
| 254 |
+
"loss": 1.3304,
|
| 255 |
+
"step": 350
|
| 256 |
+
},
|
| 257 |
+
{
|
| 258 |
+
"epoch": 0.28560095200317337,
|
| 259 |
+
"grad_norm": 5.001664767324889,
|
| 260 |
+
"learning_rate": 3.5579781962338955e-05,
|
| 261 |
+
"loss": 1.2886,
|
| 262 |
+
"step": 360
|
| 263 |
+
},
|
| 264 |
+
{
|
| 265 |
+
"epoch": 0.2935343117810393,
|
| 266 |
+
"grad_norm": 4.035122456050597,
|
| 267 |
+
"learning_rate": 3.657086223984143e-05,
|
| 268 |
+
"loss": 1.2441,
|
| 269 |
+
"step": 370
|
| 270 |
+
},
|
| 271 |
+
{
|
| 272 |
+
"epoch": 0.3014676715589052,
|
| 273 |
+
"grad_norm": 4.3131477681191015,
|
| 274 |
+
"learning_rate": 3.7561942517343904e-05,
|
| 275 |
+
"loss": 1.1973,
|
| 276 |
+
"step": 380
|
| 277 |
+
},
|
| 278 |
+
{
|
| 279 |
+
"epoch": 0.3094010313367711,
|
| 280 |
+
"grad_norm": 8.188203831814278,
|
| 281 |
+
"learning_rate": 3.855302279484638e-05,
|
| 282 |
+
"loss": 1.175,
|
| 283 |
+
"step": 390
|
| 284 |
+
},
|
| 285 |
+
{
|
| 286 |
+
"epoch": 0.31733439111463707,
|
| 287 |
+
"grad_norm": 4.755933403654607,
|
| 288 |
+
"learning_rate": 3.9544103072348866e-05,
|
| 289 |
+
"loss": 1.1342,
|
| 290 |
+
"step": 400
|
| 291 |
+
},
|
| 292 |
+
{
|
| 293 |
+
"epoch": 0.325267750892503,
|
| 294 |
+
"grad_norm": 6.57796715979494,
|
| 295 |
+
"learning_rate": 4.0535183349851344e-05,
|
| 296 |
+
"loss": 1.0971,
|
| 297 |
+
"step": 410
|
| 298 |
+
},
|
| 299 |
+
{
|
| 300 |
+
"epoch": 0.3332011106703689,
|
| 301 |
+
"grad_norm": 4.216167153902993,
|
| 302 |
+
"learning_rate": 4.1526263627353815e-05,
|
| 303 |
+
"loss": 1.0513,
|
| 304 |
+
"step": 420
|
| 305 |
+
},
|
| 306 |
+
{
|
| 307 |
+
"epoch": 0.3411344704482348,
|
| 308 |
+
"grad_norm": 8.418073662598946,
|
| 309 |
+
"learning_rate": 4.251734390485629e-05,
|
| 310 |
+
"loss": 1.0278,
|
| 311 |
+
"step": 430
|
| 312 |
+
},
|
| 313 |
+
{
|
| 314 |
+
"epoch": 0.3490678302261008,
|
| 315 |
+
"grad_norm": 4.293832971560352,
|
| 316 |
+
"learning_rate": 4.350842418235878e-05,
|
| 317 |
+
"loss": 0.9989,
|
| 318 |
+
"step": 440
|
| 319 |
+
},
|
| 320 |
+
{
|
| 321 |
+
"epoch": 0.3570011900039667,
|
| 322 |
+
"grad_norm": 4.738542701520305,
|
| 323 |
+
"learning_rate": 4.4499504459861255e-05,
|
| 324 |
+
"loss": 0.9646,
|
| 325 |
+
"step": 450
|
| 326 |
+
},
|
| 327 |
+
{
|
| 328 |
+
"epoch": 0.3649345497818326,
|
| 329 |
+
"grad_norm": 3.836235781620353,
|
| 330 |
+
"learning_rate": 4.5490584737363726e-05,
|
| 331 |
+
"loss": 0.9282,
|
| 332 |
+
"step": 460
|
| 333 |
+
},
|
| 334 |
+
{
|
| 335 |
+
"epoch": 0.3728679095596985,
|
| 336 |
+
"grad_norm": 4.385628595929679,
|
| 337 |
+
"learning_rate": 4.6481665014866204e-05,
|
| 338 |
+
"loss": 0.9069,
|
| 339 |
+
"step": 470
|
| 340 |
+
},
|
| 341 |
+
{
|
| 342 |
+
"epoch": 0.3808012693375645,
|
| 343 |
+
"grad_norm": 4.795190260431852,
|
| 344 |
+
"learning_rate": 4.747274529236869e-05,
|
| 345 |
+
"loss": 0.8767,
|
| 346 |
+
"step": 480
|
| 347 |
+
},
|
| 348 |
+
{
|
| 349 |
+
"epoch": 0.3887346291154304,
|
| 350 |
+
"grad_norm": 5.15771361003216,
|
| 351 |
+
"learning_rate": 4.846382556987116e-05,
|
| 352 |
+
"loss": 0.848,
|
| 353 |
+
"step": 490
|
| 354 |
+
},
|
| 355 |
+
{
|
| 356 |
+
"epoch": 0.3966679888932963,
|
| 357 |
+
"grad_norm": 5.00866147389186,
|
| 358 |
+
"learning_rate": 4.945490584737364e-05,
|
| 359 |
+
"loss": 0.832,
|
| 360 |
+
"step": 500
|
| 361 |
+
},
|
| 362 |
+
{
|
| 363 |
+
"epoch": 0.40460134867116226,
|
| 364 |
+
"grad_norm": 3.3749478011742617,
|
| 365 |
+
"learning_rate": 5.044598612487612e-05,
|
| 366 |
+
"loss": 0.801,
|
| 367 |
+
"step": 510
|
| 368 |
+
},
|
| 369 |
+
{
|
| 370 |
+
"epoch": 0.4125347084490282,
|
| 371 |
+
"grad_norm": 5.648100107599949,
|
| 372 |
+
"learning_rate": 5.14370664023786e-05,
|
| 373 |
+
"loss": 0.7907,
|
| 374 |
+
"step": 520
|
| 375 |
+
},
|
| 376 |
+
{
|
| 377 |
+
"epoch": 0.4204680682268941,
|
| 378 |
+
"grad_norm": 5.180183508710987,
|
| 379 |
+
"learning_rate": 5.242814667988107e-05,
|
| 380 |
+
"loss": 0.7641,
|
| 381 |
+
"step": 530
|
| 382 |
+
},
|
| 383 |
+
{
|
| 384 |
+
"epoch": 0.42840142800476,
|
| 385 |
+
"grad_norm": 4.795683231567134,
|
| 386 |
+
"learning_rate": 5.341922695738355e-05,
|
| 387 |
+
"loss": 0.7428,
|
| 388 |
+
"step": 540
|
| 389 |
+
},
|
| 390 |
+
{
|
| 391 |
+
"epoch": 0.43633478778262597,
|
| 392 |
+
"grad_norm": 4.474423469358904,
|
| 393 |
+
"learning_rate": 5.4410307234886027e-05,
|
| 394 |
+
"loss": 0.7178,
|
| 395 |
+
"step": 550
|
| 396 |
+
},
|
| 397 |
+
{
|
| 398 |
+
"epoch": 0.4442681475604919,
|
| 399 |
+
"grad_norm": 5.965818045702901,
|
| 400 |
+
"learning_rate": 5.5401387512388504e-05,
|
| 401 |
+
"loss": 0.7008,
|
| 402 |
+
"step": 560
|
| 403 |
+
},
|
| 404 |
+
{
|
| 405 |
+
"epoch": 0.4522015073383578,
|
| 406 |
+
"grad_norm": 4.4181008640757256,
|
| 407 |
+
"learning_rate": 5.639246778989098e-05,
|
| 408 |
+
"loss": 0.678,
|
| 409 |
+
"step": 570
|
| 410 |
+
},
|
| 411 |
+
{
|
| 412 |
+
"epoch": 0.4601348671162237,
|
| 413 |
+
"grad_norm": 4.925348313364693,
|
| 414 |
+
"learning_rate": 5.738354806739347e-05,
|
| 415 |
+
"loss": 0.6516,
|
| 416 |
+
"step": 580
|
| 417 |
+
},
|
| 418 |
+
{
|
| 419 |
+
"epoch": 0.46806822689408967,
|
| 420 |
+
"grad_norm": 4.995771437156914,
|
| 421 |
+
"learning_rate": 5.8374628344895944e-05,
|
| 422 |
+
"loss": 0.6261,
|
| 423 |
+
"step": 590
|
| 424 |
+
},
|
| 425 |
+
{
|
| 426 |
+
"epoch": 0.4760015866719556,
|
| 427 |
+
"grad_norm": 4.3400967454545185,
|
| 428 |
+
"learning_rate": 5.936570862239842e-05,
|
| 429 |
+
"loss": 0.6099,
|
| 430 |
+
"step": 600
|
| 431 |
+
},
|
| 432 |
+
{
|
| 433 |
+
"epoch": 0.4839349464498215,
|
| 434 |
+
"grad_norm": 3.8437851403940013,
|
| 435 |
+
"learning_rate": 6.035678889990089e-05,
|
| 436 |
+
"loss": 0.5905,
|
| 437 |
+
"step": 610
|
| 438 |
+
},
|
| 439 |
+
{
|
| 440 |
+
"epoch": 0.4918683062276874,
|
| 441 |
+
"grad_norm": 4.673291642824443,
|
| 442 |
+
"learning_rate": 6.134786917740337e-05,
|
| 443 |
+
"loss": 0.5739,
|
| 444 |
+
"step": 620
|
| 445 |
+
},
|
| 446 |
+
{
|
| 447 |
+
"epoch": 0.49980166600555337,
|
| 448 |
+
"grad_norm": 4.691220121981611,
|
| 449 |
+
"learning_rate": 6.233894945490585e-05,
|
| 450 |
+
"loss": 0.5543,
|
| 451 |
+
"step": 630
|
| 452 |
+
},
|
| 453 |
+
{
|
| 454 |
+
"epoch": 0.5077350257834192,
|
| 455 |
+
"grad_norm": 4.609768045302665,
|
| 456 |
+
"learning_rate": 6.333002973240833e-05,
|
| 457 |
+
"loss": 0.5265,
|
| 458 |
+
"step": 640
|
| 459 |
+
},
|
| 460 |
+
{
|
| 461 |
+
"epoch": 0.5156683855612852,
|
| 462 |
+
"grad_norm": 3.2621270260578967,
|
| 463 |
+
"learning_rate": 6.43211100099108e-05,
|
| 464 |
+
"loss": 0.5175,
|
| 465 |
+
"step": 650
|
| 466 |
+
},
|
| 467 |
+
{
|
| 468 |
+
"epoch": 0.5236017453391512,
|
| 469 |
+
"grad_norm": 3.5703471859456806,
|
| 470 |
+
"learning_rate": 6.531219028741328e-05,
|
| 471 |
+
"loss": 0.4972,
|
| 472 |
+
"step": 660
|
| 473 |
+
},
|
| 474 |
+
{
|
| 475 |
+
"epoch": 0.5315351051170171,
|
| 476 |
+
"grad_norm": 3.530519835264167,
|
| 477 |
+
"learning_rate": 6.630327056491576e-05,
|
| 478 |
+
"loss": 0.4717,
|
| 479 |
+
"step": 670
|
| 480 |
+
},
|
| 481 |
+
{
|
| 482 |
+
"epoch": 0.539468464894883,
|
| 483 |
+
"grad_norm": 2.392046097161058,
|
| 484 |
+
"learning_rate": 6.729435084241824e-05,
|
| 485 |
+
"loss": 0.4583,
|
| 486 |
+
"step": 680
|
| 487 |
+
},
|
| 488 |
+
{
|
| 489 |
+
"epoch": 0.5474018246727489,
|
| 490 |
+
"grad_norm": 3.149236166223247,
|
| 491 |
+
"learning_rate": 6.828543111992072e-05,
|
| 492 |
+
"loss": 0.4412,
|
| 493 |
+
"step": 690
|
| 494 |
+
},
|
| 495 |
+
{
|
| 496 |
+
"epoch": 0.5553351844506148,
|
| 497 |
+
"grad_norm": 2.9354688050931586,
|
| 498 |
+
"learning_rate": 6.92765113974232e-05,
|
| 499 |
+
"loss": 0.423,
|
| 500 |
+
"step": 700
|
| 501 |
+
},
|
| 502 |
+
{
|
| 503 |
+
"epoch": 0.5632685442284807,
|
| 504 |
+
"grad_norm": 2.7127436235307125,
|
| 505 |
+
"learning_rate": 7.026759167492567e-05,
|
| 506 |
+
"loss": 0.4062,
|
| 507 |
+
"step": 710
|
| 508 |
+
},
|
| 509 |
+
{
|
| 510 |
+
"epoch": 0.5712019040063467,
|
| 511 |
+
"grad_norm": 2.6410601628390777,
|
| 512 |
+
"learning_rate": 7.125867195242815e-05,
|
| 513 |
+
"loss": 0.3923,
|
| 514 |
+
"step": 720
|
| 515 |
+
},
|
| 516 |
+
{
|
| 517 |
+
"epoch": 0.5791352637842127,
|
| 518 |
+
"grad_norm": 2.7996527641631386,
|
| 519 |
+
"learning_rate": 7.224975222993063e-05,
|
| 520 |
+
"loss": 0.3739,
|
| 521 |
+
"step": 730
|
| 522 |
+
},
|
| 523 |
+
{
|
| 524 |
+
"epoch": 0.5870686235620786,
|
| 525 |
+
"grad_norm": 3.53218129873484,
|
| 526 |
+
"learning_rate": 7.32408325074331e-05,
|
| 527 |
+
"loss": 0.3558,
|
| 528 |
+
"step": 740
|
| 529 |
+
},
|
| 530 |
+
{
|
| 531 |
+
"epoch": 0.5950019833399445,
|
| 532 |
+
"grad_norm": 3.256471177277615,
|
| 533 |
+
"learning_rate": 7.423191278493558e-05,
|
| 534 |
+
"loss": 0.3378,
|
| 535 |
+
"step": 750
|
| 536 |
+
},
|
| 537 |
+
{
|
| 538 |
+
"epoch": 0.6029353431178104,
|
| 539 |
+
"grad_norm": 2.448855772936633,
|
| 540 |
+
"learning_rate": 7.522299306243806e-05,
|
| 541 |
+
"loss": 0.3318,
|
| 542 |
+
"step": 760
|
| 543 |
+
},
|
| 544 |
+
{
|
| 545 |
+
"epoch": 0.6108687028956763,
|
| 546 |
+
"grad_norm": 2.8206966008765355,
|
| 547 |
+
"learning_rate": 7.621407333994054e-05,
|
| 548 |
+
"loss": 0.3145,
|
| 549 |
+
"step": 770
|
| 550 |
+
},
|
| 551 |
+
{
|
| 552 |
+
"epoch": 0.6188020626735422,
|
| 553 |
+
"grad_norm": 2.262400615934418,
|
| 554 |
+
"learning_rate": 7.720515361744302e-05,
|
| 555 |
+
"loss": 0.2981,
|
| 556 |
+
"step": 780
|
| 557 |
+
},
|
| 558 |
+
{
|
| 559 |
+
"epoch": 0.6267354224514081,
|
| 560 |
+
"grad_norm": 2.6691511066584894,
|
| 561 |
+
"learning_rate": 7.81962338949455e-05,
|
| 562 |
+
"loss": 0.2878,
|
| 563 |
+
"step": 790
|
| 564 |
+
},
|
| 565 |
+
{
|
| 566 |
+
"epoch": 0.6346687822292741,
|
| 567 |
+
"grad_norm": 2.1502320771866494,
|
| 568 |
+
"learning_rate": 7.918731417244797e-05,
|
| 569 |
+
"loss": 0.2724,
|
| 570 |
+
"step": 800
|
| 571 |
+
},
|
| 572 |
+
{
|
| 573 |
+
"epoch": 0.64260214200714,
|
| 574 |
+
"grad_norm": 2.1869290626436806,
|
| 575 |
+
"learning_rate": 8.017839444995045e-05,
|
| 576 |
+
"loss": 0.2589,
|
| 577 |
+
"step": 810
|
| 578 |
+
},
|
| 579 |
+
{
|
| 580 |
+
"epoch": 0.650535501785006,
|
| 581 |
+
"grad_norm": 2.6277829837493476,
|
| 582 |
+
"learning_rate": 8.116947472745293e-05,
|
| 583 |
+
"loss": 0.2481,
|
| 584 |
+
"step": 820
|
| 585 |
+
},
|
| 586 |
+
{
|
| 587 |
+
"epoch": 0.6584688615628719,
|
| 588 |
+
"grad_norm": 1.8401496589236042,
|
| 589 |
+
"learning_rate": 8.21605550049554e-05,
|
| 590 |
+
"loss": 0.2361,
|
| 591 |
+
"step": 830
|
| 592 |
+
},
|
| 593 |
+
{
|
| 594 |
+
"epoch": 0.6664022213407378,
|
| 595 |
+
"grad_norm": 1.7862155383026863,
|
| 596 |
+
"learning_rate": 8.315163528245788e-05,
|
| 597 |
+
"loss": 0.222,
|
| 598 |
+
"step": 840
|
| 599 |
+
},
|
| 600 |
+
{
|
| 601 |
+
"epoch": 0.6743355811186037,
|
| 602 |
+
"grad_norm": 1.9765212410305726,
|
| 603 |
+
"learning_rate": 8.414271555996036e-05,
|
| 604 |
+
"loss": 0.2103,
|
| 605 |
+
"step": 850
|
| 606 |
+
},
|
| 607 |
+
{
|
| 608 |
+
"epoch": 0.6822689408964696,
|
| 609 |
+
"grad_norm": 2.4397348134664556,
|
| 610 |
+
"learning_rate": 8.513379583746284e-05,
|
| 611 |
+
"loss": 0.2021,
|
| 612 |
+
"step": 860
|
| 613 |
+
},
|
| 614 |
+
{
|
| 615 |
+
"epoch": 0.6902023006743356,
|
| 616 |
+
"grad_norm": 2.296758487647809,
|
| 617 |
+
"learning_rate": 8.612487611496532e-05,
|
| 618 |
+
"loss": 0.1924,
|
| 619 |
+
"step": 870
|
| 620 |
+
},
|
| 621 |
+
{
|
| 622 |
+
"epoch": 0.6981356604522015,
|
| 623 |
+
"grad_norm": 2.104859870669743,
|
| 624 |
+
"learning_rate": 8.71159563924678e-05,
|
| 625 |
+
"loss": 0.1817,
|
| 626 |
+
"step": 880
|
| 627 |
+
},
|
| 628 |
+
{
|
| 629 |
+
"epoch": 0.7060690202300675,
|
| 630 |
+
"grad_norm": 1.3003724019670297,
|
| 631 |
+
"learning_rate": 8.810703666997027e-05,
|
| 632 |
+
"loss": 0.1732,
|
| 633 |
+
"step": 890
|
| 634 |
+
},
|
| 635 |
+
{
|
| 636 |
+
"epoch": 0.7140023800079334,
|
| 637 |
+
"grad_norm": 1.3709932626535646,
|
| 638 |
+
"learning_rate": 8.909811694747275e-05,
|
| 639 |
+
"loss": 0.1593,
|
| 640 |
+
"step": 900
|
| 641 |
+
},
|
| 642 |
+
{
|
| 643 |
+
"epoch": 0.7219357397857993,
|
| 644 |
+
"grad_norm": 1.8342489151171628,
|
| 645 |
+
"learning_rate": 9.008919722497523e-05,
|
| 646 |
+
"loss": 0.1551,
|
| 647 |
+
"step": 910
|
| 648 |
+
},
|
| 649 |
+
{
|
| 650 |
+
"epoch": 0.7298690995636652,
|
| 651 |
+
"grad_norm": 1.762934595290678,
|
| 652 |
+
"learning_rate": 9.10802775024777e-05,
|
| 653 |
+
"loss": 0.1489,
|
| 654 |
+
"step": 920
|
| 655 |
+
},
|
| 656 |
+
{
|
| 657 |
+
"epoch": 0.7378024593415311,
|
| 658 |
+
"grad_norm": 1.4075349928757257,
|
| 659 |
+
"learning_rate": 9.207135777998018e-05,
|
| 660 |
+
"loss": 0.1391,
|
| 661 |
+
"step": 930
|
| 662 |
+
},
|
| 663 |
+
{
|
| 664 |
+
"epoch": 0.745735819119397,
|
| 665 |
+
"grad_norm": 1.3962752805987728,
|
| 666 |
+
"learning_rate": 9.306243805748266e-05,
|
| 667 |
+
"loss": 0.1332,
|
| 668 |
+
"step": 940
|
| 669 |
+
},
|
| 670 |
+
{
|
| 671 |
+
"epoch": 0.753669178897263,
|
| 672 |
+
"grad_norm": 3.5595641910775173,
|
| 673 |
+
"learning_rate": 9.405351833498514e-05,
|
| 674 |
+
"loss": 0.1507,
|
| 675 |
+
"step": 950
|
| 676 |
+
},
|
| 677 |
+
{
|
| 678 |
+
"epoch": 0.761602538675129,
|
| 679 |
+
"grad_norm": 1.2799346026465042,
|
| 680 |
+
"learning_rate": 9.504459861248762e-05,
|
| 681 |
+
"loss": 0.1706,
|
| 682 |
+
"step": 960
|
| 683 |
+
},
|
| 684 |
+
{
|
| 685 |
+
"epoch": 0.7695358984529949,
|
| 686 |
+
"grad_norm": 0.9866434355012127,
|
| 687 |
+
"learning_rate": 9.60356788899901e-05,
|
| 688 |
+
"loss": 0.1241,
|
| 689 |
+
"step": 970
|
| 690 |
+
},
|
| 691 |
+
{
|
| 692 |
+
"epoch": 0.7774692582308608,
|
| 693 |
+
"grad_norm": 1.2605525685265588,
|
| 694 |
+
"learning_rate": 9.702675916749257e-05,
|
| 695 |
+
"loss": 0.1133,
|
| 696 |
+
"step": 980
|
| 697 |
+
},
|
| 698 |
+
{
|
| 699 |
+
"epoch": 0.7854026180087267,
|
| 700 |
+
"grad_norm": 1.0654588120046724,
|
| 701 |
+
"learning_rate": 9.801783944499505e-05,
|
| 702 |
+
"loss": 0.1066,
|
| 703 |
+
"step": 990
|
| 704 |
+
},
|
| 705 |
+
{
|
| 706 |
+
"epoch": 0.7933359777865926,
|
| 707 |
+
"grad_norm": 0.761881486606698,
|
| 708 |
+
"learning_rate": 9.900891972249753e-05,
|
| 709 |
+
"loss": 0.1015,
|
| 710 |
+
"step": 1000
|
| 711 |
+
},
|
| 712 |
+
{
|
| 713 |
+
"epoch": 0.8012693375644585,
|
| 714 |
+
"grad_norm": 1.0601392633604383,
|
| 715 |
+
"learning_rate": 0.0001,
|
| 716 |
+
"loss": 0.0953,
|
| 717 |
+
"step": 1010
|
| 718 |
+
},
|
| 719 |
+
{
|
| 720 |
+
"epoch": 0.8092026973423245,
|
| 721 |
+
"grad_norm": 0.8224204361157935,
|
| 722 |
+
"learning_rate": 9.999970066101032e-05,
|
| 723 |
+
"loss": 0.0906,
|
| 724 |
+
"step": 1020
|
| 725 |
+
},
|
| 726 |
+
{
|
| 727 |
+
"epoch": 0.8171360571201904,
|
| 728 |
+
"grad_norm": 0.7080480512831684,
|
| 729 |
+
"learning_rate": 9.99988026476254e-05,
|
| 730 |
+
"loss": 0.0866,
|
| 731 |
+
"step": 1030
|
| 732 |
+
},
|
| 733 |
+
{
|
| 734 |
+
"epoch": 0.8250694168980564,
|
| 735 |
+
"grad_norm": 0.7547638843757503,
|
| 736 |
+
"learning_rate": 9.999730597059764e-05,
|
| 737 |
+
"loss": 0.0809,
|
| 738 |
+
"step": 1040
|
| 739 |
+
},
|
| 740 |
+
{
|
| 741 |
+
"epoch": 0.8330027766759223,
|
| 742 |
+
"grad_norm": 1.1588422105514353,
|
| 743 |
+
"learning_rate": 9.999521064784764e-05,
|
| 744 |
+
"loss": 0.0844,
|
| 745 |
+
"step": 1050
|
| 746 |
+
},
|
| 747 |
+
{
|
| 748 |
+
"epoch": 0.8409361364537882,
|
| 749 |
+
"grad_norm": 0.6385523334282596,
|
| 750 |
+
"learning_rate": 9.999251670446386e-05,
|
| 751 |
+
"loss": 0.0804,
|
| 752 |
+
"step": 1060
|
| 753 |
+
},
|
| 754 |
+
{
|
| 755 |
+
"epoch": 0.8488694962316541,
|
| 756 |
+
"grad_norm": 0.5623095327146973,
|
| 757 |
+
"learning_rate": 9.998922417270235e-05,
|
| 758 |
+
"loss": 0.072,
|
| 759 |
+
"step": 1070
|
| 760 |
+
},
|
| 761 |
+
{
|
| 762 |
+
"epoch": 0.85680285600952,
|
| 763 |
+
"grad_norm": 0.43678572122080983,
|
| 764 |
+
"learning_rate": 9.99853330919865e-05,
|
| 765 |
+
"loss": 0.0679,
|
| 766 |
+
"step": 1080
|
| 767 |
+
},
|
| 768 |
+
{
|
| 769 |
+
"epoch": 0.8647362157873859,
|
| 770 |
+
"grad_norm": 0.4972727247312438,
|
| 771 |
+
"learning_rate": 9.998084350890633e-05,
|
| 772 |
+
"loss": 0.065,
|
| 773 |
+
"step": 1090
|
| 774 |
+
},
|
| 775 |
+
{
|
| 776 |
+
"epoch": 0.8726695755652519,
|
| 777 |
+
"grad_norm": 0.5035917631799433,
|
| 778 |
+
"learning_rate": 9.997575547721818e-05,
|
| 779 |
+
"loss": 0.0627,
|
| 780 |
+
"step": 1100
|
| 781 |
+
},
|
| 782 |
+
{
|
| 783 |
+
"epoch": 0.8806029353431178,
|
| 784 |
+
"grad_norm": 0.5182605794595788,
|
| 785 |
+
"learning_rate": 9.997006905784388e-05,
|
| 786 |
+
"loss": 0.0602,
|
| 787 |
+
"step": 1110
|
| 788 |
+
},
|
| 789 |
+
{
|
| 790 |
+
"epoch": 0.8885362951209838,
|
| 791 |
+
"grad_norm": 0.41920699625736624,
|
| 792 |
+
"learning_rate": 9.996378431887012e-05,
|
| 793 |
+
"loss": 0.0579,
|
| 794 |
+
"step": 1120
|
| 795 |
+
},
|
| 796 |
+
{
|
| 797 |
+
"epoch": 0.8964696548988497,
|
| 798 |
+
"grad_norm": 0.414907297365465,
|
| 799 |
+
"learning_rate": 9.995690133554758e-05,
|
| 800 |
+
"loss": 0.0546,
|
| 801 |
+
"step": 1130
|
| 802 |
+
},
|
| 803 |
+
{
|
| 804 |
+
"epoch": 0.9044030146767156,
|
| 805 |
+
"grad_norm": 0.3702005062489374,
|
| 806 |
+
"learning_rate": 9.99494201902901e-05,
|
| 807 |
+
"loss": 0.0523,
|
| 808 |
+
"step": 1140
|
| 809 |
+
},
|
| 810 |
+
{
|
| 811 |
+
"epoch": 0.9123363744545815,
|
| 812 |
+
"grad_norm": 0.42037834689750175,
|
| 813 |
+
"learning_rate": 9.994134097267358e-05,
|
| 814 |
+
"loss": 0.0516,
|
| 815 |
+
"step": 1150
|
| 816 |
+
},
|
| 817 |
+
{
|
| 818 |
+
"epoch": 0.9202697342324474,
|
| 819 |
+
"grad_norm": 0.38928699328314564,
|
| 820 |
+
"learning_rate": 9.993266377943504e-05,
|
| 821 |
+
"loss": 0.0505,
|
| 822 |
+
"step": 1160
|
| 823 |
+
},
|
| 824 |
+
{
|
| 825 |
+
"epoch": 0.9282030940103134,
|
| 826 |
+
"grad_norm": 0.3941895011795836,
|
| 827 |
+
"learning_rate": 9.992338871447138e-05,
|
| 828 |
+
"loss": 0.0495,
|
| 829 |
+
"step": 1170
|
| 830 |
+
},
|
| 831 |
+
{
|
| 832 |
+
"epoch": 0.9361364537881793,
|
| 833 |
+
"grad_norm": 0.37452365492894446,
|
| 834 |
+
"learning_rate": 9.991351588883811e-05,
|
| 835 |
+
"loss": 0.0474,
|
| 836 |
+
"step": 1180
|
| 837 |
+
},
|
| 838 |
+
{
|
| 839 |
+
"epoch": 0.9440698135660452,
|
| 840 |
+
"grad_norm": 0.4395025405712049,
|
| 841 |
+
"learning_rate": 9.99030454207481e-05,
|
| 842 |
+
"loss": 0.0467,
|
| 843 |
+
"step": 1190
|
| 844 |
+
},
|
| 845 |
+
{
|
| 846 |
+
"epoch": 0.9520031733439112,
|
| 847 |
+
"grad_norm": 0.4639716702934286,
|
| 848 |
+
"learning_rate": 9.989197743557017e-05,
|
| 849 |
+
"loss": 0.045,
|
| 850 |
+
"step": 1200
|
| 851 |
+
},
|
| 852 |
+
{
|
| 853 |
+
"epoch": 0.9599365331217771,
|
| 854 |
+
"grad_norm": 0.34908102760028864,
|
| 855 |
+
"learning_rate": 9.988031206582746e-05,
|
| 856 |
+
"loss": 0.0439,
|
| 857 |
+
"step": 1210
|
| 858 |
+
},
|
| 859 |
+
{
|
| 860 |
+
"epoch": 0.967869892899643,
|
| 861 |
+
"grad_norm": 0.29211915294831964,
|
| 862 |
+
"learning_rate": 9.986804945119597e-05,
|
| 863 |
+
"loss": 0.0429,
|
| 864 |
+
"step": 1220
|
| 865 |
+
},
|
| 866 |
+
{
|
| 867 |
+
"epoch": 0.9758032526775089,
|
| 868 |
+
"grad_norm": 0.2801643208729514,
|
| 869 |
+
"learning_rate": 9.985518973850285e-05,
|
| 870 |
+
"loss": 0.041,
|
| 871 |
+
"step": 1230
|
| 872 |
+
},
|
| 873 |
+
{
|
| 874 |
+
"epoch": 0.9837366124553748,
|
| 875 |
+
"grad_norm": 0.3013251709022931,
|
| 876 |
+
"learning_rate": 9.984173308172465e-05,
|
| 877 |
+
"loss": 0.0403,
|
| 878 |
+
"step": 1240
|
| 879 |
+
},
|
| 880 |
+
{
|
| 881 |
+
"epoch": 0.9916699722332408,
|
| 882 |
+
"grad_norm": 0.28830525385588507,
|
| 883 |
+
"learning_rate": 9.982767964198542e-05,
|
| 884 |
+
"loss": 0.04,
|
| 885 |
+
"step": 1250
|
| 886 |
+
},
|
| 887 |
+
{
|
| 888 |
+
"epoch": 0.9996033320111067,
|
| 889 |
+
"grad_norm": 0.2756668810883745,
|
| 890 |
+
"learning_rate": 9.98130295875549e-05,
|
| 891 |
+
"loss": 0.0386,
|
| 892 |
+
"step": 1260
|
| 893 |
+
},
|
| 894 |
+
{
|
| 895 |
+
"epoch": 1.0071400238000794,
|
| 896 |
+
"grad_norm": 0.2673295493921112,
|
| 897 |
+
"learning_rate": 9.979778309384638e-05,
|
| 898 |
+
"loss": 0.0369,
|
| 899 |
+
"step": 1270
|
| 900 |
+
},
|
| 901 |
+
{
|
| 902 |
+
"epoch": 1.0150733835779453,
|
| 903 |
+
"grad_norm": 0.28881709817045276,
|
| 904 |
+
"learning_rate": 9.978194034341462e-05,
|
| 905 |
+
"loss": 0.0366,
|
| 906 |
+
"step": 1280
|
| 907 |
+
},
|
| 908 |
+
{
|
| 909 |
+
"epoch": 1.0230067433558112,
|
| 910 |
+
"grad_norm": 0.2443064331996704,
|
| 911 |
+
"learning_rate": 9.976550152595379e-05,
|
| 912 |
+
"loss": 0.035,
|
| 913 |
+
"step": 1290
|
| 914 |
+
},
|
| 915 |
+
{
|
| 916 |
+
"epoch": 1.0309401031336771,
|
| 917 |
+
"grad_norm": 0.2586793532004466,
|
| 918 |
+
"learning_rate": 9.974846683829501e-05,
|
| 919 |
+
"loss": 0.0351,
|
| 920 |
+
"step": 1300
|
| 921 |
+
},
|
| 922 |
+
{
|
| 923 |
+
"epoch": 1.038873462911543,
|
| 924 |
+
"grad_norm": 0.6237223735436187,
|
| 925 |
+
"learning_rate": 9.973083648440416e-05,
|
| 926 |
+
"loss": 0.0495,
|
| 927 |
+
"step": 1310
|
| 928 |
+
},
|
| 929 |
+
{
|
| 930 |
+
"epoch": 1.046806822689409,
|
| 931 |
+
"grad_norm": 0.982888186735296,
|
| 932 |
+
"learning_rate": 9.971261067537932e-05,
|
| 933 |
+
"loss": 0.0428,
|
| 934 |
+
"step": 1320
|
| 935 |
+
},
|
| 936 |
+
{
|
| 937 |
+
"epoch": 1.0547401824672749,
|
| 938 |
+
"grad_norm": 0.3721310946214082,
|
| 939 |
+
"learning_rate": 9.969378962944829e-05,
|
| 940 |
+
"loss": 0.0396,
|
| 941 |
+
"step": 1330
|
| 942 |
+
},
|
| 943 |
+
{
|
| 944 |
+
"epoch": 1.0626735422451408,
|
| 945 |
+
"grad_norm": 0.3035794519690265,
|
| 946 |
+
"learning_rate": 9.9674373571966e-05,
|
| 947 |
+
"loss": 0.0352,
|
| 948 |
+
"step": 1340
|
| 949 |
+
},
|
| 950 |
+
{
|
| 951 |
+
"epoch": 1.0706069020230067,
|
| 952 |
+
"grad_norm": 0.2564381074742219,
|
| 953 |
+
"learning_rate": 9.965436273541175e-05,
|
| 954 |
+
"loss": 0.034,
|
| 955 |
+
"step": 1350
|
| 956 |
+
},
|
| 957 |
+
{
|
| 958 |
+
"epoch": 1.0785402618008726,
|
| 959 |
+
"grad_norm": 0.2627161314780947,
|
| 960 |
+
"learning_rate": 9.963375735938651e-05,
|
| 961 |
+
"loss": 0.0328,
|
| 962 |
+
"step": 1360
|
| 963 |
+
},
|
| 964 |
+
{
|
| 965 |
+
"epoch": 1.0864736215787385,
|
| 966 |
+
"grad_norm": 0.23965693542315172,
|
| 967 |
+
"learning_rate": 9.961255769060997e-05,
|
| 968 |
+
"loss": 0.0318,
|
| 969 |
+
"step": 1370
|
| 970 |
+
},
|
| 971 |
+
{
|
| 972 |
+
"epoch": 1.0944069813566046,
|
| 973 |
+
"grad_norm": 0.24468200257324066,
|
| 974 |
+
"learning_rate": 9.95907639829176e-05,
|
| 975 |
+
"loss": 0.0312,
|
| 976 |
+
"step": 1380
|
| 977 |
+
},
|
| 978 |
+
{
|
| 979 |
+
"epoch": 1.1023403411344705,
|
| 980 |
+
"grad_norm": 0.2090058523545183,
|
| 981 |
+
"learning_rate": 9.95683764972577e-05,
|
| 982 |
+
"loss": 0.0301,
|
| 983 |
+
"step": 1390
|
| 984 |
+
},
|
| 985 |
+
{
|
| 986 |
+
"epoch": 1.1102737009123365,
|
| 987 |
+
"grad_norm": 0.2583539701995491,
|
| 988 |
+
"learning_rate": 9.954539550168816e-05,
|
| 989 |
+
"loss": 0.0301,
|
| 990 |
+
"step": 1400
|
| 991 |
+
},
|
| 992 |
+
{
|
| 993 |
+
"epoch": 1.1182070606902024,
|
| 994 |
+
"grad_norm": 0.22716588824206302,
|
| 995 |
+
"learning_rate": 9.952182127137327e-05,
|
| 996 |
+
"loss": 0.0293,
|
| 997 |
+
"step": 1410
|
| 998 |
+
},
|
| 999 |
+
{
|
| 1000 |
+
"epoch": 1.1261404204680683,
|
| 1001 |
+
"grad_norm": 0.24087202301512833,
|
| 1002 |
+
"learning_rate": 9.94976540885805e-05,
|
| 1003 |
+
"loss": 0.0287,
|
| 1004 |
+
"step": 1420
|
| 1005 |
+
},
|
| 1006 |
+
{
|
| 1007 |
+
"epoch": 1.1340737802459342,
|
| 1008 |
+
"grad_norm": 0.23388912586015503,
|
| 1009 |
+
"learning_rate": 9.947289424267703e-05,
|
| 1010 |
+
"loss": 0.0283,
|
| 1011 |
+
"step": 1430
|
| 1012 |
+
},
|
| 1013 |
+
{
|
| 1014 |
+
"epoch": 1.1420071400238,
|
| 1015 |
+
"grad_norm": 0.21775302377697844,
|
| 1016 |
+
"learning_rate": 9.944754203012639e-05,
|
| 1017 |
+
"loss": 0.0273,
|
| 1018 |
+
"step": 1440
|
| 1019 |
+
},
|
| 1020 |
+
{
|
| 1021 |
+
"epoch": 1.149940499801666,
|
| 1022 |
+
"grad_norm": 0.2120232453240847,
|
| 1023 |
+
"learning_rate": 9.942159775448478e-05,
|
| 1024 |
+
"loss": 0.0273,
|
| 1025 |
+
"step": 1450
|
| 1026 |
+
},
|
| 1027 |
+
{
|
| 1028 |
+
"epoch": 1.157873859579532,
|
| 1029 |
+
"grad_norm": 0.2055361894837688,
|
| 1030 |
+
"learning_rate": 9.939506172639753e-05,
|
| 1031 |
+
"loss": 0.027,
|
| 1032 |
+
"step": 1460
|
| 1033 |
+
},
|
| 1034 |
+
{
|
| 1035 |
+
"epoch": 1.1658072193573978,
|
| 1036 |
+
"grad_norm": 0.22045835966574195,
|
| 1037 |
+
"learning_rate": 9.936793426359536e-05,
|
| 1038 |
+
"loss": 0.0264,
|
| 1039 |
+
"step": 1470
|
| 1040 |
+
},
|
| 1041 |
+
{
|
| 1042 |
+
"epoch": 1.1737405791352638,
|
| 1043 |
+
"grad_norm": 0.19936431245788608,
|
| 1044 |
+
"learning_rate": 9.934021569089058e-05,
|
| 1045 |
+
"loss": 0.0262,
|
| 1046 |
+
"step": 1480
|
| 1047 |
+
},
|
| 1048 |
+
{
|
| 1049 |
+
"epoch": 1.1816739389131297,
|
| 1050 |
+
"grad_norm": 0.2182877537167163,
|
| 1051 |
+
"learning_rate": 9.931190634017314e-05,
|
| 1052 |
+
"loss": 0.0256,
|
| 1053 |
+
"step": 1490
|
| 1054 |
+
},
|
| 1055 |
+
{
|
| 1056 |
+
"epoch": 1.1896072986909956,
|
| 1057 |
+
"grad_norm": 0.20894413268256554,
|
| 1058 |
+
"learning_rate": 9.928300655040676e-05,
|
| 1059 |
+
"loss": 0.0255,
|
| 1060 |
+
"step": 1500
|
| 1061 |
+
},
|
| 1062 |
+
{
|
| 1063 |
+
"epoch": 1.1975406584688615,
|
| 1064 |
+
"grad_norm": 0.217040492922683,
|
| 1065 |
+
"learning_rate": 9.925351666762479e-05,
|
| 1066 |
+
"loss": 0.0248,
|
| 1067 |
+
"step": 1510
|
| 1068 |
+
},
|
| 1069 |
+
{
|
| 1070 |
+
"epoch": 1.2054740182467274,
|
| 1071 |
+
"grad_norm": 0.2080798426449999,
|
| 1072 |
+
"learning_rate": 9.92234370449261e-05,
|
| 1073 |
+
"loss": 0.0247,
|
| 1074 |
+
"step": 1520
|
| 1075 |
+
},
|
| 1076 |
+
{
|
| 1077 |
+
"epoch": 1.2134073780245935,
|
| 1078 |
+
"grad_norm": 0.21422582180243674,
|
| 1079 |
+
"learning_rate": 9.919276804247084e-05,
|
| 1080 |
+
"loss": 0.0246,
|
| 1081 |
+
"step": 1530
|
| 1082 |
+
},
|
| 1083 |
+
{
|
| 1084 |
+
"epoch": 1.2213407378024592,
|
| 1085 |
+
"grad_norm": 0.22109698138341372,
|
| 1086 |
+
"learning_rate": 9.916151002747614e-05,
|
| 1087 |
+
"loss": 0.0242,
|
| 1088 |
+
"step": 1540
|
| 1089 |
+
},
|
| 1090 |
+
{
|
| 1091 |
+
"epoch": 1.2292740975803254,
|
| 1092 |
+
"grad_norm": 0.2191386707736148,
|
| 1093 |
+
"learning_rate": 9.912966337421169e-05,
|
| 1094 |
+
"loss": 0.0241,
|
| 1095 |
+
"step": 1550
|
| 1096 |
+
},
|
| 1097 |
+
{
|
| 1098 |
+
"epoch": 1.2372074573581913,
|
| 1099 |
+
"grad_norm": 0.21285746880929995,
|
| 1100 |
+
"learning_rate": 9.909722846399531e-05,
|
| 1101 |
+
"loss": 0.0235,
|
| 1102 |
+
"step": 1560
|
| 1103 |
+
},
|
| 1104 |
+
{
|
| 1105 |
+
"epoch": 1.2451408171360572,
|
| 1106 |
+
"grad_norm": 0.2060608693515892,
|
| 1107 |
+
"learning_rate": 9.906420568518832e-05,
|
| 1108 |
+
"loss": 0.0235,
|
| 1109 |
+
"step": 1570
|
| 1110 |
+
},
|
| 1111 |
+
{
|
| 1112 |
+
"epoch": 1.253074176913923,
|
| 1113 |
+
"grad_norm": 0.2165821829643641,
|
| 1114 |
+
"learning_rate": 9.903059543319096e-05,
|
| 1115 |
+
"loss": 0.0228,
|
| 1116 |
+
"step": 1580
|
| 1117 |
+
},
|
| 1118 |
+
{
|
| 1119 |
+
"epoch": 1.261007536691789,
|
| 1120 |
+
"grad_norm": 0.20784176983531677,
|
| 1121 |
+
"learning_rate": 9.899639811043755e-05,
|
| 1122 |
+
"loss": 0.0224,
|
| 1123 |
+
"step": 1590
|
| 1124 |
+
},
|
| 1125 |
+
{
|
| 1126 |
+
"epoch": 1.268940896469655,
|
| 1127 |
+
"grad_norm": 0.1941327009060369,
|
| 1128 |
+
"learning_rate": 9.896161412639178e-05,
|
| 1129 |
+
"loss": 0.0224,
|
| 1130 |
+
"step": 1600
|
| 1131 |
+
},
|
| 1132 |
+
{
|
| 1133 |
+
"epoch": 1.2768742562475208,
|
| 1134 |
+
"grad_norm": 0.19719426217271882,
|
| 1135 |
+
"learning_rate": 9.892624389754175e-05,
|
| 1136 |
+
"loss": 0.0218,
|
| 1137 |
+
"step": 1610
|
| 1138 |
+
},
|
| 1139 |
+
{
|
| 1140 |
+
"epoch": 1.2848076160253867,
|
| 1141 |
+
"grad_norm": 0.1987404294569414,
|
| 1142 |
+
"learning_rate": 9.889028784739502e-05,
|
| 1143 |
+
"loss": 0.0217,
|
| 1144 |
+
"step": 1620
|
| 1145 |
+
},
|
| 1146 |
+
{
|
| 1147 |
+
"epoch": 1.2927409758032526,
|
| 1148 |
+
"grad_norm": 0.36446508935174654,
|
| 1149 |
+
"learning_rate": 9.885374640647349e-05,
|
| 1150 |
+
"loss": 0.0218,
|
| 1151 |
+
"step": 1630
|
| 1152 |
+
},
|
| 1153 |
+
{
|
| 1154 |
+
"epoch": 1.3006743355811186,
|
| 1155 |
+
"grad_norm": 0.2153981685427444,
|
| 1156 |
+
"learning_rate": 9.881662001230827e-05,
|
| 1157 |
+
"loss": 0.0219,
|
| 1158 |
+
"step": 1640
|
| 1159 |
+
},
|
| 1160 |
+
{
|
| 1161 |
+
"epoch": 1.3086076953589845,
|
| 1162 |
+
"grad_norm": 0.23124643936791714,
|
| 1163 |
+
"learning_rate": 9.877890910943448e-05,
|
| 1164 |
+
"loss": 0.0215,
|
| 1165 |
+
"step": 1650
|
| 1166 |
+
},
|
| 1167 |
+
{
|
| 1168 |
+
"epoch": 1.3165410551368504,
|
| 1169 |
+
"grad_norm": 0.19794561936484395,
|
| 1170 |
+
"learning_rate": 9.874061414938583e-05,
|
| 1171 |
+
"loss": 0.0209,
|
| 1172 |
+
"step": 1660
|
| 1173 |
+
},
|
| 1174 |
+
{
|
| 1175 |
+
"epoch": 1.3244744149147163,
|
| 1176 |
+
"grad_norm": 0.1939423361167064,
|
| 1177 |
+
"learning_rate": 9.870173559068933e-05,
|
| 1178 |
+
"loss": 0.0204,
|
| 1179 |
+
"step": 1670
|
| 1180 |
+
},
|
| 1181 |
+
{
|
| 1182 |
+
"epoch": 1.3324077746925824,
|
| 1183 |
+
"grad_norm": 0.2157419658468051,
|
| 1184 |
+
"learning_rate": 9.86622738988597e-05,
|
| 1185 |
+
"loss": 0.02,
|
| 1186 |
+
"step": 1680
|
| 1187 |
+
},
|
| 1188 |
+
{
|
| 1189 |
+
"epoch": 1.3403411344704481,
|
| 1190 |
+
"grad_norm": 0.19391991216609314,
|
| 1191 |
+
"learning_rate": 9.862222954639389e-05,
|
| 1192 |
+
"loss": 0.0196,
|
| 1193 |
+
"step": 1690
|
| 1194 |
+
},
|
| 1195 |
+
{
|
| 1196 |
+
"epoch": 1.3482744942483142,
|
| 1197 |
+
"grad_norm": 0.19977705728669262,
|
| 1198 |
+
"learning_rate": 9.858160301276529e-05,
|
| 1199 |
+
"loss": 0.0191,
|
| 1200 |
+
"step": 1700
|
| 1201 |
+
},
|
| 1202 |
+
{
|
| 1203 |
+
"epoch": 1.3562078540261802,
|
| 1204 |
+
"grad_norm": 0.1658808248673058,
|
| 1205 |
+
"learning_rate": 9.854039478441817e-05,
|
| 1206 |
+
"loss": 0.0194,
|
| 1207 |
+
"step": 1710
|
| 1208 |
+
},
|
| 1209 |
+
{
|
| 1210 |
+
"epoch": 1.364141213804046,
|
| 1211 |
+
"grad_norm": 0.18655747400847833,
|
| 1212 |
+
"learning_rate": 9.849860535476168e-05,
|
| 1213 |
+
"loss": 0.0193,
|
| 1214 |
+
"step": 1720
|
| 1215 |
+
},
|
| 1216 |
+
{
|
| 1217 |
+
"epoch": 1.372074573581912,
|
| 1218 |
+
"grad_norm": 0.18591273213926962,
|
| 1219 |
+
"learning_rate": 9.845623522416407e-05,
|
| 1220 |
+
"loss": 0.0191,
|
| 1221 |
+
"step": 1730
|
| 1222 |
+
},
|
| 1223 |
+
{
|
| 1224 |
+
"epoch": 1.380007933359778,
|
| 1225 |
+
"grad_norm": 0.19943417251594622,
|
| 1226 |
+
"learning_rate": 9.841328489994659e-05,
|
| 1227 |
+
"loss": 0.0189,
|
| 1228 |
+
"step": 1740
|
| 1229 |
+
},
|
| 1230 |
+
{
|
| 1231 |
+
"epoch": 1.3879412931376438,
|
| 1232 |
+
"grad_norm": 0.20534200797212154,
|
| 1233 |
+
"learning_rate": 9.836975489637754e-05,
|
| 1234 |
+
"loss": 0.0184,
|
| 1235 |
+
"step": 1750
|
| 1236 |
+
},
|
| 1237 |
+
{
|
| 1238 |
+
"epoch": 1.3958746529155097,
|
| 1239 |
+
"grad_norm": 0.1911015742718259,
|
| 1240 |
+
"learning_rate": 9.832564573466598e-05,
|
| 1241 |
+
"loss": 0.0182,
|
| 1242 |
+
"step": 1760
|
| 1243 |
+
},
|
| 1244 |
+
{
|
| 1245 |
+
"epoch": 1.4038080126933756,
|
| 1246 |
+
"grad_norm": 0.18521458075594055,
|
| 1247 |
+
"learning_rate": 9.82809579429556e-05,
|
| 1248 |
+
"loss": 0.0178,
|
| 1249 |
+
"step": 1770
|
| 1250 |
+
},
|
| 1251 |
+
{
|
| 1252 |
+
"epoch": 1.4117413724712415,
|
| 1253 |
+
"grad_norm": 0.19914721587866432,
|
| 1254 |
+
"learning_rate": 9.823569205631836e-05,
|
| 1255 |
+
"loss": 0.0179,
|
| 1256 |
+
"step": 1780
|
| 1257 |
+
},
|
| 1258 |
+
{
|
| 1259 |
+
"epoch": 1.4196747322491075,
|
| 1260 |
+
"grad_norm": 0.19845608937674608,
|
| 1261 |
+
"learning_rate": 9.818984861674803e-05,
|
| 1262 |
+
"loss": 0.0176,
|
| 1263 |
+
"step": 1790
|
| 1264 |
+
},
|
| 1265 |
+
{
|
| 1266 |
+
"epoch": 1.4276080920269734,
|
| 1267 |
+
"grad_norm": 0.19674825577684443,
|
| 1268 |
+
"learning_rate": 9.814342817315374e-05,
|
| 1269 |
+
"loss": 0.0177,
|
| 1270 |
+
"step": 1800
|
| 1271 |
+
}
|
| 1272 |
+
],
|
| 1273 |
+
"logging_steps": 10,
|
| 1274 |
+
"max_steps": 10088,
|
| 1275 |
+
"num_input_tokens_seen": 0,
|
| 1276 |
+
"num_train_epochs": 8,
|
| 1277 |
+
"save_steps": 1800,
|
| 1278 |
+
"stateful_callbacks": {
|
| 1279 |
+
"TrainerControl": {
|
| 1280 |
+
"args": {
|
| 1281 |
+
"should_epoch_stop": false,
|
| 1282 |
+
"should_evaluate": false,
|
| 1283 |
+
"should_log": false,
|
| 1284 |
+
"should_save": true,
|
| 1285 |
+
"should_training_stop": false
|
| 1286 |
+
},
|
| 1287 |
+
"attributes": {}
|
| 1288 |
+
}
|
| 1289 |
+
},
|
| 1290 |
+
"total_flos": 129584125181952.0,
|
| 1291 |
+
"train_batch_size": 128,
|
| 1292 |
+
"trial_name": null,
|
| 1293 |
+
"trial_params": null
|
| 1294 |
+
}
|