Upload multi_student_4/checkpoint-1800/trainer_state.json with huggingface_hub
Browse files
multi_student_4/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": 1.1842105263157894,
|
| 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": 0.006578947368421052,
|
| 14 |
+
"grad_norm": 19.201007081270827,
|
| 15 |
+
"learning_rate": 8.458646616541353e-07,
|
| 16 |
+
"loss": 8.2522,
|
| 17 |
+
"step": 10
|
| 18 |
+
},
|
| 19 |
+
{
|
| 20 |
+
"epoch": 0.013157894736842105,
|
| 21 |
+
"grad_norm": 18.114759300569585,
|
| 22 |
+
"learning_rate": 1.7857142857142857e-06,
|
| 23 |
+
"loss": 8.1346,
|
| 24 |
+
"step": 20
|
| 25 |
+
},
|
| 26 |
+
{
|
| 27 |
+
"epoch": 0.019736842105263157,
|
| 28 |
+
"grad_norm": 9.479492840003324,
|
| 29 |
+
"learning_rate": 2.725563909774436e-06,
|
| 30 |
+
"loss": 7.5119,
|
| 31 |
+
"step": 30
|
| 32 |
+
},
|
| 33 |
+
{
|
| 34 |
+
"epoch": 0.02631578947368421,
|
| 35 |
+
"grad_norm": 4.976331973381574,
|
| 36 |
+
"learning_rate": 3.6654135338345866e-06,
|
| 37 |
+
"loss": 6.7788,
|
| 38 |
+
"step": 40
|
| 39 |
+
},
|
| 40 |
+
{
|
| 41 |
+
"epoch": 0.03289473684210526,
|
| 42 |
+
"grad_norm": 3.2804593438080754,
|
| 43 |
+
"learning_rate": 4.605263157894737e-06,
|
| 44 |
+
"loss": 6.2892,
|
| 45 |
+
"step": 50
|
| 46 |
+
},
|
| 47 |
+
{
|
| 48 |
+
"epoch": 0.039473684210526314,
|
| 49 |
+
"grad_norm": 2.8583967169035027,
|
| 50 |
+
"learning_rate": 5.5451127819548875e-06,
|
| 51 |
+
"loss": 5.9528,
|
| 52 |
+
"step": 60
|
| 53 |
+
},
|
| 54 |
+
{
|
| 55 |
+
"epoch": 0.046052631578947366,
|
| 56 |
+
"grad_norm": 3.5512069733934597,
|
| 57 |
+
"learning_rate": 6.484962406015038e-06,
|
| 58 |
+
"loss": 5.5755,
|
| 59 |
+
"step": 70
|
| 60 |
+
},
|
| 61 |
+
{
|
| 62 |
+
"epoch": 0.05263157894736842,
|
| 63 |
+
"grad_norm": 3.731444843289522,
|
| 64 |
+
"learning_rate": 7.424812030075189e-06,
|
| 65 |
+
"loss": 4.9914,
|
| 66 |
+
"step": 80
|
| 67 |
+
},
|
| 68 |
+
{
|
| 69 |
+
"epoch": 0.05921052631578947,
|
| 70 |
+
"grad_norm": 5.60127901105382,
|
| 71 |
+
"learning_rate": 8.364661654135339e-06,
|
| 72 |
+
"loss": 4.3233,
|
| 73 |
+
"step": 90
|
| 74 |
+
},
|
| 75 |
+
{
|
| 76 |
+
"epoch": 0.06578947368421052,
|
| 77 |
+
"grad_norm": 3.7839180004348996,
|
| 78 |
+
"learning_rate": 9.30451127819549e-06,
|
| 79 |
+
"loss": 3.7305,
|
| 80 |
+
"step": 100
|
| 81 |
+
},
|
| 82 |
+
{
|
| 83 |
+
"epoch": 0.07236842105263158,
|
| 84 |
+
"grad_norm": 5.421177691842472,
|
| 85 |
+
"learning_rate": 1.024436090225564e-05,
|
| 86 |
+
"loss": 3.3248,
|
| 87 |
+
"step": 110
|
| 88 |
+
},
|
| 89 |
+
{
|
| 90 |
+
"epoch": 0.07894736842105263,
|
| 91 |
+
"grad_norm": 3.7158079343320694,
|
| 92 |
+
"learning_rate": 1.118421052631579e-05,
|
| 93 |
+
"loss": 3.092,
|
| 94 |
+
"step": 120
|
| 95 |
+
},
|
| 96 |
+
{
|
| 97 |
+
"epoch": 0.08552631578947369,
|
| 98 |
+
"grad_norm": 5.042680618782553,
|
| 99 |
+
"learning_rate": 1.212406015037594e-05,
|
| 100 |
+
"loss": 2.9383,
|
| 101 |
+
"step": 130
|
| 102 |
+
},
|
| 103 |
+
{
|
| 104 |
+
"epoch": 0.09210526315789473,
|
| 105 |
+
"grad_norm": 5.8029975922287615,
|
| 106 |
+
"learning_rate": 1.3063909774436093e-05,
|
| 107 |
+
"loss": 2.8108,
|
| 108 |
+
"step": 140
|
| 109 |
+
},
|
| 110 |
+
{
|
| 111 |
+
"epoch": 0.09868421052631579,
|
| 112 |
+
"grad_norm": 7.443887556478386,
|
| 113 |
+
"learning_rate": 1.4003759398496241e-05,
|
| 114 |
+
"loss": 2.7256,
|
| 115 |
+
"step": 150
|
| 116 |
+
},
|
| 117 |
+
{
|
| 118 |
+
"epoch": 0.10526315789473684,
|
| 119 |
+
"grad_norm": 6.942055828457074,
|
| 120 |
+
"learning_rate": 1.4943609022556393e-05,
|
| 121 |
+
"loss": 2.6465,
|
| 122 |
+
"step": 160
|
| 123 |
+
},
|
| 124 |
+
{
|
| 125 |
+
"epoch": 0.1118421052631579,
|
| 126 |
+
"grad_norm": 20.211444588304172,
|
| 127 |
+
"learning_rate": 1.5883458646616542e-05,
|
| 128 |
+
"loss": 2.5756,
|
| 129 |
+
"step": 170
|
| 130 |
+
},
|
| 131 |
+
{
|
| 132 |
+
"epoch": 0.11842105263157894,
|
| 133 |
+
"grad_norm": 23.482614308148037,
|
| 134 |
+
"learning_rate": 1.6823308270676694e-05,
|
| 135 |
+
"loss": 2.5262,
|
| 136 |
+
"step": 180
|
| 137 |
+
},
|
| 138 |
+
{
|
| 139 |
+
"epoch": 0.125,
|
| 140 |
+
"grad_norm": 13.648396628618737,
|
| 141 |
+
"learning_rate": 1.7763157894736842e-05,
|
| 142 |
+
"loss": 2.4699,
|
| 143 |
+
"step": 190
|
| 144 |
+
},
|
| 145 |
+
{
|
| 146 |
+
"epoch": 0.13157894736842105,
|
| 147 |
+
"grad_norm": 15.205800265082752,
|
| 148 |
+
"learning_rate": 1.8703007518796993e-05,
|
| 149 |
+
"loss": 2.4115,
|
| 150 |
+
"step": 200
|
| 151 |
+
},
|
| 152 |
+
{
|
| 153 |
+
"epoch": 0.13815789473684212,
|
| 154 |
+
"grad_norm": 16.093588917694532,
|
| 155 |
+
"learning_rate": 1.9642857142857145e-05,
|
| 156 |
+
"loss": 2.3548,
|
| 157 |
+
"step": 210
|
| 158 |
+
},
|
| 159 |
+
{
|
| 160 |
+
"epoch": 0.14473684210526316,
|
| 161 |
+
"grad_norm": 19.61654573509106,
|
| 162 |
+
"learning_rate": 2.0582706766917297e-05,
|
| 163 |
+
"loss": 2.3042,
|
| 164 |
+
"step": 220
|
| 165 |
+
},
|
| 166 |
+
{
|
| 167 |
+
"epoch": 0.1513157894736842,
|
| 168 |
+
"grad_norm": 12.268305710278183,
|
| 169 |
+
"learning_rate": 2.1522556390977445e-05,
|
| 170 |
+
"loss": 2.2494,
|
| 171 |
+
"step": 230
|
| 172 |
+
},
|
| 173 |
+
{
|
| 174 |
+
"epoch": 0.15789473684210525,
|
| 175 |
+
"grad_norm": 12.399788763360808,
|
| 176 |
+
"learning_rate": 2.2462406015037596e-05,
|
| 177 |
+
"loss": 2.1853,
|
| 178 |
+
"step": 240
|
| 179 |
+
},
|
| 180 |
+
{
|
| 181 |
+
"epoch": 0.16447368421052633,
|
| 182 |
+
"grad_norm": 7.463420275410263,
|
| 183 |
+
"learning_rate": 2.3402255639097744e-05,
|
| 184 |
+
"loss": 2.0998,
|
| 185 |
+
"step": 250
|
| 186 |
+
},
|
| 187 |
+
{
|
| 188 |
+
"epoch": 0.17105263157894737,
|
| 189 |
+
"grad_norm": 12.087535187997679,
|
| 190 |
+
"learning_rate": 2.4342105263157896e-05,
|
| 191 |
+
"loss": 2.0075,
|
| 192 |
+
"step": 260
|
| 193 |
+
},
|
| 194 |
+
{
|
| 195 |
+
"epoch": 0.17763157894736842,
|
| 196 |
+
"grad_norm": 6.62910919540051,
|
| 197 |
+
"learning_rate": 2.5281954887218044e-05,
|
| 198 |
+
"loss": 1.8974,
|
| 199 |
+
"step": 270
|
| 200 |
+
},
|
| 201 |
+
{
|
| 202 |
+
"epoch": 0.18421052631578946,
|
| 203 |
+
"grad_norm": 8.989470792407845,
|
| 204 |
+
"learning_rate": 2.62218045112782e-05,
|
| 205 |
+
"loss": 1.7928,
|
| 206 |
+
"step": 280
|
| 207 |
+
},
|
| 208 |
+
{
|
| 209 |
+
"epoch": 0.19078947368421054,
|
| 210 |
+
"grad_norm": 6.928524369383665,
|
| 211 |
+
"learning_rate": 2.716165413533835e-05,
|
| 212 |
+
"loss": 1.7057,
|
| 213 |
+
"step": 290
|
| 214 |
+
},
|
| 215 |
+
{
|
| 216 |
+
"epoch": 0.19736842105263158,
|
| 217 |
+
"grad_norm": 10.311746680652767,
|
| 218 |
+
"learning_rate": 2.8101503759398495e-05,
|
| 219 |
+
"loss": 1.6251,
|
| 220 |
+
"step": 300
|
| 221 |
+
},
|
| 222 |
+
{
|
| 223 |
+
"epoch": 0.20394736842105263,
|
| 224 |
+
"grad_norm": 7.107313244759928,
|
| 225 |
+
"learning_rate": 2.9041353383458647e-05,
|
| 226 |
+
"loss": 1.5522,
|
| 227 |
+
"step": 310
|
| 228 |
+
},
|
| 229 |
+
{
|
| 230 |
+
"epoch": 0.21052631578947367,
|
| 231 |
+
"grad_norm": 5.8857302431981156,
|
| 232 |
+
"learning_rate": 2.9981203007518798e-05,
|
| 233 |
+
"loss": 1.4974,
|
| 234 |
+
"step": 320
|
| 235 |
+
},
|
| 236 |
+
{
|
| 237 |
+
"epoch": 0.21710526315789475,
|
| 238 |
+
"grad_norm": 6.775917238582609,
|
| 239 |
+
"learning_rate": 3.092105263157895e-05,
|
| 240 |
+
"loss": 1.4355,
|
| 241 |
+
"step": 330
|
| 242 |
+
},
|
| 243 |
+
{
|
| 244 |
+
"epoch": 0.2236842105263158,
|
| 245 |
+
"grad_norm": 9.938119729849578,
|
| 246 |
+
"learning_rate": 3.1860902255639094e-05,
|
| 247 |
+
"loss": 1.3895,
|
| 248 |
+
"step": 340
|
| 249 |
+
},
|
| 250 |
+
{
|
| 251 |
+
"epoch": 0.23026315789473684,
|
| 252 |
+
"grad_norm": 4.870796794495703,
|
| 253 |
+
"learning_rate": 3.2800751879699246e-05,
|
| 254 |
+
"loss": 1.3432,
|
| 255 |
+
"step": 350
|
| 256 |
+
},
|
| 257 |
+
{
|
| 258 |
+
"epoch": 0.23684210526315788,
|
| 259 |
+
"grad_norm": 5.353011232164181,
|
| 260 |
+
"learning_rate": 3.37406015037594e-05,
|
| 261 |
+
"loss": 1.3002,
|
| 262 |
+
"step": 360
|
| 263 |
+
},
|
| 264 |
+
{
|
| 265 |
+
"epoch": 0.24342105263157895,
|
| 266 |
+
"grad_norm": 7.078794515741611,
|
| 267 |
+
"learning_rate": 3.468045112781955e-05,
|
| 268 |
+
"loss": 1.2641,
|
| 269 |
+
"step": 370
|
| 270 |
+
},
|
| 271 |
+
{
|
| 272 |
+
"epoch": 0.25,
|
| 273 |
+
"grad_norm": 5.419600483372193,
|
| 274 |
+
"learning_rate": 3.56203007518797e-05,
|
| 275 |
+
"loss": 1.209,
|
| 276 |
+
"step": 380
|
| 277 |
+
},
|
| 278 |
+
{
|
| 279 |
+
"epoch": 0.2565789473684211,
|
| 280 |
+
"grad_norm": 3.5327527407457926,
|
| 281 |
+
"learning_rate": 3.656015037593985e-05,
|
| 282 |
+
"loss": 1.1644,
|
| 283 |
+
"step": 390
|
| 284 |
+
},
|
| 285 |
+
{
|
| 286 |
+
"epoch": 0.2631578947368421,
|
| 287 |
+
"grad_norm": 5.894523313050658,
|
| 288 |
+
"learning_rate": 3.7500000000000003e-05,
|
| 289 |
+
"loss": 1.1319,
|
| 290 |
+
"step": 400
|
| 291 |
+
},
|
| 292 |
+
{
|
| 293 |
+
"epoch": 0.26973684210526316,
|
| 294 |
+
"grad_norm": 3.9863193805342316,
|
| 295 |
+
"learning_rate": 3.8439849624060155e-05,
|
| 296 |
+
"loss": 1.094,
|
| 297 |
+
"step": 410
|
| 298 |
+
},
|
| 299 |
+
{
|
| 300 |
+
"epoch": 0.27631578947368424,
|
| 301 |
+
"grad_norm": 4.225678742387179,
|
| 302 |
+
"learning_rate": 3.9379699248120307e-05,
|
| 303 |
+
"loss": 1.0651,
|
| 304 |
+
"step": 420
|
| 305 |
+
},
|
| 306 |
+
{
|
| 307 |
+
"epoch": 0.28289473684210525,
|
| 308 |
+
"grad_norm": 5.586389531542015,
|
| 309 |
+
"learning_rate": 4.031954887218045e-05,
|
| 310 |
+
"loss": 1.0247,
|
| 311 |
+
"step": 430
|
| 312 |
+
},
|
| 313 |
+
{
|
| 314 |
+
"epoch": 0.2894736842105263,
|
| 315 |
+
"grad_norm": 7.0928476432315675,
|
| 316 |
+
"learning_rate": 4.12593984962406e-05,
|
| 317 |
+
"loss": 0.9973,
|
| 318 |
+
"step": 440
|
| 319 |
+
},
|
| 320 |
+
{
|
| 321 |
+
"epoch": 0.29605263157894735,
|
| 322 |
+
"grad_norm": 5.797737179453841,
|
| 323 |
+
"learning_rate": 4.2199248120300754e-05,
|
| 324 |
+
"loss": 0.9738,
|
| 325 |
+
"step": 450
|
| 326 |
+
},
|
| 327 |
+
{
|
| 328 |
+
"epoch": 0.3026315789473684,
|
| 329 |
+
"grad_norm": 5.436160406452568,
|
| 330 |
+
"learning_rate": 4.3139097744360906e-05,
|
| 331 |
+
"loss": 0.9539,
|
| 332 |
+
"step": 460
|
| 333 |
+
},
|
| 334 |
+
{
|
| 335 |
+
"epoch": 0.3092105263157895,
|
| 336 |
+
"grad_norm": 4.940473111611711,
|
| 337 |
+
"learning_rate": 4.407894736842105e-05,
|
| 338 |
+
"loss": 0.9197,
|
| 339 |
+
"step": 470
|
| 340 |
+
},
|
| 341 |
+
{
|
| 342 |
+
"epoch": 0.3157894736842105,
|
| 343 |
+
"grad_norm": 5.844247327503255,
|
| 344 |
+
"learning_rate": 4.50187969924812e-05,
|
| 345 |
+
"loss": 0.8862,
|
| 346 |
+
"step": 480
|
| 347 |
+
},
|
| 348 |
+
{
|
| 349 |
+
"epoch": 0.3223684210526316,
|
| 350 |
+
"grad_norm": 4.744124381908569,
|
| 351 |
+
"learning_rate": 4.5958646616541354e-05,
|
| 352 |
+
"loss": 0.8668,
|
| 353 |
+
"step": 490
|
| 354 |
+
},
|
| 355 |
+
{
|
| 356 |
+
"epoch": 0.32894736842105265,
|
| 357 |
+
"grad_norm": 4.689028836800058,
|
| 358 |
+
"learning_rate": 4.6898496240601505e-05,
|
| 359 |
+
"loss": 0.8476,
|
| 360 |
+
"step": 500
|
| 361 |
+
},
|
| 362 |
+
{
|
| 363 |
+
"epoch": 0.3355263157894737,
|
| 364 |
+
"grad_norm": 4.398308995209079,
|
| 365 |
+
"learning_rate": 4.783834586466166e-05,
|
| 366 |
+
"loss": 0.82,
|
| 367 |
+
"step": 510
|
| 368 |
+
},
|
| 369 |
+
{
|
| 370 |
+
"epoch": 0.34210526315789475,
|
| 371 |
+
"grad_norm": 5.554811082819198,
|
| 372 |
+
"learning_rate": 4.877819548872181e-05,
|
| 373 |
+
"loss": 0.7974,
|
| 374 |
+
"step": 520
|
| 375 |
+
},
|
| 376 |
+
{
|
| 377 |
+
"epoch": 0.34868421052631576,
|
| 378 |
+
"grad_norm": 3.7243976034524584,
|
| 379 |
+
"learning_rate": 4.971804511278196e-05,
|
| 380 |
+
"loss": 0.7715,
|
| 381 |
+
"step": 530
|
| 382 |
+
},
|
| 383 |
+
{
|
| 384 |
+
"epoch": 0.35526315789473684,
|
| 385 |
+
"grad_norm": 4.879577435611285,
|
| 386 |
+
"learning_rate": 5.0657894736842104e-05,
|
| 387 |
+
"loss": 0.7514,
|
| 388 |
+
"step": 540
|
| 389 |
+
},
|
| 390 |
+
{
|
| 391 |
+
"epoch": 0.3618421052631579,
|
| 392 |
+
"grad_norm": 5.516075231521318,
|
| 393 |
+
"learning_rate": 5.1597744360902256e-05,
|
| 394 |
+
"loss": 0.7291,
|
| 395 |
+
"step": 550
|
| 396 |
+
},
|
| 397 |
+
{
|
| 398 |
+
"epoch": 0.3684210526315789,
|
| 399 |
+
"grad_norm": 5.494655015332771,
|
| 400 |
+
"learning_rate": 5.253759398496241e-05,
|
| 401 |
+
"loss": 0.6974,
|
| 402 |
+
"step": 560
|
| 403 |
+
},
|
| 404 |
+
{
|
| 405 |
+
"epoch": 0.375,
|
| 406 |
+
"grad_norm": 5.0610034411984826,
|
| 407 |
+
"learning_rate": 5.347744360902256e-05,
|
| 408 |
+
"loss": 0.6779,
|
| 409 |
+
"step": 570
|
| 410 |
+
},
|
| 411 |
+
{
|
| 412 |
+
"epoch": 0.3815789473684211,
|
| 413 |
+
"grad_norm": 6.241132985448258,
|
| 414 |
+
"learning_rate": 5.441729323308271e-05,
|
| 415 |
+
"loss": 0.6723,
|
| 416 |
+
"step": 580
|
| 417 |
+
},
|
| 418 |
+
{
|
| 419 |
+
"epoch": 0.3881578947368421,
|
| 420 |
+
"grad_norm": 4.582816961569543,
|
| 421 |
+
"learning_rate": 5.535714285714286e-05,
|
| 422 |
+
"loss": 0.652,
|
| 423 |
+
"step": 590
|
| 424 |
+
},
|
| 425 |
+
{
|
| 426 |
+
"epoch": 0.39473684210526316,
|
| 427 |
+
"grad_norm": 3.5845708789821042,
|
| 428 |
+
"learning_rate": 5.6296992481203013e-05,
|
| 429 |
+
"loss": 0.6154,
|
| 430 |
+
"step": 600
|
| 431 |
+
},
|
| 432 |
+
{
|
| 433 |
+
"epoch": 0.40131578947368424,
|
| 434 |
+
"grad_norm": 4.328768044098044,
|
| 435 |
+
"learning_rate": 5.723684210526316e-05,
|
| 436 |
+
"loss": 0.6042,
|
| 437 |
+
"step": 610
|
| 438 |
+
},
|
| 439 |
+
{
|
| 440 |
+
"epoch": 0.40789473684210525,
|
| 441 |
+
"grad_norm": 3.5473602851751793,
|
| 442 |
+
"learning_rate": 5.817669172932331e-05,
|
| 443 |
+
"loss": 0.5803,
|
| 444 |
+
"step": 620
|
| 445 |
+
},
|
| 446 |
+
{
|
| 447 |
+
"epoch": 0.4144736842105263,
|
| 448 |
+
"grad_norm": 4.233181045605189,
|
| 449 |
+
"learning_rate": 5.911654135338346e-05,
|
| 450 |
+
"loss": 0.5688,
|
| 451 |
+
"step": 630
|
| 452 |
+
},
|
| 453 |
+
{
|
| 454 |
+
"epoch": 0.42105263157894735,
|
| 455 |
+
"grad_norm": 4.0831509210740835,
|
| 456 |
+
"learning_rate": 6.005639097744361e-05,
|
| 457 |
+
"loss": 0.5385,
|
| 458 |
+
"step": 640
|
| 459 |
+
},
|
| 460 |
+
{
|
| 461 |
+
"epoch": 0.4276315789473684,
|
| 462 |
+
"grad_norm": 3.8705204303765814,
|
| 463 |
+
"learning_rate": 6.0996240601503764e-05,
|
| 464 |
+
"loss": 0.5201,
|
| 465 |
+
"step": 650
|
| 466 |
+
},
|
| 467 |
+
{
|
| 468 |
+
"epoch": 0.4342105263157895,
|
| 469 |
+
"grad_norm": 3.8269827275761337,
|
| 470 |
+
"learning_rate": 6.193609022556392e-05,
|
| 471 |
+
"loss": 0.504,
|
| 472 |
+
"step": 660
|
| 473 |
+
},
|
| 474 |
+
{
|
| 475 |
+
"epoch": 0.4407894736842105,
|
| 476 |
+
"grad_norm": 3.7042333223437476,
|
| 477 |
+
"learning_rate": 6.287593984962407e-05,
|
| 478 |
+
"loss": 0.485,
|
| 479 |
+
"step": 670
|
| 480 |
+
},
|
| 481 |
+
{
|
| 482 |
+
"epoch": 0.4473684210526316,
|
| 483 |
+
"grad_norm": 3.1454584376141725,
|
| 484 |
+
"learning_rate": 6.381578947368422e-05,
|
| 485 |
+
"loss": 0.4692,
|
| 486 |
+
"step": 680
|
| 487 |
+
},
|
| 488 |
+
{
|
| 489 |
+
"epoch": 0.45394736842105265,
|
| 490 |
+
"grad_norm": 3.6715281367823698,
|
| 491 |
+
"learning_rate": 6.475563909774437e-05,
|
| 492 |
+
"loss": 0.4474,
|
| 493 |
+
"step": 690
|
| 494 |
+
},
|
| 495 |
+
{
|
| 496 |
+
"epoch": 0.4605263157894737,
|
| 497 |
+
"grad_norm": 3.3520165299765203,
|
| 498 |
+
"learning_rate": 6.569548872180451e-05,
|
| 499 |
+
"loss": 0.4335,
|
| 500 |
+
"step": 700
|
| 501 |
+
},
|
| 502 |
+
{
|
| 503 |
+
"epoch": 0.46710526315789475,
|
| 504 |
+
"grad_norm": 3.4830856151143554,
|
| 505 |
+
"learning_rate": 6.663533834586466e-05,
|
| 506 |
+
"loss": 0.4156,
|
| 507 |
+
"step": 710
|
| 508 |
+
},
|
| 509 |
+
{
|
| 510 |
+
"epoch": 0.47368421052631576,
|
| 511 |
+
"grad_norm": 3.4130338264227835,
|
| 512 |
+
"learning_rate": 6.757518796992481e-05,
|
| 513 |
+
"loss": 0.3984,
|
| 514 |
+
"step": 720
|
| 515 |
+
},
|
| 516 |
+
{
|
| 517 |
+
"epoch": 0.48026315789473684,
|
| 518 |
+
"grad_norm": 3.6699720598135834,
|
| 519 |
+
"learning_rate": 6.851503759398496e-05,
|
| 520 |
+
"loss": 0.3884,
|
| 521 |
+
"step": 730
|
| 522 |
+
},
|
| 523 |
+
{
|
| 524 |
+
"epoch": 0.4868421052631579,
|
| 525 |
+
"grad_norm": 3.2598730307422192,
|
| 526 |
+
"learning_rate": 6.945488721804511e-05,
|
| 527 |
+
"loss": 0.3727,
|
| 528 |
+
"step": 740
|
| 529 |
+
},
|
| 530 |
+
{
|
| 531 |
+
"epoch": 0.4934210526315789,
|
| 532 |
+
"grad_norm": 3.0065146921519097,
|
| 533 |
+
"learning_rate": 7.039473684210527e-05,
|
| 534 |
+
"loss": 0.3521,
|
| 535 |
+
"step": 750
|
| 536 |
+
},
|
| 537 |
+
{
|
| 538 |
+
"epoch": 0.5,
|
| 539 |
+
"grad_norm": 2.3801132893266956,
|
| 540 |
+
"learning_rate": 7.133458646616542e-05,
|
| 541 |
+
"loss": 0.3415,
|
| 542 |
+
"step": 760
|
| 543 |
+
},
|
| 544 |
+
{
|
| 545 |
+
"epoch": 0.506578947368421,
|
| 546 |
+
"grad_norm": 2.5040976105308466,
|
| 547 |
+
"learning_rate": 7.227443609022557e-05,
|
| 548 |
+
"loss": 0.3303,
|
| 549 |
+
"step": 770
|
| 550 |
+
},
|
| 551 |
+
{
|
| 552 |
+
"epoch": 0.5131578947368421,
|
| 553 |
+
"grad_norm": 2.3095873713136004,
|
| 554 |
+
"learning_rate": 7.321428571428571e-05,
|
| 555 |
+
"loss": 0.3155,
|
| 556 |
+
"step": 780
|
| 557 |
+
},
|
| 558 |
+
{
|
| 559 |
+
"epoch": 0.5197368421052632,
|
| 560 |
+
"grad_norm": 2.221909098699267,
|
| 561 |
+
"learning_rate": 7.415413533834586e-05,
|
| 562 |
+
"loss": 0.3,
|
| 563 |
+
"step": 790
|
| 564 |
+
},
|
| 565 |
+
{
|
| 566 |
+
"epoch": 0.5263157894736842,
|
| 567 |
+
"grad_norm": 2.3650008130107008,
|
| 568 |
+
"learning_rate": 7.509398496240601e-05,
|
| 569 |
+
"loss": 0.2881,
|
| 570 |
+
"step": 800
|
| 571 |
+
},
|
| 572 |
+
{
|
| 573 |
+
"epoch": 0.5328947368421053,
|
| 574 |
+
"grad_norm": 2.92228181953594,
|
| 575 |
+
"learning_rate": 7.603383458646618e-05,
|
| 576 |
+
"loss": 0.2759,
|
| 577 |
+
"step": 810
|
| 578 |
+
},
|
| 579 |
+
{
|
| 580 |
+
"epoch": 0.5394736842105263,
|
| 581 |
+
"grad_norm": 2.39576316770683,
|
| 582 |
+
"learning_rate": 7.697368421052633e-05,
|
| 583 |
+
"loss": 0.2611,
|
| 584 |
+
"step": 820
|
| 585 |
+
},
|
| 586 |
+
{
|
| 587 |
+
"epoch": 0.5460526315789473,
|
| 588 |
+
"grad_norm": 1.9345517046210943,
|
| 589 |
+
"learning_rate": 7.791353383458648e-05,
|
| 590 |
+
"loss": 0.2472,
|
| 591 |
+
"step": 830
|
| 592 |
+
},
|
| 593 |
+
{
|
| 594 |
+
"epoch": 0.5526315789473685,
|
| 595 |
+
"grad_norm": 2.321485582474392,
|
| 596 |
+
"learning_rate": 7.885338345864663e-05,
|
| 597 |
+
"loss": 0.2409,
|
| 598 |
+
"step": 840
|
| 599 |
+
},
|
| 600 |
+
{
|
| 601 |
+
"epoch": 0.5592105263157895,
|
| 602 |
+
"grad_norm": 1.8690657258212422,
|
| 603 |
+
"learning_rate": 7.979323308270678e-05,
|
| 604 |
+
"loss": 0.2308,
|
| 605 |
+
"step": 850
|
| 606 |
+
},
|
| 607 |
+
{
|
| 608 |
+
"epoch": 0.5657894736842105,
|
| 609 |
+
"grad_norm": 1.901049780468097,
|
| 610 |
+
"learning_rate": 8.073308270676692e-05,
|
| 611 |
+
"loss": 0.2171,
|
| 612 |
+
"step": 860
|
| 613 |
+
},
|
| 614 |
+
{
|
| 615 |
+
"epoch": 0.5723684210526315,
|
| 616 |
+
"grad_norm": 2.2229311039801596,
|
| 617 |
+
"learning_rate": 8.167293233082707e-05,
|
| 618 |
+
"loss": 0.2072,
|
| 619 |
+
"step": 870
|
| 620 |
+
},
|
| 621 |
+
{
|
| 622 |
+
"epoch": 0.5789473684210527,
|
| 623 |
+
"grad_norm": 2.213221418124993,
|
| 624 |
+
"learning_rate": 8.261278195488722e-05,
|
| 625 |
+
"loss": 0.1965,
|
| 626 |
+
"step": 880
|
| 627 |
+
},
|
| 628 |
+
{
|
| 629 |
+
"epoch": 0.5855263157894737,
|
| 630 |
+
"grad_norm": 1.870540842703205,
|
| 631 |
+
"learning_rate": 8.355263157894737e-05,
|
| 632 |
+
"loss": 0.1877,
|
| 633 |
+
"step": 890
|
| 634 |
+
},
|
| 635 |
+
{
|
| 636 |
+
"epoch": 0.5921052631578947,
|
| 637 |
+
"grad_norm": 1.311680351399902,
|
| 638 |
+
"learning_rate": 8.449248120300753e-05,
|
| 639 |
+
"loss": 0.1792,
|
| 640 |
+
"step": 900
|
| 641 |
+
},
|
| 642 |
+
{
|
| 643 |
+
"epoch": 0.5986842105263158,
|
| 644 |
+
"grad_norm": 1.5798663811679752,
|
| 645 |
+
"learning_rate": 8.543233082706768e-05,
|
| 646 |
+
"loss": 0.1716,
|
| 647 |
+
"step": 910
|
| 648 |
+
},
|
| 649 |
+
{
|
| 650 |
+
"epoch": 0.6052631578947368,
|
| 651 |
+
"grad_norm": 7.169134004434136,
|
| 652 |
+
"learning_rate": 8.637218045112783e-05,
|
| 653 |
+
"loss": 0.1721,
|
| 654 |
+
"step": 920
|
| 655 |
+
},
|
| 656 |
+
{
|
| 657 |
+
"epoch": 0.6118421052631579,
|
| 658 |
+
"grad_norm": 1.5044155665130965,
|
| 659 |
+
"learning_rate": 8.731203007518798e-05,
|
| 660 |
+
"loss": 0.193,
|
| 661 |
+
"step": 930
|
| 662 |
+
},
|
| 663 |
+
{
|
| 664 |
+
"epoch": 0.618421052631579,
|
| 665 |
+
"grad_norm": 1.305309283144098,
|
| 666 |
+
"learning_rate": 8.825187969924813e-05,
|
| 667 |
+
"loss": 0.1566,
|
| 668 |
+
"step": 940
|
| 669 |
+
},
|
| 670 |
+
{
|
| 671 |
+
"epoch": 0.625,
|
| 672 |
+
"grad_norm": 1.23686939558907,
|
| 673 |
+
"learning_rate": 8.919172932330827e-05,
|
| 674 |
+
"loss": 0.1445,
|
| 675 |
+
"step": 950
|
| 676 |
+
},
|
| 677 |
+
{
|
| 678 |
+
"epoch": 0.631578947368421,
|
| 679 |
+
"grad_norm": 1.0445490067006857,
|
| 680 |
+
"learning_rate": 9.013157894736842e-05,
|
| 681 |
+
"loss": 0.1356,
|
| 682 |
+
"step": 960
|
| 683 |
+
},
|
| 684 |
+
{
|
| 685 |
+
"epoch": 0.6381578947368421,
|
| 686 |
+
"grad_norm": 1.3136634908844078,
|
| 687 |
+
"learning_rate": 9.107142857142857e-05,
|
| 688 |
+
"loss": 0.1308,
|
| 689 |
+
"step": 970
|
| 690 |
+
},
|
| 691 |
+
{
|
| 692 |
+
"epoch": 0.6447368421052632,
|
| 693 |
+
"grad_norm": 1.7716809252302665,
|
| 694 |
+
"learning_rate": 9.201127819548872e-05,
|
| 695 |
+
"loss": 0.124,
|
| 696 |
+
"step": 980
|
| 697 |
+
},
|
| 698 |
+
{
|
| 699 |
+
"epoch": 0.6513157894736842,
|
| 700 |
+
"grad_norm": 1.0772175167269311,
|
| 701 |
+
"learning_rate": 9.295112781954888e-05,
|
| 702 |
+
"loss": 0.1202,
|
| 703 |
+
"step": 990
|
| 704 |
+
},
|
| 705 |
+
{
|
| 706 |
+
"epoch": 0.6578947368421053,
|
| 707 |
+
"grad_norm": 1.1878390185089365,
|
| 708 |
+
"learning_rate": 9.389097744360903e-05,
|
| 709 |
+
"loss": 0.114,
|
| 710 |
+
"step": 1000
|
| 711 |
+
},
|
| 712 |
+
{
|
| 713 |
+
"epoch": 0.6644736842105263,
|
| 714 |
+
"grad_norm": 0.8915150933559234,
|
| 715 |
+
"learning_rate": 9.483082706766918e-05,
|
| 716 |
+
"loss": 0.1054,
|
| 717 |
+
"step": 1010
|
| 718 |
+
},
|
| 719 |
+
{
|
| 720 |
+
"epoch": 0.6710526315789473,
|
| 721 |
+
"grad_norm": 0.7818448157917527,
|
| 722 |
+
"learning_rate": 9.577067669172933e-05,
|
| 723 |
+
"loss": 0.0996,
|
| 724 |
+
"step": 1020
|
| 725 |
+
},
|
| 726 |
+
{
|
| 727 |
+
"epoch": 0.6776315789473685,
|
| 728 |
+
"grad_norm": 0.990430969229284,
|
| 729 |
+
"learning_rate": 9.671052631578947e-05,
|
| 730 |
+
"loss": 0.0974,
|
| 731 |
+
"step": 1030
|
| 732 |
+
},
|
| 733 |
+
{
|
| 734 |
+
"epoch": 0.6842105263157895,
|
| 735 |
+
"grad_norm": 0.9276518666370865,
|
| 736 |
+
"learning_rate": 9.765037593984962e-05,
|
| 737 |
+
"loss": 0.0933,
|
| 738 |
+
"step": 1040
|
| 739 |
+
},
|
| 740 |
+
{
|
| 741 |
+
"epoch": 0.6907894736842105,
|
| 742 |
+
"grad_norm": 0.6846215892309722,
|
| 743 |
+
"learning_rate": 9.859022556390977e-05,
|
| 744 |
+
"loss": 0.0874,
|
| 745 |
+
"step": 1050
|
| 746 |
+
},
|
| 747 |
+
{
|
| 748 |
+
"epoch": 0.6973684210526315,
|
| 749 |
+
"grad_norm": 0.714553023496577,
|
| 750 |
+
"learning_rate": 9.953007518796992e-05,
|
| 751 |
+
"loss": 0.0814,
|
| 752 |
+
"step": 1060
|
| 753 |
+
},
|
| 754 |
+
{
|
| 755 |
+
"epoch": 0.7039473684210527,
|
| 756 |
+
"grad_norm": 0.7546346945965402,
|
| 757 |
+
"learning_rate": 9.999993273155467e-05,
|
| 758 |
+
"loss": 0.0774,
|
| 759 |
+
"step": 1070
|
| 760 |
+
},
|
| 761 |
+
{
|
| 762 |
+
"epoch": 0.7105263157894737,
|
| 763 |
+
"grad_norm": 0.571192300635228,
|
| 764 |
+
"learning_rate": 9.999939458507816e-05,
|
| 765 |
+
"loss": 0.0762,
|
| 766 |
+
"step": 1080
|
| 767 |
+
},
|
| 768 |
+
{
|
| 769 |
+
"epoch": 0.7171052631578947,
|
| 770 |
+
"grad_norm": 0.8584193322720889,
|
| 771 |
+
"learning_rate": 9.999831829791713e-05,
|
| 772 |
+
"loss": 0.0756,
|
| 773 |
+
"step": 1090
|
| 774 |
+
},
|
| 775 |
+
{
|
| 776 |
+
"epoch": 0.7236842105263158,
|
| 777 |
+
"grad_norm": 1.3952479453359226,
|
| 778 |
+
"learning_rate": 9.999670388165563e-05,
|
| 779 |
+
"loss": 0.075,
|
| 780 |
+
"step": 1100
|
| 781 |
+
},
|
| 782 |
+
{
|
| 783 |
+
"epoch": 0.7302631578947368,
|
| 784 |
+
"grad_norm": 0.751465926189691,
|
| 785 |
+
"learning_rate": 9.999455135366951e-05,
|
| 786 |
+
"loss": 0.069,
|
| 787 |
+
"step": 1110
|
| 788 |
+
},
|
| 789 |
+
{
|
| 790 |
+
"epoch": 0.7368421052631579,
|
| 791 |
+
"grad_norm": 0.4873110858150168,
|
| 792 |
+
"learning_rate": 9.999186073712635e-05,
|
| 793 |
+
"loss": 0.064,
|
| 794 |
+
"step": 1120
|
| 795 |
+
},
|
| 796 |
+
{
|
| 797 |
+
"epoch": 0.743421052631579,
|
| 798 |
+
"grad_norm": 0.4495642582063532,
|
| 799 |
+
"learning_rate": 9.998863206098507e-05,
|
| 800 |
+
"loss": 0.0623,
|
| 801 |
+
"step": 1130
|
| 802 |
+
},
|
| 803 |
+
{
|
| 804 |
+
"epoch": 0.75,
|
| 805 |
+
"grad_norm": 0.45684969056385805,
|
| 806 |
+
"learning_rate": 9.998486535999573e-05,
|
| 807 |
+
"loss": 0.059,
|
| 808 |
+
"step": 1140
|
| 809 |
+
},
|
| 810 |
+
{
|
| 811 |
+
"epoch": 0.756578947368421,
|
| 812 |
+
"grad_norm": 0.4304228051311271,
|
| 813 |
+
"learning_rate": 9.998056067469915e-05,
|
| 814 |
+
"loss": 0.0574,
|
| 815 |
+
"step": 1150
|
| 816 |
+
},
|
| 817 |
+
{
|
| 818 |
+
"epoch": 0.7631578947368421,
|
| 819 |
+
"grad_norm": 0.38692754350368835,
|
| 820 |
+
"learning_rate": 9.997571805142639e-05,
|
| 821 |
+
"loss": 0.0568,
|
| 822 |
+
"step": 1160
|
| 823 |
+
},
|
| 824 |
+
{
|
| 825 |
+
"epoch": 0.7697368421052632,
|
| 826 |
+
"grad_norm": 0.44697102978905795,
|
| 827 |
+
"learning_rate": 9.997033754229835e-05,
|
| 828 |
+
"loss": 0.054,
|
| 829 |
+
"step": 1170
|
| 830 |
+
},
|
| 831 |
+
{
|
| 832 |
+
"epoch": 0.7763157894736842,
|
| 833 |
+
"grad_norm": 0.3475319362475707,
|
| 834 |
+
"learning_rate": 9.996441920522512e-05,
|
| 835 |
+
"loss": 0.0512,
|
| 836 |
+
"step": 1180
|
| 837 |
+
},
|
| 838 |
+
{
|
| 839 |
+
"epoch": 0.7828947368421053,
|
| 840 |
+
"grad_norm": 0.3776964171490181,
|
| 841 |
+
"learning_rate": 9.995796310390547e-05,
|
| 842 |
+
"loss": 0.0496,
|
| 843 |
+
"step": 1190
|
| 844 |
+
},
|
| 845 |
+
{
|
| 846 |
+
"epoch": 0.7894736842105263,
|
| 847 |
+
"grad_norm": 0.3599383764053137,
|
| 848 |
+
"learning_rate": 9.995096930782604e-05,
|
| 849 |
+
"loss": 0.049,
|
| 850 |
+
"step": 1200
|
| 851 |
+
},
|
| 852 |
+
{
|
| 853 |
+
"epoch": 0.7960526315789473,
|
| 854 |
+
"grad_norm": 0.4004701605242826,
|
| 855 |
+
"learning_rate": 9.994343789226066e-05,
|
| 856 |
+
"loss": 0.0468,
|
| 857 |
+
"step": 1210
|
| 858 |
+
},
|
| 859 |
+
{
|
| 860 |
+
"epoch": 0.8026315789473685,
|
| 861 |
+
"grad_norm": 0.3822329385724864,
|
| 862 |
+
"learning_rate": 9.993536893826954e-05,
|
| 863 |
+
"loss": 0.0457,
|
| 864 |
+
"step": 1220
|
| 865 |
+
},
|
| 866 |
+
{
|
| 867 |
+
"epoch": 0.8092105263157895,
|
| 868 |
+
"grad_norm": 0.3620248875103048,
|
| 869 |
+
"learning_rate": 9.992676253269837e-05,
|
| 870 |
+
"loss": 0.0446,
|
| 871 |
+
"step": 1230
|
| 872 |
+
},
|
| 873 |
+
{
|
| 874 |
+
"epoch": 0.8157894736842105,
|
| 875 |
+
"grad_norm": 0.28358957764617276,
|
| 876 |
+
"learning_rate": 9.991761876817743e-05,
|
| 877 |
+
"loss": 0.0429,
|
| 878 |
+
"step": 1240
|
| 879 |
+
},
|
| 880 |
+
{
|
| 881 |
+
"epoch": 0.8223684210526315,
|
| 882 |
+
"grad_norm": 0.32929903528232773,
|
| 883 |
+
"learning_rate": 9.990793774312053e-05,
|
| 884 |
+
"loss": 0.0421,
|
| 885 |
+
"step": 1250
|
| 886 |
+
},
|
| 887 |
+
{
|
| 888 |
+
"epoch": 0.8289473684210527,
|
| 889 |
+
"grad_norm": 0.34141322063560897,
|
| 890 |
+
"learning_rate": 9.9897719561724e-05,
|
| 891 |
+
"loss": 0.0418,
|
| 892 |
+
"step": 1260
|
| 893 |
+
},
|
| 894 |
+
{
|
| 895 |
+
"epoch": 0.8355263157894737,
|
| 896 |
+
"grad_norm": 0.28296527941759625,
|
| 897 |
+
"learning_rate": 9.988696433396557e-05,
|
| 898 |
+
"loss": 0.0403,
|
| 899 |
+
"step": 1270
|
| 900 |
+
},
|
| 901 |
+
{
|
| 902 |
+
"epoch": 0.8421052631578947,
|
| 903 |
+
"grad_norm": 0.3686897659419288,
|
| 904 |
+
"learning_rate": 9.987567217560314e-05,
|
| 905 |
+
"loss": 0.0395,
|
| 906 |
+
"step": 1280
|
| 907 |
+
},
|
| 908 |
+
{
|
| 909 |
+
"epoch": 0.8486842105263158,
|
| 910 |
+
"grad_norm": 0.3066999773216274,
|
| 911 |
+
"learning_rate": 9.986384320817359e-05,
|
| 912 |
+
"loss": 0.0395,
|
| 913 |
+
"step": 1290
|
| 914 |
+
},
|
| 915 |
+
{
|
| 916 |
+
"epoch": 0.8552631578947368,
|
| 917 |
+
"grad_norm": 0.33225917789321974,
|
| 918 |
+
"learning_rate": 9.985147755899142e-05,
|
| 919 |
+
"loss": 0.0385,
|
| 920 |
+
"step": 1300
|
| 921 |
+
},
|
| 922 |
+
{
|
| 923 |
+
"epoch": 0.8618421052631579,
|
| 924 |
+
"grad_norm": 0.3004075266809671,
|
| 925 |
+
"learning_rate": 9.983857536114743e-05,
|
| 926 |
+
"loss": 0.0377,
|
| 927 |
+
"step": 1310
|
| 928 |
+
},
|
| 929 |
+
{
|
| 930 |
+
"epoch": 0.868421052631579,
|
| 931 |
+
"grad_norm": 0.2698302147467991,
|
| 932 |
+
"learning_rate": 9.982513675350724e-05,
|
| 933 |
+
"loss": 0.0367,
|
| 934 |
+
"step": 1320
|
| 935 |
+
},
|
| 936 |
+
{
|
| 937 |
+
"epoch": 0.875,
|
| 938 |
+
"grad_norm": 0.26807280624783797,
|
| 939 |
+
"learning_rate": 9.981116188070984e-05,
|
| 940 |
+
"loss": 0.0356,
|
| 941 |
+
"step": 1330
|
| 942 |
+
},
|
| 943 |
+
{
|
| 944 |
+
"epoch": 0.881578947368421,
|
| 945 |
+
"grad_norm": 0.3127969801843736,
|
| 946 |
+
"learning_rate": 9.979665089316602e-05,
|
| 947 |
+
"loss": 0.035,
|
| 948 |
+
"step": 1340
|
| 949 |
+
},
|
| 950 |
+
{
|
| 951 |
+
"epoch": 0.8881578947368421,
|
| 952 |
+
"grad_norm": 0.28653732054861003,
|
| 953 |
+
"learning_rate": 9.978160394705668e-05,
|
| 954 |
+
"loss": 0.0343,
|
| 955 |
+
"step": 1350
|
| 956 |
+
},
|
| 957 |
+
{
|
| 958 |
+
"epoch": 0.8947368421052632,
|
| 959 |
+
"grad_norm": 0.2504478087548026,
|
| 960 |
+
"learning_rate": 9.976602120433133e-05,
|
| 961 |
+
"loss": 0.0341,
|
| 962 |
+
"step": 1360
|
| 963 |
+
},
|
| 964 |
+
{
|
| 965 |
+
"epoch": 0.9013157894736842,
|
| 966 |
+
"grad_norm": 0.24532062538161709,
|
| 967 |
+
"learning_rate": 9.97499028327061e-05,
|
| 968 |
+
"loss": 0.0331,
|
| 969 |
+
"step": 1370
|
| 970 |
+
},
|
| 971 |
+
{
|
| 972 |
+
"epoch": 0.9078947368421053,
|
| 973 |
+
"grad_norm": 0.5595503828448865,
|
| 974 |
+
"learning_rate": 9.973324900566213e-05,
|
| 975 |
+
"loss": 0.041,
|
| 976 |
+
"step": 1380
|
| 977 |
+
},
|
| 978 |
+
{
|
| 979 |
+
"epoch": 0.9144736842105263,
|
| 980 |
+
"grad_norm": 0.32103277633584737,
|
| 981 |
+
"learning_rate": 9.971605990244365e-05,
|
| 982 |
+
"loss": 0.0396,
|
| 983 |
+
"step": 1390
|
| 984 |
+
},
|
| 985 |
+
{
|
| 986 |
+
"epoch": 0.9210526315789473,
|
| 987 |
+
"grad_norm": 0.39243105455264066,
|
| 988 |
+
"learning_rate": 9.969833570805599e-05,
|
| 989 |
+
"loss": 0.0345,
|
| 990 |
+
"step": 1400
|
| 991 |
+
},
|
| 992 |
+
{
|
| 993 |
+
"epoch": 0.9276315789473685,
|
| 994 |
+
"grad_norm": 0.24537350120004972,
|
| 995 |
+
"learning_rate": 9.968007661326369e-05,
|
| 996 |
+
"loss": 0.0323,
|
| 997 |
+
"step": 1410
|
| 998 |
+
},
|
| 999 |
+
{
|
| 1000 |
+
"epoch": 0.9342105263157895,
|
| 1001 |
+
"grad_norm": 0.21282115870790244,
|
| 1002 |
+
"learning_rate": 9.966128281458832e-05,
|
| 1003 |
+
"loss": 0.0309,
|
| 1004 |
+
"step": 1420
|
| 1005 |
+
},
|
| 1006 |
+
{
|
| 1007 |
+
"epoch": 0.9407894736842105,
|
| 1008 |
+
"grad_norm": 0.22792100873990018,
|
| 1009 |
+
"learning_rate": 9.964195451430653e-05,
|
| 1010 |
+
"loss": 0.0302,
|
| 1011 |
+
"step": 1430
|
| 1012 |
+
},
|
| 1013 |
+
{
|
| 1014 |
+
"epoch": 0.9473684210526315,
|
| 1015 |
+
"grad_norm": 0.2513211642537743,
|
| 1016 |
+
"learning_rate": 9.962209192044772e-05,
|
| 1017 |
+
"loss": 0.0298,
|
| 1018 |
+
"step": 1440
|
| 1019 |
+
},
|
| 1020 |
+
{
|
| 1021 |
+
"epoch": 0.9539473684210527,
|
| 1022 |
+
"grad_norm": 0.25009559541845094,
|
| 1023 |
+
"learning_rate": 9.960169524679187e-05,
|
| 1024 |
+
"loss": 0.0293,
|
| 1025 |
+
"step": 1450
|
| 1026 |
+
},
|
| 1027 |
+
{
|
| 1028 |
+
"epoch": 0.9605263157894737,
|
| 1029 |
+
"grad_norm": 0.23293995789061867,
|
| 1030 |
+
"learning_rate": 9.958076471286724e-05,
|
| 1031 |
+
"loss": 0.0289,
|
| 1032 |
+
"step": 1460
|
| 1033 |
+
},
|
| 1034 |
+
{
|
| 1035 |
+
"epoch": 0.9671052631578947,
|
| 1036 |
+
"grad_norm": 0.22361624312785255,
|
| 1037 |
+
"learning_rate": 9.955930054394798e-05,
|
| 1038 |
+
"loss": 0.0282,
|
| 1039 |
+
"step": 1470
|
| 1040 |
+
},
|
| 1041 |
+
{
|
| 1042 |
+
"epoch": 0.9736842105263158,
|
| 1043 |
+
"grad_norm": 0.23906361425178743,
|
| 1044 |
+
"learning_rate": 9.953730297105175e-05,
|
| 1045 |
+
"loss": 0.0281,
|
| 1046 |
+
"step": 1480
|
| 1047 |
+
},
|
| 1048 |
+
{
|
| 1049 |
+
"epoch": 0.9802631578947368,
|
| 1050 |
+
"grad_norm": 0.23143391534241473,
|
| 1051 |
+
"learning_rate": 9.951477223093722e-05,
|
| 1052 |
+
"loss": 0.0275,
|
| 1053 |
+
"step": 1490
|
| 1054 |
+
},
|
| 1055 |
+
{
|
| 1056 |
+
"epoch": 0.9868421052631579,
|
| 1057 |
+
"grad_norm": 0.2147052689574608,
|
| 1058 |
+
"learning_rate": 9.949170856610144e-05,
|
| 1059 |
+
"loss": 0.0271,
|
| 1060 |
+
"step": 1500
|
| 1061 |
+
},
|
| 1062 |
+
{
|
| 1063 |
+
"epoch": 0.993421052631579,
|
| 1064 |
+
"grad_norm": 0.20377711124649098,
|
| 1065 |
+
"learning_rate": 9.946811222477737e-05,
|
| 1066 |
+
"loss": 0.0263,
|
| 1067 |
+
"step": 1510
|
| 1068 |
+
},
|
| 1069 |
+
{
|
| 1070 |
+
"epoch": 1.0,
|
| 1071 |
+
"grad_norm": 0.2135903768894096,
|
| 1072 |
+
"learning_rate": 9.944398346093109e-05,
|
| 1073 |
+
"loss": 0.0262,
|
| 1074 |
+
"step": 1520
|
| 1075 |
+
},
|
| 1076 |
+
{
|
| 1077 |
+
"epoch": 1.006578947368421,
|
| 1078 |
+
"grad_norm": 0.20254766138767108,
|
| 1079 |
+
"learning_rate": 9.941932253425917e-05,
|
| 1080 |
+
"loss": 0.0248,
|
| 1081 |
+
"step": 1530
|
| 1082 |
+
},
|
| 1083 |
+
{
|
| 1084 |
+
"epoch": 1.013157894736842,
|
| 1085 |
+
"grad_norm": 0.2197030932035537,
|
| 1086 |
+
"learning_rate": 9.939412971018574e-05,
|
| 1087 |
+
"loss": 0.0246,
|
| 1088 |
+
"step": 1540
|
| 1089 |
+
},
|
| 1090 |
+
{
|
| 1091 |
+
"epoch": 1.019736842105263,
|
| 1092 |
+
"grad_norm": 0.1933628837351623,
|
| 1093 |
+
"learning_rate": 9.936840525985978e-05,
|
| 1094 |
+
"loss": 0.0242,
|
| 1095 |
+
"step": 1550
|
| 1096 |
+
},
|
| 1097 |
+
{
|
| 1098 |
+
"epoch": 1.0263157894736843,
|
| 1099 |
+
"grad_norm": 0.19721213696918533,
|
| 1100 |
+
"learning_rate": 9.934214946015213e-05,
|
| 1101 |
+
"loss": 0.0238,
|
| 1102 |
+
"step": 1560
|
| 1103 |
+
},
|
| 1104 |
+
{
|
| 1105 |
+
"epoch": 1.0328947368421053,
|
| 1106 |
+
"grad_norm": 0.20774618691091581,
|
| 1107 |
+
"learning_rate": 9.931536259365244e-05,
|
| 1108 |
+
"loss": 0.0234,
|
| 1109 |
+
"step": 1570
|
| 1110 |
+
},
|
| 1111 |
+
{
|
| 1112 |
+
"epoch": 1.0394736842105263,
|
| 1113 |
+
"grad_norm": 0.19785742831408606,
|
| 1114 |
+
"learning_rate": 9.928804494866629e-05,
|
| 1115 |
+
"loss": 0.0228,
|
| 1116 |
+
"step": 1580
|
| 1117 |
+
},
|
| 1118 |
+
{
|
| 1119 |
+
"epoch": 1.0460526315789473,
|
| 1120 |
+
"grad_norm": 0.2035358859489929,
|
| 1121 |
+
"learning_rate": 9.926019681921196e-05,
|
| 1122 |
+
"loss": 0.0227,
|
| 1123 |
+
"step": 1590
|
| 1124 |
+
},
|
| 1125 |
+
{
|
| 1126 |
+
"epoch": 1.0526315789473684,
|
| 1127 |
+
"grad_norm": 0.20629560858059492,
|
| 1128 |
+
"learning_rate": 9.92318185050173e-05,
|
| 1129 |
+
"loss": 0.0227,
|
| 1130 |
+
"step": 1600
|
| 1131 |
+
},
|
| 1132 |
+
{
|
| 1133 |
+
"epoch": 1.0592105263157894,
|
| 1134 |
+
"grad_norm": 0.17943582514083126,
|
| 1135 |
+
"learning_rate": 9.920291031151652e-05,
|
| 1136 |
+
"loss": 0.0223,
|
| 1137 |
+
"step": 1610
|
| 1138 |
+
},
|
| 1139 |
+
{
|
| 1140 |
+
"epoch": 1.0657894736842106,
|
| 1141 |
+
"grad_norm": 0.2069047937737935,
|
| 1142 |
+
"learning_rate": 9.917347254984688e-05,
|
| 1143 |
+
"loss": 0.0219,
|
| 1144 |
+
"step": 1620
|
| 1145 |
+
},
|
| 1146 |
+
{
|
| 1147 |
+
"epoch": 1.0723684210526316,
|
| 1148 |
+
"grad_norm": 0.18149990279040348,
|
| 1149 |
+
"learning_rate": 9.91435055368454e-05,
|
| 1150 |
+
"loss": 0.0215,
|
| 1151 |
+
"step": 1630
|
| 1152 |
+
},
|
| 1153 |
+
{
|
| 1154 |
+
"epoch": 1.0789473684210527,
|
| 1155 |
+
"grad_norm": 0.21588259501861692,
|
| 1156 |
+
"learning_rate": 9.91130095950453e-05,
|
| 1157 |
+
"loss": 0.0216,
|
| 1158 |
+
"step": 1640
|
| 1159 |
+
},
|
| 1160 |
+
{
|
| 1161 |
+
"epoch": 1.0855263157894737,
|
| 1162 |
+
"grad_norm": 0.19248955273518795,
|
| 1163 |
+
"learning_rate": 9.908198505267273e-05,
|
| 1164 |
+
"loss": 0.0211,
|
| 1165 |
+
"step": 1650
|
| 1166 |
+
},
|
| 1167 |
+
{
|
| 1168 |
+
"epoch": 1.0921052631578947,
|
| 1169 |
+
"grad_norm": 0.21323425974286558,
|
| 1170 |
+
"learning_rate": 9.90504322436431e-05,
|
| 1171 |
+
"loss": 0.0212,
|
| 1172 |
+
"step": 1660
|
| 1173 |
+
},
|
| 1174 |
+
{
|
| 1175 |
+
"epoch": 1.0986842105263157,
|
| 1176 |
+
"grad_norm": 0.19659430552879165,
|
| 1177 |
+
"learning_rate": 9.901835150755754e-05,
|
| 1178 |
+
"loss": 0.0204,
|
| 1179 |
+
"step": 1670
|
| 1180 |
+
},
|
| 1181 |
+
{
|
| 1182 |
+
"epoch": 1.1052631578947367,
|
| 1183 |
+
"grad_norm": 0.18723287056746327,
|
| 1184 |
+
"learning_rate": 9.898574318969918e-05,
|
| 1185 |
+
"loss": 0.0202,
|
| 1186 |
+
"step": 1680
|
| 1187 |
+
},
|
| 1188 |
+
{
|
| 1189 |
+
"epoch": 1.111842105263158,
|
| 1190 |
+
"grad_norm": 0.18512625335755417,
|
| 1191 |
+
"learning_rate": 9.895260764102956e-05,
|
| 1192 |
+
"loss": 0.0198,
|
| 1193 |
+
"step": 1690
|
| 1194 |
+
},
|
| 1195 |
+
{
|
| 1196 |
+
"epoch": 1.118421052631579,
|
| 1197 |
+
"grad_norm": 0.19124114649058113,
|
| 1198 |
+
"learning_rate": 9.89189452181847e-05,
|
| 1199 |
+
"loss": 0.0201,
|
| 1200 |
+
"step": 1700
|
| 1201 |
+
},
|
| 1202 |
+
{
|
| 1203 |
+
"epoch": 1.125,
|
| 1204 |
+
"grad_norm": 0.18895302686535878,
|
| 1205 |
+
"learning_rate": 9.888475628347142e-05,
|
| 1206 |
+
"loss": 0.0195,
|
| 1207 |
+
"step": 1710
|
| 1208 |
+
},
|
| 1209 |
+
{
|
| 1210 |
+
"epoch": 1.131578947368421,
|
| 1211 |
+
"grad_norm": 0.1772267382799342,
|
| 1212 |
+
"learning_rate": 9.885004120486326e-05,
|
| 1213 |
+
"loss": 0.0191,
|
| 1214 |
+
"step": 1720
|
| 1215 |
+
},
|
| 1216 |
+
{
|
| 1217 |
+
"epoch": 1.138157894736842,
|
| 1218 |
+
"grad_norm": 0.17908235775562148,
|
| 1219 |
+
"learning_rate": 9.881480035599667e-05,
|
| 1220 |
+
"loss": 0.0189,
|
| 1221 |
+
"step": 1730
|
| 1222 |
+
},
|
| 1223 |
+
{
|
| 1224 |
+
"epoch": 1.1447368421052633,
|
| 1225 |
+
"grad_norm": 0.17580586627136524,
|
| 1226 |
+
"learning_rate": 9.877903411616697e-05,
|
| 1227 |
+
"loss": 0.0189,
|
| 1228 |
+
"step": 1740
|
| 1229 |
+
},
|
| 1230 |
+
{
|
| 1231 |
+
"epoch": 1.1513157894736843,
|
| 1232 |
+
"grad_norm": 0.18518601084678493,
|
| 1233 |
+
"learning_rate": 9.874274287032417e-05,
|
| 1234 |
+
"loss": 0.0186,
|
| 1235 |
+
"step": 1750
|
| 1236 |
+
},
|
| 1237 |
+
{
|
| 1238 |
+
"epoch": 1.1578947368421053,
|
| 1239 |
+
"grad_norm": 0.18048608002452143,
|
| 1240 |
+
"learning_rate": 9.870592700906892e-05,
|
| 1241 |
+
"loss": 0.0183,
|
| 1242 |
+
"step": 1760
|
| 1243 |
+
},
|
| 1244 |
+
{
|
| 1245 |
+
"epoch": 1.1644736842105263,
|
| 1246 |
+
"grad_norm": 0.15777383982627308,
|
| 1247 |
+
"learning_rate": 9.866858692864829e-05,
|
| 1248 |
+
"loss": 0.018,
|
| 1249 |
+
"step": 1770
|
| 1250 |
+
},
|
| 1251 |
+
{
|
| 1252 |
+
"epoch": 1.1710526315789473,
|
| 1253 |
+
"grad_norm": 0.20607208990685497,
|
| 1254 |
+
"learning_rate": 9.863072303095144e-05,
|
| 1255 |
+
"loss": 0.0175,
|
| 1256 |
+
"step": 1780
|
| 1257 |
+
},
|
| 1258 |
+
{
|
| 1259 |
+
"epoch": 1.1776315789473684,
|
| 1260 |
+
"grad_norm": 0.1611318675087161,
|
| 1261 |
+
"learning_rate": 9.859233572350541e-05,
|
| 1262 |
+
"loss": 0.0171,
|
| 1263 |
+
"step": 1790
|
| 1264 |
+
},
|
| 1265 |
+
{
|
| 1266 |
+
"epoch": 1.1842105263157894,
|
| 1267 |
+
"grad_norm": 0.17589801094486104,
|
| 1268 |
+
"learning_rate": 9.855342541947065e-05,
|
| 1269 |
+
"loss": 0.0171,
|
| 1270 |
+
"step": 1800
|
| 1271 |
+
}
|
| 1272 |
+
],
|
| 1273 |
+
"logging_steps": 10,
|
| 1274 |
+
"max_steps": 10640,
|
| 1275 |
+
"num_input_tokens_seen": 0,
|
| 1276 |
+
"num_train_epochs": 7,
|
| 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": 143324589588480.0,
|
| 1291 |
+
"train_batch_size": 128,
|
| 1292 |
+
"trial_name": null,
|
| 1293 |
+
"trial_params": null
|
| 1294 |
+
}
|