Upload multi_student_7/checkpoint-1800/trainer_state.json with huggingface_hub
Browse files
multi_student_7/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": 0.9266409266409267,
|
| 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": 0.005148005148005148,
|
| 14 |
+
"grad_norm": 19.51059658759793,
|
| 15 |
+
"learning_rate": 7.718696397941682e-07,
|
| 16 |
+
"loss": 8.2552,
|
| 17 |
+
"step": 10
|
| 18 |
+
},
|
| 19 |
+
{
|
| 20 |
+
"epoch": 0.010296010296010296,
|
| 21 |
+
"grad_norm": 19.70146681963939,
|
| 22 |
+
"learning_rate": 1.6295025728987994e-06,
|
| 23 |
+
"loss": 8.168,
|
| 24 |
+
"step": 20
|
| 25 |
+
},
|
| 26 |
+
{
|
| 27 |
+
"epoch": 0.015444015444015444,
|
| 28 |
+
"grad_norm": 9.795523599582376,
|
| 29 |
+
"learning_rate": 2.4871355060034306e-06,
|
| 30 |
+
"loss": 7.5953,
|
| 31 |
+
"step": 30
|
| 32 |
+
},
|
| 33 |
+
{
|
| 34 |
+
"epoch": 0.02059202059202059,
|
| 35 |
+
"grad_norm": 5.9480235723668144,
|
| 36 |
+
"learning_rate": 3.344768439108062e-06,
|
| 37 |
+
"loss": 6.8507,
|
| 38 |
+
"step": 40
|
| 39 |
+
},
|
| 40 |
+
{
|
| 41 |
+
"epoch": 0.02574002574002574,
|
| 42 |
+
"grad_norm": 3.5099463285447223,
|
| 43 |
+
"learning_rate": 4.202401372212693e-06,
|
| 44 |
+
"loss": 6.3398,
|
| 45 |
+
"step": 50
|
| 46 |
+
},
|
| 47 |
+
{
|
| 48 |
+
"epoch": 0.03088803088803089,
|
| 49 |
+
"grad_norm": 2.90890000793416,
|
| 50 |
+
"learning_rate": 5.060034305317325e-06,
|
| 51 |
+
"loss": 5.9988,
|
| 52 |
+
"step": 60
|
| 53 |
+
},
|
| 54 |
+
{
|
| 55 |
+
"epoch": 0.036036036036036036,
|
| 56 |
+
"grad_norm": 3.3978017812772445,
|
| 57 |
+
"learning_rate": 5.9176672384219555e-06,
|
| 58 |
+
"loss": 5.6503,
|
| 59 |
+
"step": 70
|
| 60 |
+
},
|
| 61 |
+
{
|
| 62 |
+
"epoch": 0.04118404118404118,
|
| 63 |
+
"grad_norm": 4.103088821630264,
|
| 64 |
+
"learning_rate": 6.775300171526587e-06,
|
| 65 |
+
"loss": 5.1289,
|
| 66 |
+
"step": 80
|
| 67 |
+
},
|
| 68 |
+
{
|
| 69 |
+
"epoch": 0.04633204633204633,
|
| 70 |
+
"grad_norm": 4.487842621805789,
|
| 71 |
+
"learning_rate": 7.632933104631218e-06,
|
| 72 |
+
"loss": 4.511,
|
| 73 |
+
"step": 90
|
| 74 |
+
},
|
| 75 |
+
{
|
| 76 |
+
"epoch": 0.05148005148005148,
|
| 77 |
+
"grad_norm": 7.491973587682773,
|
| 78 |
+
"learning_rate": 8.49056603773585e-06,
|
| 79 |
+
"loss": 3.9137,
|
| 80 |
+
"step": 100
|
| 81 |
+
},
|
| 82 |
+
{
|
| 83 |
+
"epoch": 0.05662805662805663,
|
| 84 |
+
"grad_norm": 7.323430680232261,
|
| 85 |
+
"learning_rate": 9.348198970840481e-06,
|
| 86 |
+
"loss": 3.4784,
|
| 87 |
+
"step": 110
|
| 88 |
+
},
|
| 89 |
+
{
|
| 90 |
+
"epoch": 0.06177606177606178,
|
| 91 |
+
"grad_norm": 6.619474879048035,
|
| 92 |
+
"learning_rate": 1.0205831903945113e-05,
|
| 93 |
+
"loss": 3.2067,
|
| 94 |
+
"step": 120
|
| 95 |
+
},
|
| 96 |
+
{
|
| 97 |
+
"epoch": 0.06692406692406692,
|
| 98 |
+
"grad_norm": 3.867444417417378,
|
| 99 |
+
"learning_rate": 1.1063464837049744e-05,
|
| 100 |
+
"loss": 3.0386,
|
| 101 |
+
"step": 130
|
| 102 |
+
},
|
| 103 |
+
{
|
| 104 |
+
"epoch": 0.07207207207207207,
|
| 105 |
+
"grad_norm": 4.117034943974073,
|
| 106 |
+
"learning_rate": 1.1921097770154374e-05,
|
| 107 |
+
"loss": 2.9159,
|
| 108 |
+
"step": 140
|
| 109 |
+
},
|
| 110 |
+
{
|
| 111 |
+
"epoch": 0.07722007722007722,
|
| 112 |
+
"grad_norm": 6.009222032591652,
|
| 113 |
+
"learning_rate": 1.2778730703259007e-05,
|
| 114 |
+
"loss": 2.796,
|
| 115 |
+
"step": 150
|
| 116 |
+
},
|
| 117 |
+
{
|
| 118 |
+
"epoch": 0.08236808236808237,
|
| 119 |
+
"grad_norm": 11.57505334622991,
|
| 120 |
+
"learning_rate": 1.3636363636363637e-05,
|
| 121 |
+
"loss": 2.7077,
|
| 122 |
+
"step": 160
|
| 123 |
+
},
|
| 124 |
+
{
|
| 125 |
+
"epoch": 0.08751608751608751,
|
| 126 |
+
"grad_norm": 11.428627421387615,
|
| 127 |
+
"learning_rate": 1.4493996569468268e-05,
|
| 128 |
+
"loss": 2.6403,
|
| 129 |
+
"step": 170
|
| 130 |
+
},
|
| 131 |
+
{
|
| 132 |
+
"epoch": 0.09266409266409266,
|
| 133 |
+
"grad_norm": 10.905546843716907,
|
| 134 |
+
"learning_rate": 1.53516295025729e-05,
|
| 135 |
+
"loss": 2.5735,
|
| 136 |
+
"step": 180
|
| 137 |
+
},
|
| 138 |
+
{
|
| 139 |
+
"epoch": 0.0978120978120978,
|
| 140 |
+
"grad_norm": 8.661551806496005,
|
| 141 |
+
"learning_rate": 1.620926243567753e-05,
|
| 142 |
+
"loss": 2.5078,
|
| 143 |
+
"step": 190
|
| 144 |
+
},
|
| 145 |
+
{
|
| 146 |
+
"epoch": 0.10296010296010295,
|
| 147 |
+
"grad_norm": 9.940151278870877,
|
| 148 |
+
"learning_rate": 1.7066895368782164e-05,
|
| 149 |
+
"loss": 2.4567,
|
| 150 |
+
"step": 200
|
| 151 |
+
},
|
| 152 |
+
{
|
| 153 |
+
"epoch": 0.10810810810810811,
|
| 154 |
+
"grad_norm": 13.091142904495205,
|
| 155 |
+
"learning_rate": 1.7924528301886792e-05,
|
| 156 |
+
"loss": 2.3953,
|
| 157 |
+
"step": 210
|
| 158 |
+
},
|
| 159 |
+
{
|
| 160 |
+
"epoch": 0.11325611325611326,
|
| 161 |
+
"grad_norm": 18.06611865183751,
|
| 162 |
+
"learning_rate": 1.8782161234991423e-05,
|
| 163 |
+
"loss": 2.329,
|
| 164 |
+
"step": 220
|
| 165 |
+
},
|
| 166 |
+
{
|
| 167 |
+
"epoch": 0.11840411840411841,
|
| 168 |
+
"grad_norm": 16.145152244084983,
|
| 169 |
+
"learning_rate": 1.9639794168096058e-05,
|
| 170 |
+
"loss": 2.2691,
|
| 171 |
+
"step": 230
|
| 172 |
+
},
|
| 173 |
+
{
|
| 174 |
+
"epoch": 0.12355212355212356,
|
| 175 |
+
"grad_norm": 35.1534299592601,
|
| 176 |
+
"learning_rate": 2.0497427101200686e-05,
|
| 177 |
+
"loss": 2.2524,
|
| 178 |
+
"step": 240
|
| 179 |
+
},
|
| 180 |
+
{
|
| 181 |
+
"epoch": 0.1287001287001287,
|
| 182 |
+
"grad_norm": 14.465207787142148,
|
| 183 |
+
"learning_rate": 2.1355060034305318e-05,
|
| 184 |
+
"loss": 2.1974,
|
| 185 |
+
"step": 250
|
| 186 |
+
},
|
| 187 |
+
{
|
| 188 |
+
"epoch": 0.13384813384813385,
|
| 189 |
+
"grad_norm": 11.353218963690644,
|
| 190 |
+
"learning_rate": 2.2212692967409952e-05,
|
| 191 |
+
"loss": 2.0873,
|
| 192 |
+
"step": 260
|
| 193 |
+
},
|
| 194 |
+
{
|
| 195 |
+
"epoch": 0.138996138996139,
|
| 196 |
+
"grad_norm": 12.277610068118626,
|
| 197 |
+
"learning_rate": 2.307032590051458e-05,
|
| 198 |
+
"loss": 2.0068,
|
| 199 |
+
"step": 270
|
| 200 |
+
},
|
| 201 |
+
{
|
| 202 |
+
"epoch": 0.14414414414414414,
|
| 203 |
+
"grad_norm": 10.505580540439455,
|
| 204 |
+
"learning_rate": 2.3927958833619212e-05,
|
| 205 |
+
"loss": 1.8888,
|
| 206 |
+
"step": 280
|
| 207 |
+
},
|
| 208 |
+
{
|
| 209 |
+
"epoch": 0.1492921492921493,
|
| 210 |
+
"grad_norm": 14.960676861450345,
|
| 211 |
+
"learning_rate": 2.4785591766723843e-05,
|
| 212 |
+
"loss": 1.7894,
|
| 213 |
+
"step": 290
|
| 214 |
+
},
|
| 215 |
+
{
|
| 216 |
+
"epoch": 0.15444015444015444,
|
| 217 |
+
"grad_norm": 8.613188849395,
|
| 218 |
+
"learning_rate": 2.564322469982847e-05,
|
| 219 |
+
"loss": 1.6967,
|
| 220 |
+
"step": 300
|
| 221 |
+
},
|
| 222 |
+
{
|
| 223 |
+
"epoch": 0.15958815958815958,
|
| 224 |
+
"grad_norm": 6.959841307105666,
|
| 225 |
+
"learning_rate": 2.6500857632933106e-05,
|
| 226 |
+
"loss": 1.6019,
|
| 227 |
+
"step": 310
|
| 228 |
+
},
|
| 229 |
+
{
|
| 230 |
+
"epoch": 0.16473616473616473,
|
| 231 |
+
"grad_norm": 5.864669793953841,
|
| 232 |
+
"learning_rate": 2.7358490566037738e-05,
|
| 233 |
+
"loss": 1.5326,
|
| 234 |
+
"step": 320
|
| 235 |
+
},
|
| 236 |
+
{
|
| 237 |
+
"epoch": 0.16988416988416988,
|
| 238 |
+
"grad_norm": 8.165848406130657,
|
| 239 |
+
"learning_rate": 2.8216123499142366e-05,
|
| 240 |
+
"loss": 1.4842,
|
| 241 |
+
"step": 330
|
| 242 |
+
},
|
| 243 |
+
{
|
| 244 |
+
"epoch": 0.17503217503217502,
|
| 245 |
+
"grad_norm": 6.747629987576382,
|
| 246 |
+
"learning_rate": 2.9073756432247e-05,
|
| 247 |
+
"loss": 1.426,
|
| 248 |
+
"step": 340
|
| 249 |
+
},
|
| 250 |
+
{
|
| 251 |
+
"epoch": 0.18018018018018017,
|
| 252 |
+
"grad_norm": 4.897733377675901,
|
| 253 |
+
"learning_rate": 2.9931389365351632e-05,
|
| 254 |
+
"loss": 1.3614,
|
| 255 |
+
"step": 350
|
| 256 |
+
},
|
| 257 |
+
{
|
| 258 |
+
"epoch": 0.18532818532818532,
|
| 259 |
+
"grad_norm": 5.330679969567318,
|
| 260 |
+
"learning_rate": 3.078902229845626e-05,
|
| 261 |
+
"loss": 1.3182,
|
| 262 |
+
"step": 360
|
| 263 |
+
},
|
| 264 |
+
{
|
| 265 |
+
"epoch": 0.19047619047619047,
|
| 266 |
+
"grad_norm": 4.722018507382775,
|
| 267 |
+
"learning_rate": 3.16466552315609e-05,
|
| 268 |
+
"loss": 1.26,
|
| 269 |
+
"step": 370
|
| 270 |
+
},
|
| 271 |
+
{
|
| 272 |
+
"epoch": 0.1956241956241956,
|
| 273 |
+
"grad_norm": 6.77470473058928,
|
| 274 |
+
"learning_rate": 3.250428816466552e-05,
|
| 275 |
+
"loss": 1.2128,
|
| 276 |
+
"step": 380
|
| 277 |
+
},
|
| 278 |
+
{
|
| 279 |
+
"epoch": 0.20077220077220076,
|
| 280 |
+
"grad_norm": 4.902136742839717,
|
| 281 |
+
"learning_rate": 3.3361921097770154e-05,
|
| 282 |
+
"loss": 1.1829,
|
| 283 |
+
"step": 390
|
| 284 |
+
},
|
| 285 |
+
{
|
| 286 |
+
"epoch": 0.2059202059202059,
|
| 287 |
+
"grad_norm": 5.515696505859524,
|
| 288 |
+
"learning_rate": 3.421955403087479e-05,
|
| 289 |
+
"loss": 1.1379,
|
| 290 |
+
"step": 400
|
| 291 |
+
},
|
| 292 |
+
{
|
| 293 |
+
"epoch": 0.21106821106821108,
|
| 294 |
+
"grad_norm": 3.9113735830318603,
|
| 295 |
+
"learning_rate": 3.507718696397942e-05,
|
| 296 |
+
"loss": 1.1002,
|
| 297 |
+
"step": 410
|
| 298 |
+
},
|
| 299 |
+
{
|
| 300 |
+
"epoch": 0.21621621621621623,
|
| 301 |
+
"grad_norm": 5.8061510583054545,
|
| 302 |
+
"learning_rate": 3.593481989708405e-05,
|
| 303 |
+
"loss": 1.0744,
|
| 304 |
+
"step": 420
|
| 305 |
+
},
|
| 306 |
+
{
|
| 307 |
+
"epoch": 0.22136422136422138,
|
| 308 |
+
"grad_norm": 5.927584495277828,
|
| 309 |
+
"learning_rate": 3.679245283018868e-05,
|
| 310 |
+
"loss": 1.0304,
|
| 311 |
+
"step": 430
|
| 312 |
+
},
|
| 313 |
+
{
|
| 314 |
+
"epoch": 0.22651222651222652,
|
| 315 |
+
"grad_norm": 5.920319878093127,
|
| 316 |
+
"learning_rate": 3.765008576329331e-05,
|
| 317 |
+
"loss": 1.0023,
|
| 318 |
+
"step": 440
|
| 319 |
+
},
|
| 320 |
+
{
|
| 321 |
+
"epoch": 0.23166023166023167,
|
| 322 |
+
"grad_norm": 5.302890318730487,
|
| 323 |
+
"learning_rate": 3.850771869639794e-05,
|
| 324 |
+
"loss": 0.9776,
|
| 325 |
+
"step": 450
|
| 326 |
+
},
|
| 327 |
+
{
|
| 328 |
+
"epoch": 0.23680823680823682,
|
| 329 |
+
"grad_norm": 5.653239199646973,
|
| 330 |
+
"learning_rate": 3.9365351629502574e-05,
|
| 331 |
+
"loss": 0.9482,
|
| 332 |
+
"step": 460
|
| 333 |
+
},
|
| 334 |
+
{
|
| 335 |
+
"epoch": 0.24195624195624196,
|
| 336 |
+
"grad_norm": 4.852345392975872,
|
| 337 |
+
"learning_rate": 4.0222984562607206e-05,
|
| 338 |
+
"loss": 0.9155,
|
| 339 |
+
"step": 470
|
| 340 |
+
},
|
| 341 |
+
{
|
| 342 |
+
"epoch": 0.2471042471042471,
|
| 343 |
+
"grad_norm": 4.499279652551338,
|
| 344 |
+
"learning_rate": 4.108061749571184e-05,
|
| 345 |
+
"loss": 0.8867,
|
| 346 |
+
"step": 480
|
| 347 |
+
},
|
| 348 |
+
{
|
| 349 |
+
"epoch": 0.25225225225225223,
|
| 350 |
+
"grad_norm": 5.181168954723113,
|
| 351 |
+
"learning_rate": 4.193825042881647e-05,
|
| 352 |
+
"loss": 0.8756,
|
| 353 |
+
"step": 490
|
| 354 |
+
},
|
| 355 |
+
{
|
| 356 |
+
"epoch": 0.2574002574002574,
|
| 357 |
+
"grad_norm": 6.255177791414999,
|
| 358 |
+
"learning_rate": 4.27958833619211e-05,
|
| 359 |
+
"loss": 0.8506,
|
| 360 |
+
"step": 500
|
| 361 |
+
},
|
| 362 |
+
{
|
| 363 |
+
"epoch": 0.2625482625482625,
|
| 364 |
+
"grad_norm": 5.396447906024421,
|
| 365 |
+
"learning_rate": 4.365351629502573e-05,
|
| 366 |
+
"loss": 0.8203,
|
| 367 |
+
"step": 510
|
| 368 |
+
},
|
| 369 |
+
{
|
| 370 |
+
"epoch": 0.2676962676962677,
|
| 371 |
+
"grad_norm": 7.371015885385536,
|
| 372 |
+
"learning_rate": 4.451114922813036e-05,
|
| 373 |
+
"loss": 0.8028,
|
| 374 |
+
"step": 520
|
| 375 |
+
},
|
| 376 |
+
{
|
| 377 |
+
"epoch": 0.2728442728442728,
|
| 378 |
+
"grad_norm": 6.198537777815005,
|
| 379 |
+
"learning_rate": 4.5368782161234994e-05,
|
| 380 |
+
"loss": 0.7808,
|
| 381 |
+
"step": 530
|
| 382 |
+
},
|
| 383 |
+
{
|
| 384 |
+
"epoch": 0.277992277992278,
|
| 385 |
+
"grad_norm": 4.485699286277638,
|
| 386 |
+
"learning_rate": 4.6226415094339625e-05,
|
| 387 |
+
"loss": 0.7639,
|
| 388 |
+
"step": 540
|
| 389 |
+
},
|
| 390 |
+
{
|
| 391 |
+
"epoch": 0.28314028314028317,
|
| 392 |
+
"grad_norm": 6.865323396700289,
|
| 393 |
+
"learning_rate": 4.708404802744426e-05,
|
| 394 |
+
"loss": 0.745,
|
| 395 |
+
"step": 550
|
| 396 |
+
},
|
| 397 |
+
{
|
| 398 |
+
"epoch": 0.2882882882882883,
|
| 399 |
+
"grad_norm": 4.718888462403314,
|
| 400 |
+
"learning_rate": 4.794168096054888e-05,
|
| 401 |
+
"loss": 0.7209,
|
| 402 |
+
"step": 560
|
| 403 |
+
},
|
| 404 |
+
{
|
| 405 |
+
"epoch": 0.29343629343629346,
|
| 406 |
+
"grad_norm": 4.887504164875748,
|
| 407 |
+
"learning_rate": 4.879931389365352e-05,
|
| 408 |
+
"loss": 0.7015,
|
| 409 |
+
"step": 570
|
| 410 |
+
},
|
| 411 |
+
{
|
| 412 |
+
"epoch": 0.2985842985842986,
|
| 413 |
+
"grad_norm": 5.372842102076737,
|
| 414 |
+
"learning_rate": 4.965694682675815e-05,
|
| 415 |
+
"loss": 0.6767,
|
| 416 |
+
"step": 580
|
| 417 |
+
},
|
| 418 |
+
{
|
| 419 |
+
"epoch": 0.30373230373230375,
|
| 420 |
+
"grad_norm": 6.766330192939623,
|
| 421 |
+
"learning_rate": 5.0514579759862776e-05,
|
| 422 |
+
"loss": 0.6624,
|
| 423 |
+
"step": 590
|
| 424 |
+
},
|
| 425 |
+
{
|
| 426 |
+
"epoch": 0.3088803088803089,
|
| 427 |
+
"grad_norm": 4.614837742067113,
|
| 428 |
+
"learning_rate": 5.1372212692967414e-05,
|
| 429 |
+
"loss": 0.6405,
|
| 430 |
+
"step": 600
|
| 431 |
+
},
|
| 432 |
+
{
|
| 433 |
+
"epoch": 0.31402831402831405,
|
| 434 |
+
"grad_norm": 4.201337449433587,
|
| 435 |
+
"learning_rate": 5.222984562607204e-05,
|
| 436 |
+
"loss": 0.6194,
|
| 437 |
+
"step": 610
|
| 438 |
+
},
|
| 439 |
+
{
|
| 440 |
+
"epoch": 0.31917631917631917,
|
| 441 |
+
"grad_norm": 5.51280432030841,
|
| 442 |
+
"learning_rate": 5.308747855917667e-05,
|
| 443 |
+
"loss": 0.5951,
|
| 444 |
+
"step": 620
|
| 445 |
+
},
|
| 446 |
+
{
|
| 447 |
+
"epoch": 0.32432432432432434,
|
| 448 |
+
"grad_norm": 5.129963269925849,
|
| 449 |
+
"learning_rate": 5.394511149228131e-05,
|
| 450 |
+
"loss": 0.5777,
|
| 451 |
+
"step": 630
|
| 452 |
+
},
|
| 453 |
+
{
|
| 454 |
+
"epoch": 0.32947232947232946,
|
| 455 |
+
"grad_norm": 4.412209442955385,
|
| 456 |
+
"learning_rate": 5.480274442538593e-05,
|
| 457 |
+
"loss": 0.5642,
|
| 458 |
+
"step": 640
|
| 459 |
+
},
|
| 460 |
+
{
|
| 461 |
+
"epoch": 0.33462033462033464,
|
| 462 |
+
"grad_norm": 4.336638446576286,
|
| 463 |
+
"learning_rate": 5.5660377358490564e-05,
|
| 464 |
+
"loss": 0.5426,
|
| 465 |
+
"step": 650
|
| 466 |
+
},
|
| 467 |
+
{
|
| 468 |
+
"epoch": 0.33976833976833976,
|
| 469 |
+
"grad_norm": 3.556283909809826,
|
| 470 |
+
"learning_rate": 5.65180102915952e-05,
|
| 471 |
+
"loss": 0.5112,
|
| 472 |
+
"step": 660
|
| 473 |
+
},
|
| 474 |
+
{
|
| 475 |
+
"epoch": 0.34491634491634493,
|
| 476 |
+
"grad_norm": 5.225584058992361,
|
| 477 |
+
"learning_rate": 5.737564322469983e-05,
|
| 478 |
+
"loss": 0.5044,
|
| 479 |
+
"step": 670
|
| 480 |
+
},
|
| 481 |
+
{
|
| 482 |
+
"epoch": 0.35006435006435005,
|
| 483 |
+
"grad_norm": 3.1697056899798186,
|
| 484 |
+
"learning_rate": 5.823327615780446e-05,
|
| 485 |
+
"loss": 0.4857,
|
| 486 |
+
"step": 680
|
| 487 |
+
},
|
| 488 |
+
{
|
| 489 |
+
"epoch": 0.3552123552123552,
|
| 490 |
+
"grad_norm": 3.5258824608431003,
|
| 491 |
+
"learning_rate": 5.90909090909091e-05,
|
| 492 |
+
"loss": 0.4747,
|
| 493 |
+
"step": 690
|
| 494 |
+
},
|
| 495 |
+
{
|
| 496 |
+
"epoch": 0.36036036036036034,
|
| 497 |
+
"grad_norm": 3.5240284197467155,
|
| 498 |
+
"learning_rate": 5.994854202401372e-05,
|
| 499 |
+
"loss": 0.4563,
|
| 500 |
+
"step": 700
|
| 501 |
+
},
|
| 502 |
+
{
|
| 503 |
+
"epoch": 0.3655083655083655,
|
| 504 |
+
"grad_norm": 3.787588207851851,
|
| 505 |
+
"learning_rate": 6.080617495711835e-05,
|
| 506 |
+
"loss": 0.4382,
|
| 507 |
+
"step": 710
|
| 508 |
+
},
|
| 509 |
+
{
|
| 510 |
+
"epoch": 0.37065637065637064,
|
| 511 |
+
"grad_norm": 3.28493026844208,
|
| 512 |
+
"learning_rate": 6.166380789022299e-05,
|
| 513 |
+
"loss": 0.4216,
|
| 514 |
+
"step": 720
|
| 515 |
+
},
|
| 516 |
+
{
|
| 517 |
+
"epoch": 0.3758043758043758,
|
| 518 |
+
"grad_norm": 3.144221026231138,
|
| 519 |
+
"learning_rate": 6.252144082332761e-05,
|
| 520 |
+
"loss": 0.417,
|
| 521 |
+
"step": 730
|
| 522 |
+
},
|
| 523 |
+
{
|
| 524 |
+
"epoch": 0.38095238095238093,
|
| 525 |
+
"grad_norm": 2.780102841682132,
|
| 526 |
+
"learning_rate": 6.337907375643225e-05,
|
| 527 |
+
"loss": 0.3942,
|
| 528 |
+
"step": 740
|
| 529 |
+
},
|
| 530 |
+
{
|
| 531 |
+
"epoch": 0.3861003861003861,
|
| 532 |
+
"grad_norm": 3.1996820882351713,
|
| 533 |
+
"learning_rate": 6.423670668953689e-05,
|
| 534 |
+
"loss": 0.3787,
|
| 535 |
+
"step": 750
|
| 536 |
+
},
|
| 537 |
+
{
|
| 538 |
+
"epoch": 0.3912483912483912,
|
| 539 |
+
"grad_norm": 2.8374055946677923,
|
| 540 |
+
"learning_rate": 6.50943396226415e-05,
|
| 541 |
+
"loss": 0.3714,
|
| 542 |
+
"step": 760
|
| 543 |
+
},
|
| 544 |
+
{
|
| 545 |
+
"epoch": 0.3963963963963964,
|
| 546 |
+
"grad_norm": 3.0094036046611454,
|
| 547 |
+
"learning_rate": 6.595197255574615e-05,
|
| 548 |
+
"loss": 0.3606,
|
| 549 |
+
"step": 770
|
| 550 |
+
},
|
| 551 |
+
{
|
| 552 |
+
"epoch": 0.4015444015444015,
|
| 553 |
+
"grad_norm": 2.6679117500690204,
|
| 554 |
+
"learning_rate": 6.680960548885078e-05,
|
| 555 |
+
"loss": 0.3379,
|
| 556 |
+
"step": 780
|
| 557 |
+
},
|
| 558 |
+
{
|
| 559 |
+
"epoch": 0.4066924066924067,
|
| 560 |
+
"grad_norm": 3.101181469988925,
|
| 561 |
+
"learning_rate": 6.76672384219554e-05,
|
| 562 |
+
"loss": 0.3271,
|
| 563 |
+
"step": 790
|
| 564 |
+
},
|
| 565 |
+
{
|
| 566 |
+
"epoch": 0.4118404118404118,
|
| 567 |
+
"grad_norm": 2.421653916410696,
|
| 568 |
+
"learning_rate": 6.852487135506004e-05,
|
| 569 |
+
"loss": 0.3132,
|
| 570 |
+
"step": 800
|
| 571 |
+
},
|
| 572 |
+
{
|
| 573 |
+
"epoch": 0.416988416988417,
|
| 574 |
+
"grad_norm": 2.1110958003741174,
|
| 575 |
+
"learning_rate": 6.938250428816467e-05,
|
| 576 |
+
"loss": 0.3013,
|
| 577 |
+
"step": 810
|
| 578 |
+
},
|
| 579 |
+
{
|
| 580 |
+
"epoch": 0.42213642213642216,
|
| 581 |
+
"grad_norm": 1.9804288224535966,
|
| 582 |
+
"learning_rate": 7.024013722126929e-05,
|
| 583 |
+
"loss": 0.2892,
|
| 584 |
+
"step": 820
|
| 585 |
+
},
|
| 586 |
+
{
|
| 587 |
+
"epoch": 0.4272844272844273,
|
| 588 |
+
"grad_norm": 2.0965122713743725,
|
| 589 |
+
"learning_rate": 7.109777015437394e-05,
|
| 590 |
+
"loss": 0.2802,
|
| 591 |
+
"step": 830
|
| 592 |
+
},
|
| 593 |
+
{
|
| 594 |
+
"epoch": 0.43243243243243246,
|
| 595 |
+
"grad_norm": 2.0662383619585323,
|
| 596 |
+
"learning_rate": 7.195540308747857e-05,
|
| 597 |
+
"loss": 0.2666,
|
| 598 |
+
"step": 840
|
| 599 |
+
},
|
| 600 |
+
{
|
| 601 |
+
"epoch": 0.4375804375804376,
|
| 602 |
+
"grad_norm": 1.9080845115386118,
|
| 603 |
+
"learning_rate": 7.281303602058319e-05,
|
| 604 |
+
"loss": 0.2557,
|
| 605 |
+
"step": 850
|
| 606 |
+
},
|
| 607 |
+
{
|
| 608 |
+
"epoch": 0.44272844272844275,
|
| 609 |
+
"grad_norm": 2.435492621224063,
|
| 610 |
+
"learning_rate": 7.367066895368783e-05,
|
| 611 |
+
"loss": 0.2441,
|
| 612 |
+
"step": 860
|
| 613 |
+
},
|
| 614 |
+
{
|
| 615 |
+
"epoch": 0.44787644787644787,
|
| 616 |
+
"grad_norm": 2.3478908862806707,
|
| 617 |
+
"learning_rate": 7.452830188679245e-05,
|
| 618 |
+
"loss": 0.2302,
|
| 619 |
+
"step": 870
|
| 620 |
+
},
|
| 621 |
+
{
|
| 622 |
+
"epoch": 0.45302445302445304,
|
| 623 |
+
"grad_norm": 2.033330789035785,
|
| 624 |
+
"learning_rate": 7.538593481989708e-05,
|
| 625 |
+
"loss": 0.222,
|
| 626 |
+
"step": 880
|
| 627 |
+
},
|
| 628 |
+
{
|
| 629 |
+
"epoch": 0.45817245817245816,
|
| 630 |
+
"grad_norm": 1.8373992771294045,
|
| 631 |
+
"learning_rate": 7.624356775300173e-05,
|
| 632 |
+
"loss": 0.2079,
|
| 633 |
+
"step": 890
|
| 634 |
+
},
|
| 635 |
+
{
|
| 636 |
+
"epoch": 0.46332046332046334,
|
| 637 |
+
"grad_norm": 1.9518420548134319,
|
| 638 |
+
"learning_rate": 7.710120068610634e-05,
|
| 639 |
+
"loss": 0.1979,
|
| 640 |
+
"step": 900
|
| 641 |
+
},
|
| 642 |
+
{
|
| 643 |
+
"epoch": 0.46846846846846846,
|
| 644 |
+
"grad_norm": 1.6103496909038504,
|
| 645 |
+
"learning_rate": 7.795883361921097e-05,
|
| 646 |
+
"loss": 0.1936,
|
| 647 |
+
"step": 910
|
| 648 |
+
},
|
| 649 |
+
{
|
| 650 |
+
"epoch": 0.47361647361647363,
|
| 651 |
+
"grad_norm": 1.3694934489721085,
|
| 652 |
+
"learning_rate": 7.881646655231562e-05,
|
| 653 |
+
"loss": 0.1841,
|
| 654 |
+
"step": 920
|
| 655 |
+
},
|
| 656 |
+
{
|
| 657 |
+
"epoch": 0.47876447876447875,
|
| 658 |
+
"grad_norm": 1.2736699951465489,
|
| 659 |
+
"learning_rate": 7.967409948542024e-05,
|
| 660 |
+
"loss": 0.1766,
|
| 661 |
+
"step": 930
|
| 662 |
+
},
|
| 663 |
+
{
|
| 664 |
+
"epoch": 0.4839124839124839,
|
| 665 |
+
"grad_norm": 2.1829922501226444,
|
| 666 |
+
"learning_rate": 8.053173241852487e-05,
|
| 667 |
+
"loss": 0.1675,
|
| 668 |
+
"step": 940
|
| 669 |
+
},
|
| 670 |
+
{
|
| 671 |
+
"epoch": 0.48906048906048905,
|
| 672 |
+
"grad_norm": 1.659964942426995,
|
| 673 |
+
"learning_rate": 8.138936535162951e-05,
|
| 674 |
+
"loss": 0.161,
|
| 675 |
+
"step": 950
|
| 676 |
+
},
|
| 677 |
+
{
|
| 678 |
+
"epoch": 0.4942084942084942,
|
| 679 |
+
"grad_norm": 1.5767630911351294,
|
| 680 |
+
"learning_rate": 8.224699828473413e-05,
|
| 681 |
+
"loss": 0.1494,
|
| 682 |
+
"step": 960
|
| 683 |
+
},
|
| 684 |
+
{
|
| 685 |
+
"epoch": 0.49935649935649934,
|
| 686 |
+
"grad_norm": 1.537782758985047,
|
| 687 |
+
"learning_rate": 8.310463121783876e-05,
|
| 688 |
+
"loss": 0.1445,
|
| 689 |
+
"step": 970
|
| 690 |
+
},
|
| 691 |
+
{
|
| 692 |
+
"epoch": 0.5045045045045045,
|
| 693 |
+
"grad_norm": 1.604101659458472,
|
| 694 |
+
"learning_rate": 8.396226415094341e-05,
|
| 695 |
+
"loss": 0.1402,
|
| 696 |
+
"step": 980
|
| 697 |
+
},
|
| 698 |
+
{
|
| 699 |
+
"epoch": 0.5096525096525096,
|
| 700 |
+
"grad_norm": 2.256805925155236,
|
| 701 |
+
"learning_rate": 8.481989708404803e-05,
|
| 702 |
+
"loss": 0.1405,
|
| 703 |
+
"step": 990
|
| 704 |
+
},
|
| 705 |
+
{
|
| 706 |
+
"epoch": 0.5148005148005148,
|
| 707 |
+
"grad_norm": 1.4857578289963005,
|
| 708 |
+
"learning_rate": 8.567753001715266e-05,
|
| 709 |
+
"loss": 0.1321,
|
| 710 |
+
"step": 1000
|
| 711 |
+
},
|
| 712 |
+
{
|
| 713 |
+
"epoch": 0.51994851994852,
|
| 714 |
+
"grad_norm": 1.0326983623178083,
|
| 715 |
+
"learning_rate": 8.65351629502573e-05,
|
| 716 |
+
"loss": 0.1224,
|
| 717 |
+
"step": 1010
|
| 718 |
+
},
|
| 719 |
+
{
|
| 720 |
+
"epoch": 0.525096525096525,
|
| 721 |
+
"grad_norm": 1.1180976317039228,
|
| 722 |
+
"learning_rate": 8.739279588336192e-05,
|
| 723 |
+
"loss": 0.1189,
|
| 724 |
+
"step": 1020
|
| 725 |
+
},
|
| 726 |
+
{
|
| 727 |
+
"epoch": 0.5302445302445302,
|
| 728 |
+
"grad_norm": 1.1721351234336,
|
| 729 |
+
"learning_rate": 8.825042881646655e-05,
|
| 730 |
+
"loss": 0.1092,
|
| 731 |
+
"step": 1030
|
| 732 |
+
},
|
| 733 |
+
{
|
| 734 |
+
"epoch": 0.5353925353925354,
|
| 735 |
+
"grad_norm": 0.9640093738745075,
|
| 736 |
+
"learning_rate": 8.91080617495712e-05,
|
| 737 |
+
"loss": 0.1021,
|
| 738 |
+
"step": 1040
|
| 739 |
+
},
|
| 740 |
+
{
|
| 741 |
+
"epoch": 0.5405405405405406,
|
| 742 |
+
"grad_norm": 1.0306043643976568,
|
| 743 |
+
"learning_rate": 8.996569468267581e-05,
|
| 744 |
+
"loss": 0.0982,
|
| 745 |
+
"step": 1050
|
| 746 |
+
},
|
| 747 |
+
{
|
| 748 |
+
"epoch": 0.5456885456885456,
|
| 749 |
+
"grad_norm": 1.4999347000799717,
|
| 750 |
+
"learning_rate": 9.082332761578045e-05,
|
| 751 |
+
"loss": 0.0936,
|
| 752 |
+
"step": 1060
|
| 753 |
+
},
|
| 754 |
+
{
|
| 755 |
+
"epoch": 0.5508365508365508,
|
| 756 |
+
"grad_norm": 1.0452281614309868,
|
| 757 |
+
"learning_rate": 9.168096054888509e-05,
|
| 758 |
+
"loss": 0.0878,
|
| 759 |
+
"step": 1070
|
| 760 |
+
},
|
| 761 |
+
{
|
| 762 |
+
"epoch": 0.555984555984556,
|
| 763 |
+
"grad_norm": 1.2088207649167284,
|
| 764 |
+
"learning_rate": 9.253859348198971e-05,
|
| 765 |
+
"loss": 0.0824,
|
| 766 |
+
"step": 1080
|
| 767 |
+
},
|
| 768 |
+
{
|
| 769 |
+
"epoch": 0.5611325611325612,
|
| 770 |
+
"grad_norm": 0.6373898488496081,
|
| 771 |
+
"learning_rate": 9.339622641509434e-05,
|
| 772 |
+
"loss": 0.0809,
|
| 773 |
+
"step": 1090
|
| 774 |
+
},
|
| 775 |
+
{
|
| 776 |
+
"epoch": 0.5662805662805663,
|
| 777 |
+
"grad_norm": 0.5917955695640877,
|
| 778 |
+
"learning_rate": 9.425385934819899e-05,
|
| 779 |
+
"loss": 0.0742,
|
| 780 |
+
"step": 1100
|
| 781 |
+
},
|
| 782 |
+
{
|
| 783 |
+
"epoch": 0.5714285714285714,
|
| 784 |
+
"grad_norm": 0.6146276278340692,
|
| 785 |
+
"learning_rate": 9.51114922813036e-05,
|
| 786 |
+
"loss": 0.0706,
|
| 787 |
+
"step": 1110
|
| 788 |
+
},
|
| 789 |
+
{
|
| 790 |
+
"epoch": 0.5765765765765766,
|
| 791 |
+
"grad_norm": 0.8766352471472525,
|
| 792 |
+
"learning_rate": 9.596912521440823e-05,
|
| 793 |
+
"loss": 0.0672,
|
| 794 |
+
"step": 1120
|
| 795 |
+
},
|
| 796 |
+
{
|
| 797 |
+
"epoch": 0.5817245817245817,
|
| 798 |
+
"grad_norm": 0.6678260178109867,
|
| 799 |
+
"learning_rate": 9.682675814751287e-05,
|
| 800 |
+
"loss": 0.0655,
|
| 801 |
+
"step": 1130
|
| 802 |
+
},
|
| 803 |
+
{
|
| 804 |
+
"epoch": 0.5868725868725869,
|
| 805 |
+
"grad_norm": 0.49817357185278055,
|
| 806 |
+
"learning_rate": 9.76843910806175e-05,
|
| 807 |
+
"loss": 0.0611,
|
| 808 |
+
"step": 1140
|
| 809 |
+
},
|
| 810 |
+
{
|
| 811 |
+
"epoch": 0.592020592020592,
|
| 812 |
+
"grad_norm": 1.0087468424307855,
|
| 813 |
+
"learning_rate": 9.854202401372213e-05,
|
| 814 |
+
"loss": 0.0611,
|
| 815 |
+
"step": 1150
|
| 816 |
+
},
|
| 817 |
+
{
|
| 818 |
+
"epoch": 0.5971685971685972,
|
| 819 |
+
"grad_norm": 0.5362523083451194,
|
| 820 |
+
"learning_rate": 9.939965694682676e-05,
|
| 821 |
+
"loss": 0.0579,
|
| 822 |
+
"step": 1160
|
| 823 |
+
},
|
| 824 |
+
{
|
| 825 |
+
"epoch": 0.6023166023166023,
|
| 826 |
+
"grad_norm": 0.4534299709647709,
|
| 827 |
+
"learning_rate": 9.99999798272238e-05,
|
| 828 |
+
"loss": 0.0549,
|
| 829 |
+
"step": 1170
|
| 830 |
+
},
|
| 831 |
+
{
|
| 832 |
+
"epoch": 0.6074646074646075,
|
| 833 |
+
"grad_norm": 0.6406776231632558,
|
| 834 |
+
"learning_rate": 9.999962120054416e-05,
|
| 835 |
+
"loss": 0.0534,
|
| 836 |
+
"step": 1180
|
| 837 |
+
},
|
| 838 |
+
{
|
| 839 |
+
"epoch": 0.6126126126126126,
|
| 840 |
+
"grad_norm": 0.5792838241902996,
|
| 841 |
+
"learning_rate": 9.999881429364988e-05,
|
| 842 |
+
"loss": 0.0512,
|
| 843 |
+
"step": 1190
|
| 844 |
+
},
|
| 845 |
+
{
|
| 846 |
+
"epoch": 0.6177606177606177,
|
| 847 |
+
"grad_norm": 0.4275795011357426,
|
| 848 |
+
"learning_rate": 9.999755911377544e-05,
|
| 849 |
+
"loss": 0.0488,
|
| 850 |
+
"step": 1200
|
| 851 |
+
},
|
| 852 |
+
{
|
| 853 |
+
"epoch": 0.6229086229086229,
|
| 854 |
+
"grad_norm": 0.33615545845119615,
|
| 855 |
+
"learning_rate": 9.999585567217439e-05,
|
| 856 |
+
"loss": 0.0476,
|
| 857 |
+
"step": 1210
|
| 858 |
+
},
|
| 859 |
+
{
|
| 860 |
+
"epoch": 0.6280566280566281,
|
| 861 |
+
"grad_norm": 0.33655008372027123,
|
| 862 |
+
"learning_rate": 9.99937039841192e-05,
|
| 863 |
+
"loss": 0.0444,
|
| 864 |
+
"step": 1220
|
| 865 |
+
},
|
| 866 |
+
{
|
| 867 |
+
"epoch": 0.6332046332046332,
|
| 868 |
+
"grad_norm": 0.43244236090431704,
|
| 869 |
+
"learning_rate": 9.999110406890121e-05,
|
| 870 |
+
"loss": 0.0438,
|
| 871 |
+
"step": 1230
|
| 872 |
+
},
|
| 873 |
+
{
|
| 874 |
+
"epoch": 0.6383526383526383,
|
| 875 |
+
"grad_norm": 0.41139634543075815,
|
| 876 |
+
"learning_rate": 9.998805594983043e-05,
|
| 877 |
+
"loss": 0.042,
|
| 878 |
+
"step": 1240
|
| 879 |
+
},
|
| 880 |
+
{
|
| 881 |
+
"epoch": 0.6435006435006435,
|
| 882 |
+
"grad_norm": 0.5206972896361842,
|
| 883 |
+
"learning_rate": 9.998455965423525e-05,
|
| 884 |
+
"loss": 0.0431,
|
| 885 |
+
"step": 1250
|
| 886 |
+
},
|
| 887 |
+
{
|
| 888 |
+
"epoch": 0.6486486486486487,
|
| 889 |
+
"grad_norm": 0.3937459395109014,
|
| 890 |
+
"learning_rate": 9.998061521346236e-05,
|
| 891 |
+
"loss": 0.041,
|
| 892 |
+
"step": 1260
|
| 893 |
+
},
|
| 894 |
+
{
|
| 895 |
+
"epoch": 0.6537966537966537,
|
| 896 |
+
"grad_norm": 0.2963678010736684,
|
| 897 |
+
"learning_rate": 9.997622266287629e-05,
|
| 898 |
+
"loss": 0.0393,
|
| 899 |
+
"step": 1270
|
| 900 |
+
},
|
| 901 |
+
{
|
| 902 |
+
"epoch": 0.6589446589446589,
|
| 903 |
+
"grad_norm": 0.26754815540836463,
|
| 904 |
+
"learning_rate": 9.997138204185923e-05,
|
| 905 |
+
"loss": 0.0376,
|
| 906 |
+
"step": 1280
|
| 907 |
+
},
|
| 908 |
+
{
|
| 909 |
+
"epoch": 0.6640926640926641,
|
| 910 |
+
"grad_norm": 0.29045330821553156,
|
| 911 |
+
"learning_rate": 9.996609339381057e-05,
|
| 912 |
+
"loss": 0.036,
|
| 913 |
+
"step": 1290
|
| 914 |
+
},
|
| 915 |
+
{
|
| 916 |
+
"epoch": 0.6692406692406693,
|
| 917 |
+
"grad_norm": 0.3089797983346258,
|
| 918 |
+
"learning_rate": 9.996035676614663e-05,
|
| 919 |
+
"loss": 0.0356,
|
| 920 |
+
"step": 1300
|
| 921 |
+
},
|
| 922 |
+
{
|
| 923 |
+
"epoch": 0.6743886743886743,
|
| 924 |
+
"grad_norm": 0.26639822964562176,
|
| 925 |
+
"learning_rate": 9.99541722103001e-05,
|
| 926 |
+
"loss": 0.0349,
|
| 927 |
+
"step": 1310
|
| 928 |
+
},
|
| 929 |
+
{
|
| 930 |
+
"epoch": 0.6795366795366795,
|
| 931 |
+
"grad_norm": 0.45121223083554485,
|
| 932 |
+
"learning_rate": 9.994753978171971e-05,
|
| 933 |
+
"loss": 0.0351,
|
| 934 |
+
"step": 1320
|
| 935 |
+
},
|
| 936 |
+
{
|
| 937 |
+
"epoch": 0.6846846846846847,
|
| 938 |
+
"grad_norm": 0.3667468018122777,
|
| 939 |
+
"learning_rate": 9.994045953986964e-05,
|
| 940 |
+
"loss": 0.0338,
|
| 941 |
+
"step": 1330
|
| 942 |
+
},
|
| 943 |
+
{
|
| 944 |
+
"epoch": 0.6898326898326899,
|
| 945 |
+
"grad_norm": 0.24841154017565364,
|
| 946 |
+
"learning_rate": 9.993293154822901e-05,
|
| 947 |
+
"loss": 0.0328,
|
| 948 |
+
"step": 1340
|
| 949 |
+
},
|
| 950 |
+
{
|
| 951 |
+
"epoch": 0.694980694980695,
|
| 952 |
+
"grad_norm": 0.283360706082952,
|
| 953 |
+
"learning_rate": 9.992495587429129e-05,
|
| 954 |
+
"loss": 0.0315,
|
| 955 |
+
"step": 1350
|
| 956 |
+
},
|
| 957 |
+
{
|
| 958 |
+
"epoch": 0.7001287001287001,
|
| 959 |
+
"grad_norm": 0.24413900296621255,
|
| 960 |
+
"learning_rate": 9.991653258956381e-05,
|
| 961 |
+
"loss": 0.0311,
|
| 962 |
+
"step": 1360
|
| 963 |
+
},
|
| 964 |
+
{
|
| 965 |
+
"epoch": 0.7052767052767053,
|
| 966 |
+
"grad_norm": 0.28975076041601167,
|
| 967 |
+
"learning_rate": 9.990766176956697e-05,
|
| 968 |
+
"loss": 0.0303,
|
| 969 |
+
"step": 1370
|
| 970 |
+
},
|
| 971 |
+
{
|
| 972 |
+
"epoch": 0.7104247104247104,
|
| 973 |
+
"grad_norm": 0.26351763932473593,
|
| 974 |
+
"learning_rate": 9.989834349383361e-05,
|
| 975 |
+
"loss": 0.03,
|
| 976 |
+
"step": 1380
|
| 977 |
+
},
|
| 978 |
+
{
|
| 979 |
+
"epoch": 0.7155727155727156,
|
| 980 |
+
"grad_norm": 0.5331124737621773,
|
| 981 |
+
"learning_rate": 9.988857784590836e-05,
|
| 982 |
+
"loss": 0.0291,
|
| 983 |
+
"step": 1390
|
| 984 |
+
},
|
| 985 |
+
{
|
| 986 |
+
"epoch": 0.7207207207207207,
|
| 987 |
+
"grad_norm": 0.24186391898474244,
|
| 988 |
+
"learning_rate": 9.98783649133468e-05,
|
| 989 |
+
"loss": 0.0284,
|
| 990 |
+
"step": 1400
|
| 991 |
+
},
|
| 992 |
+
{
|
| 993 |
+
"epoch": 0.7258687258687259,
|
| 994 |
+
"grad_norm": 0.21615992134341075,
|
| 995 |
+
"learning_rate": 9.986770478771475e-05,
|
| 996 |
+
"loss": 0.0273,
|
| 997 |
+
"step": 1410
|
| 998 |
+
},
|
| 999 |
+
{
|
| 1000 |
+
"epoch": 0.731016731016731,
|
| 1001 |
+
"grad_norm": 0.21264572442174073,
|
| 1002 |
+
"learning_rate": 9.985659756458741e-05,
|
| 1003 |
+
"loss": 0.0267,
|
| 1004 |
+
"step": 1420
|
| 1005 |
+
},
|
| 1006 |
+
{
|
| 1007 |
+
"epoch": 0.7361647361647362,
|
| 1008 |
+
"grad_norm": 0.2320920564652625,
|
| 1009 |
+
"learning_rate": 9.984504334354847e-05,
|
| 1010 |
+
"loss": 0.0262,
|
| 1011 |
+
"step": 1430
|
| 1012 |
+
},
|
| 1013 |
+
{
|
| 1014 |
+
"epoch": 0.7413127413127413,
|
| 1015 |
+
"grad_norm": 0.2342099433653173,
|
| 1016 |
+
"learning_rate": 9.983304222818929e-05,
|
| 1017 |
+
"loss": 0.0252,
|
| 1018 |
+
"step": 1440
|
| 1019 |
+
},
|
| 1020 |
+
{
|
| 1021 |
+
"epoch": 0.7464607464607464,
|
| 1022 |
+
"grad_norm": 0.240834162772985,
|
| 1023 |
+
"learning_rate": 9.9820594326108e-05,
|
| 1024 |
+
"loss": 0.0245,
|
| 1025 |
+
"step": 1450
|
| 1026 |
+
},
|
| 1027 |
+
{
|
| 1028 |
+
"epoch": 0.7516087516087516,
|
| 1029 |
+
"grad_norm": 0.2161277200650416,
|
| 1030 |
+
"learning_rate": 9.980769974890833e-05,
|
| 1031 |
+
"loss": 0.0238,
|
| 1032 |
+
"step": 1460
|
| 1033 |
+
},
|
| 1034 |
+
{
|
| 1035 |
+
"epoch": 0.7567567567567568,
|
| 1036 |
+
"grad_norm": 0.24019983928841215,
|
| 1037 |
+
"learning_rate": 9.979435861219888e-05,
|
| 1038 |
+
"loss": 0.0233,
|
| 1039 |
+
"step": 1470
|
| 1040 |
+
},
|
| 1041 |
+
{
|
| 1042 |
+
"epoch": 0.7619047619047619,
|
| 1043 |
+
"grad_norm": 0.2970865898808262,
|
| 1044 |
+
"learning_rate": 9.978057103559188e-05,
|
| 1045 |
+
"loss": 0.0226,
|
| 1046 |
+
"step": 1480
|
| 1047 |
+
},
|
| 1048 |
+
{
|
| 1049 |
+
"epoch": 0.767052767052767,
|
| 1050 |
+
"grad_norm": 0.24780480301331292,
|
| 1051 |
+
"learning_rate": 9.976633714270226e-05,
|
| 1052 |
+
"loss": 0.0221,
|
| 1053 |
+
"step": 1490
|
| 1054 |
+
},
|
| 1055 |
+
{
|
| 1056 |
+
"epoch": 0.7722007722007722,
|
| 1057 |
+
"grad_norm": 0.25727613523882426,
|
| 1058 |
+
"learning_rate": 9.975165706114639e-05,
|
| 1059 |
+
"loss": 0.0214,
|
| 1060 |
+
"step": 1500
|
| 1061 |
+
},
|
| 1062 |
+
{
|
| 1063 |
+
"epoch": 0.7773487773487774,
|
| 1064 |
+
"grad_norm": 0.24359971967254024,
|
| 1065 |
+
"learning_rate": 9.973653092254113e-05,
|
| 1066 |
+
"loss": 0.0209,
|
| 1067 |
+
"step": 1510
|
| 1068 |
+
},
|
| 1069 |
+
{
|
| 1070 |
+
"epoch": 0.7824967824967825,
|
| 1071 |
+
"grad_norm": 0.2067746721639629,
|
| 1072 |
+
"learning_rate": 9.97209588625024e-05,
|
| 1073 |
+
"loss": 0.0202,
|
| 1074 |
+
"step": 1520
|
| 1075 |
+
},
|
| 1076 |
+
{
|
| 1077 |
+
"epoch": 0.7876447876447876,
|
| 1078 |
+
"grad_norm": 0.19304855106218685,
|
| 1079 |
+
"learning_rate": 9.970494102064427e-05,
|
| 1080 |
+
"loss": 0.0195,
|
| 1081 |
+
"step": 1530
|
| 1082 |
+
},
|
| 1083 |
+
{
|
| 1084 |
+
"epoch": 0.7927927927927928,
|
| 1085 |
+
"grad_norm": 0.22693160915077307,
|
| 1086 |
+
"learning_rate": 9.968847754057742e-05,
|
| 1087 |
+
"loss": 0.019,
|
| 1088 |
+
"step": 1540
|
| 1089 |
+
},
|
| 1090 |
+
{
|
| 1091 |
+
"epoch": 0.797940797940798,
|
| 1092 |
+
"grad_norm": 0.23060204821027103,
|
| 1093 |
+
"learning_rate": 9.967156856990801e-05,
|
| 1094 |
+
"loss": 0.0186,
|
| 1095 |
+
"step": 1550
|
| 1096 |
+
},
|
| 1097 |
+
{
|
| 1098 |
+
"epoch": 0.803088803088803,
|
| 1099 |
+
"grad_norm": 2.572239383211685,
|
| 1100 |
+
"learning_rate": 9.965421426023631e-05,
|
| 1101 |
+
"loss": 0.1139,
|
| 1102 |
+
"step": 1560
|
| 1103 |
+
},
|
| 1104 |
+
{
|
| 1105 |
+
"epoch": 0.8082368082368082,
|
| 1106 |
+
"grad_norm": 0.5437887934638203,
|
| 1107 |
+
"learning_rate": 9.963641476715541e-05,
|
| 1108 |
+
"loss": 0.0343,
|
| 1109 |
+
"step": 1570
|
| 1110 |
+
},
|
| 1111 |
+
{
|
| 1112 |
+
"epoch": 0.8133848133848134,
|
| 1113 |
+
"grad_norm": 0.23796250360876128,
|
| 1114 |
+
"learning_rate": 9.961817025024974e-05,
|
| 1115 |
+
"loss": 0.0235,
|
| 1116 |
+
"step": 1580
|
| 1117 |
+
},
|
| 1118 |
+
{
|
| 1119 |
+
"epoch": 0.8185328185328186,
|
| 1120 |
+
"grad_norm": 0.1764844020309923,
|
| 1121 |
+
"learning_rate": 9.959948087309362e-05,
|
| 1122 |
+
"loss": 0.0195,
|
| 1123 |
+
"step": 1590
|
| 1124 |
+
},
|
| 1125 |
+
{
|
| 1126 |
+
"epoch": 0.8236808236808236,
|
| 1127 |
+
"grad_norm": 0.17189189871306304,
|
| 1128 |
+
"learning_rate": 9.95803468032499e-05,
|
| 1129 |
+
"loss": 0.0177,
|
| 1130 |
+
"step": 1600
|
| 1131 |
+
},
|
| 1132 |
+
{
|
| 1133 |
+
"epoch": 0.8288288288288288,
|
| 1134 |
+
"grad_norm": 0.17751312710898323,
|
| 1135 |
+
"learning_rate": 9.956076821226836e-05,
|
| 1136 |
+
"loss": 0.0171,
|
| 1137 |
+
"step": 1610
|
| 1138 |
+
},
|
| 1139 |
+
{
|
| 1140 |
+
"epoch": 0.833976833976834,
|
| 1141 |
+
"grad_norm": 0.16437143945406296,
|
| 1142 |
+
"learning_rate": 9.954074527568426e-05,
|
| 1143 |
+
"loss": 0.0161,
|
| 1144 |
+
"step": 1620
|
| 1145 |
+
},
|
| 1146 |
+
{
|
| 1147 |
+
"epoch": 0.8391248391248392,
|
| 1148 |
+
"grad_norm": 0.15832400304822497,
|
| 1149 |
+
"learning_rate": 9.952027817301667e-05,
|
| 1150 |
+
"loss": 0.0157,
|
| 1151 |
+
"step": 1630
|
| 1152 |
+
},
|
| 1153 |
+
{
|
| 1154 |
+
"epoch": 0.8442728442728443,
|
| 1155 |
+
"grad_norm": 0.16117748572605023,
|
| 1156 |
+
"learning_rate": 9.949936708776691e-05,
|
| 1157 |
+
"loss": 0.0148,
|
| 1158 |
+
"step": 1640
|
| 1159 |
+
},
|
| 1160 |
+
{
|
| 1161 |
+
"epoch": 0.8494208494208494,
|
| 1162 |
+
"grad_norm": 0.16984391534714557,
|
| 1163 |
+
"learning_rate": 9.947801220741695e-05,
|
| 1164 |
+
"loss": 0.0143,
|
| 1165 |
+
"step": 1650
|
| 1166 |
+
},
|
| 1167 |
+
{
|
| 1168 |
+
"epoch": 0.8545688545688546,
|
| 1169 |
+
"grad_norm": 0.1696223065684197,
|
| 1170 |
+
"learning_rate": 9.945621372342762e-05,
|
| 1171 |
+
"loss": 0.0137,
|
| 1172 |
+
"step": 1660
|
| 1173 |
+
},
|
| 1174 |
+
{
|
| 1175 |
+
"epoch": 0.8597168597168597,
|
| 1176 |
+
"grad_norm": 0.1800035242900872,
|
| 1177 |
+
"learning_rate": 9.943397183123697e-05,
|
| 1178 |
+
"loss": 0.0133,
|
| 1179 |
+
"step": 1670
|
| 1180 |
+
},
|
| 1181 |
+
{
|
| 1182 |
+
"epoch": 0.8648648648648649,
|
| 1183 |
+
"grad_norm": 0.15249239519006993,
|
| 1184 |
+
"learning_rate": 9.941128673025854e-05,
|
| 1185 |
+
"loss": 0.0133,
|
| 1186 |
+
"step": 1680
|
| 1187 |
+
},
|
| 1188 |
+
{
|
| 1189 |
+
"epoch": 0.87001287001287,
|
| 1190 |
+
"grad_norm": 0.1551659031233157,
|
| 1191 |
+
"learning_rate": 9.938815862387953e-05,
|
| 1192 |
+
"loss": 0.0132,
|
| 1193 |
+
"step": 1690
|
| 1194 |
+
},
|
| 1195 |
+
{
|
| 1196 |
+
"epoch": 0.8751608751608752,
|
| 1197 |
+
"grad_norm": 0.14348975402940561,
|
| 1198 |
+
"learning_rate": 9.936458771945891e-05,
|
| 1199 |
+
"loss": 0.0131,
|
| 1200 |
+
"step": 1700
|
| 1201 |
+
},
|
| 1202 |
+
{
|
| 1203 |
+
"epoch": 0.8803088803088803,
|
| 1204 |
+
"grad_norm": 0.140601933336413,
|
| 1205 |
+
"learning_rate": 9.934057422832572e-05,
|
| 1206 |
+
"loss": 0.0123,
|
| 1207 |
+
"step": 1710
|
| 1208 |
+
},
|
| 1209 |
+
{
|
| 1210 |
+
"epoch": 0.8854568854568855,
|
| 1211 |
+
"grad_norm": 0.1275666846983677,
|
| 1212 |
+
"learning_rate": 9.931611836577703e-05,
|
| 1213 |
+
"loss": 0.012,
|
| 1214 |
+
"step": 1720
|
| 1215 |
+
},
|
| 1216 |
+
{
|
| 1217 |
+
"epoch": 0.8906048906048906,
|
| 1218 |
+
"grad_norm": 0.13327069275308084,
|
| 1219 |
+
"learning_rate": 9.92912203510761e-05,
|
| 1220 |
+
"loss": 0.012,
|
| 1221 |
+
"step": 1730
|
| 1222 |
+
},
|
| 1223 |
+
{
|
| 1224 |
+
"epoch": 0.8957528957528957,
|
| 1225 |
+
"grad_norm": 0.1658754837807317,
|
| 1226 |
+
"learning_rate": 9.926588040745035e-05,
|
| 1227 |
+
"loss": 0.0117,
|
| 1228 |
+
"step": 1740
|
| 1229 |
+
},
|
| 1230 |
+
{
|
| 1231 |
+
"epoch": 0.9009009009009009,
|
| 1232 |
+
"grad_norm": 0.14050126640493185,
|
| 1233 |
+
"learning_rate": 9.924009876208944e-05,
|
| 1234 |
+
"loss": 0.0116,
|
| 1235 |
+
"step": 1750
|
| 1236 |
+
},
|
| 1237 |
+
{
|
| 1238 |
+
"epoch": 0.9060489060489061,
|
| 1239 |
+
"grad_norm": 0.1413904681709387,
|
| 1240 |
+
"learning_rate": 9.921387564614313e-05,
|
| 1241 |
+
"loss": 0.0111,
|
| 1242 |
+
"step": 1760
|
| 1243 |
+
},
|
| 1244 |
+
{
|
| 1245 |
+
"epoch": 0.9111969111969112,
|
| 1246 |
+
"grad_norm": 0.13671787855200412,
|
| 1247 |
+
"learning_rate": 9.918721129471927e-05,
|
| 1248 |
+
"loss": 0.0113,
|
| 1249 |
+
"step": 1770
|
| 1250 |
+
},
|
| 1251 |
+
{
|
| 1252 |
+
"epoch": 0.9163449163449163,
|
| 1253 |
+
"grad_norm": 0.1300287060008007,
|
| 1254 |
+
"learning_rate": 9.916010594688173e-05,
|
| 1255 |
+
"loss": 0.011,
|
| 1256 |
+
"step": 1780
|
| 1257 |
+
},
|
| 1258 |
+
{
|
| 1259 |
+
"epoch": 0.9214929214929215,
|
| 1260 |
+
"grad_norm": 0.12272142768731135,
|
| 1261 |
+
"learning_rate": 9.913255984564814e-05,
|
| 1262 |
+
"loss": 0.0104,
|
| 1263 |
+
"step": 1790
|
| 1264 |
+
},
|
| 1265 |
+
{
|
| 1266 |
+
"epoch": 0.9266409266409267,
|
| 1267 |
+
"grad_norm": 0.13585333911080333,
|
| 1268 |
+
"learning_rate": 9.910457323798783e-05,
|
| 1269 |
+
"loss": 0.0101,
|
| 1270 |
+
"step": 1800
|
| 1271 |
+
}
|
| 1272 |
+
],
|
| 1273 |
+
"logging_steps": 10,
|
| 1274 |
+
"max_steps": 11658,
|
| 1275 |
+
"num_input_tokens_seen": 0,
|
| 1276 |
+
"num_train_epochs": 6,
|
| 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": 161574079168512.0,
|
| 1291 |
+
"train_batch_size": 128,
|
| 1292 |
+
"trial_name": null,
|
| 1293 |
+
"trial_params": null
|
| 1294 |
+
}
|