Upload ensemble_teacher2_1/checkpoint-1800/trainer_state.json with huggingface_hub
Browse files
ensemble_teacher2_1/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": 0.3561183104164606,
|
| 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": 0.0019784350578692256,
|
| 14 |
+
"grad_norm": 19.637782613049417,
|
| 15 |
+
"learning_rate": 5.932762030323006e-07,
|
| 16 |
+
"loss": 8.2273,
|
| 17 |
+
"step": 10
|
| 18 |
+
},
|
| 19 |
+
{
|
| 20 |
+
"epoch": 0.003956870115738451,
|
| 21 |
+
"grad_norm": 19.83226249327557,
|
| 22 |
+
"learning_rate": 1.2524719841793013e-06,
|
| 23 |
+
"loss": 8.167,
|
| 24 |
+
"step": 20
|
| 25 |
+
},
|
| 26 |
+
{
|
| 27 |
+
"epoch": 0.005935305173607676,
|
| 28 |
+
"grad_norm": 17.787035299527364,
|
| 29 |
+
"learning_rate": 1.9116677653263018e-06,
|
| 30 |
+
"loss": 7.8304,
|
| 31 |
+
"step": 30
|
| 32 |
+
},
|
| 33 |
+
{
|
| 34 |
+
"epoch": 0.007913740231476903,
|
| 35 |
+
"grad_norm": 6.953509182433938,
|
| 36 |
+
"learning_rate": 2.570863546473303e-06,
|
| 37 |
+
"loss": 7.0706,
|
| 38 |
+
"step": 40
|
| 39 |
+
},
|
| 40 |
+
{
|
| 41 |
+
"epoch": 0.009892175289346126,
|
| 42 |
+
"grad_norm": 4.3091682221623735,
|
| 43 |
+
"learning_rate": 3.2300593276203036e-06,
|
| 44 |
+
"loss": 6.5597,
|
| 45 |
+
"step": 50
|
| 46 |
+
},
|
| 47 |
+
{
|
| 48 |
+
"epoch": 0.011870610347215352,
|
| 49 |
+
"grad_norm": 3.0516932716653735,
|
| 50 |
+
"learning_rate": 3.889255108767304e-06,
|
| 51 |
+
"loss": 6.2279,
|
| 52 |
+
"step": 60
|
| 53 |
+
},
|
| 54 |
+
{
|
| 55 |
+
"epoch": 0.013849045405084578,
|
| 56 |
+
"grad_norm": 2.7480308952739625,
|
| 57 |
+
"learning_rate": 4.548450889914305e-06,
|
| 58 |
+
"loss": 5.9488,
|
| 59 |
+
"step": 70
|
| 60 |
+
},
|
| 61 |
+
{
|
| 62 |
+
"epoch": 0.015827480462953805,
|
| 63 |
+
"grad_norm": 3.533312550024619,
|
| 64 |
+
"learning_rate": 5.207646671061305e-06,
|
| 65 |
+
"loss": 5.6328,
|
| 66 |
+
"step": 80
|
| 67 |
+
},
|
| 68 |
+
{
|
| 69 |
+
"epoch": 0.01780591552082303,
|
| 70 |
+
"grad_norm": 3.654444836350219,
|
| 71 |
+
"learning_rate": 5.866842452208306e-06,
|
| 72 |
+
"loss": 5.1732,
|
| 73 |
+
"step": 90
|
| 74 |
+
},
|
| 75 |
+
{
|
| 76 |
+
"epoch": 0.019784350578692253,
|
| 77 |
+
"grad_norm": 4.086440552060253,
|
| 78 |
+
"learning_rate": 6.5260382333553066e-06,
|
| 79 |
+
"loss": 4.6653,
|
| 80 |
+
"step": 100
|
| 81 |
+
},
|
| 82 |
+
{
|
| 83 |
+
"epoch": 0.02176278563656148,
|
| 84 |
+
"grad_norm": 4.138046221068996,
|
| 85 |
+
"learning_rate": 7.1852340145023085e-06,
|
| 86 |
+
"loss": 4.2061,
|
| 87 |
+
"step": 110
|
| 88 |
+
},
|
| 89 |
+
{
|
| 90 |
+
"epoch": 0.023741220694430704,
|
| 91 |
+
"grad_norm": 5.7178401280922735,
|
| 92 |
+
"learning_rate": 7.844429795649308e-06,
|
| 93 |
+
"loss": 3.8123,
|
| 94 |
+
"step": 120
|
| 95 |
+
},
|
| 96 |
+
{
|
| 97 |
+
"epoch": 0.02571965575229993,
|
| 98 |
+
"grad_norm": 5.106742130319203,
|
| 99 |
+
"learning_rate": 8.503625576796309e-06,
|
| 100 |
+
"loss": 3.512,
|
| 101 |
+
"step": 130
|
| 102 |
+
},
|
| 103 |
+
{
|
| 104 |
+
"epoch": 0.027698090810169156,
|
| 105 |
+
"grad_norm": 4.573421695623524,
|
| 106 |
+
"learning_rate": 9.162821357943308e-06,
|
| 107 |
+
"loss": 3.2543,
|
| 108 |
+
"step": 140
|
| 109 |
+
},
|
| 110 |
+
{
|
| 111 |
+
"epoch": 0.029676525868038383,
|
| 112 |
+
"grad_norm": 5.128687988140343,
|
| 113 |
+
"learning_rate": 9.82201713909031e-06,
|
| 114 |
+
"loss": 3.0517,
|
| 115 |
+
"step": 150
|
| 116 |
+
},
|
| 117 |
+
{
|
| 118 |
+
"epoch": 0.03165496092590761,
|
| 119 |
+
"grad_norm": 4.880606343802238,
|
| 120 |
+
"learning_rate": 1.048121292023731e-05,
|
| 121 |
+
"loss": 2.8845,
|
| 122 |
+
"step": 160
|
| 123 |
+
},
|
| 124 |
+
{
|
| 125 |
+
"epoch": 0.033633395983776834,
|
| 126 |
+
"grad_norm": 8.127044152874786,
|
| 127 |
+
"learning_rate": 1.1140408701384312e-05,
|
| 128 |
+
"loss": 2.748,
|
| 129 |
+
"step": 170
|
| 130 |
+
},
|
| 131 |
+
{
|
| 132 |
+
"epoch": 0.03561183104164606,
|
| 133 |
+
"grad_norm": 8.485194090796208,
|
| 134 |
+
"learning_rate": 1.1799604482531313e-05,
|
| 135 |
+
"loss": 2.6414,
|
| 136 |
+
"step": 180
|
| 137 |
+
},
|
| 138 |
+
{
|
| 139 |
+
"epoch": 0.03759026609951528,
|
| 140 |
+
"grad_norm": 6.612294272350823,
|
| 141 |
+
"learning_rate": 1.2458800263678314e-05,
|
| 142 |
+
"loss": 2.5598,
|
| 143 |
+
"step": 190
|
| 144 |
+
},
|
| 145 |
+
{
|
| 146 |
+
"epoch": 0.039568701157384506,
|
| 147 |
+
"grad_norm": 12.16444744207085,
|
| 148 |
+
"learning_rate": 1.3117996044825315e-05,
|
| 149 |
+
"loss": 2.4886,
|
| 150 |
+
"step": 200
|
| 151 |
+
},
|
| 152 |
+
{
|
| 153 |
+
"epoch": 0.04154713621525374,
|
| 154 |
+
"grad_norm": 18.63069790020457,
|
| 155 |
+
"learning_rate": 1.3777191825972313e-05,
|
| 156 |
+
"loss": 2.4395,
|
| 157 |
+
"step": 210
|
| 158 |
+
},
|
| 159 |
+
{
|
| 160 |
+
"epoch": 0.04352557127312296,
|
| 161 |
+
"grad_norm": 11.412955192769877,
|
| 162 |
+
"learning_rate": 1.4436387607119314e-05,
|
| 163 |
+
"loss": 2.383,
|
| 164 |
+
"step": 220
|
| 165 |
+
},
|
| 166 |
+
{
|
| 167 |
+
"epoch": 0.045504006330992185,
|
| 168 |
+
"grad_norm": 11.202485180848445,
|
| 169 |
+
"learning_rate": 1.5095583388266315e-05,
|
| 170 |
+
"loss": 2.3341,
|
| 171 |
+
"step": 230
|
| 172 |
+
},
|
| 173 |
+
{
|
| 174 |
+
"epoch": 0.04748244138886141,
|
| 175 |
+
"grad_norm": 31.060419144311027,
|
| 176 |
+
"learning_rate": 1.5754779169413318e-05,
|
| 177 |
+
"loss": 2.3041,
|
| 178 |
+
"step": 240
|
| 179 |
+
},
|
| 180 |
+
{
|
| 181 |
+
"epoch": 0.04946087644673064,
|
| 182 |
+
"grad_norm": 22.03166390729224,
|
| 183 |
+
"learning_rate": 1.6413974950560317e-05,
|
| 184 |
+
"loss": 2.2843,
|
| 185 |
+
"step": 250
|
| 186 |
+
},
|
| 187 |
+
{
|
| 188 |
+
"epoch": 0.05143931150459986,
|
| 189 |
+
"grad_norm": 32.84287275099665,
|
| 190 |
+
"learning_rate": 1.707317073170732e-05,
|
| 191 |
+
"loss": 2.2874,
|
| 192 |
+
"step": 260
|
| 193 |
+
},
|
| 194 |
+
{
|
| 195 |
+
"epoch": 0.05341774656246909,
|
| 196 |
+
"grad_norm": 13.792898956806326,
|
| 197 |
+
"learning_rate": 1.7732366512854316e-05,
|
| 198 |
+
"loss": 2.2134,
|
| 199 |
+
"step": 270
|
| 200 |
+
},
|
| 201 |
+
{
|
| 202 |
+
"epoch": 0.05539618162033831,
|
| 203 |
+
"grad_norm": 9.703481654155626,
|
| 204 |
+
"learning_rate": 1.839156229400132e-05,
|
| 205 |
+
"loss": 2.1511,
|
| 206 |
+
"step": 280
|
| 207 |
+
},
|
| 208 |
+
{
|
| 209 |
+
"epoch": 0.057374616678207535,
|
| 210 |
+
"grad_norm": 18.160904957620193,
|
| 211 |
+
"learning_rate": 1.905075807514832e-05,
|
| 212 |
+
"loss": 2.1061,
|
| 213 |
+
"step": 290
|
| 214 |
+
},
|
| 215 |
+
{
|
| 216 |
+
"epoch": 0.059353051736076766,
|
| 217 |
+
"grad_norm": 31.18164400199025,
|
| 218 |
+
"learning_rate": 1.970995385629532e-05,
|
| 219 |
+
"loss": 2.0923,
|
| 220 |
+
"step": 300
|
| 221 |
+
},
|
| 222 |
+
{
|
| 223 |
+
"epoch": 0.06133148679394599,
|
| 224 |
+
"grad_norm": 18.352406637861762,
|
| 225 |
+
"learning_rate": 2.0369149637442324e-05,
|
| 226 |
+
"loss": 2.0124,
|
| 227 |
+
"step": 310
|
| 228 |
+
},
|
| 229 |
+
{
|
| 230 |
+
"epoch": 0.06330992185181522,
|
| 231 |
+
"grad_norm": 13.168827033502875,
|
| 232 |
+
"learning_rate": 2.102834541858932e-05,
|
| 233 |
+
"loss": 1.9455,
|
| 234 |
+
"step": 320
|
| 235 |
+
},
|
| 236 |
+
{
|
| 237 |
+
"epoch": 0.06528835690968444,
|
| 238 |
+
"grad_norm": 12.98291760357499,
|
| 239 |
+
"learning_rate": 2.1687541199736323e-05,
|
| 240 |
+
"loss": 1.8702,
|
| 241 |
+
"step": 330
|
| 242 |
+
},
|
| 243 |
+
{
|
| 244 |
+
"epoch": 0.06726679196755367,
|
| 245 |
+
"grad_norm": 11.475153571817428,
|
| 246 |
+
"learning_rate": 2.2346736980883322e-05,
|
| 247 |
+
"loss": 1.7829,
|
| 248 |
+
"step": 340
|
| 249 |
+
},
|
| 250 |
+
{
|
| 251 |
+
"epoch": 0.06924522702542289,
|
| 252 |
+
"grad_norm": 14.551237589997115,
|
| 253 |
+
"learning_rate": 2.3005932762030325e-05,
|
| 254 |
+
"loss": 1.6901,
|
| 255 |
+
"step": 350
|
| 256 |
+
},
|
| 257 |
+
{
|
| 258 |
+
"epoch": 0.07122366208329212,
|
| 259 |
+
"grad_norm": 9.667096995303945,
|
| 260 |
+
"learning_rate": 2.3665128543177324e-05,
|
| 261 |
+
"loss": 1.5745,
|
| 262 |
+
"step": 360
|
| 263 |
+
},
|
| 264 |
+
{
|
| 265 |
+
"epoch": 0.07320209714116134,
|
| 266 |
+
"grad_norm": 13.163411291543234,
|
| 267 |
+
"learning_rate": 2.4324324324324327e-05,
|
| 268 |
+
"loss": 1.4658,
|
| 269 |
+
"step": 370
|
| 270 |
+
},
|
| 271 |
+
{
|
| 272 |
+
"epoch": 0.07518053219903056,
|
| 273 |
+
"grad_norm": 9.940351335922,
|
| 274 |
+
"learning_rate": 2.4983520105471327e-05,
|
| 275 |
+
"loss": 1.3668,
|
| 276 |
+
"step": 380
|
| 277 |
+
},
|
| 278 |
+
{
|
| 279 |
+
"epoch": 0.07715896725689979,
|
| 280 |
+
"grad_norm": 13.992341034680026,
|
| 281 |
+
"learning_rate": 2.564271588661833e-05,
|
| 282 |
+
"loss": 1.2807,
|
| 283 |
+
"step": 390
|
| 284 |
+
},
|
| 285 |
+
{
|
| 286 |
+
"epoch": 0.07913740231476901,
|
| 287 |
+
"grad_norm": 7.4831699972814105,
|
| 288 |
+
"learning_rate": 2.630191166776533e-05,
|
| 289 |
+
"loss": 1.211,
|
| 290 |
+
"step": 400
|
| 291 |
+
},
|
| 292 |
+
{
|
| 293 |
+
"epoch": 0.08111583737263825,
|
| 294 |
+
"grad_norm": 9.56614415520462,
|
| 295 |
+
"learning_rate": 2.6961107448912325e-05,
|
| 296 |
+
"loss": 1.1591,
|
| 297 |
+
"step": 410
|
| 298 |
+
},
|
| 299 |
+
{
|
| 300 |
+
"epoch": 0.08309427243050747,
|
| 301 |
+
"grad_norm": 9.870662837309114,
|
| 302 |
+
"learning_rate": 2.762030323005933e-05,
|
| 303 |
+
"loss": 1.1007,
|
| 304 |
+
"step": 420
|
| 305 |
+
},
|
| 306 |
+
{
|
| 307 |
+
"epoch": 0.0850727074883767,
|
| 308 |
+
"grad_norm": 7.296161468394743,
|
| 309 |
+
"learning_rate": 2.8279499011206327e-05,
|
| 310 |
+
"loss": 1.0605,
|
| 311 |
+
"step": 430
|
| 312 |
+
},
|
| 313 |
+
{
|
| 314 |
+
"epoch": 0.08705114254624592,
|
| 315 |
+
"grad_norm": 9.934379350610987,
|
| 316 |
+
"learning_rate": 2.8938694792353333e-05,
|
| 317 |
+
"loss": 1.0221,
|
| 318 |
+
"step": 440
|
| 319 |
+
},
|
| 320 |
+
{
|
| 321 |
+
"epoch": 0.08902957760411515,
|
| 322 |
+
"grad_norm": 8.322773668200096,
|
| 323 |
+
"learning_rate": 2.959789057350033e-05,
|
| 324 |
+
"loss": 0.9699,
|
| 325 |
+
"step": 450
|
| 326 |
+
},
|
| 327 |
+
{
|
| 328 |
+
"epoch": 0.09100801266198437,
|
| 329 |
+
"grad_norm": 11.476438775344631,
|
| 330 |
+
"learning_rate": 3.025708635464733e-05,
|
| 331 |
+
"loss": 0.9331,
|
| 332 |
+
"step": 460
|
| 333 |
+
},
|
| 334 |
+
{
|
| 335 |
+
"epoch": 0.0929864477198536,
|
| 336 |
+
"grad_norm": 7.972860756568456,
|
| 337 |
+
"learning_rate": 3.0916282135794335e-05,
|
| 338 |
+
"loss": 0.9203,
|
| 339 |
+
"step": 470
|
| 340 |
+
},
|
| 341 |
+
{
|
| 342 |
+
"epoch": 0.09496488277772282,
|
| 343 |
+
"grad_norm": 11.520251589320083,
|
| 344 |
+
"learning_rate": 3.1575477916941334e-05,
|
| 345 |
+
"loss": 0.8832,
|
| 346 |
+
"step": 480
|
| 347 |
+
},
|
| 348 |
+
{
|
| 349 |
+
"epoch": 0.09694331783559204,
|
| 350 |
+
"grad_norm": 9.465237203788785,
|
| 351 |
+
"learning_rate": 3.2234673698088334e-05,
|
| 352 |
+
"loss": 0.8484,
|
| 353 |
+
"step": 490
|
| 354 |
+
},
|
| 355 |
+
{
|
| 356 |
+
"epoch": 0.09892175289346128,
|
| 357 |
+
"grad_norm": 13.258709856888437,
|
| 358 |
+
"learning_rate": 3.289386947923533e-05,
|
| 359 |
+
"loss": 0.8216,
|
| 360 |
+
"step": 500
|
| 361 |
+
},
|
| 362 |
+
{
|
| 363 |
+
"epoch": 0.1009001879513305,
|
| 364 |
+
"grad_norm": 9.067024221441216,
|
| 365 |
+
"learning_rate": 3.355306526038233e-05,
|
| 366 |
+
"loss": 0.806,
|
| 367 |
+
"step": 510
|
| 368 |
+
},
|
| 369 |
+
{
|
| 370 |
+
"epoch": 0.10287862300919973,
|
| 371 |
+
"grad_norm": 9.967066995989914,
|
| 372 |
+
"learning_rate": 3.421226104152934e-05,
|
| 373 |
+
"loss": 0.7768,
|
| 374 |
+
"step": 520
|
| 375 |
+
},
|
| 376 |
+
{
|
| 377 |
+
"epoch": 0.10485705806706895,
|
| 378 |
+
"grad_norm": 6.990448616699275,
|
| 379 |
+
"learning_rate": 3.487145682267634e-05,
|
| 380 |
+
"loss": 0.7528,
|
| 381 |
+
"step": 530
|
| 382 |
+
},
|
| 383 |
+
{
|
| 384 |
+
"epoch": 0.10683549312493817,
|
| 385 |
+
"grad_norm": 8.863709916752113,
|
| 386 |
+
"learning_rate": 3.553065260382334e-05,
|
| 387 |
+
"loss": 0.7225,
|
| 388 |
+
"step": 540
|
| 389 |
+
},
|
| 390 |
+
{
|
| 391 |
+
"epoch": 0.1088139281828074,
|
| 392 |
+
"grad_norm": 8.219483179811172,
|
| 393 |
+
"learning_rate": 3.618984838497034e-05,
|
| 394 |
+
"loss": 0.6902,
|
| 395 |
+
"step": 550
|
| 396 |
+
},
|
| 397 |
+
{
|
| 398 |
+
"epoch": 0.11079236324067662,
|
| 399 |
+
"grad_norm": 8.18954618815931,
|
| 400 |
+
"learning_rate": 3.684904416611734e-05,
|
| 401 |
+
"loss": 0.6664,
|
| 402 |
+
"step": 560
|
| 403 |
+
},
|
| 404 |
+
{
|
| 405 |
+
"epoch": 0.11277079829854585,
|
| 406 |
+
"grad_norm": 6.99844977800662,
|
| 407 |
+
"learning_rate": 3.750823994726434e-05,
|
| 408 |
+
"loss": 0.6444,
|
| 409 |
+
"step": 570
|
| 410 |
+
},
|
| 411 |
+
{
|
| 412 |
+
"epoch": 0.11474923335641507,
|
| 413 |
+
"grad_norm": 7.503046833377609,
|
| 414 |
+
"learning_rate": 3.8167435728411335e-05,
|
| 415 |
+
"loss": 0.6149,
|
| 416 |
+
"step": 580
|
| 417 |
+
},
|
| 418 |
+
{
|
| 419 |
+
"epoch": 0.11672766841428431,
|
| 420 |
+
"grad_norm": 6.722656169187286,
|
| 421 |
+
"learning_rate": 3.882663150955834e-05,
|
| 422 |
+
"loss": 0.6168,
|
| 423 |
+
"step": 590
|
| 424 |
+
},
|
| 425 |
+
{
|
| 426 |
+
"epoch": 0.11870610347215353,
|
| 427 |
+
"grad_norm": 7.030946593625458,
|
| 428 |
+
"learning_rate": 3.948582729070534e-05,
|
| 429 |
+
"loss": 0.5881,
|
| 430 |
+
"step": 600
|
| 431 |
+
},
|
| 432 |
+
{
|
| 433 |
+
"epoch": 0.12068453853002276,
|
| 434 |
+
"grad_norm": 9.774246488301465,
|
| 435 |
+
"learning_rate": 4.014502307185235e-05,
|
| 436 |
+
"loss": 0.5607,
|
| 437 |
+
"step": 610
|
| 438 |
+
},
|
| 439 |
+
{
|
| 440 |
+
"epoch": 0.12266297358789198,
|
| 441 |
+
"grad_norm": 7.078903913735779,
|
| 442 |
+
"learning_rate": 4.0804218852999346e-05,
|
| 443 |
+
"loss": 0.5405,
|
| 444 |
+
"step": 620
|
| 445 |
+
},
|
| 446 |
+
{
|
| 447 |
+
"epoch": 0.1246414086457612,
|
| 448 |
+
"grad_norm": 7.402770055900705,
|
| 449 |
+
"learning_rate": 4.146341463414634e-05,
|
| 450 |
+
"loss": 0.5212,
|
| 451 |
+
"step": 630
|
| 452 |
+
},
|
| 453 |
+
{
|
| 454 |
+
"epoch": 0.12661984370363044,
|
| 455 |
+
"grad_norm": 7.182019494715211,
|
| 456 |
+
"learning_rate": 4.2122610415293345e-05,
|
| 457 |
+
"loss": 0.4983,
|
| 458 |
+
"step": 640
|
| 459 |
+
},
|
| 460 |
+
{
|
| 461 |
+
"epoch": 0.12859827876149965,
|
| 462 |
+
"grad_norm": 5.276303108063209,
|
| 463 |
+
"learning_rate": 4.2781806196440345e-05,
|
| 464 |
+
"loss": 0.4771,
|
| 465 |
+
"step": 650
|
| 466 |
+
},
|
| 467 |
+
{
|
| 468 |
+
"epoch": 0.1305767138193689,
|
| 469 |
+
"grad_norm": 4.7737357936486235,
|
| 470 |
+
"learning_rate": 4.344100197758735e-05,
|
| 471 |
+
"loss": 0.4604,
|
| 472 |
+
"step": 660
|
| 473 |
+
},
|
| 474 |
+
{
|
| 475 |
+
"epoch": 0.1325551488772381,
|
| 476 |
+
"grad_norm": 5.4871112446079655,
|
| 477 |
+
"learning_rate": 4.4100197758734343e-05,
|
| 478 |
+
"loss": 0.4532,
|
| 479 |
+
"step": 670
|
| 480 |
+
},
|
| 481 |
+
{
|
| 482 |
+
"epoch": 0.13453358393510734,
|
| 483 |
+
"grad_norm": 6.595412809013484,
|
| 484 |
+
"learning_rate": 4.475939353988134e-05,
|
| 485 |
+
"loss": 0.4311,
|
| 486 |
+
"step": 680
|
| 487 |
+
},
|
| 488 |
+
{
|
| 489 |
+
"epoch": 0.13651201899297655,
|
| 490 |
+
"grad_norm": 4.841829206426501,
|
| 491 |
+
"learning_rate": 4.541858932102835e-05,
|
| 492 |
+
"loss": 0.4181,
|
| 493 |
+
"step": 690
|
| 494 |
+
},
|
| 495 |
+
{
|
| 496 |
+
"epoch": 0.13849045405084578,
|
| 497 |
+
"grad_norm": 4.7217934333031435,
|
| 498 |
+
"learning_rate": 4.607778510217535e-05,
|
| 499 |
+
"loss": 0.3928,
|
| 500 |
+
"step": 700
|
| 501 |
+
},
|
| 502 |
+
{
|
| 503 |
+
"epoch": 0.140468889108715,
|
| 504 |
+
"grad_norm": 6.026086612922775,
|
| 505 |
+
"learning_rate": 4.673698088332235e-05,
|
| 506 |
+
"loss": 0.3897,
|
| 507 |
+
"step": 710
|
| 508 |
+
},
|
| 509 |
+
{
|
| 510 |
+
"epoch": 0.14244732416658423,
|
| 511 |
+
"grad_norm": 4.396456746375336,
|
| 512 |
+
"learning_rate": 4.739617666446935e-05,
|
| 513 |
+
"loss": 0.3701,
|
| 514 |
+
"step": 720
|
| 515 |
+
},
|
| 516 |
+
{
|
| 517 |
+
"epoch": 0.14442575922445347,
|
| 518 |
+
"grad_norm": 6.14790077100577,
|
| 519 |
+
"learning_rate": 4.805537244561635e-05,
|
| 520 |
+
"loss": 0.3488,
|
| 521 |
+
"step": 730
|
| 522 |
+
},
|
| 523 |
+
{
|
| 524 |
+
"epoch": 0.14640419428232268,
|
| 525 |
+
"grad_norm": 5.390368007008413,
|
| 526 |
+
"learning_rate": 4.871456822676335e-05,
|
| 527 |
+
"loss": 0.3357,
|
| 528 |
+
"step": 740
|
| 529 |
+
},
|
| 530 |
+
{
|
| 531 |
+
"epoch": 0.14838262934019192,
|
| 532 |
+
"grad_norm": 4.441570962528891,
|
| 533 |
+
"learning_rate": 4.937376400791035e-05,
|
| 534 |
+
"loss": 0.3203,
|
| 535 |
+
"step": 750
|
| 536 |
+
},
|
| 537 |
+
{
|
| 538 |
+
"epoch": 0.15036106439806113,
|
| 539 |
+
"grad_norm": 5.270754499685052,
|
| 540 |
+
"learning_rate": 5.003295978905735e-05,
|
| 541 |
+
"loss": 0.3071,
|
| 542 |
+
"step": 760
|
| 543 |
+
},
|
| 544 |
+
{
|
| 545 |
+
"epoch": 0.15233949945593037,
|
| 546 |
+
"grad_norm": 4.071551636960703,
|
| 547 |
+
"learning_rate": 5.069215557020436e-05,
|
| 548 |
+
"loss": 0.2991,
|
| 549 |
+
"step": 770
|
| 550 |
+
},
|
| 551 |
+
{
|
| 552 |
+
"epoch": 0.15431793451379958,
|
| 553 |
+
"grad_norm": 3.0561647726741783,
|
| 554 |
+
"learning_rate": 5.135135135135135e-05,
|
| 555 |
+
"loss": 0.2832,
|
| 556 |
+
"step": 780
|
| 557 |
+
},
|
| 558 |
+
{
|
| 559 |
+
"epoch": 0.15629636957166881,
|
| 560 |
+
"grad_norm": 2.9516027116400427,
|
| 561 |
+
"learning_rate": 5.201054713249836e-05,
|
| 562 |
+
"loss": 0.2681,
|
| 563 |
+
"step": 790
|
| 564 |
+
},
|
| 565 |
+
{
|
| 566 |
+
"epoch": 0.15827480462953802,
|
| 567 |
+
"grad_norm": 4.621200447731901,
|
| 568 |
+
"learning_rate": 5.2669742913645356e-05,
|
| 569 |
+
"loss": 0.26,
|
| 570 |
+
"step": 800
|
| 571 |
+
},
|
| 572 |
+
{
|
| 573 |
+
"epoch": 0.16025323968740726,
|
| 574 |
+
"grad_norm": 3.041429947319794,
|
| 575 |
+
"learning_rate": 5.332893869479235e-05,
|
| 576 |
+
"loss": 0.2471,
|
| 577 |
+
"step": 810
|
| 578 |
+
},
|
| 579 |
+
{
|
| 580 |
+
"epoch": 0.1622316747452765,
|
| 581 |
+
"grad_norm": 2.995343942067045,
|
| 582 |
+
"learning_rate": 5.3988134475939355e-05,
|
| 583 |
+
"loss": 0.2369,
|
| 584 |
+
"step": 820
|
| 585 |
+
},
|
| 586 |
+
{
|
| 587 |
+
"epoch": 0.1642101098031457,
|
| 588 |
+
"grad_norm": 3.1918861691745795,
|
| 589 |
+
"learning_rate": 5.464733025708636e-05,
|
| 590 |
+
"loss": 0.2279,
|
| 591 |
+
"step": 830
|
| 592 |
+
},
|
| 593 |
+
{
|
| 594 |
+
"epoch": 0.16618854486101495,
|
| 595 |
+
"grad_norm": 2.1160708161237194,
|
| 596 |
+
"learning_rate": 5.5306526038233354e-05,
|
| 597 |
+
"loss": 0.2168,
|
| 598 |
+
"step": 840
|
| 599 |
+
},
|
| 600 |
+
{
|
| 601 |
+
"epoch": 0.16816697991888416,
|
| 602 |
+
"grad_norm": 2.849132153355752,
|
| 603 |
+
"learning_rate": 5.596572181938036e-05,
|
| 604 |
+
"loss": 0.2107,
|
| 605 |
+
"step": 850
|
| 606 |
+
},
|
| 607 |
+
{
|
| 608 |
+
"epoch": 0.1701454149767534,
|
| 609 |
+
"grad_norm": 2.5306963483355673,
|
| 610 |
+
"learning_rate": 5.662491760052736e-05,
|
| 611 |
+
"loss": 0.1974,
|
| 612 |
+
"step": 860
|
| 613 |
+
},
|
| 614 |
+
{
|
| 615 |
+
"epoch": 0.1721238500346226,
|
| 616 |
+
"grad_norm": 3.134461921573245,
|
| 617 |
+
"learning_rate": 5.7284113381674366e-05,
|
| 618 |
+
"loss": 0.1898,
|
| 619 |
+
"step": 870
|
| 620 |
+
},
|
| 621 |
+
{
|
| 622 |
+
"epoch": 0.17410228509249184,
|
| 623 |
+
"grad_norm": 4.467213018542406,
|
| 624 |
+
"learning_rate": 5.794330916282136e-05,
|
| 625 |
+
"loss": 0.1918,
|
| 626 |
+
"step": 880
|
| 627 |
+
},
|
| 628 |
+
{
|
| 629 |
+
"epoch": 0.17608072015036105,
|
| 630 |
+
"grad_norm": 1.9384617338789845,
|
| 631 |
+
"learning_rate": 5.8602504943968364e-05,
|
| 632 |
+
"loss": 0.1846,
|
| 633 |
+
"step": 890
|
| 634 |
+
},
|
| 635 |
+
{
|
| 636 |
+
"epoch": 0.1780591552082303,
|
| 637 |
+
"grad_norm": 2.22914110082071,
|
| 638 |
+
"learning_rate": 5.9261700725115364e-05,
|
| 639 |
+
"loss": 0.176,
|
| 640 |
+
"step": 900
|
| 641 |
+
},
|
| 642 |
+
{
|
| 643 |
+
"epoch": 0.18003759026609953,
|
| 644 |
+
"grad_norm": 2.6195670505855104,
|
| 645 |
+
"learning_rate": 5.9920896506262356e-05,
|
| 646 |
+
"loss": 0.1645,
|
| 647 |
+
"step": 910
|
| 648 |
+
},
|
| 649 |
+
{
|
| 650 |
+
"epoch": 0.18201602532396874,
|
| 651 |
+
"grad_norm": 5.623481408126527,
|
| 652 |
+
"learning_rate": 6.058009228740936e-05,
|
| 653 |
+
"loss": 0.1614,
|
| 654 |
+
"step": 920
|
| 655 |
+
},
|
| 656 |
+
{
|
| 657 |
+
"epoch": 0.18399446038183798,
|
| 658 |
+
"grad_norm": 2.541850203721254,
|
| 659 |
+
"learning_rate": 6.123928806855637e-05,
|
| 660 |
+
"loss": 0.1562,
|
| 661 |
+
"step": 930
|
| 662 |
+
},
|
| 663 |
+
{
|
| 664 |
+
"epoch": 0.1859728954397072,
|
| 665 |
+
"grad_norm": 2.530270886035028,
|
| 666 |
+
"learning_rate": 6.189848384970337e-05,
|
| 667 |
+
"loss": 0.1498,
|
| 668 |
+
"step": 940
|
| 669 |
+
},
|
| 670 |
+
{
|
| 671 |
+
"epoch": 0.18795133049757642,
|
| 672 |
+
"grad_norm": 1.8747639348929863,
|
| 673 |
+
"learning_rate": 6.255767963085037e-05,
|
| 674 |
+
"loss": 0.1418,
|
| 675 |
+
"step": 950
|
| 676 |
+
},
|
| 677 |
+
{
|
| 678 |
+
"epoch": 0.18992976555544563,
|
| 679 |
+
"grad_norm": 1.6302710016776405,
|
| 680 |
+
"learning_rate": 6.321687541199737e-05,
|
| 681 |
+
"loss": 0.1353,
|
| 682 |
+
"step": 960
|
| 683 |
+
},
|
| 684 |
+
{
|
| 685 |
+
"epoch": 0.19190820061331487,
|
| 686 |
+
"grad_norm": 1.374949112823051,
|
| 687 |
+
"learning_rate": 6.387607119314437e-05,
|
| 688 |
+
"loss": 0.1303,
|
| 689 |
+
"step": 970
|
| 690 |
+
},
|
| 691 |
+
{
|
| 692 |
+
"epoch": 0.19388663567118408,
|
| 693 |
+
"grad_norm": 2.300554535314371,
|
| 694 |
+
"learning_rate": 6.453526697429137e-05,
|
| 695 |
+
"loss": 0.1262,
|
| 696 |
+
"step": 980
|
| 697 |
+
},
|
| 698 |
+
{
|
| 699 |
+
"epoch": 0.19586507072905332,
|
| 700 |
+
"grad_norm": 1.6596926456190018,
|
| 701 |
+
"learning_rate": 6.519446275543837e-05,
|
| 702 |
+
"loss": 0.1338,
|
| 703 |
+
"step": 990
|
| 704 |
+
},
|
| 705 |
+
{
|
| 706 |
+
"epoch": 0.19784350578692256,
|
| 707 |
+
"grad_norm": 1.6448857589126298,
|
| 708 |
+
"learning_rate": 6.585365853658538e-05,
|
| 709 |
+
"loss": 0.1282,
|
| 710 |
+
"step": 1000
|
| 711 |
+
},
|
| 712 |
+
{
|
| 713 |
+
"epoch": 0.19982194084479177,
|
| 714 |
+
"grad_norm": 1.4480849991974878,
|
| 715 |
+
"learning_rate": 6.651285431773236e-05,
|
| 716 |
+
"loss": 0.116,
|
| 717 |
+
"step": 1010
|
| 718 |
+
},
|
| 719 |
+
{
|
| 720 |
+
"epoch": 0.201800375902661,
|
| 721 |
+
"grad_norm": 1.8869991728949147,
|
| 722 |
+
"learning_rate": 6.717205009887936e-05,
|
| 723 |
+
"loss": 0.1104,
|
| 724 |
+
"step": 1020
|
| 725 |
+
},
|
| 726 |
+
{
|
| 727 |
+
"epoch": 0.20377881096053022,
|
| 728 |
+
"grad_norm": 1.6293318832625847,
|
| 729 |
+
"learning_rate": 6.783124588002638e-05,
|
| 730 |
+
"loss": 0.1061,
|
| 731 |
+
"step": 1030
|
| 732 |
+
},
|
| 733 |
+
{
|
| 734 |
+
"epoch": 0.20575724601839945,
|
| 735 |
+
"grad_norm": 1.2427464212813581,
|
| 736 |
+
"learning_rate": 6.849044166117338e-05,
|
| 737 |
+
"loss": 0.1022,
|
| 738 |
+
"step": 1040
|
| 739 |
+
},
|
| 740 |
+
{
|
| 741 |
+
"epoch": 0.20773568107626866,
|
| 742 |
+
"grad_norm": 1.2404060214766448,
|
| 743 |
+
"learning_rate": 6.914963744232038e-05,
|
| 744 |
+
"loss": 0.0974,
|
| 745 |
+
"step": 1050
|
| 746 |
+
},
|
| 747 |
+
{
|
| 748 |
+
"epoch": 0.2097141161341379,
|
| 749 |
+
"grad_norm": 6.9065034819409625,
|
| 750 |
+
"learning_rate": 6.980883322346737e-05,
|
| 751 |
+
"loss": 0.0977,
|
| 752 |
+
"step": 1060
|
| 753 |
+
},
|
| 754 |
+
{
|
| 755 |
+
"epoch": 0.2116925511920071,
|
| 756 |
+
"grad_norm": 2.190513083820012,
|
| 757 |
+
"learning_rate": 7.046802900461437e-05,
|
| 758 |
+
"loss": 0.1328,
|
| 759 |
+
"step": 1070
|
| 760 |
+
},
|
| 761 |
+
{
|
| 762 |
+
"epoch": 0.21367098624987635,
|
| 763 |
+
"grad_norm": 1.263188586172193,
|
| 764 |
+
"learning_rate": 7.112722478576137e-05,
|
| 765 |
+
"loss": 0.0917,
|
| 766 |
+
"step": 1080
|
| 767 |
+
},
|
| 768 |
+
{
|
| 769 |
+
"epoch": 0.2156494213077456,
|
| 770 |
+
"grad_norm": 1.9784069526825092,
|
| 771 |
+
"learning_rate": 7.178642056690837e-05,
|
| 772 |
+
"loss": 0.0871,
|
| 773 |
+
"step": 1090
|
| 774 |
+
},
|
| 775 |
+
{
|
| 776 |
+
"epoch": 0.2176278563656148,
|
| 777 |
+
"grad_norm": 0.8523617558822155,
|
| 778 |
+
"learning_rate": 7.244561634805537e-05,
|
| 779 |
+
"loss": 0.085,
|
| 780 |
+
"step": 1100
|
| 781 |
+
},
|
| 782 |
+
{
|
| 783 |
+
"epoch": 0.21960629142348403,
|
| 784 |
+
"grad_norm": 0.8144860765656456,
|
| 785 |
+
"learning_rate": 7.310481212920237e-05,
|
| 786 |
+
"loss": 0.0775,
|
| 787 |
+
"step": 1110
|
| 788 |
+
},
|
| 789 |
+
{
|
| 790 |
+
"epoch": 0.22158472648135324,
|
| 791 |
+
"grad_norm": 0.8322446467787168,
|
| 792 |
+
"learning_rate": 7.376400791034937e-05,
|
| 793 |
+
"loss": 0.072,
|
| 794 |
+
"step": 1120
|
| 795 |
+
},
|
| 796 |
+
{
|
| 797 |
+
"epoch": 0.22356316153922248,
|
| 798 |
+
"grad_norm": 0.8840235823881915,
|
| 799 |
+
"learning_rate": 7.442320369149638e-05,
|
| 800 |
+
"loss": 0.0781,
|
| 801 |
+
"step": 1130
|
| 802 |
+
},
|
| 803 |
+
{
|
| 804 |
+
"epoch": 0.2255415965970917,
|
| 805 |
+
"grad_norm": 0.7406476619212958,
|
| 806 |
+
"learning_rate": 7.508239947264338e-05,
|
| 807 |
+
"loss": 0.067,
|
| 808 |
+
"step": 1140
|
| 809 |
+
},
|
| 810 |
+
{
|
| 811 |
+
"epoch": 0.22752003165496093,
|
| 812 |
+
"grad_norm": 0.6135846645545082,
|
| 813 |
+
"learning_rate": 7.574159525379037e-05,
|
| 814 |
+
"loss": 0.0625,
|
| 815 |
+
"step": 1150
|
| 816 |
+
},
|
| 817 |
+
{
|
| 818 |
+
"epoch": 0.22949846671283014,
|
| 819 |
+
"grad_norm": 0.8497087947713994,
|
| 820 |
+
"learning_rate": 7.640079103493738e-05,
|
| 821 |
+
"loss": 0.0588,
|
| 822 |
+
"step": 1160
|
| 823 |
+
},
|
| 824 |
+
{
|
| 825 |
+
"epoch": 0.23147690177069938,
|
| 826 |
+
"grad_norm": 0.5684521607051449,
|
| 827 |
+
"learning_rate": 7.705998681608438e-05,
|
| 828 |
+
"loss": 0.0559,
|
| 829 |
+
"step": 1170
|
| 830 |
+
},
|
| 831 |
+
{
|
| 832 |
+
"epoch": 0.23345533682856862,
|
| 833 |
+
"grad_norm": 0.44642133269859746,
|
| 834 |
+
"learning_rate": 7.771918259723138e-05,
|
| 835 |
+
"loss": 0.0524,
|
| 836 |
+
"step": 1180
|
| 837 |
+
},
|
| 838 |
+
{
|
| 839 |
+
"epoch": 0.23543377188643783,
|
| 840 |
+
"grad_norm": 0.7417253246437843,
|
| 841 |
+
"learning_rate": 7.837837837837838e-05,
|
| 842 |
+
"loss": 0.049,
|
| 843 |
+
"step": 1190
|
| 844 |
+
},
|
| 845 |
+
{
|
| 846 |
+
"epoch": 0.23741220694430706,
|
| 847 |
+
"grad_norm": 0.6699824985744847,
|
| 848 |
+
"learning_rate": 7.903757415952538e-05,
|
| 849 |
+
"loss": 0.0489,
|
| 850 |
+
"step": 1200
|
| 851 |
+
},
|
| 852 |
+
{
|
| 853 |
+
"epoch": 0.23939064200217627,
|
| 854 |
+
"grad_norm": 0.4584409365864605,
|
| 855 |
+
"learning_rate": 7.969676994067239e-05,
|
| 856 |
+
"loss": 0.0459,
|
| 857 |
+
"step": 1210
|
| 858 |
+
},
|
| 859 |
+
{
|
| 860 |
+
"epoch": 0.2413690770600455,
|
| 861 |
+
"grad_norm": 0.44232745755322544,
|
| 862 |
+
"learning_rate": 8.035596572181938e-05,
|
| 863 |
+
"loss": 0.0436,
|
| 864 |
+
"step": 1220
|
| 865 |
+
},
|
| 866 |
+
{
|
| 867 |
+
"epoch": 0.24334751211791472,
|
| 868 |
+
"grad_norm": 0.3867957221399885,
|
| 869 |
+
"learning_rate": 8.101516150296639e-05,
|
| 870 |
+
"loss": 0.0407,
|
| 871 |
+
"step": 1230
|
| 872 |
+
},
|
| 873 |
+
{
|
| 874 |
+
"epoch": 0.24532594717578396,
|
| 875 |
+
"grad_norm": 0.41432418911839947,
|
| 876 |
+
"learning_rate": 8.167435728411339e-05,
|
| 877 |
+
"loss": 0.0401,
|
| 878 |
+
"step": 1240
|
| 879 |
+
},
|
| 880 |
+
{
|
| 881 |
+
"epoch": 0.24730438223365317,
|
| 882 |
+
"grad_norm": 0.43958767606215066,
|
| 883 |
+
"learning_rate": 8.233355306526038e-05,
|
| 884 |
+
"loss": 0.0385,
|
| 885 |
+
"step": 1250
|
| 886 |
+
},
|
| 887 |
+
{
|
| 888 |
+
"epoch": 0.2492828172915224,
|
| 889 |
+
"grad_norm": 2.3478935687105587,
|
| 890 |
+
"learning_rate": 8.299274884640739e-05,
|
| 891 |
+
"loss": 0.0391,
|
| 892 |
+
"step": 1260
|
| 893 |
+
},
|
| 894 |
+
{
|
| 895 |
+
"epoch": 0.2512612523493916,
|
| 896 |
+
"grad_norm": 0.7921257388969846,
|
| 897 |
+
"learning_rate": 8.365194462755439e-05,
|
| 898 |
+
"loss": 0.0391,
|
| 899 |
+
"step": 1270
|
| 900 |
+
},
|
| 901 |
+
{
|
| 902 |
+
"epoch": 0.2532396874072609,
|
| 903 |
+
"grad_norm": 0.4159095258664668,
|
| 904 |
+
"learning_rate": 8.431114040870139e-05,
|
| 905 |
+
"loss": 0.0364,
|
| 906 |
+
"step": 1280
|
| 907 |
+
},
|
| 908 |
+
{
|
| 909 |
+
"epoch": 0.2552181224651301,
|
| 910 |
+
"grad_norm": 0.3574794027980809,
|
| 911 |
+
"learning_rate": 8.497033618984839e-05,
|
| 912 |
+
"loss": 0.0347,
|
| 913 |
+
"step": 1290
|
| 914 |
+
},
|
| 915 |
+
{
|
| 916 |
+
"epoch": 0.2571965575229993,
|
| 917 |
+
"grad_norm": 0.3793357762723568,
|
| 918 |
+
"learning_rate": 8.562953197099539e-05,
|
| 919 |
+
"loss": 0.0338,
|
| 920 |
+
"step": 1300
|
| 921 |
+
},
|
| 922 |
+
{
|
| 923 |
+
"epoch": 0.2591749925808685,
|
| 924 |
+
"grad_norm": 0.39061129654899224,
|
| 925 |
+
"learning_rate": 8.62887277521424e-05,
|
| 926 |
+
"loss": 0.0322,
|
| 927 |
+
"step": 1310
|
| 928 |
+
},
|
| 929 |
+
{
|
| 930 |
+
"epoch": 0.2611534276387378,
|
| 931 |
+
"grad_norm": 0.3507422200583872,
|
| 932 |
+
"learning_rate": 8.694792353328939e-05,
|
| 933 |
+
"loss": 0.0318,
|
| 934 |
+
"step": 1320
|
| 935 |
+
},
|
| 936 |
+
{
|
| 937 |
+
"epoch": 0.263131862696607,
|
| 938 |
+
"grad_norm": 0.3316497467819614,
|
| 939 |
+
"learning_rate": 8.760711931443639e-05,
|
| 940 |
+
"loss": 0.0309,
|
| 941 |
+
"step": 1330
|
| 942 |
+
},
|
| 943 |
+
{
|
| 944 |
+
"epoch": 0.2651102977544762,
|
| 945 |
+
"grad_norm": 0.4422978758582334,
|
| 946 |
+
"learning_rate": 8.82663150955834e-05,
|
| 947 |
+
"loss": 0.0302,
|
| 948 |
+
"step": 1340
|
| 949 |
+
},
|
| 950 |
+
{
|
| 951 |
+
"epoch": 0.2670887328123454,
|
| 952 |
+
"grad_norm": 0.42532269482645996,
|
| 953 |
+
"learning_rate": 8.892551087673038e-05,
|
| 954 |
+
"loss": 0.0292,
|
| 955 |
+
"step": 1350
|
| 956 |
+
},
|
| 957 |
+
{
|
| 958 |
+
"epoch": 0.2690671678702147,
|
| 959 |
+
"grad_norm": 1.2751835724526872,
|
| 960 |
+
"learning_rate": 8.95847066578774e-05,
|
| 961 |
+
"loss": 0.0322,
|
| 962 |
+
"step": 1360
|
| 963 |
+
},
|
| 964 |
+
{
|
| 965 |
+
"epoch": 0.2710456029280839,
|
| 966 |
+
"grad_norm": 0.41974258023847216,
|
| 967 |
+
"learning_rate": 9.02439024390244e-05,
|
| 968 |
+
"loss": 0.0299,
|
| 969 |
+
"step": 1370
|
| 970 |
+
},
|
| 971 |
+
{
|
| 972 |
+
"epoch": 0.2730240379859531,
|
| 973 |
+
"grad_norm": 0.32580553825660563,
|
| 974 |
+
"learning_rate": 9.09030982201714e-05,
|
| 975 |
+
"loss": 0.0285,
|
| 976 |
+
"step": 1380
|
| 977 |
+
},
|
| 978 |
+
{
|
| 979 |
+
"epoch": 0.27500247304382236,
|
| 980 |
+
"grad_norm": 0.362899277572937,
|
| 981 |
+
"learning_rate": 9.15622940013184e-05,
|
| 982 |
+
"loss": 0.0272,
|
| 983 |
+
"step": 1390
|
| 984 |
+
},
|
| 985 |
+
{
|
| 986 |
+
"epoch": 0.27698090810169157,
|
| 987 |
+
"grad_norm": 0.31286682503756197,
|
| 988 |
+
"learning_rate": 9.22214897824654e-05,
|
| 989 |
+
"loss": 0.0263,
|
| 990 |
+
"step": 1400
|
| 991 |
+
},
|
| 992 |
+
{
|
| 993 |
+
"epoch": 0.2789593431595608,
|
| 994 |
+
"grad_norm": 0.26773424358272674,
|
| 995 |
+
"learning_rate": 9.288068556361241e-05,
|
| 996 |
+
"loss": 0.0255,
|
| 997 |
+
"step": 1410
|
| 998 |
+
},
|
| 999 |
+
{
|
| 1000 |
+
"epoch": 0.28093777821743,
|
| 1001 |
+
"grad_norm": 0.40448389034455734,
|
| 1002 |
+
"learning_rate": 9.35398813447594e-05,
|
| 1003 |
+
"loss": 0.0252,
|
| 1004 |
+
"step": 1420
|
| 1005 |
+
},
|
| 1006 |
+
{
|
| 1007 |
+
"epoch": 0.28291621327529926,
|
| 1008 |
+
"grad_norm": 0.2783819383288086,
|
| 1009 |
+
"learning_rate": 9.41990771259064e-05,
|
| 1010 |
+
"loss": 0.0241,
|
| 1011 |
+
"step": 1430
|
| 1012 |
+
},
|
| 1013 |
+
{
|
| 1014 |
+
"epoch": 0.28489464833316847,
|
| 1015 |
+
"grad_norm": 0.29961272536024675,
|
| 1016 |
+
"learning_rate": 9.48582729070534e-05,
|
| 1017 |
+
"loss": 0.0236,
|
| 1018 |
+
"step": 1440
|
| 1019 |
+
},
|
| 1020 |
+
{
|
| 1021 |
+
"epoch": 0.2868730833910377,
|
| 1022 |
+
"grad_norm": 0.24347322009574,
|
| 1023 |
+
"learning_rate": 9.551746868820039e-05,
|
| 1024 |
+
"loss": 0.0228,
|
| 1025 |
+
"step": 1450
|
| 1026 |
+
},
|
| 1027 |
+
{
|
| 1028 |
+
"epoch": 0.28885151844890694,
|
| 1029 |
+
"grad_norm": 0.23839323438206578,
|
| 1030 |
+
"learning_rate": 9.61766644693474e-05,
|
| 1031 |
+
"loss": 0.0226,
|
| 1032 |
+
"step": 1460
|
| 1033 |
+
},
|
| 1034 |
+
{
|
| 1035 |
+
"epoch": 0.29082995350677615,
|
| 1036 |
+
"grad_norm": 0.26473311552442963,
|
| 1037 |
+
"learning_rate": 9.68358602504944e-05,
|
| 1038 |
+
"loss": 0.0217,
|
| 1039 |
+
"step": 1470
|
| 1040 |
+
},
|
| 1041 |
+
{
|
| 1042 |
+
"epoch": 0.29280838856464536,
|
| 1043 |
+
"grad_norm": 0.25862281572683987,
|
| 1044 |
+
"learning_rate": 9.74950560316414e-05,
|
| 1045 |
+
"loss": 0.0211,
|
| 1046 |
+
"step": 1480
|
| 1047 |
+
},
|
| 1048 |
+
{
|
| 1049 |
+
"epoch": 0.29478682362251457,
|
| 1050 |
+
"grad_norm": 0.2277270114727975,
|
| 1051 |
+
"learning_rate": 9.81542518127884e-05,
|
| 1052 |
+
"loss": 0.0205,
|
| 1053 |
+
"step": 1490
|
| 1054 |
+
},
|
| 1055 |
+
{
|
| 1056 |
+
"epoch": 0.29676525868038384,
|
| 1057 |
+
"grad_norm": 0.24087395405359055,
|
| 1058 |
+
"learning_rate": 9.88134475939354e-05,
|
| 1059 |
+
"loss": 0.02,
|
| 1060 |
+
"step": 1500
|
| 1061 |
+
},
|
| 1062 |
+
{
|
| 1063 |
+
"epoch": 0.29874369373825305,
|
| 1064 |
+
"grad_norm": 0.22063387099417836,
|
| 1065 |
+
"learning_rate": 9.94726433750824e-05,
|
| 1066 |
+
"loss": 0.0193,
|
| 1067 |
+
"step": 1510
|
| 1068 |
+
},
|
| 1069 |
+
{
|
| 1070 |
+
"epoch": 0.30072212879612226,
|
| 1071 |
+
"grad_norm": 0.20714882477980437,
|
| 1072 |
+
"learning_rate": 9.999999470139298e-05,
|
| 1073 |
+
"loss": 0.0187,
|
| 1074 |
+
"step": 1520
|
| 1075 |
+
},
|
| 1076 |
+
{
|
| 1077 |
+
"epoch": 0.30270056385399147,
|
| 1078 |
+
"grad_norm": 0.2302853547965489,
|
| 1079 |
+
"learning_rate": 9.99998092502654e-05,
|
| 1080 |
+
"loss": 0.0184,
|
| 1081 |
+
"step": 1530
|
| 1082 |
+
},
|
| 1083 |
+
{
|
| 1084 |
+
"epoch": 0.30467899891186073,
|
| 1085 |
+
"grad_norm": 0.24697567028746728,
|
| 1086 |
+
"learning_rate": 9.999935886991012e-05,
|
| 1087 |
+
"loss": 0.0184,
|
| 1088 |
+
"step": 1540
|
| 1089 |
+
},
|
| 1090 |
+
{
|
| 1091 |
+
"epoch": 0.30665743396972994,
|
| 1092 |
+
"grad_norm": 0.2162360843440974,
|
| 1093 |
+
"learning_rate": 9.99986435627135e-05,
|
| 1094 |
+
"loss": 0.0179,
|
| 1095 |
+
"step": 1550
|
| 1096 |
+
},
|
| 1097 |
+
{
|
| 1098 |
+
"epoch": 0.30863586902759915,
|
| 1099 |
+
"grad_norm": 0.22785434831280413,
|
| 1100 |
+
"learning_rate": 9.99976633324657e-05,
|
| 1101 |
+
"loss": 0.0174,
|
| 1102 |
+
"step": 1560
|
| 1103 |
+
},
|
| 1104 |
+
{
|
| 1105 |
+
"epoch": 0.3106143040854684,
|
| 1106 |
+
"grad_norm": 0.21011992583277403,
|
| 1107 |
+
"learning_rate": 9.999641818436056e-05,
|
| 1108 |
+
"loss": 0.0169,
|
| 1109 |
+
"step": 1570
|
| 1110 |
+
},
|
| 1111 |
+
{
|
| 1112 |
+
"epoch": 0.31259273914333763,
|
| 1113 |
+
"grad_norm": 0.23882667469933117,
|
| 1114 |
+
"learning_rate": 9.999490812499562e-05,
|
| 1115 |
+
"loss": 0.0169,
|
| 1116 |
+
"step": 1580
|
| 1117 |
+
},
|
| 1118 |
+
{
|
| 1119 |
+
"epoch": 0.31457117420120684,
|
| 1120 |
+
"grad_norm": 0.22869025276994928,
|
| 1121 |
+
"learning_rate": 9.999313316237211e-05,
|
| 1122 |
+
"loss": 0.0162,
|
| 1123 |
+
"step": 1590
|
| 1124 |
+
},
|
| 1125 |
+
{
|
| 1126 |
+
"epoch": 0.31654960925907605,
|
| 1127 |
+
"grad_norm": 0.19936558362244672,
|
| 1128 |
+
"learning_rate": 9.999109330589484e-05,
|
| 1129 |
+
"loss": 0.0158,
|
| 1130 |
+
"step": 1600
|
| 1131 |
+
},
|
| 1132 |
+
{
|
| 1133 |
+
"epoch": 0.3185280443169453,
|
| 1134 |
+
"grad_norm": 0.18776511028399503,
|
| 1135 |
+
"learning_rate": 9.998878856637221e-05,
|
| 1136 |
+
"loss": 0.0157,
|
| 1137 |
+
"step": 1610
|
| 1138 |
+
},
|
| 1139 |
+
{
|
| 1140 |
+
"epoch": 0.3205064793748145,
|
| 1141 |
+
"grad_norm": 0.22047831859103678,
|
| 1142 |
+
"learning_rate": 9.998621895601612e-05,
|
| 1143 |
+
"loss": 0.0151,
|
| 1144 |
+
"step": 1620
|
| 1145 |
+
},
|
| 1146 |
+
{
|
| 1147 |
+
"epoch": 0.32248491443268373,
|
| 1148 |
+
"grad_norm": 0.21909712933711048,
|
| 1149 |
+
"learning_rate": 9.998338448844192e-05,
|
| 1150 |
+
"loss": 0.0153,
|
| 1151 |
+
"step": 1630
|
| 1152 |
+
},
|
| 1153 |
+
{
|
| 1154 |
+
"epoch": 0.324463349490553,
|
| 1155 |
+
"grad_norm": 0.18342955578512224,
|
| 1156 |
+
"learning_rate": 9.998028517866833e-05,
|
| 1157 |
+
"loss": 0.0147,
|
| 1158 |
+
"step": 1640
|
| 1159 |
+
},
|
| 1160 |
+
{
|
| 1161 |
+
"epoch": 0.3264417845484222,
|
| 1162 |
+
"grad_norm": 0.19614528373745874,
|
| 1163 |
+
"learning_rate": 9.997692104311738e-05,
|
| 1164 |
+
"loss": 0.0144,
|
| 1165 |
+
"step": 1650
|
| 1166 |
+
},
|
| 1167 |
+
{
|
| 1168 |
+
"epoch": 0.3284202196062914,
|
| 1169 |
+
"grad_norm": 0.1910028678576799,
|
| 1170 |
+
"learning_rate": 9.997329209961429e-05,
|
| 1171 |
+
"loss": 0.0141,
|
| 1172 |
+
"step": 1660
|
| 1173 |
+
},
|
| 1174 |
+
{
|
| 1175 |
+
"epoch": 0.33039865466416063,
|
| 1176 |
+
"grad_norm": 0.1875581912159569,
|
| 1177 |
+
"learning_rate": 9.996939836738737e-05,
|
| 1178 |
+
"loss": 0.0138,
|
| 1179 |
+
"step": 1670
|
| 1180 |
+
},
|
| 1181 |
+
{
|
| 1182 |
+
"epoch": 0.3323770897220299,
|
| 1183 |
+
"grad_norm": 0.21318059322872368,
|
| 1184 |
+
"learning_rate": 9.9965239867068e-05,
|
| 1185 |
+
"loss": 0.0138,
|
| 1186 |
+
"step": 1680
|
| 1187 |
+
},
|
| 1188 |
+
{
|
| 1189 |
+
"epoch": 0.3343555247798991,
|
| 1190 |
+
"grad_norm": 0.15100583828371544,
|
| 1191 |
+
"learning_rate": 9.996081662069043e-05,
|
| 1192 |
+
"loss": 0.0138,
|
| 1193 |
+
"step": 1690
|
| 1194 |
+
},
|
| 1195 |
+
{
|
| 1196 |
+
"epoch": 0.3363339598377683,
|
| 1197 |
+
"grad_norm": 0.16665345852934302,
|
| 1198 |
+
"learning_rate": 9.995612865169168e-05,
|
| 1199 |
+
"loss": 0.0136,
|
| 1200 |
+
"step": 1700
|
| 1201 |
+
},
|
| 1202 |
+
{
|
| 1203 |
+
"epoch": 0.3383123948956375,
|
| 1204 |
+
"grad_norm": 0.9719045950170202,
|
| 1205 |
+
"learning_rate": 9.995117598491145e-05,
|
| 1206 |
+
"loss": 0.0277,
|
| 1207 |
+
"step": 1710
|
| 1208 |
+
},
|
| 1209 |
+
{
|
| 1210 |
+
"epoch": 0.3402908299535068,
|
| 1211 |
+
"grad_norm": 3.161728848159066,
|
| 1212 |
+
"learning_rate": 9.994595864659197e-05,
|
| 1213 |
+
"loss": 0.0406,
|
| 1214 |
+
"step": 1720
|
| 1215 |
+
},
|
| 1216 |
+
{
|
| 1217 |
+
"epoch": 0.342269265011376,
|
| 1218 |
+
"grad_norm": 3.6147429552746417,
|
| 1219 |
+
"learning_rate": 9.994047666437785e-05,
|
| 1220 |
+
"loss": 0.0542,
|
| 1221 |
+
"step": 1730
|
| 1222 |
+
},
|
| 1223 |
+
{
|
| 1224 |
+
"epoch": 0.3442477000692452,
|
| 1225 |
+
"grad_norm": 0.2735534000567824,
|
| 1226 |
+
"learning_rate": 9.993473006731592e-05,
|
| 1227 |
+
"loss": 0.022,
|
| 1228 |
+
"step": 1740
|
| 1229 |
+
},
|
| 1230 |
+
{
|
| 1231 |
+
"epoch": 0.3462261351271145,
|
| 1232 |
+
"grad_norm": 0.1810803504663682,
|
| 1233 |
+
"learning_rate": 9.992871888585518e-05,
|
| 1234 |
+
"loss": 0.016,
|
| 1235 |
+
"step": 1750
|
| 1236 |
+
},
|
| 1237 |
+
{
|
| 1238 |
+
"epoch": 0.3482045701849837,
|
| 1239 |
+
"grad_norm": 0.14853009791786845,
|
| 1240 |
+
"learning_rate": 9.992244315184647e-05,
|
| 1241 |
+
"loss": 0.0143,
|
| 1242 |
+
"step": 1760
|
| 1243 |
+
},
|
| 1244 |
+
{
|
| 1245 |
+
"epoch": 0.3501830052428529,
|
| 1246 |
+
"grad_norm": 0.1508942913302995,
|
| 1247 |
+
"learning_rate": 9.991590289854243e-05,
|
| 1248 |
+
"loss": 0.0136,
|
| 1249 |
+
"step": 1770
|
| 1250 |
+
},
|
| 1251 |
+
{
|
| 1252 |
+
"epoch": 0.3521614403007221,
|
| 1253 |
+
"grad_norm": 0.14224740250692022,
|
| 1254 |
+
"learning_rate": 9.990909816059725e-05,
|
| 1255 |
+
"loss": 0.0126,
|
| 1256 |
+
"step": 1780
|
| 1257 |
+
},
|
| 1258 |
+
{
|
| 1259 |
+
"epoch": 0.35413987535859137,
|
| 1260 |
+
"grad_norm": 0.13888816460678802,
|
| 1261 |
+
"learning_rate": 9.99020289740666e-05,
|
| 1262 |
+
"loss": 0.0124,
|
| 1263 |
+
"step": 1790
|
| 1264 |
+
},
|
| 1265 |
+
{
|
| 1266 |
+
"epoch": 0.3561183104164606,
|
| 1267 |
+
"grad_norm": 0.14941569438589294,
|
| 1268 |
+
"learning_rate": 9.989469537640728e-05,
|
| 1269 |
+
"loss": 0.0122,
|
| 1270 |
+
"step": 1800
|
| 1271 |
+
}
|
| 1272 |
+
],
|
| 1273 |
+
"logging_steps": 10,
|
| 1274 |
+
"max_steps": 15165,
|
| 1275 |
+
"num_input_tokens_seen": 0,
|
| 1276 |
+
"num_train_epochs": 3,
|
| 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": 163099529183232.0,
|
| 1291 |
+
"train_batch_size": 128,
|
| 1292 |
+
"trial_name": null,
|
| 1293 |
+
"trial_params": null
|
| 1294 |
+
}
|