Upload trainer_state.json
Browse filesEssential checkpoint files (13/17) - excluding optimizer states
- trainer_state.json +1261 -0
trainer_state.json
ADDED
|
@@ -0,0 +1,1261 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"best_global_step": null,
|
| 3 |
+
"best_metric": null,
|
| 4 |
+
"best_model_checkpoint": null,
|
| 5 |
+
"epoch": 1.7818654775176515,
|
| 6 |
+
"eval_steps": 200,
|
| 7 |
+
"global_step": 600,
|
| 8 |
+
"is_hyper_param_search": false,
|
| 9 |
+
"is_local_process_zero": true,
|
| 10 |
+
"is_world_process_zero": true,
|
| 11 |
+
"log_history": [
|
| 12 |
+
{
|
| 13 |
+
"epoch": 0.014864362690449646,
|
| 14 |
+
"grad_norm": 2.1816246446736707,
|
| 15 |
+
"learning_rate": 1.6666666666666667e-06,
|
| 16 |
+
"loss": 1.5437,
|
| 17 |
+
"num_input_tokens_seen": 30040,
|
| 18 |
+
"step": 5,
|
| 19 |
+
"train_runtime": 185.9609,
|
| 20 |
+
"train_tokens_per_second": 161.539
|
| 21 |
+
},
|
| 22 |
+
{
|
| 23 |
+
"epoch": 0.029728725380899292,
|
| 24 |
+
"grad_norm": 1.297173579671946,
|
| 25 |
+
"learning_rate": 3.75e-06,
|
| 26 |
+
"loss": 1.3114,
|
| 27 |
+
"num_input_tokens_seen": 65616,
|
| 28 |
+
"step": 10,
|
| 29 |
+
"train_runtime": 371.0314,
|
| 30 |
+
"train_tokens_per_second": 176.848
|
| 31 |
+
},
|
| 32 |
+
{
|
| 33 |
+
"epoch": 0.044593088071348944,
|
| 34 |
+
"grad_norm": 1.3154044523767012,
|
| 35 |
+
"learning_rate": 5.833333333333334e-06,
|
| 36 |
+
"loss": 1.5722,
|
| 37 |
+
"num_input_tokens_seen": 92696,
|
| 38 |
+
"step": 15,
|
| 39 |
+
"train_runtime": 537.3955,
|
| 40 |
+
"train_tokens_per_second": 172.491
|
| 41 |
+
},
|
| 42 |
+
{
|
| 43 |
+
"epoch": 0.059457450761798585,
|
| 44 |
+
"grad_norm": 1.901062718640284,
|
| 45 |
+
"learning_rate": 7.916666666666667e-06,
|
| 46 |
+
"loss": 1.4077,
|
| 47 |
+
"num_input_tokens_seen": 123944,
|
| 48 |
+
"step": 20,
|
| 49 |
+
"train_runtime": 720.0196,
|
| 50 |
+
"train_tokens_per_second": 172.14
|
| 51 |
+
},
|
| 52 |
+
{
|
| 53 |
+
"epoch": 0.07432181345224824,
|
| 54 |
+
"grad_norm": 0.865431080037828,
|
| 55 |
+
"learning_rate": 1e-05,
|
| 56 |
+
"loss": 1.2642,
|
| 57 |
+
"num_input_tokens_seen": 154432,
|
| 58 |
+
"step": 25,
|
| 59 |
+
"train_runtime": 892.3433,
|
| 60 |
+
"train_tokens_per_second": 173.063
|
| 61 |
+
},
|
| 62 |
+
{
|
| 63 |
+
"epoch": 0.08918617614269789,
|
| 64 |
+
"grad_norm": 0.7781390780834406,
|
| 65 |
+
"learning_rate": 1.2083333333333333e-05,
|
| 66 |
+
"loss": 1.1572,
|
| 67 |
+
"num_input_tokens_seen": 181960,
|
| 68 |
+
"step": 30,
|
| 69 |
+
"train_runtime": 1057.0334,
|
| 70 |
+
"train_tokens_per_second": 172.142
|
| 71 |
+
},
|
| 72 |
+
{
|
| 73 |
+
"epoch": 0.10405053883314754,
|
| 74 |
+
"grad_norm": 0.665108652040459,
|
| 75 |
+
"learning_rate": 1.4166666666666668e-05,
|
| 76 |
+
"loss": 0.9421,
|
| 77 |
+
"num_input_tokens_seen": 212640,
|
| 78 |
+
"step": 35,
|
| 79 |
+
"train_runtime": 1231.9846,
|
| 80 |
+
"train_tokens_per_second": 172.6
|
| 81 |
+
},
|
| 82 |
+
{
|
| 83 |
+
"epoch": 0.11891490152359717,
|
| 84 |
+
"grad_norm": 0.2825351078495791,
|
| 85 |
+
"learning_rate": 1.6250000000000002e-05,
|
| 86 |
+
"loss": 0.8661,
|
| 87 |
+
"num_input_tokens_seen": 245808,
|
| 88 |
+
"step": 40,
|
| 89 |
+
"train_runtime": 1416.5893,
|
| 90 |
+
"train_tokens_per_second": 173.521
|
| 91 |
+
},
|
| 92 |
+
{
|
| 93 |
+
"epoch": 0.13377926421404682,
|
| 94 |
+
"grad_norm": 0.36334526479148993,
|
| 95 |
+
"learning_rate": 1.8333333333333333e-05,
|
| 96 |
+
"loss": 0.9087,
|
| 97 |
+
"num_input_tokens_seen": 272224,
|
| 98 |
+
"step": 45,
|
| 99 |
+
"train_runtime": 1582.5403,
|
| 100 |
+
"train_tokens_per_second": 172.017
|
| 101 |
+
},
|
| 102 |
+
{
|
| 103 |
+
"epoch": 0.14864362690449648,
|
| 104 |
+
"grad_norm": 0.44642145260416205,
|
| 105 |
+
"learning_rate": 2.0416666666666667e-05,
|
| 106 |
+
"loss": 0.8855,
|
| 107 |
+
"num_input_tokens_seen": 305560,
|
| 108 |
+
"step": 50,
|
| 109 |
+
"train_runtime": 1768.3526,
|
| 110 |
+
"train_tokens_per_second": 172.794
|
| 111 |
+
},
|
| 112 |
+
{
|
| 113 |
+
"epoch": 0.1635079895949461,
|
| 114 |
+
"grad_norm": 0.41738282561467666,
|
| 115 |
+
"learning_rate": 2.25e-05,
|
| 116 |
+
"loss": 0.8737,
|
| 117 |
+
"num_input_tokens_seen": 332024,
|
| 118 |
+
"step": 55,
|
| 119 |
+
"train_runtime": 1927.929,
|
| 120 |
+
"train_tokens_per_second": 172.218
|
| 121 |
+
},
|
| 122 |
+
{
|
| 123 |
+
"epoch": 0.17837235228539577,
|
| 124 |
+
"grad_norm": 0.39634079236288305,
|
| 125 |
+
"learning_rate": 2.4583333333333332e-05,
|
| 126 |
+
"loss": 0.8043,
|
| 127 |
+
"num_input_tokens_seen": 360856,
|
| 128 |
+
"step": 60,
|
| 129 |
+
"train_runtime": 2097.4995,
|
| 130 |
+
"train_tokens_per_second": 172.041
|
| 131 |
+
},
|
| 132 |
+
{
|
| 133 |
+
"epoch": 0.1932367149758454,
|
| 134 |
+
"grad_norm": 0.35491219061131213,
|
| 135 |
+
"learning_rate": 2.6666666666666667e-05,
|
| 136 |
+
"loss": 0.7943,
|
| 137 |
+
"num_input_tokens_seen": 398608,
|
| 138 |
+
"step": 65,
|
| 139 |
+
"train_runtime": 2304.6738,
|
| 140 |
+
"train_tokens_per_second": 172.956
|
| 141 |
+
},
|
| 142 |
+
{
|
| 143 |
+
"epoch": 0.20810107766629507,
|
| 144 |
+
"grad_norm": 0.35737550120310063,
|
| 145 |
+
"learning_rate": 2.8749999999999997e-05,
|
| 146 |
+
"loss": 0.8492,
|
| 147 |
+
"num_input_tokens_seen": 425552,
|
| 148 |
+
"step": 70,
|
| 149 |
+
"train_runtime": 2460.8004,
|
| 150 |
+
"train_tokens_per_second": 172.932
|
| 151 |
+
},
|
| 152 |
+
{
|
| 153 |
+
"epoch": 0.2229654403567447,
|
| 154 |
+
"grad_norm": 0.48693276619814596,
|
| 155 |
+
"learning_rate": 3.0833333333333335e-05,
|
| 156 |
+
"loss": 0.6925,
|
| 157 |
+
"num_input_tokens_seen": 457392,
|
| 158 |
+
"step": 75,
|
| 159 |
+
"train_runtime": 2638.2785,
|
| 160 |
+
"train_tokens_per_second": 173.368
|
| 161 |
+
},
|
| 162 |
+
{
|
| 163 |
+
"epoch": 0.23782980304719434,
|
| 164 |
+
"grad_norm": 0.5078391701662902,
|
| 165 |
+
"learning_rate": 3.291666666666667e-05,
|
| 166 |
+
"loss": 0.8085,
|
| 167 |
+
"num_input_tokens_seen": 483824,
|
| 168 |
+
"step": 80,
|
| 169 |
+
"train_runtime": 2806.7407,
|
| 170 |
+
"train_tokens_per_second": 172.379
|
| 171 |
+
},
|
| 172 |
+
{
|
| 173 |
+
"epoch": 0.25269416573764397,
|
| 174 |
+
"grad_norm": 0.3024390891532131,
|
| 175 |
+
"learning_rate": 3.5e-05,
|
| 176 |
+
"loss": 0.7564,
|
| 177 |
+
"num_input_tokens_seen": 512864,
|
| 178 |
+
"step": 85,
|
| 179 |
+
"train_runtime": 2979.1533,
|
| 180 |
+
"train_tokens_per_second": 172.151
|
| 181 |
+
},
|
| 182 |
+
{
|
| 183 |
+
"epoch": 0.26755852842809363,
|
| 184 |
+
"grad_norm": 0.37806969672351337,
|
| 185 |
+
"learning_rate": 3.708333333333334e-05,
|
| 186 |
+
"loss": 0.8039,
|
| 187 |
+
"num_input_tokens_seen": 538720,
|
| 188 |
+
"step": 90,
|
| 189 |
+
"train_runtime": 3137.4212,
|
| 190 |
+
"train_tokens_per_second": 171.708
|
| 191 |
+
},
|
| 192 |
+
{
|
| 193 |
+
"epoch": 0.2824228911185433,
|
| 194 |
+
"grad_norm": 0.3125837610721519,
|
| 195 |
+
"learning_rate": 3.9166666666666665e-05,
|
| 196 |
+
"loss": 0.7889,
|
| 197 |
+
"num_input_tokens_seen": 569688,
|
| 198 |
+
"step": 95,
|
| 199 |
+
"train_runtime": 3317.1304,
|
| 200 |
+
"train_tokens_per_second": 171.741
|
| 201 |
+
},
|
| 202 |
+
{
|
| 203 |
+
"epoch": 0.29728725380899296,
|
| 204 |
+
"grad_norm": 0.2612296186361751,
|
| 205 |
+
"learning_rate": 4.125e-05,
|
| 206 |
+
"loss": 0.8101,
|
| 207 |
+
"num_input_tokens_seen": 603480,
|
| 208 |
+
"step": 100,
|
| 209 |
+
"train_runtime": 3508.249,
|
| 210 |
+
"train_tokens_per_second": 172.017
|
| 211 |
+
},
|
| 212 |
+
{
|
| 213 |
+
"epoch": 0.31215161649944256,
|
| 214 |
+
"grad_norm": 0.4664200778768721,
|
| 215 |
+
"learning_rate": 4.3333333333333334e-05,
|
| 216 |
+
"loss": 0.7587,
|
| 217 |
+
"num_input_tokens_seen": 632928,
|
| 218 |
+
"step": 105,
|
| 219 |
+
"train_runtime": 3681.3714,
|
| 220 |
+
"train_tokens_per_second": 171.927
|
| 221 |
+
},
|
| 222 |
+
{
|
| 223 |
+
"epoch": 0.3270159791898922,
|
| 224 |
+
"grad_norm": 0.4122820082467493,
|
| 225 |
+
"learning_rate": 4.541666666666667e-05,
|
| 226 |
+
"loss": 0.7925,
|
| 227 |
+
"num_input_tokens_seen": 659840,
|
| 228 |
+
"step": 110,
|
| 229 |
+
"train_runtime": 3850.7584,
|
| 230 |
+
"train_tokens_per_second": 171.353
|
| 231 |
+
},
|
| 232 |
+
{
|
| 233 |
+
"epoch": 0.3418803418803419,
|
| 234 |
+
"grad_norm": 0.29086455774181913,
|
| 235 |
+
"learning_rate": 4.75e-05,
|
| 236 |
+
"loss": 0.8142,
|
| 237 |
+
"num_input_tokens_seen": 690432,
|
| 238 |
+
"step": 115,
|
| 239 |
+
"train_runtime": 4029.493,
|
| 240 |
+
"train_tokens_per_second": 171.345
|
| 241 |
+
},
|
| 242 |
+
{
|
| 243 |
+
"epoch": 0.35674470457079155,
|
| 244 |
+
"grad_norm": 0.49406027919264844,
|
| 245 |
+
"learning_rate": 4.958333333333334e-05,
|
| 246 |
+
"loss": 0.7088,
|
| 247 |
+
"num_input_tokens_seen": 726264,
|
| 248 |
+
"step": 120,
|
| 249 |
+
"train_runtime": 4222.1079,
|
| 250 |
+
"train_tokens_per_second": 172.015
|
| 251 |
+
},
|
| 252 |
+
{
|
| 253 |
+
"epoch": 0.37160906726124115,
|
| 254 |
+
"grad_norm": 0.29533667352828036,
|
| 255 |
+
"learning_rate": 4.9999194067399066e-05,
|
| 256 |
+
"loss": 0.6237,
|
| 257 |
+
"num_input_tokens_seen": 763328,
|
| 258 |
+
"step": 125,
|
| 259 |
+
"train_runtime": 4420.0509,
|
| 260 |
+
"train_tokens_per_second": 172.697
|
| 261 |
+
},
|
| 262 |
+
{
|
| 263 |
+
"epoch": 0.3864734299516908,
|
| 264 |
+
"grad_norm": 0.554389641306832,
|
| 265 |
+
"learning_rate": 4.999592005526383e-05,
|
| 266 |
+
"loss": 0.6609,
|
| 267 |
+
"num_input_tokens_seen": 792888,
|
| 268 |
+
"step": 130,
|
| 269 |
+
"train_runtime": 4590.9234,
|
| 270 |
+
"train_tokens_per_second": 172.708
|
| 271 |
+
},
|
| 272 |
+
{
|
| 273 |
+
"epoch": 0.4013377926421405,
|
| 274 |
+
"grad_norm": 0.3913197863670873,
|
| 275 |
+
"learning_rate": 4.999012792238118e-05,
|
| 276 |
+
"loss": 0.7873,
|
| 277 |
+
"num_input_tokens_seen": 830784,
|
| 278 |
+
"step": 135,
|
| 279 |
+
"train_runtime": 4780.7961,
|
| 280 |
+
"train_tokens_per_second": 173.775
|
| 281 |
+
},
|
| 282 |
+
{
|
| 283 |
+
"epoch": 0.41620215533259014,
|
| 284 |
+
"grad_norm": 0.7292785043191194,
|
| 285 |
+
"learning_rate": 4.998181825225791e-05,
|
| 286 |
+
"loss": 0.7047,
|
| 287 |
+
"num_input_tokens_seen": 869304,
|
| 288 |
+
"step": 140,
|
| 289 |
+
"train_runtime": 4979.1373,
|
| 290 |
+
"train_tokens_per_second": 174.589
|
| 291 |
+
},
|
| 292 |
+
{
|
| 293 |
+
"epoch": 0.43106651802303975,
|
| 294 |
+
"grad_norm": 0.3288366913256035,
|
| 295 |
+
"learning_rate": 4.997099188202077e-05,
|
| 296 |
+
"loss": 0.6608,
|
| 297 |
+
"num_input_tokens_seen": 903752,
|
| 298 |
+
"step": 145,
|
| 299 |
+
"train_runtime": 5170.5438,
|
| 300 |
+
"train_tokens_per_second": 174.789
|
| 301 |
+
},
|
| 302 |
+
{
|
| 303 |
+
"epoch": 0.4459308807134894,
|
| 304 |
+
"grad_norm": 0.32850031143064623,
|
| 305 |
+
"learning_rate": 4.995764990233205e-05,
|
| 306 |
+
"loss": 0.7203,
|
| 307 |
+
"num_input_tokens_seen": 935672,
|
| 308 |
+
"step": 150,
|
| 309 |
+
"train_runtime": 5350.7854,
|
| 310 |
+
"train_tokens_per_second": 174.866
|
| 311 |
+
},
|
| 312 |
+
{
|
| 313 |
+
"epoch": 0.46079524340393907,
|
| 314 |
+
"grad_norm": 0.4093406468089712,
|
| 315 |
+
"learning_rate": 4.994179365727973e-05,
|
| 316 |
+
"loss": 0.7793,
|
| 317 |
+
"num_input_tokens_seen": 967928,
|
| 318 |
+
"step": 155,
|
| 319 |
+
"train_runtime": 5526.3425,
|
| 320 |
+
"train_tokens_per_second": 175.148
|
| 321 |
+
},
|
| 322 |
+
{
|
| 323 |
+
"epoch": 0.4756596060943887,
|
| 324 |
+
"grad_norm": 0.26538290553046884,
|
| 325 |
+
"learning_rate": 4.992342474424209e-05,
|
| 326 |
+
"loss": 0.6941,
|
| 327 |
+
"num_input_tokens_seen": 1006856,
|
| 328 |
+
"step": 160,
|
| 329 |
+
"train_runtime": 5732.4064,
|
| 330 |
+
"train_tokens_per_second": 175.643
|
| 331 |
+
},
|
| 332 |
+
{
|
| 333 |
+
"epoch": 0.49052396878483834,
|
| 334 |
+
"grad_norm": 0.3293209375345381,
|
| 335 |
+
"learning_rate": 4.990254501372677e-05,
|
| 336 |
+
"loss": 0.7154,
|
| 337 |
+
"num_input_tokens_seen": 1040320,
|
| 338 |
+
"step": 165,
|
| 339 |
+
"train_runtime": 5919.4184,
|
| 340 |
+
"train_tokens_per_second": 175.747
|
| 341 |
+
},
|
| 342 |
+
{
|
| 343 |
+
"epoch": 0.5053883314752879,
|
| 344 |
+
"grad_norm": 0.37872717401082634,
|
| 345 |
+
"learning_rate": 4.987915656918435e-05,
|
| 346 |
+
"loss": 0.7354,
|
| 347 |
+
"num_input_tokens_seen": 1071792,
|
| 348 |
+
"step": 170,
|
| 349 |
+
"train_runtime": 6093.1775,
|
| 350 |
+
"train_tokens_per_second": 175.9
|
| 351 |
+
},
|
| 352 |
+
{
|
| 353 |
+
"epoch": 0.5202526941657376,
|
| 354 |
+
"grad_norm": 0.34111517332292757,
|
| 355 |
+
"learning_rate": 4.985326176679645e-05,
|
| 356 |
+
"loss": 0.7598,
|
| 357 |
+
"num_input_tokens_seen": 1103216,
|
| 358 |
+
"step": 175,
|
| 359 |
+
"train_runtime": 6268.9044,
|
| 360 |
+
"train_tokens_per_second": 175.982
|
| 361 |
+
},
|
| 362 |
+
{
|
| 363 |
+
"epoch": 0.5351170568561873,
|
| 364 |
+
"grad_norm": 0.44888012700873,
|
| 365 |
+
"learning_rate": 4.9824863215238373e-05,
|
| 366 |
+
"loss": 0.7502,
|
| 367 |
+
"num_input_tokens_seen": 1131576,
|
| 368 |
+
"step": 180,
|
| 369 |
+
"train_runtime": 6435.3922,
|
| 370 |
+
"train_tokens_per_second": 175.836
|
| 371 |
+
},
|
| 372 |
+
{
|
| 373 |
+
"epoch": 0.5499814195466369,
|
| 374 |
+
"grad_norm": 0.4802566147521983,
|
| 375 |
+
"learning_rate": 4.979396377541628e-05,
|
| 376 |
+
"loss": 0.6844,
|
| 377 |
+
"num_input_tokens_seen": 1160040,
|
| 378 |
+
"step": 185,
|
| 379 |
+
"train_runtime": 6605.4368,
|
| 380 |
+
"train_tokens_per_second": 175.619
|
| 381 |
+
},
|
| 382 |
+
{
|
| 383 |
+
"epoch": 0.5648457822370866,
|
| 384 |
+
"grad_norm": 0.4821989259826241,
|
| 385 |
+
"learning_rate": 4.976056656017901e-05,
|
| 386 |
+
"loss": 0.8291,
|
| 387 |
+
"num_input_tokens_seen": 1190376,
|
| 388 |
+
"step": 190,
|
| 389 |
+
"train_runtime": 6782.0935,
|
| 390 |
+
"train_tokens_per_second": 175.517
|
| 391 |
+
},
|
| 392 |
+
{
|
| 393 |
+
"epoch": 0.5797101449275363,
|
| 394 |
+
"grad_norm": 0.386641731129637,
|
| 395 |
+
"learning_rate": 4.972467493400445e-05,
|
| 396 |
+
"loss": 0.7762,
|
| 397 |
+
"num_input_tokens_seen": 1217904,
|
| 398 |
+
"step": 195,
|
| 399 |
+
"train_runtime": 6948.2037,
|
| 400 |
+
"train_tokens_per_second": 175.283
|
| 401 |
+
},
|
| 402 |
+
{
|
| 403 |
+
"epoch": 0.5945745076179859,
|
| 404 |
+
"grad_norm": 0.4418388818648707,
|
| 405 |
+
"learning_rate": 4.968629251266064e-05,
|
| 406 |
+
"loss": 0.6688,
|
| 407 |
+
"num_input_tokens_seen": 1256312,
|
| 408 |
+
"step": 200,
|
| 409 |
+
"train_runtime": 7158.3945,
|
| 410 |
+
"train_tokens_per_second": 175.502
|
| 411 |
+
},
|
| 412 |
+
{
|
| 413 |
+
"epoch": 0.5945745076179859,
|
| 414 |
+
"eval_loss": 0.8953876495361328,
|
| 415 |
+
"eval_runtime": 353.0879,
|
| 416 |
+
"eval_samples_per_second": 1.325,
|
| 417 |
+
"eval_steps_per_second": 0.663,
|
| 418 |
+
"num_input_tokens_seen": 1256312,
|
| 419 |
+
"step": 200
|
| 420 |
+
},
|
| 421 |
+
{
|
| 422 |
+
"epoch": 0.6094388703084356,
|
| 423 |
+
"grad_norm": 0.5521465457074078,
|
| 424 |
+
"learning_rate": 4.964542316284147e-05,
|
| 425 |
+
"loss": 0.7398,
|
| 426 |
+
"num_input_tokens_seen": 1290880,
|
| 427 |
+
"step": 205,
|
| 428 |
+
"train_runtime": 7728.2908,
|
| 429 |
+
"train_tokens_per_second": 167.033
|
| 430 |
+
},
|
| 431 |
+
{
|
| 432 |
+
"epoch": 0.6243032329988851,
|
| 433 |
+
"grad_norm": 0.4773991938686689,
|
| 434 |
+
"learning_rate": 4.960207100177716e-05,
|
| 435 |
+
"loss": 0.7169,
|
| 436 |
+
"num_input_tokens_seen": 1320288,
|
| 437 |
+
"step": 210,
|
| 438 |
+
"train_runtime": 7894.6486,
|
| 439 |
+
"train_tokens_per_second": 167.238
|
| 440 |
+
},
|
| 441 |
+
{
|
| 442 |
+
"epoch": 0.6391675956893348,
|
| 443 |
+
"grad_norm": 0.51804635767658,
|
| 444 |
+
"learning_rate": 4.955624039681952e-05,
|
| 445 |
+
"loss": 0.7618,
|
| 446 |
+
"num_input_tokens_seen": 1346816,
|
| 447 |
+
"step": 215,
|
| 448 |
+
"train_runtime": 8060.6928,
|
| 449 |
+
"train_tokens_per_second": 167.084
|
| 450 |
+
},
|
| 451 |
+
{
|
| 452 |
+
"epoch": 0.6540319583797845,
|
| 453 |
+
"grad_norm": 0.5900793472953364,
|
| 454 |
+
"learning_rate": 4.950793596500192e-05,
|
| 455 |
+
"loss": 0.7563,
|
| 456 |
+
"num_input_tokens_seen": 1382472,
|
| 457 |
+
"step": 220,
|
| 458 |
+
"train_runtime": 8252.4478,
|
| 459 |
+
"train_tokens_per_second": 167.523
|
| 460 |
+
},
|
| 461 |
+
{
|
| 462 |
+
"epoch": 0.6688963210702341,
|
| 463 |
+
"grad_norm": 0.39058107426349015,
|
| 464 |
+
"learning_rate": 4.94571625725742e-05,
|
| 465 |
+
"loss": 0.6626,
|
| 466 |
+
"num_input_tokens_seen": 1420832,
|
| 467 |
+
"step": 225,
|
| 468 |
+
"train_runtime": 8450.5909,
|
| 469 |
+
"train_tokens_per_second": 168.134
|
| 470 |
+
},
|
| 471 |
+
{
|
| 472 |
+
"epoch": 0.6837606837606838,
|
| 473 |
+
"grad_norm": 0.5663525201259818,
|
| 474 |
+
"learning_rate": 4.940392533451244e-05,
|
| 475 |
+
"loss": 0.7741,
|
| 476 |
+
"num_input_tokens_seen": 1451496,
|
| 477 |
+
"step": 230,
|
| 478 |
+
"train_runtime": 8624.0331,
|
| 479 |
+
"train_tokens_per_second": 168.308
|
| 480 |
+
},
|
| 481 |
+
{
|
| 482 |
+
"epoch": 0.6986250464511334,
|
| 483 |
+
"grad_norm": 0.3852996317746423,
|
| 484 |
+
"learning_rate": 4.9348229614003615e-05,
|
| 485 |
+
"loss": 0.7749,
|
| 486 |
+
"num_input_tokens_seen": 1483136,
|
| 487 |
+
"step": 235,
|
| 488 |
+
"train_runtime": 8805.4306,
|
| 489 |
+
"train_tokens_per_second": 168.434
|
| 490 |
+
},
|
| 491 |
+
{
|
| 492 |
+
"epoch": 0.7134894091415831,
|
| 493 |
+
"grad_norm": 0.4546419228429511,
|
| 494 |
+
"learning_rate": 4.9290081021905416e-05,
|
| 495 |
+
"loss": 0.7186,
|
| 496 |
+
"num_input_tokens_seen": 1522176,
|
| 497 |
+
"step": 240,
|
| 498 |
+
"train_runtime": 9011.7223,
|
| 499 |
+
"train_tokens_per_second": 168.911
|
| 500 |
+
},
|
| 501 |
+
{
|
| 502 |
+
"epoch": 0.7283537718320326,
|
| 503 |
+
"grad_norm": 0.27319864084916956,
|
| 504 |
+
"learning_rate": 4.9229485416180876e-05,
|
| 505 |
+
"loss": 0.659,
|
| 506 |
+
"num_input_tokens_seen": 1562400,
|
| 507 |
+
"step": 245,
|
| 508 |
+
"train_runtime": 9221.6215,
|
| 509 |
+
"train_tokens_per_second": 169.428
|
| 510 |
+
},
|
| 511 |
+
{
|
| 512 |
+
"epoch": 0.7432181345224823,
|
| 513 |
+
"grad_norm": 0.5971581275751557,
|
| 514 |
+
"learning_rate": 4.916644890130831e-05,
|
| 515 |
+
"loss": 0.747,
|
| 516 |
+
"num_input_tokens_seen": 1587136,
|
| 517 |
+
"step": 250,
|
| 518 |
+
"train_runtime": 9378.073,
|
| 519 |
+
"train_tokens_per_second": 169.239
|
| 520 |
+
},
|
| 521 |
+
{
|
| 522 |
+
"epoch": 0.758082497212932,
|
| 523 |
+
"grad_norm": 0.3725130213516166,
|
| 524 |
+
"learning_rate": 4.9100977827666345e-05,
|
| 525 |
+
"loss": 0.698,
|
| 526 |
+
"num_input_tokens_seen": 1614920,
|
| 527 |
+
"step": 255,
|
| 528 |
+
"train_runtime": 9541.7299,
|
| 529 |
+
"train_tokens_per_second": 169.248
|
| 530 |
+
},
|
| 531 |
+
{
|
| 532 |
+
"epoch": 0.7729468599033816,
|
| 533 |
+
"grad_norm": 0.4907840552388026,
|
| 534 |
+
"learning_rate": 4.903307879089411e-05,
|
| 535 |
+
"loss": 0.6477,
|
| 536 |
+
"num_input_tokens_seen": 1645688,
|
| 537 |
+
"step": 260,
|
| 538 |
+
"train_runtime": 9716.0007,
|
| 539 |
+
"train_tokens_per_second": 169.379
|
| 540 |
+
},
|
| 541 |
+
{
|
| 542 |
+
"epoch": 0.7878112225938313,
|
| 543 |
+
"grad_norm": 0.5241784073891005,
|
| 544 |
+
"learning_rate": 4.896275863122685e-05,
|
| 545 |
+
"loss": 0.7262,
|
| 546 |
+
"num_input_tokens_seen": 1681888,
|
| 547 |
+
"step": 265,
|
| 548 |
+
"train_runtime": 9905.9233,
|
| 549 |
+
"train_tokens_per_second": 169.786
|
| 550 |
+
},
|
| 551 |
+
{
|
| 552 |
+
"epoch": 0.802675585284281,
|
| 553 |
+
"grad_norm": 0.4747498444810908,
|
| 554 |
+
"learning_rate": 4.8890024432806806e-05,
|
| 555 |
+
"loss": 0.7729,
|
| 556 |
+
"num_input_tokens_seen": 1709480,
|
| 557 |
+
"step": 270,
|
| 558 |
+
"train_runtime": 10078.2583,
|
| 559 |
+
"train_tokens_per_second": 169.621
|
| 560 |
+
},
|
| 561 |
+
{
|
| 562 |
+
"epoch": 0.8175399479747306,
|
| 563 |
+
"grad_norm": 0.40268303548713125,
|
| 564 |
+
"learning_rate": 4.8814883522969545e-05,
|
| 565 |
+
"loss": 0.653,
|
| 566 |
+
"num_input_tokens_seen": 1744560,
|
| 567 |
+
"step": 275,
|
| 568 |
+
"train_runtime": 10262.7966,
|
| 569 |
+
"train_tokens_per_second": 169.989
|
| 570 |
+
},
|
| 571 |
+
{
|
| 572 |
+
"epoch": 0.8324043106651803,
|
| 573 |
+
"grad_norm": 0.5361180434189863,
|
| 574 |
+
"learning_rate": 4.8737343471505806e-05,
|
| 575 |
+
"loss": 0.75,
|
| 576 |
+
"num_input_tokens_seen": 1773576,
|
| 577 |
+
"step": 280,
|
| 578 |
+
"train_runtime": 10433.1959,
|
| 579 |
+
"train_tokens_per_second": 169.994
|
| 580 |
+
},
|
| 581 |
+
{
|
| 582 |
+
"epoch": 0.8472686733556298,
|
| 583 |
+
"grad_norm": 0.4150141248761779,
|
| 584 |
+
"learning_rate": 4.865741208989888e-05,
|
| 585 |
+
"loss": 0.6366,
|
| 586 |
+
"num_input_tokens_seen": 1807248,
|
| 587 |
+
"step": 285,
|
| 588 |
+
"train_runtime": 10621.0739,
|
| 589 |
+
"train_tokens_per_second": 170.157
|
| 590 |
+
},
|
| 591 |
+
{
|
| 592 |
+
"epoch": 0.8621330360460795,
|
| 593 |
+
"grad_norm": 0.557076443271248,
|
| 594 |
+
"learning_rate": 4.857509743053774e-05,
|
| 595 |
+
"loss": 0.7992,
|
| 596 |
+
"num_input_tokens_seen": 1836512,
|
| 597 |
+
"step": 290,
|
| 598 |
+
"train_runtime": 10790.2552,
|
| 599 |
+
"train_tokens_per_second": 170.201
|
| 600 |
+
},
|
| 601 |
+
{
|
| 602 |
+
"epoch": 0.8769973987365292,
|
| 603 |
+
"grad_norm": 0.622394894811983,
|
| 604 |
+
"learning_rate": 4.8490407785905756e-05,
|
| 605 |
+
"loss": 0.7259,
|
| 606 |
+
"num_input_tokens_seen": 1864768,
|
| 607 |
+
"step": 295,
|
| 608 |
+
"train_runtime": 10958.7213,
|
| 609 |
+
"train_tokens_per_second": 170.163
|
| 610 |
+
},
|
| 611 |
+
{
|
| 612 |
+
"epoch": 0.8918617614269788,
|
| 613 |
+
"grad_norm": 0.32132112389269674,
|
| 614 |
+
"learning_rate": 4.840335168774532e-05,
|
| 615 |
+
"loss": 0.6569,
|
| 616 |
+
"num_input_tokens_seen": 1897416,
|
| 617 |
+
"step": 300,
|
| 618 |
+
"train_runtime": 11132.0894,
|
| 619 |
+
"train_tokens_per_second": 170.446
|
| 620 |
+
},
|
| 621 |
+
{
|
| 622 |
+
"epoch": 0.9067261241174285,
|
| 623 |
+
"grad_norm": 0.5699931721537229,
|
| 624 |
+
"learning_rate": 4.8313937906198415e-05,
|
| 625 |
+
"loss": 0.7109,
|
| 626 |
+
"num_input_tokens_seen": 1925464,
|
| 627 |
+
"step": 305,
|
| 628 |
+
"train_runtime": 11302.0412,
|
| 629 |
+
"train_tokens_per_second": 170.364
|
| 630 |
+
},
|
| 631 |
+
{
|
| 632 |
+
"epoch": 0.9215904868078781,
|
| 633 |
+
"grad_norm": 0.49216652434145286,
|
| 634 |
+
"learning_rate": 4.822217544892298e-05,
|
| 635 |
+
"loss": 0.6374,
|
| 636 |
+
"num_input_tokens_seen": 1956064,
|
| 637 |
+
"step": 310,
|
| 638 |
+
"train_runtime": 11477.605,
|
| 639 |
+
"train_tokens_per_second": 170.424
|
| 640 |
+
},
|
| 641 |
+
{
|
| 642 |
+
"epoch": 0.9364548494983278,
|
| 643 |
+
"grad_norm": 0.289252551788521,
|
| 644 |
+
"learning_rate": 4.812807356018556e-05,
|
| 645 |
+
"loss": 0.7321,
|
| 646 |
+
"num_input_tokens_seen": 1986504,
|
| 647 |
+
"step": 315,
|
| 648 |
+
"train_runtime": 11660.7035,
|
| 649 |
+
"train_tokens_per_second": 170.359
|
| 650 |
+
},
|
| 651 |
+
{
|
| 652 |
+
"epoch": 0.9513192121887774,
|
| 653 |
+
"grad_norm": 0.4471738070997589,
|
| 654 |
+
"learning_rate": 4.803164171993001e-05,
|
| 655 |
+
"loss": 0.6904,
|
| 656 |
+
"num_input_tokens_seen": 2012976,
|
| 657 |
+
"step": 320,
|
| 658 |
+
"train_runtime": 11819.4269,
|
| 659 |
+
"train_tokens_per_second": 170.311
|
| 660 |
+
},
|
| 661 |
+
{
|
| 662 |
+
"epoch": 0.966183574879227,
|
| 663 |
+
"grad_norm": 0.4988862477404023,
|
| 664 |
+
"learning_rate": 4.793288964282244e-05,
|
| 665 |
+
"loss": 0.8029,
|
| 666 |
+
"num_input_tokens_seen": 2041512,
|
| 667 |
+
"step": 325,
|
| 668 |
+
"train_runtime": 11993.4775,
|
| 669 |
+
"train_tokens_per_second": 170.219
|
| 670 |
+
},
|
| 671 |
+
{
|
| 672 |
+
"epoch": 0.9810479375696767,
|
| 673 |
+
"grad_norm": 0.4455586736978044,
|
| 674 |
+
"learning_rate": 4.783182727727258e-05,
|
| 675 |
+
"loss": 0.7208,
|
| 676 |
+
"num_input_tokens_seen": 2065864,
|
| 677 |
+
"step": 330,
|
| 678 |
+
"train_runtime": 12150.1754,
|
| 679 |
+
"train_tokens_per_second": 170.028
|
| 680 |
+
},
|
| 681 |
+
{
|
| 682 |
+
"epoch": 0.9959123002601263,
|
| 683 |
+
"grad_norm": 0.45313304199376386,
|
| 684 |
+
"learning_rate": 4.772846480443154e-05,
|
| 685 |
+
"loss": 0.6892,
|
| 686 |
+
"num_input_tokens_seen": 2099192,
|
| 687 |
+
"step": 335,
|
| 688 |
+
"train_runtime": 12332.4171,
|
| 689 |
+
"train_tokens_per_second": 170.217
|
| 690 |
+
},
|
| 691 |
+
{
|
| 692 |
+
"epoch": 1.0089186176142697,
|
| 693 |
+
"grad_norm": 0.34932689166112835,
|
| 694 |
+
"learning_rate": 4.762281263716619e-05,
|
| 695 |
+
"loss": 0.7558,
|
| 696 |
+
"num_input_tokens_seen": 2125816,
|
| 697 |
+
"step": 340,
|
| 698 |
+
"train_runtime": 12484.2733,
|
| 699 |
+
"train_tokens_per_second": 170.28
|
| 700 |
+
},
|
| 701 |
+
{
|
| 702 |
+
"epoch": 1.0237829803047194,
|
| 703 |
+
"grad_norm": 0.4092397826706536,
|
| 704 |
+
"learning_rate": 4.751488141901009e-05,
|
| 705 |
+
"loss": 0.5846,
|
| 706 |
+
"num_input_tokens_seen": 2160160,
|
| 707 |
+
"step": 345,
|
| 708 |
+
"train_runtime": 12674.5489,
|
| 709 |
+
"train_tokens_per_second": 170.433
|
| 710 |
+
},
|
| 711 |
+
{
|
| 712 |
+
"epoch": 1.038647342995169,
|
| 713 |
+
"grad_norm": 0.31168294599405477,
|
| 714 |
+
"learning_rate": 4.740468202309132e-05,
|
| 715 |
+
"loss": 0.5396,
|
| 716 |
+
"num_input_tokens_seen": 2202880,
|
| 717 |
+
"step": 350,
|
| 718 |
+
"train_runtime": 12878.5359,
|
| 719 |
+
"train_tokens_per_second": 171.05
|
| 720 |
+
},
|
| 721 |
+
{
|
| 722 |
+
"epoch": 1.0535117056856187,
|
| 723 |
+
"grad_norm": 0.4562243418713546,
|
| 724 |
+
"learning_rate": 4.729222555103703e-05,
|
| 725 |
+
"loss": 0.6491,
|
| 726 |
+
"num_input_tokens_seen": 2232216,
|
| 727 |
+
"step": 355,
|
| 728 |
+
"train_runtime": 13048.5779,
|
| 729 |
+
"train_tokens_per_second": 171.07
|
| 730 |
+
},
|
| 731 |
+
{
|
| 732 |
+
"epoch": 1.0683760683760684,
|
| 733 |
+
"grad_norm": 0.37853300240110366,
|
| 734 |
+
"learning_rate": 4.717752333185511e-05,
|
| 735 |
+
"loss": 0.5624,
|
| 736 |
+
"num_input_tokens_seen": 2255520,
|
| 737 |
+
"step": 360,
|
| 738 |
+
"train_runtime": 13200.2201,
|
| 739 |
+
"train_tokens_per_second": 170.87
|
| 740 |
+
},
|
| 741 |
+
{
|
| 742 |
+
"epoch": 1.083240431066518,
|
| 743 |
+
"grad_norm": 0.505365165589364,
|
| 744 |
+
"learning_rate": 4.706058692079288e-05,
|
| 745 |
+
"loss": 0.5882,
|
| 746 |
+
"num_input_tokens_seen": 2295400,
|
| 747 |
+
"step": 365,
|
| 748 |
+
"train_runtime": 13416.4241,
|
| 749 |
+
"train_tokens_per_second": 171.089
|
| 750 |
+
},
|
| 751 |
+
{
|
| 752 |
+
"epoch": 1.0981047937569677,
|
| 753 |
+
"grad_norm": 0.7984507974057846,
|
| 754 |
+
"learning_rate": 4.6941428098172956e-05,
|
| 755 |
+
"loss": 0.6382,
|
| 756 |
+
"num_input_tokens_seen": 2322496,
|
| 757 |
+
"step": 370,
|
| 758 |
+
"train_runtime": 13576.7515,
|
| 759 |
+
"train_tokens_per_second": 171.064
|
| 760 |
+
},
|
| 761 |
+
{
|
| 762 |
+
"epoch": 1.1129691564474173,
|
| 763 |
+
"grad_norm": 0.5722349779883125,
|
| 764 |
+
"learning_rate": 4.682005886820656e-05,
|
| 765 |
+
"loss": 0.6791,
|
| 766 |
+
"num_input_tokens_seen": 2353904,
|
| 767 |
+
"step": 375,
|
| 768 |
+
"train_runtime": 13756.8344,
|
| 769 |
+
"train_tokens_per_second": 171.108
|
| 770 |
+
},
|
| 771 |
+
{
|
| 772 |
+
"epoch": 1.127833519137867,
|
| 773 |
+
"grad_norm": 0.826155965272619,
|
| 774 |
+
"learning_rate": 4.669649145778412e-05,
|
| 775 |
+
"loss": 0.6277,
|
| 776 |
+
"num_input_tokens_seen": 2389640,
|
| 777 |
+
"step": 380,
|
| 778 |
+
"train_runtime": 13955.4804,
|
| 779 |
+
"train_tokens_per_second": 171.233
|
| 780 |
+
},
|
| 781 |
+
{
|
| 782 |
+
"epoch": 1.1426978818283167,
|
| 783 |
+
"grad_norm": 0.48208601176965793,
|
| 784 |
+
"learning_rate": 4.657073831524358e-05,
|
| 785 |
+
"loss": 0.5038,
|
| 786 |
+
"num_input_tokens_seen": 2427872,
|
| 787 |
+
"step": 385,
|
| 788 |
+
"train_runtime": 14154.7296,
|
| 789 |
+
"train_tokens_per_second": 171.524
|
| 790 |
+
},
|
| 791 |
+
{
|
| 792 |
+
"epoch": 1.1575622445187663,
|
| 793 |
+
"grad_norm": 0.862951567640207,
|
| 794 |
+
"learning_rate": 4.644281210911631e-05,
|
| 795 |
+
"loss": 0.6581,
|
| 796 |
+
"num_input_tokens_seen": 2453984,
|
| 797 |
+
"step": 390,
|
| 798 |
+
"train_runtime": 14320.9438,
|
| 799 |
+
"train_tokens_per_second": 171.356
|
| 800 |
+
},
|
| 801 |
+
{
|
| 802 |
+
"epoch": 1.172426607209216,
|
| 803 |
+
"grad_norm": 0.6446091778736193,
|
| 804 |
+
"learning_rate": 4.631272572685086e-05,
|
| 805 |
+
"loss": 0.5966,
|
| 806 |
+
"num_input_tokens_seen": 2482696,
|
| 807 |
+
"step": 395,
|
| 808 |
+
"train_runtime": 14484.582,
|
| 809 |
+
"train_tokens_per_second": 171.403
|
| 810 |
+
},
|
| 811 |
+
{
|
| 812 |
+
"epoch": 1.1872909698996654,
|
| 813 |
+
"grad_norm": 0.5940985307805492,
|
| 814 |
+
"learning_rate": 4.618049227351467e-05,
|
| 815 |
+
"loss": 0.6238,
|
| 816 |
+
"num_input_tokens_seen": 2515024,
|
| 817 |
+
"step": 400,
|
| 818 |
+
"train_runtime": 14670.3277,
|
| 819 |
+
"train_tokens_per_second": 171.436
|
| 820 |
+
},
|
| 821 |
+
{
|
| 822 |
+
"epoch": 1.1872909698996654,
|
| 823 |
+
"eval_loss": 0.8729309439659119,
|
| 824 |
+
"eval_runtime": 355.7054,
|
| 825 |
+
"eval_samples_per_second": 1.316,
|
| 826 |
+
"eval_steps_per_second": 0.658,
|
| 827 |
+
"num_input_tokens_seen": 2515024,
|
| 828 |
+
"step": 400
|
| 829 |
+
},
|
| 830 |
+
{
|
| 831 |
+
"epoch": 1.2021553325901153,
|
| 832 |
+
"grad_norm": 0.42070342961969176,
|
| 833 |
+
"learning_rate": 4.6046125070473854e-05,
|
| 834 |
+
"loss": 0.6013,
|
| 835 |
+
"num_input_tokens_seen": 2544048,
|
| 836 |
+
"step": 405,
|
| 837 |
+
"train_runtime": 15222.6559,
|
| 838 |
+
"train_tokens_per_second": 167.122
|
| 839 |
+
},
|
| 840 |
+
{
|
| 841 |
+
"epoch": 1.2170196952805648,
|
| 842 |
+
"grad_norm": 0.5065404783086159,
|
| 843 |
+
"learning_rate": 4.5909637654051194e-05,
|
| 844 |
+
"loss": 0.6597,
|
| 845 |
+
"num_input_tokens_seen": 2577080,
|
| 846 |
+
"step": 410,
|
| 847 |
+
"train_runtime": 15412.9664,
|
| 848 |
+
"train_tokens_per_second": 167.202
|
| 849 |
+
},
|
| 850 |
+
{
|
| 851 |
+
"epoch": 1.2318840579710144,
|
| 852 |
+
"grad_norm": 0.7043622561243575,
|
| 853 |
+
"learning_rate": 4.577104377416243e-05,
|
| 854 |
+
"loss": 0.6459,
|
| 855 |
+
"num_input_tokens_seen": 2605240,
|
| 856 |
+
"step": 415,
|
| 857 |
+
"train_runtime": 15585.0317,
|
| 858 |
+
"train_tokens_per_second": 167.163
|
| 859 |
+
},
|
| 860 |
+
{
|
| 861 |
+
"epoch": 1.246748420661464,
|
| 862 |
+
"grad_norm": 0.5815684563195597,
|
| 863 |
+
"learning_rate": 4.5630357392931136e-05,
|
| 864 |
+
"loss": 0.5973,
|
| 865 |
+
"num_input_tokens_seen": 2634016,
|
| 866 |
+
"step": 420,
|
| 867 |
+
"train_runtime": 15757.2057,
|
| 868 |
+
"train_tokens_per_second": 167.163
|
| 869 |
+
},
|
| 870 |
+
{
|
| 871 |
+
"epoch": 1.2616127833519137,
|
| 872 |
+
"grad_norm": 0.5571563336020446,
|
| 873 |
+
"learning_rate": 4.548759268328211e-05,
|
| 874 |
+
"loss": 0.6341,
|
| 875 |
+
"num_input_tokens_seen": 2663784,
|
| 876 |
+
"step": 425,
|
| 877 |
+
"train_runtime": 15930.8715,
|
| 878 |
+
"train_tokens_per_second": 167.209
|
| 879 |
+
},
|
| 880 |
+
{
|
| 881 |
+
"epoch": 1.2764771460423634,
|
| 882 |
+
"grad_norm": 0.6460743117704272,
|
| 883 |
+
"learning_rate": 4.534276402751361e-05,
|
| 884 |
+
"loss": 0.651,
|
| 885 |
+
"num_input_tokens_seen": 2697024,
|
| 886 |
+
"step": 430,
|
| 887 |
+
"train_runtime": 16122.3767,
|
| 888 |
+
"train_tokens_per_second": 167.285
|
| 889 |
+
},
|
| 890 |
+
{
|
| 891 |
+
"epoch": 1.291341508732813,
|
| 892 |
+
"grad_norm": 0.6482175581649149,
|
| 893 |
+
"learning_rate": 4.5195886015848454e-05,
|
| 894 |
+
"loss": 0.6475,
|
| 895 |
+
"num_input_tokens_seen": 2719400,
|
| 896 |
+
"step": 435,
|
| 897 |
+
"train_runtime": 16266.13,
|
| 898 |
+
"train_tokens_per_second": 167.182
|
| 899 |
+
},
|
| 900 |
+
{
|
| 901 |
+
"epoch": 1.3062058714232627,
|
| 902 |
+
"grad_norm": 0.6094464333253732,
|
| 903 |
+
"learning_rate": 4.5046973444964165e-05,
|
| 904 |
+
"loss": 0.6373,
|
| 905 |
+
"num_input_tokens_seen": 2751744,
|
| 906 |
+
"step": 440,
|
| 907 |
+
"train_runtime": 16434.9886,
|
| 908 |
+
"train_tokens_per_second": 167.432
|
| 909 |
+
},
|
| 910 |
+
{
|
| 911 |
+
"epoch": 1.3210702341137124,
|
| 912 |
+
"grad_norm": 0.8126749512005293,
|
| 913 |
+
"learning_rate": 4.4896041316502335e-05,
|
| 914 |
+
"loss": 0.6434,
|
| 915 |
+
"num_input_tokens_seen": 2791080,
|
| 916 |
+
"step": 445,
|
| 917 |
+
"train_runtime": 16641.4496,
|
| 918 |
+
"train_tokens_per_second": 167.719
|
| 919 |
+
},
|
| 920 |
+
{
|
| 921 |
+
"epoch": 1.335934596804162,
|
| 922 |
+
"grad_norm": 0.5026965778291799,
|
| 923 |
+
"learning_rate": 4.474310483555739e-05,
|
| 924 |
+
"loss": 0.5877,
|
| 925 |
+
"num_input_tokens_seen": 2823104,
|
| 926 |
+
"step": 450,
|
| 927 |
+
"train_runtime": 16826.3481,
|
| 928 |
+
"train_tokens_per_second": 167.779
|
| 929 |
+
},
|
| 930 |
+
{
|
| 931 |
+
"epoch": 1.3507989594946117,
|
| 932 |
+
"grad_norm": 0.824948158879483,
|
| 933 |
+
"learning_rate": 4.4588179409144734e-05,
|
| 934 |
+
"loss": 0.6944,
|
| 935 |
+
"num_input_tokens_seen": 2850480,
|
| 936 |
+
"step": 455,
|
| 937 |
+
"train_runtime": 16993.1957,
|
| 938 |
+
"train_tokens_per_second": 167.742
|
| 939 |
+
},
|
| 940 |
+
{
|
| 941 |
+
"epoch": 1.3656633221850614,
|
| 942 |
+
"grad_norm": 0.6935578787290407,
|
| 943 |
+
"learning_rate": 4.4431280644648676e-05,
|
| 944 |
+
"loss": 0.6677,
|
| 945 |
+
"num_input_tokens_seen": 2877920,
|
| 946 |
+
"step": 460,
|
| 947 |
+
"train_runtime": 17162.7566,
|
| 948 |
+
"train_tokens_per_second": 167.684
|
| 949 |
+
},
|
| 950 |
+
{
|
| 951 |
+
"epoch": 1.380527684875511,
|
| 952 |
+
"grad_norm": 0.775256808439057,
|
| 953 |
+
"learning_rate": 4.427242434825013e-05,
|
| 954 |
+
"loss": 0.6185,
|
| 955 |
+
"num_input_tokens_seen": 2910104,
|
| 956 |
+
"step": 465,
|
| 957 |
+
"train_runtime": 17349.4341,
|
| 958 |
+
"train_tokens_per_second": 167.735
|
| 959 |
+
},
|
| 960 |
+
{
|
| 961 |
+
"epoch": 1.3953920475659607,
|
| 962 |
+
"grad_norm": 0.5290136444612117,
|
| 963 |
+
"learning_rate": 4.4111626523334235e-05,
|
| 964 |
+
"loss": 0.4792,
|
| 965 |
+
"num_input_tokens_seen": 2943600,
|
| 966 |
+
"step": 470,
|
| 967 |
+
"train_runtime": 17535.2111,
|
| 968 |
+
"train_tokens_per_second": 167.868
|
| 969 |
+
},
|
| 970 |
+
{
|
| 971 |
+
"epoch": 1.4102564102564101,
|
| 972 |
+
"grad_norm": 0.431851575535354,
|
| 973 |
+
"learning_rate": 4.394890336887819e-05,
|
| 974 |
+
"loss": 0.5479,
|
| 975 |
+
"num_input_tokens_seen": 2983616,
|
| 976 |
+
"step": 475,
|
| 977 |
+
"train_runtime": 17735.6146,
|
| 978 |
+
"train_tokens_per_second": 168.227
|
| 979 |
+
},
|
| 980 |
+
{
|
| 981 |
+
"epoch": 1.42512077294686,
|
| 982 |
+
"grad_norm": 0.707234404224925,
|
| 983 |
+
"learning_rate": 4.378427127781935e-05,
|
| 984 |
+
"loss": 0.6563,
|
| 985 |
+
"num_input_tokens_seen": 3010896,
|
| 986 |
+
"step": 480,
|
| 987 |
+
"train_runtime": 17897.1321,
|
| 988 |
+
"train_tokens_per_second": 168.233
|
| 989 |
+
},
|
| 990 |
+
{
|
| 991 |
+
"epoch": 1.4399851356373095,
|
| 992 |
+
"grad_norm": 0.5496990766753803,
|
| 993 |
+
"learning_rate": 4.361774683540375e-05,
|
| 994 |
+
"loss": 0.5839,
|
| 995 |
+
"num_input_tokens_seen": 3040128,
|
| 996 |
+
"step": 485,
|
| 997 |
+
"train_runtime": 18074.4912,
|
| 998 |
+
"train_tokens_per_second": 168.2
|
| 999 |
+
},
|
| 1000 |
+
{
|
| 1001 |
+
"epoch": 1.4548494983277591,
|
| 1002 |
+
"grad_norm": 0.40637808524319596,
|
| 1003 |
+
"learning_rate": 4.34493468175153e-05,
|
| 1004 |
+
"loss": 0.5499,
|
| 1005 |
+
"num_input_tokens_seen": 3070192,
|
| 1006 |
+
"step": 490,
|
| 1007 |
+
"train_runtime": 18243.0065,
|
| 1008 |
+
"train_tokens_per_second": 168.294
|
| 1009 |
+
},
|
| 1010 |
+
{
|
| 1011 |
+
"epoch": 1.4697138610182088,
|
| 1012 |
+
"grad_norm": 0.5939003521421573,
|
| 1013 |
+
"learning_rate": 4.327908818898581e-05,
|
| 1014 |
+
"loss": 0.6602,
|
| 1015 |
+
"num_input_tokens_seen": 3100544,
|
| 1016 |
+
"step": 495,
|
| 1017 |
+
"train_runtime": 18419.7661,
|
| 1018 |
+
"train_tokens_per_second": 168.327
|
| 1019 |
+
},
|
| 1020 |
+
{
|
| 1021 |
+
"epoch": 1.4845782237086584,
|
| 1022 |
+
"grad_norm": 0.5699315589577931,
|
| 1023 |
+
"learning_rate": 4.3106988101885825e-05,
|
| 1024 |
+
"loss": 0.6172,
|
| 1025 |
+
"num_input_tokens_seen": 3134864,
|
| 1026 |
+
"step": 500,
|
| 1027 |
+
"train_runtime": 18605.4997,
|
| 1028 |
+
"train_tokens_per_second": 168.491
|
| 1029 |
+
},
|
| 1030 |
+
{
|
| 1031 |
+
"epoch": 1.499442586399108,
|
| 1032 |
+
"grad_norm": 0.5176180044018102,
|
| 1033 |
+
"learning_rate": 4.293306389379682e-05,
|
| 1034 |
+
"loss": 0.6932,
|
| 1035 |
+
"num_input_tokens_seen": 3164496,
|
| 1036 |
+
"step": 505,
|
| 1037 |
+
"train_runtime": 18782.3882,
|
| 1038 |
+
"train_tokens_per_second": 168.482
|
| 1039 |
+
},
|
| 1040 |
+
{
|
| 1041 |
+
"epoch": 1.5143069490895578,
|
| 1042 |
+
"grad_norm": 0.8125909060302873,
|
| 1043 |
+
"learning_rate": 4.275733308606452e-05,
|
| 1044 |
+
"loss": 0.6075,
|
| 1045 |
+
"num_input_tokens_seen": 3201592,
|
| 1046 |
+
"step": 510,
|
| 1047 |
+
"train_runtime": 18974.7401,
|
| 1048 |
+
"train_tokens_per_second": 168.729
|
| 1049 |
+
},
|
| 1050 |
+
{
|
| 1051 |
+
"epoch": 1.5291713117800074,
|
| 1052 |
+
"grad_norm": 0.4238803109744304,
|
| 1053 |
+
"learning_rate": 4.2579813382033764e-05,
|
| 1054 |
+
"loss": 0.6465,
|
| 1055 |
+
"num_input_tokens_seen": 3242824,
|
| 1056 |
+
"step": 515,
|
| 1057 |
+
"train_runtime": 19182.5699,
|
| 1058 |
+
"train_tokens_per_second": 169.051
|
| 1059 |
+
},
|
| 1060 |
+
{
|
| 1061 |
+
"epoch": 1.544035674470457,
|
| 1062 |
+
"grad_norm": 0.5568097005839904,
|
| 1063 |
+
"learning_rate": 4.240052266526512e-05,
|
| 1064 |
+
"loss": 0.6159,
|
| 1065 |
+
"num_input_tokens_seen": 3273600,
|
| 1066 |
+
"step": 520,
|
| 1067 |
+
"train_runtime": 19367.6233,
|
| 1068 |
+
"train_tokens_per_second": 169.024
|
| 1069 |
+
},
|
| 1070 |
+
{
|
| 1071 |
+
"epoch": 1.5589000371609067,
|
| 1072 |
+
"grad_norm": 0.5943741262587303,
|
| 1073 |
+
"learning_rate": 4.22194789977332e-05,
|
| 1074 |
+
"loss": 0.6439,
|
| 1075 |
+
"num_input_tokens_seen": 3308472,
|
| 1076 |
+
"step": 525,
|
| 1077 |
+
"train_runtime": 19562.3706,
|
| 1078 |
+
"train_tokens_per_second": 169.124
|
| 1079 |
+
},
|
| 1080 |
+
{
|
| 1081 |
+
"epoch": 1.5737643998513564,
|
| 1082 |
+
"grad_norm": 0.8079971411544303,
|
| 1083 |
+
"learning_rate": 4.203670061800712e-05,
|
| 1084 |
+
"loss": 0.6874,
|
| 1085 |
+
"num_input_tokens_seen": 3332568,
|
| 1086 |
+
"step": 530,
|
| 1087 |
+
"train_runtime": 19717.9765,
|
| 1088 |
+
"train_tokens_per_second": 169.012
|
| 1089 |
+
},
|
| 1090 |
+
{
|
| 1091 |
+
"epoch": 1.588628762541806,
|
| 1092 |
+
"grad_norm": 0.551762199364396,
|
| 1093 |
+
"learning_rate": 4.1852205939413104e-05,
|
| 1094 |
+
"loss": 0.6013,
|
| 1095 |
+
"num_input_tokens_seen": 3366712,
|
| 1096 |
+
"step": 535,
|
| 1097 |
+
"train_runtime": 19903.6186,
|
| 1098 |
+
"train_tokens_per_second": 169.151
|
| 1099 |
+
},
|
| 1100 |
+
{
|
| 1101 |
+
"epoch": 1.6034931252322555,
|
| 1102 |
+
"grad_norm": 0.5068985496527462,
|
| 1103 |
+
"learning_rate": 4.1666013548179496e-05,
|
| 1104 |
+
"loss": 0.6359,
|
| 1105 |
+
"num_input_tokens_seen": 3391608,
|
| 1106 |
+
"step": 540,
|
| 1107 |
+
"train_runtime": 20057.39,
|
| 1108 |
+
"train_tokens_per_second": 169.095
|
| 1109 |
+
},
|
| 1110 |
+
{
|
| 1111 |
+
"epoch": 1.6183574879227054,
|
| 1112 |
+
"grad_norm": 0.6855669469937492,
|
| 1113 |
+
"learning_rate": 4.147814220156437e-05,
|
| 1114 |
+
"loss": 0.6127,
|
| 1115 |
+
"num_input_tokens_seen": 3421400,
|
| 1116 |
+
"step": 545,
|
| 1117 |
+
"train_runtime": 20228.9847,
|
| 1118 |
+
"train_tokens_per_second": 169.134
|
| 1119 |
+
},
|
| 1120 |
+
{
|
| 1121 |
+
"epoch": 1.6332218506131548,
|
| 1122 |
+
"grad_norm": 0.4879299108867579,
|
| 1123 |
+
"learning_rate": 4.128861082596592e-05,
|
| 1124 |
+
"loss": 0.5612,
|
| 1125 |
+
"num_input_tokens_seen": 3455584,
|
| 1126 |
+
"step": 550,
|
| 1127 |
+
"train_runtime": 20414.7402,
|
| 1128 |
+
"train_tokens_per_second": 169.269
|
| 1129 |
+
},
|
| 1130 |
+
{
|
| 1131 |
+
"epoch": 1.6480862133036047,
|
| 1132 |
+
"grad_norm": 0.8129316983564234,
|
| 1133 |
+
"learning_rate": 4.109743851501573e-05,
|
| 1134 |
+
"loss": 0.6068,
|
| 1135 |
+
"num_input_tokens_seen": 3480392,
|
| 1136 |
+
"step": 555,
|
| 1137 |
+
"train_runtime": 20570.2335,
|
| 1138 |
+
"train_tokens_per_second": 169.196
|
| 1139 |
+
},
|
| 1140 |
+
{
|
| 1141 |
+
"epoch": 1.6629505759940542,
|
| 1142 |
+
"grad_norm": 0.5732379522903723,
|
| 1143 |
+
"learning_rate": 4.090464452765535e-05,
|
| 1144 |
+
"loss": 0.6331,
|
| 1145 |
+
"num_input_tokens_seen": 3515664,
|
| 1146 |
+
"step": 560,
|
| 1147 |
+
"train_runtime": 20754.2066,
|
| 1148 |
+
"train_tokens_per_second": 169.395
|
| 1149 |
+
},
|
| 1150 |
+
{
|
| 1151 |
+
"epoch": 1.677814938684504,
|
| 1152 |
+
"grad_norm": 0.8129257440384582,
|
| 1153 |
+
"learning_rate": 4.0710248286195994e-05,
|
| 1154 |
+
"loss": 0.6944,
|
| 1155 |
+
"num_input_tokens_seen": 3538880,
|
| 1156 |
+
"step": 565,
|
| 1157 |
+
"train_runtime": 20901.2916,
|
| 1158 |
+
"train_tokens_per_second": 169.314
|
| 1159 |
+
},
|
| 1160 |
+
{
|
| 1161 |
+
"epoch": 1.6926793013749535,
|
| 1162 |
+
"grad_norm": 0.5743351437424525,
|
| 1163 |
+
"learning_rate": 4.051426937436207e-05,
|
| 1164 |
+
"loss": 0.6007,
|
| 1165 |
+
"num_input_tokens_seen": 3576168,
|
| 1166 |
+
"step": 570,
|
| 1167 |
+
"train_runtime": 21091.5246,
|
| 1168 |
+
"train_tokens_per_second": 169.555
|
| 1169 |
+
},
|
| 1170 |
+
{
|
| 1171 |
+
"epoch": 1.7075436640654031,
|
| 1172 |
+
"grad_norm": 1.7481870694933204,
|
| 1173 |
+
"learning_rate": 4.0316727535318175e-05,
|
| 1174 |
+
"loss": 0.6374,
|
| 1175 |
+
"num_input_tokens_seen": 3606944,
|
| 1176 |
+
"step": 575,
|
| 1177 |
+
"train_runtime": 21266.3906,
|
| 1178 |
+
"train_tokens_per_second": 169.608
|
| 1179 |
+
},
|
| 1180 |
+
{
|
| 1181 |
+
"epoch": 1.7224080267558528,
|
| 1182 |
+
"grad_norm": 0.5813131410466363,
|
| 1183 |
+
"learning_rate": 4.0117642669680164e-05,
|
| 1184 |
+
"loss": 0.5499,
|
| 1185 |
+
"num_input_tokens_seen": 3637888,
|
| 1186 |
+
"step": 580,
|
| 1187 |
+
"train_runtime": 21440.8922,
|
| 1188 |
+
"train_tokens_per_second": 169.671
|
| 1189 |
+
},
|
| 1190 |
+
{
|
| 1191 |
+
"epoch": 1.7372723894463025,
|
| 1192 |
+
"grad_norm": 0.8484268316277586,
|
| 1193 |
+
"learning_rate": 3.991703483351039e-05,
|
| 1194 |
+
"loss": 0.6668,
|
| 1195 |
+
"num_input_tokens_seen": 3661216,
|
| 1196 |
+
"step": 585,
|
| 1197 |
+
"train_runtime": 21594.0452,
|
| 1198 |
+
"train_tokens_per_second": 169.547
|
| 1199 |
+
},
|
| 1200 |
+
{
|
| 1201 |
+
"epoch": 1.7521367521367521,
|
| 1202 |
+
"grad_norm": 0.34272148077746284,
|
| 1203 |
+
"learning_rate": 3.9714924236297155e-05,
|
| 1204 |
+
"loss": 0.6033,
|
| 1205 |
+
"num_input_tokens_seen": 3700064,
|
| 1206 |
+
"step": 590,
|
| 1207 |
+
"train_runtime": 21798.0768,
|
| 1208 |
+
"train_tokens_per_second": 169.743
|
| 1209 |
+
},
|
| 1210 |
+
{
|
| 1211 |
+
"epoch": 1.7670011148272018,
|
| 1212 |
+
"grad_norm": 0.588668446023156,
|
| 1213 |
+
"learning_rate": 3.9511331238918837e-05,
|
| 1214 |
+
"loss": 0.5136,
|
| 1215 |
+
"num_input_tokens_seen": 3730992,
|
| 1216 |
+
"step": 595,
|
| 1217 |
+
"train_runtime": 21979.8348,
|
| 1218 |
+
"train_tokens_per_second": 169.746
|
| 1219 |
+
},
|
| 1220 |
+
{
|
| 1221 |
+
"epoch": 1.7818654775176515,
|
| 1222 |
+
"grad_norm": 0.5461111827190066,
|
| 1223 |
+
"learning_rate": 3.9306276351592685e-05,
|
| 1224 |
+
"loss": 0.5867,
|
| 1225 |
+
"num_input_tokens_seen": 3766176,
|
| 1226 |
+
"step": 600,
|
| 1227 |
+
"train_runtime": 22165.8728,
|
| 1228 |
+
"train_tokens_per_second": 169.909
|
| 1229 |
+
},
|
| 1230 |
+
{
|
| 1231 |
+
"epoch": 1.7818654775176515,
|
| 1232 |
+
"eval_loss": 0.8493290543556213,
|
| 1233 |
+
"eval_runtime": 354.1156,
|
| 1234 |
+
"eval_samples_per_second": 1.322,
|
| 1235 |
+
"eval_steps_per_second": 0.661,
|
| 1236 |
+
"num_input_tokens_seen": 3766176,
|
| 1237 |
+
"step": 600
|
| 1238 |
+
}
|
| 1239 |
+
],
|
| 1240 |
+
"logging_steps": 5,
|
| 1241 |
+
"max_steps": 1685,
|
| 1242 |
+
"num_input_tokens_seen": 3766176,
|
| 1243 |
+
"num_train_epochs": 5,
|
| 1244 |
+
"save_steps": 200,
|
| 1245 |
+
"stateful_callbacks": {
|
| 1246 |
+
"TrainerControl": {
|
| 1247 |
+
"args": {
|
| 1248 |
+
"should_epoch_stop": false,
|
| 1249 |
+
"should_evaluate": false,
|
| 1250 |
+
"should_log": false,
|
| 1251 |
+
"should_save": true,
|
| 1252 |
+
"should_training_stop": false
|
| 1253 |
+
},
|
| 1254 |
+
"attributes": {}
|
| 1255 |
+
}
|
| 1256 |
+
},
|
| 1257 |
+
"total_flos": 166671066071040.0,
|
| 1258 |
+
"train_batch_size": 1,
|
| 1259 |
+
"trial_name": null,
|
| 1260 |
+
"trial_params": null
|
| 1261 |
+
}
|