Upload trainer_state.json
Browse files- trainer_state.json +1296 -0
trainer_state.json
ADDED
|
@@ -0,0 +1,1296 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"best_metric": null,
|
| 3 |
+
"best_model_checkpoint": null,
|
| 4 |
+
"epoch": 0.25033952297803896,
|
| 5 |
+
"global_step": 40000,
|
| 6 |
+
"is_hyper_param_search": false,
|
| 7 |
+
"is_local_process_zero": true,
|
| 8 |
+
"is_world_process_zero": true,
|
| 9 |
+
"log_history": [
|
| 10 |
+
{
|
| 11 |
+
"epoch": 0.0,
|
| 12 |
+
"learning_rate": 4.984353779813873e-05,
|
| 13 |
+
"loss": 2.5206,
|
| 14 |
+
"step": 500
|
| 15 |
+
},
|
| 16 |
+
{
|
| 17 |
+
"epoch": 0.0,
|
| 18 |
+
"eval_bleu": 38.3843,
|
| 19 |
+
"eval_gen_len": 23.8603,
|
| 20 |
+
"eval_loss": 1.8556830883026123,
|
| 21 |
+
"eval_runtime": 1197.8495,
|
| 22 |
+
"eval_samples_per_second": 2.515,
|
| 23 |
+
"eval_steps_per_second": 0.629,
|
| 24 |
+
"step": 500
|
| 25 |
+
},
|
| 26 |
+
{
|
| 27 |
+
"epoch": 0.01,
|
| 28 |
+
"learning_rate": 4.9687075596277455e-05,
|
| 29 |
+
"loss": 1.9572,
|
| 30 |
+
"step": 1000
|
| 31 |
+
},
|
| 32 |
+
{
|
| 33 |
+
"epoch": 0.01,
|
| 34 |
+
"eval_bleu": 22.3733,
|
| 35 |
+
"eval_gen_len": 18.9091,
|
| 36 |
+
"eval_loss": 2.9118094444274902,
|
| 37 |
+
"eval_runtime": 1006.4293,
|
| 38 |
+
"eval_samples_per_second": 2.994,
|
| 39 |
+
"eval_steps_per_second": 0.749,
|
| 40 |
+
"step": 1000
|
| 41 |
+
},
|
| 42 |
+
{
|
| 43 |
+
"epoch": 0.01,
|
| 44 |
+
"learning_rate": 4.953061339441618e-05,
|
| 45 |
+
"loss": 1.7653,
|
| 46 |
+
"step": 1500
|
| 47 |
+
},
|
| 48 |
+
{
|
| 49 |
+
"epoch": 0.01,
|
| 50 |
+
"eval_bleu": 38.119,
|
| 51 |
+
"eval_gen_len": 23.6728,
|
| 52 |
+
"eval_loss": 1.9158709049224854,
|
| 53 |
+
"eval_runtime": 1104.4939,
|
| 54 |
+
"eval_samples_per_second": 2.728,
|
| 55 |
+
"eval_steps_per_second": 0.683,
|
| 56 |
+
"step": 1500
|
| 57 |
+
},
|
| 58 |
+
{
|
| 59 |
+
"epoch": 0.01,
|
| 60 |
+
"learning_rate": 4.937415119255491e-05,
|
| 61 |
+
"loss": 1.715,
|
| 62 |
+
"step": 2000
|
| 63 |
+
},
|
| 64 |
+
{
|
| 65 |
+
"epoch": 0.01,
|
| 66 |
+
"eval_bleu": 37.8451,
|
| 67 |
+
"eval_gen_len": 23.7046,
|
| 68 |
+
"eval_loss": 1.9664654731750488,
|
| 69 |
+
"eval_runtime": 1119.4046,
|
| 70 |
+
"eval_samples_per_second": 2.692,
|
| 71 |
+
"eval_steps_per_second": 0.674,
|
| 72 |
+
"step": 2000
|
| 73 |
+
},
|
| 74 |
+
{
|
| 75 |
+
"epoch": 0.02,
|
| 76 |
+
"learning_rate": 4.921768899069363e-05,
|
| 77 |
+
"loss": 1.7164,
|
| 78 |
+
"step": 2500
|
| 79 |
+
},
|
| 80 |
+
{
|
| 81 |
+
"epoch": 0.02,
|
| 82 |
+
"eval_bleu": 36.315,
|
| 83 |
+
"eval_gen_len": 23.1172,
|
| 84 |
+
"eval_loss": 1.8593918085098267,
|
| 85 |
+
"eval_runtime": 1061.9031,
|
| 86 |
+
"eval_samples_per_second": 2.837,
|
| 87 |
+
"eval_steps_per_second": 0.71,
|
| 88 |
+
"step": 2500
|
| 89 |
+
},
|
| 90 |
+
{
|
| 91 |
+
"epoch": 0.02,
|
| 92 |
+
"learning_rate": 4.906122678883235e-05,
|
| 93 |
+
"loss": 1.616,
|
| 94 |
+
"step": 3000
|
| 95 |
+
},
|
| 96 |
+
{
|
| 97 |
+
"epoch": 0.02,
|
| 98 |
+
"eval_bleu": 37.6941,
|
| 99 |
+
"eval_gen_len": 23.617,
|
| 100 |
+
"eval_loss": 1.816091537475586,
|
| 101 |
+
"eval_runtime": 1096.7235,
|
| 102 |
+
"eval_samples_per_second": 2.747,
|
| 103 |
+
"eval_steps_per_second": 0.688,
|
| 104 |
+
"step": 3000
|
| 105 |
+
},
|
| 106 |
+
{
|
| 107 |
+
"epoch": 0.02,
|
| 108 |
+
"learning_rate": 4.890476458697108e-05,
|
| 109 |
+
"loss": 1.6131,
|
| 110 |
+
"step": 3500
|
| 111 |
+
},
|
| 112 |
+
{
|
| 113 |
+
"epoch": 0.02,
|
| 114 |
+
"eval_bleu": 34.7085,
|
| 115 |
+
"eval_gen_len": 22.4826,
|
| 116 |
+
"eval_loss": 1.9029065370559692,
|
| 117 |
+
"eval_runtime": 1058.2776,
|
| 118 |
+
"eval_samples_per_second": 2.847,
|
| 119 |
+
"eval_steps_per_second": 0.712,
|
| 120 |
+
"step": 3500
|
| 121 |
+
},
|
| 122 |
+
{
|
| 123 |
+
"epoch": 0.03,
|
| 124 |
+
"learning_rate": 4.8748302385109805e-05,
|
| 125 |
+
"loss": 1.5981,
|
| 126 |
+
"step": 4000
|
| 127 |
+
},
|
| 128 |
+
{
|
| 129 |
+
"epoch": 0.03,
|
| 130 |
+
"eval_bleu": 37.0045,
|
| 131 |
+
"eval_gen_len": 24.0611,
|
| 132 |
+
"eval_loss": 1.9932270050048828,
|
| 133 |
+
"eval_runtime": 1166.939,
|
| 134 |
+
"eval_samples_per_second": 2.582,
|
| 135 |
+
"eval_steps_per_second": 0.646,
|
| 136 |
+
"step": 4000
|
| 137 |
+
},
|
| 138 |
+
{
|
| 139 |
+
"epoch": 0.03,
|
| 140 |
+
"learning_rate": 4.859184018324853e-05,
|
| 141 |
+
"loss": 1.5898,
|
| 142 |
+
"step": 4500
|
| 143 |
+
},
|
| 144 |
+
{
|
| 145 |
+
"epoch": 0.03,
|
| 146 |
+
"eval_bleu": 38.2992,
|
| 147 |
+
"eval_gen_len": 24.2121,
|
| 148 |
+
"eval_loss": 1.874466896057129,
|
| 149 |
+
"eval_runtime": 1133.0065,
|
| 150 |
+
"eval_samples_per_second": 2.659,
|
| 151 |
+
"eval_steps_per_second": 0.665,
|
| 152 |
+
"step": 4500
|
| 153 |
+
},
|
| 154 |
+
{
|
| 155 |
+
"epoch": 0.03,
|
| 156 |
+
"learning_rate": 4.843537798138726e-05,
|
| 157 |
+
"loss": 1.5582,
|
| 158 |
+
"step": 5000
|
| 159 |
+
},
|
| 160 |
+
{
|
| 161 |
+
"epoch": 0.03,
|
| 162 |
+
"eval_bleu": 36.9851,
|
| 163 |
+
"eval_gen_len": 23.4202,
|
| 164 |
+
"eval_loss": 1.8227670192718506,
|
| 165 |
+
"eval_runtime": 1093.0431,
|
| 166 |
+
"eval_samples_per_second": 2.757,
|
| 167 |
+
"eval_steps_per_second": 0.69,
|
| 168 |
+
"step": 5000
|
| 169 |
+
},
|
| 170 |
+
{
|
| 171 |
+
"epoch": 0.03,
|
| 172 |
+
"learning_rate": 4.8278915779525984e-05,
|
| 173 |
+
"loss": 1.7726,
|
| 174 |
+
"step": 5500
|
| 175 |
+
},
|
| 176 |
+
{
|
| 177 |
+
"epoch": 0.03,
|
| 178 |
+
"eval_bleu": 32.3628,
|
| 179 |
+
"eval_gen_len": 22.3083,
|
| 180 |
+
"eval_loss": 2.096940755844116,
|
| 181 |
+
"eval_runtime": 1059.9352,
|
| 182 |
+
"eval_samples_per_second": 2.843,
|
| 183 |
+
"eval_steps_per_second": 0.711,
|
| 184 |
+
"step": 5500
|
| 185 |
+
},
|
| 186 |
+
{
|
| 187 |
+
"epoch": 0.04,
|
| 188 |
+
"learning_rate": 4.812245357766471e-05,
|
| 189 |
+
"loss": 1.7102,
|
| 190 |
+
"step": 6000
|
| 191 |
+
},
|
| 192 |
+
{
|
| 193 |
+
"epoch": 0.04,
|
| 194 |
+
"eval_bleu": 36.1748,
|
| 195 |
+
"eval_gen_len": 24.1942,
|
| 196 |
+
"eval_loss": 2.018200397491455,
|
| 197 |
+
"eval_runtime": 1129.5162,
|
| 198 |
+
"eval_samples_per_second": 2.668,
|
| 199 |
+
"eval_steps_per_second": 0.668,
|
| 200 |
+
"step": 6000
|
| 201 |
+
},
|
| 202 |
+
{
|
| 203 |
+
"epoch": 0.04,
|
| 204 |
+
"learning_rate": 4.7965991375803436e-05,
|
| 205 |
+
"loss": 1.521,
|
| 206 |
+
"step": 6500
|
| 207 |
+
},
|
| 208 |
+
{
|
| 209 |
+
"epoch": 0.04,
|
| 210 |
+
"eval_bleu": 32.2516,
|
| 211 |
+
"eval_gen_len": 28.4076,
|
| 212 |
+
"eval_loss": 2.057112693786621,
|
| 213 |
+
"eval_runtime": 1610.6798,
|
| 214 |
+
"eval_samples_per_second": 1.871,
|
| 215 |
+
"eval_steps_per_second": 0.468,
|
| 216 |
+
"step": 6500
|
| 217 |
+
},
|
| 218 |
+
{
|
| 219 |
+
"epoch": 0.04,
|
| 220 |
+
"learning_rate": 4.780952917394216e-05,
|
| 221 |
+
"loss": 1.6184,
|
| 222 |
+
"step": 7000
|
| 223 |
+
},
|
| 224 |
+
{
|
| 225 |
+
"epoch": 0.04,
|
| 226 |
+
"eval_bleu": 37.8637,
|
| 227 |
+
"eval_gen_len": 23.6844,
|
| 228 |
+
"eval_loss": 1.8763169050216675,
|
| 229 |
+
"eval_runtime": 1099.6674,
|
| 230 |
+
"eval_samples_per_second": 2.74,
|
| 231 |
+
"eval_steps_per_second": 0.686,
|
| 232 |
+
"step": 7000
|
| 233 |
+
},
|
| 234 |
+
{
|
| 235 |
+
"epoch": 0.05,
|
| 236 |
+
"learning_rate": 4.765306697208088e-05,
|
| 237 |
+
"loss": 1.5384,
|
| 238 |
+
"step": 7500
|
| 239 |
+
},
|
| 240 |
+
{
|
| 241 |
+
"epoch": 0.05,
|
| 242 |
+
"eval_bleu": 37.7753,
|
| 243 |
+
"eval_gen_len": 23.8984,
|
| 244 |
+
"eval_loss": 1.8554104566574097,
|
| 245 |
+
"eval_runtime": 1111.1157,
|
| 246 |
+
"eval_samples_per_second": 2.712,
|
| 247 |
+
"eval_steps_per_second": 0.679,
|
| 248 |
+
"step": 7500
|
| 249 |
+
},
|
| 250 |
+
{
|
| 251 |
+
"epoch": 0.05,
|
| 252 |
+
"learning_rate": 4.749660477021961e-05,
|
| 253 |
+
"loss": 1.5485,
|
| 254 |
+
"step": 8000
|
| 255 |
+
},
|
| 256 |
+
{
|
| 257 |
+
"epoch": 0.05,
|
| 258 |
+
"eval_bleu": 37.1929,
|
| 259 |
+
"eval_gen_len": 24.0408,
|
| 260 |
+
"eval_loss": 1.9456806182861328,
|
| 261 |
+
"eval_runtime": 1136.0111,
|
| 262 |
+
"eval_samples_per_second": 2.652,
|
| 263 |
+
"eval_steps_per_second": 0.664,
|
| 264 |
+
"step": 8000
|
| 265 |
+
},
|
| 266 |
+
{
|
| 267 |
+
"epoch": 0.05,
|
| 268 |
+
"learning_rate": 4.7340142568358334e-05,
|
| 269 |
+
"loss": 1.4872,
|
| 270 |
+
"step": 8500
|
| 271 |
+
},
|
| 272 |
+
{
|
| 273 |
+
"epoch": 0.05,
|
| 274 |
+
"eval_bleu": 35.4276,
|
| 275 |
+
"eval_gen_len": 23.2008,
|
| 276 |
+
"eval_loss": 1.8643290996551514,
|
| 277 |
+
"eval_runtime": 1078.1924,
|
| 278 |
+
"eval_samples_per_second": 2.794,
|
| 279 |
+
"eval_steps_per_second": 0.699,
|
| 280 |
+
"step": 8500
|
| 281 |
+
},
|
| 282 |
+
{
|
| 283 |
+
"epoch": 0.06,
|
| 284 |
+
"learning_rate": 4.718368036649706e-05,
|
| 285 |
+
"loss": 1.4412,
|
| 286 |
+
"step": 9000
|
| 287 |
+
},
|
| 288 |
+
{
|
| 289 |
+
"epoch": 0.06,
|
| 290 |
+
"eval_bleu": 35.7002,
|
| 291 |
+
"eval_gen_len": 23.613,
|
| 292 |
+
"eval_loss": 1.910874366760254,
|
| 293 |
+
"eval_runtime": 1134.7549,
|
| 294 |
+
"eval_samples_per_second": 2.655,
|
| 295 |
+
"eval_steps_per_second": 0.664,
|
| 296 |
+
"step": 9000
|
| 297 |
+
},
|
| 298 |
+
{
|
| 299 |
+
"epoch": 0.06,
|
| 300 |
+
"learning_rate": 4.702721816463579e-05,
|
| 301 |
+
"loss": 1.4483,
|
| 302 |
+
"step": 9500
|
| 303 |
+
},
|
| 304 |
+
{
|
| 305 |
+
"epoch": 0.06,
|
| 306 |
+
"eval_bleu": 37.6828,
|
| 307 |
+
"eval_gen_len": 24.0441,
|
| 308 |
+
"eval_loss": 1.8838558197021484,
|
| 309 |
+
"eval_runtime": 1126.6767,
|
| 310 |
+
"eval_samples_per_second": 2.674,
|
| 311 |
+
"eval_steps_per_second": 0.669,
|
| 312 |
+
"step": 9500
|
| 313 |
+
},
|
| 314 |
+
{
|
| 315 |
+
"epoch": 0.06,
|
| 316 |
+
"learning_rate": 4.687075596277451e-05,
|
| 317 |
+
"loss": 1.4523,
|
| 318 |
+
"step": 10000
|
| 319 |
+
},
|
| 320 |
+
{
|
| 321 |
+
"epoch": 0.06,
|
| 322 |
+
"eval_bleu": 36.2473,
|
| 323 |
+
"eval_gen_len": 23.3561,
|
| 324 |
+
"eval_loss": 1.940147042274475,
|
| 325 |
+
"eval_runtime": 1087.7551,
|
| 326 |
+
"eval_samples_per_second": 2.77,
|
| 327 |
+
"eval_steps_per_second": 0.693,
|
| 328 |
+
"step": 10000
|
| 329 |
+
},
|
| 330 |
+
{
|
| 331 |
+
"epoch": 0.07,
|
| 332 |
+
"learning_rate": 4.671429376091324e-05,
|
| 333 |
+
"loss": 1.4405,
|
| 334 |
+
"step": 10500
|
| 335 |
+
},
|
| 336 |
+
{
|
| 337 |
+
"epoch": 0.07,
|
| 338 |
+
"eval_bleu": 35.6564,
|
| 339 |
+
"eval_gen_len": 22.9851,
|
| 340 |
+
"eval_loss": 1.8357064723968506,
|
| 341 |
+
"eval_runtime": 1082.157,
|
| 342 |
+
"eval_samples_per_second": 2.784,
|
| 343 |
+
"eval_steps_per_second": 0.697,
|
| 344 |
+
"step": 10500
|
| 345 |
+
},
|
| 346 |
+
{
|
| 347 |
+
"epoch": 0.07,
|
| 348 |
+
"learning_rate": 4.6557831559051965e-05,
|
| 349 |
+
"loss": 1.3966,
|
| 350 |
+
"step": 11000
|
| 351 |
+
},
|
| 352 |
+
{
|
| 353 |
+
"epoch": 0.07,
|
| 354 |
+
"eval_bleu": 37.8601,
|
| 355 |
+
"eval_gen_len": 24.1235,
|
| 356 |
+
"eval_loss": 1.8839383125305176,
|
| 357 |
+
"eval_runtime": 1137.6484,
|
| 358 |
+
"eval_samples_per_second": 2.648,
|
| 359 |
+
"eval_steps_per_second": 0.663,
|
| 360 |
+
"step": 11000
|
| 361 |
+
},
|
| 362 |
+
{
|
| 363 |
+
"epoch": 0.07,
|
| 364 |
+
"learning_rate": 4.640136935719069e-05,
|
| 365 |
+
"loss": 1.415,
|
| 366 |
+
"step": 11500
|
| 367 |
+
},
|
| 368 |
+
{
|
| 369 |
+
"epoch": 0.07,
|
| 370 |
+
"eval_bleu": 35.9576,
|
| 371 |
+
"eval_gen_len": 23.4434,
|
| 372 |
+
"eval_loss": 1.868781566619873,
|
| 373 |
+
"eval_runtime": 1187.6496,
|
| 374 |
+
"eval_samples_per_second": 2.537,
|
| 375 |
+
"eval_steps_per_second": 0.635,
|
| 376 |
+
"step": 11500
|
| 377 |
+
},
|
| 378 |
+
{
|
| 379 |
+
"epoch": 0.08,
|
| 380 |
+
"learning_rate": 4.624490715532942e-05,
|
| 381 |
+
"loss": 1.4306,
|
| 382 |
+
"step": 12000
|
| 383 |
+
},
|
| 384 |
+
{
|
| 385 |
+
"epoch": 0.08,
|
| 386 |
+
"eval_bleu": 37.8482,
|
| 387 |
+
"eval_gen_len": 24.2154,
|
| 388 |
+
"eval_loss": 1.8083645105361938,
|
| 389 |
+
"eval_runtime": 1247.7304,
|
| 390 |
+
"eval_samples_per_second": 2.415,
|
| 391 |
+
"eval_steps_per_second": 0.604,
|
| 392 |
+
"step": 12000
|
| 393 |
+
},
|
| 394 |
+
{
|
| 395 |
+
"epoch": 0.08,
|
| 396 |
+
"learning_rate": 4.6088444953468144e-05,
|
| 397 |
+
"loss": 1.3337,
|
| 398 |
+
"step": 12500
|
| 399 |
+
},
|
| 400 |
+
{
|
| 401 |
+
"epoch": 0.08,
|
| 402 |
+
"eval_bleu": 37.4047,
|
| 403 |
+
"eval_gen_len": 23.692,
|
| 404 |
+
"eval_loss": 1.798228144645691,
|
| 405 |
+
"eval_runtime": 1200.6209,
|
| 406 |
+
"eval_samples_per_second": 2.51,
|
| 407 |
+
"eval_steps_per_second": 0.628,
|
| 408 |
+
"step": 12500
|
| 409 |
+
},
|
| 410 |
+
{
|
| 411 |
+
"epoch": 0.08,
|
| 412 |
+
"learning_rate": 4.593198275160687e-05,
|
| 413 |
+
"loss": 1.4188,
|
| 414 |
+
"step": 13000
|
| 415 |
+
},
|
| 416 |
+
{
|
| 417 |
+
"epoch": 0.08,
|
| 418 |
+
"eval_bleu": 36.76,
|
| 419 |
+
"eval_gen_len": 23.5446,
|
| 420 |
+
"eval_loss": 1.8160576820373535,
|
| 421 |
+
"eval_runtime": 1178.8918,
|
| 422 |
+
"eval_samples_per_second": 2.556,
|
| 423 |
+
"eval_steps_per_second": 0.64,
|
| 424 |
+
"step": 13000
|
| 425 |
+
},
|
| 426 |
+
{
|
| 427 |
+
"epoch": 0.08,
|
| 428 |
+
"learning_rate": 4.5775520549745596e-05,
|
| 429 |
+
"loss": 1.3576,
|
| 430 |
+
"step": 13500
|
| 431 |
+
},
|
| 432 |
+
{
|
| 433 |
+
"epoch": 0.08,
|
| 434 |
+
"eval_bleu": 37.1149,
|
| 435 |
+
"eval_gen_len": 23.6963,
|
| 436 |
+
"eval_loss": 1.8919538259506226,
|
| 437 |
+
"eval_runtime": 1218.0527,
|
| 438 |
+
"eval_samples_per_second": 2.474,
|
| 439 |
+
"eval_steps_per_second": 0.619,
|
| 440 |
+
"step": 13500
|
| 441 |
+
},
|
| 442 |
+
{
|
| 443 |
+
"epoch": 0.09,
|
| 444 |
+
"learning_rate": 4.561905834788432e-05,
|
| 445 |
+
"loss": 1.4464,
|
| 446 |
+
"step": 14000
|
| 447 |
+
},
|
| 448 |
+
{
|
| 449 |
+
"epoch": 0.09,
|
| 450 |
+
"eval_bleu": 38.2523,
|
| 451 |
+
"eval_gen_len": 24.2449,
|
| 452 |
+
"eval_loss": 1.8881698846817017,
|
| 453 |
+
"eval_runtime": 1185.5541,
|
| 454 |
+
"eval_samples_per_second": 2.541,
|
| 455 |
+
"eval_steps_per_second": 0.636,
|
| 456 |
+
"step": 14000
|
| 457 |
+
},
|
| 458 |
+
{
|
| 459 |
+
"epoch": 0.09,
|
| 460 |
+
"learning_rate": 4.546259614602305e-05,
|
| 461 |
+
"loss": 1.3548,
|
| 462 |
+
"step": 14500
|
| 463 |
+
},
|
| 464 |
+
{
|
| 465 |
+
"epoch": 0.09,
|
| 466 |
+
"eval_bleu": 37.0608,
|
| 467 |
+
"eval_gen_len": 23.6077,
|
| 468 |
+
"eval_loss": 1.8448089361190796,
|
| 469 |
+
"eval_runtime": 1093.1892,
|
| 470 |
+
"eval_samples_per_second": 2.756,
|
| 471 |
+
"eval_steps_per_second": 0.69,
|
| 472 |
+
"step": 14500
|
| 473 |
+
},
|
| 474 |
+
{
|
| 475 |
+
"epoch": 0.09,
|
| 476 |
+
"learning_rate": 4.5306133944161775e-05,
|
| 477 |
+
"loss": 1.3779,
|
| 478 |
+
"step": 15000
|
| 479 |
+
},
|
| 480 |
+
{
|
| 481 |
+
"epoch": 0.09,
|
| 482 |
+
"eval_bleu": 37.2558,
|
| 483 |
+
"eval_gen_len": 23.5888,
|
| 484 |
+
"eval_loss": 1.868304967880249,
|
| 485 |
+
"eval_runtime": 1104.5613,
|
| 486 |
+
"eval_samples_per_second": 2.728,
|
| 487 |
+
"eval_steps_per_second": 0.683,
|
| 488 |
+
"step": 15000
|
| 489 |
+
},
|
| 490 |
+
{
|
| 491 |
+
"epoch": 0.1,
|
| 492 |
+
"learning_rate": 4.51496717423005e-05,
|
| 493 |
+
"loss": 1.3565,
|
| 494 |
+
"step": 15500
|
| 495 |
+
},
|
| 496 |
+
{
|
| 497 |
+
"epoch": 0.1,
|
| 498 |
+
"eval_bleu": 38.3579,
|
| 499 |
+
"eval_gen_len": 24.3107,
|
| 500 |
+
"eval_loss": 1.9323915243148804,
|
| 501 |
+
"eval_runtime": 1143.8823,
|
| 502 |
+
"eval_samples_per_second": 2.634,
|
| 503 |
+
"eval_steps_per_second": 0.659,
|
| 504 |
+
"step": 15500
|
| 505 |
+
},
|
| 506 |
+
{
|
| 507 |
+
"epoch": 0.1,
|
| 508 |
+
"learning_rate": 4.499320954043923e-05,
|
| 509 |
+
"loss": 1.3474,
|
| 510 |
+
"step": 16000
|
| 511 |
+
},
|
| 512 |
+
{
|
| 513 |
+
"epoch": 0.1,
|
| 514 |
+
"eval_bleu": 37.9665,
|
| 515 |
+
"eval_gen_len": 23.8453,
|
| 516 |
+
"eval_loss": 1.8365575075149536,
|
| 517 |
+
"eval_runtime": 1107.7008,
|
| 518 |
+
"eval_samples_per_second": 2.72,
|
| 519 |
+
"eval_steps_per_second": 0.681,
|
| 520 |
+
"step": 16000
|
| 521 |
+
},
|
| 522 |
+
{
|
| 523 |
+
"epoch": 0.1,
|
| 524 |
+
"learning_rate": 4.4836747338577954e-05,
|
| 525 |
+
"loss": 1.3492,
|
| 526 |
+
"step": 16500
|
| 527 |
+
},
|
| 528 |
+
{
|
| 529 |
+
"epoch": 0.1,
|
| 530 |
+
"eval_bleu": 37.6124,
|
| 531 |
+
"eval_gen_len": 23.7939,
|
| 532 |
+
"eval_loss": 1.8358746767044067,
|
| 533 |
+
"eval_runtime": 1110.7352,
|
| 534 |
+
"eval_samples_per_second": 2.713,
|
| 535 |
+
"eval_steps_per_second": 0.679,
|
| 536 |
+
"step": 16500
|
| 537 |
+
},
|
| 538 |
+
{
|
| 539 |
+
"epoch": 0.11,
|
| 540 |
+
"learning_rate": 4.468028513671668e-05,
|
| 541 |
+
"loss": 1.3347,
|
| 542 |
+
"step": 17000
|
| 543 |
+
},
|
| 544 |
+
{
|
| 545 |
+
"epoch": 0.11,
|
| 546 |
+
"eval_bleu": 37.7852,
|
| 547 |
+
"eval_gen_len": 23.8994,
|
| 548 |
+
"eval_loss": 1.7998509407043457,
|
| 549 |
+
"eval_runtime": 1113.2105,
|
| 550 |
+
"eval_samples_per_second": 2.707,
|
| 551 |
+
"eval_steps_per_second": 0.677,
|
| 552 |
+
"step": 17000
|
| 553 |
+
},
|
| 554 |
+
{
|
| 555 |
+
"epoch": 0.11,
|
| 556 |
+
"learning_rate": 4.45238229348554e-05,
|
| 557 |
+
"loss": 1.3361,
|
| 558 |
+
"step": 17500
|
| 559 |
+
},
|
| 560 |
+
{
|
| 561 |
+
"epoch": 0.11,
|
| 562 |
+
"eval_bleu": 37.3385,
|
| 563 |
+
"eval_gen_len": 23.6077,
|
| 564 |
+
"eval_loss": 1.849636197090149,
|
| 565 |
+
"eval_runtime": 1105.7834,
|
| 566 |
+
"eval_samples_per_second": 2.725,
|
| 567 |
+
"eval_steps_per_second": 0.682,
|
| 568 |
+
"step": 17500
|
| 569 |
+
},
|
| 570 |
+
{
|
| 571 |
+
"epoch": 0.11,
|
| 572 |
+
"learning_rate": 4.4367360732994125e-05,
|
| 573 |
+
"loss": 1.3204,
|
| 574 |
+
"step": 18000
|
| 575 |
+
},
|
| 576 |
+
{
|
| 577 |
+
"epoch": 0.11,
|
| 578 |
+
"eval_bleu": 36.8887,
|
| 579 |
+
"eval_gen_len": 23.385,
|
| 580 |
+
"eval_loss": 1.8504753112792969,
|
| 581 |
+
"eval_runtime": 1074.3508,
|
| 582 |
+
"eval_samples_per_second": 2.804,
|
| 583 |
+
"eval_steps_per_second": 0.702,
|
| 584 |
+
"step": 18000
|
| 585 |
+
},
|
| 586 |
+
{
|
| 587 |
+
"epoch": 0.12,
|
| 588 |
+
"learning_rate": 4.421089853113285e-05,
|
| 589 |
+
"loss": 1.3234,
|
| 590 |
+
"step": 18500
|
| 591 |
+
},
|
| 592 |
+
{
|
| 593 |
+
"epoch": 0.12,
|
| 594 |
+
"eval_bleu": 36.6707,
|
| 595 |
+
"eval_gen_len": 23.2629,
|
| 596 |
+
"eval_loss": 1.8434782028198242,
|
| 597 |
+
"eval_runtime": 1075.4498,
|
| 598 |
+
"eval_samples_per_second": 2.802,
|
| 599 |
+
"eval_steps_per_second": 0.701,
|
| 600 |
+
"step": 18500
|
| 601 |
+
},
|
| 602 |
+
{
|
| 603 |
+
"epoch": 0.12,
|
| 604 |
+
"learning_rate": 4.405443632927158e-05,
|
| 605 |
+
"loss": 1.3189,
|
| 606 |
+
"step": 19000
|
| 607 |
+
},
|
| 608 |
+
{
|
| 609 |
+
"epoch": 0.12,
|
| 610 |
+
"eval_bleu": 36.6831,
|
| 611 |
+
"eval_gen_len": 23.6163,
|
| 612 |
+
"eval_loss": 1.8203562498092651,
|
| 613 |
+
"eval_runtime": 1116.9517,
|
| 614 |
+
"eval_samples_per_second": 2.698,
|
| 615 |
+
"eval_steps_per_second": 0.675,
|
| 616 |
+
"step": 19000
|
| 617 |
+
},
|
| 618 |
+
{
|
| 619 |
+
"epoch": 0.12,
|
| 620 |
+
"learning_rate": 4.3897974127410304e-05,
|
| 621 |
+
"loss": 1.2984,
|
| 622 |
+
"step": 19500
|
| 623 |
+
},
|
| 624 |
+
{
|
| 625 |
+
"epoch": 0.12,
|
| 626 |
+
"eval_bleu": 36.9382,
|
| 627 |
+
"eval_gen_len": 23.699,
|
| 628 |
+
"eval_loss": 1.7791001796722412,
|
| 629 |
+
"eval_runtime": 1129.6793,
|
| 630 |
+
"eval_samples_per_second": 2.667,
|
| 631 |
+
"eval_steps_per_second": 0.667,
|
| 632 |
+
"step": 19500
|
| 633 |
+
},
|
| 634 |
+
{
|
| 635 |
+
"epoch": 0.13,
|
| 636 |
+
"learning_rate": 4.374151192554903e-05,
|
| 637 |
+
"loss": 1.2913,
|
| 638 |
+
"step": 20000
|
| 639 |
+
},
|
| 640 |
+
{
|
| 641 |
+
"epoch": 0.13,
|
| 642 |
+
"eval_bleu": 38.155,
|
| 643 |
+
"eval_gen_len": 24.1394,
|
| 644 |
+
"eval_loss": 1.8548424243927002,
|
| 645 |
+
"eval_runtime": 1126.2716,
|
| 646 |
+
"eval_samples_per_second": 2.675,
|
| 647 |
+
"eval_steps_per_second": 0.669,
|
| 648 |
+
"step": 20000
|
| 649 |
+
},
|
| 650 |
+
{
|
| 651 |
+
"epoch": 0.13,
|
| 652 |
+
"learning_rate": 4.3585049723687756e-05,
|
| 653 |
+
"loss": 1.3222,
|
| 654 |
+
"step": 20500
|
| 655 |
+
},
|
| 656 |
+
{
|
| 657 |
+
"epoch": 0.13,
|
| 658 |
+
"eval_bleu": 37.9634,
|
| 659 |
+
"eval_gen_len": 23.9492,
|
| 660 |
+
"eval_loss": 1.8165639638900757,
|
| 661 |
+
"eval_runtime": 1117.1297,
|
| 662 |
+
"eval_samples_per_second": 2.697,
|
| 663 |
+
"eval_steps_per_second": 0.675,
|
| 664 |
+
"step": 20500
|
| 665 |
+
},
|
| 666 |
+
{
|
| 667 |
+
"epoch": 0.13,
|
| 668 |
+
"learning_rate": 4.342858752182648e-05,
|
| 669 |
+
"loss": 1.3223,
|
| 670 |
+
"step": 21000
|
| 671 |
+
},
|
| 672 |
+
{
|
| 673 |
+
"epoch": 0.13,
|
| 674 |
+
"eval_bleu": 37.6942,
|
| 675 |
+
"eval_gen_len": 24.0,
|
| 676 |
+
"eval_loss": 1.8118115663528442,
|
| 677 |
+
"eval_runtime": 1118.3371,
|
| 678 |
+
"eval_samples_per_second": 2.694,
|
| 679 |
+
"eval_steps_per_second": 0.674,
|
| 680 |
+
"step": 21000
|
| 681 |
+
},
|
| 682 |
+
{
|
| 683 |
+
"epoch": 0.13,
|
| 684 |
+
"learning_rate": 4.327212531996521e-05,
|
| 685 |
+
"loss": 1.3134,
|
| 686 |
+
"step": 21500
|
| 687 |
+
},
|
| 688 |
+
{
|
| 689 |
+
"epoch": 0.13,
|
| 690 |
+
"eval_bleu": 37.7975,
|
| 691 |
+
"eval_gen_len": 23.9004,
|
| 692 |
+
"eval_loss": 1.8295111656188965,
|
| 693 |
+
"eval_runtime": 1106.5041,
|
| 694 |
+
"eval_samples_per_second": 2.723,
|
| 695 |
+
"eval_steps_per_second": 0.681,
|
| 696 |
+
"step": 21500
|
| 697 |
+
},
|
| 698 |
+
{
|
| 699 |
+
"epoch": 0.14,
|
| 700 |
+
"learning_rate": 4.3115663118103935e-05,
|
| 701 |
+
"loss": 1.3045,
|
| 702 |
+
"step": 22000
|
| 703 |
+
},
|
| 704 |
+
{
|
| 705 |
+
"epoch": 0.14,
|
| 706 |
+
"eval_bleu": 37.7737,
|
| 707 |
+
"eval_gen_len": 23.9373,
|
| 708 |
+
"eval_loss": 1.8213391304016113,
|
| 709 |
+
"eval_runtime": 1127.6192,
|
| 710 |
+
"eval_samples_per_second": 2.672,
|
| 711 |
+
"eval_steps_per_second": 0.669,
|
| 712 |
+
"step": 22000
|
| 713 |
+
},
|
| 714 |
+
{
|
| 715 |
+
"epoch": 0.14,
|
| 716 |
+
"learning_rate": 4.2959200916242654e-05,
|
| 717 |
+
"loss": 1.3111,
|
| 718 |
+
"step": 22500
|
| 719 |
+
},
|
| 720 |
+
{
|
| 721 |
+
"epoch": 0.14,
|
| 722 |
+
"eval_bleu": 37.8615,
|
| 723 |
+
"eval_gen_len": 23.9519,
|
| 724 |
+
"eval_loss": 1.8346147537231445,
|
| 725 |
+
"eval_runtime": 1143.7572,
|
| 726 |
+
"eval_samples_per_second": 2.634,
|
| 727 |
+
"eval_steps_per_second": 0.659,
|
| 728 |
+
"step": 22500
|
| 729 |
+
},
|
| 730 |
+
{
|
| 731 |
+
"epoch": 0.14,
|
| 732 |
+
"learning_rate": 4.280273871438138e-05,
|
| 733 |
+
"loss": 1.3346,
|
| 734 |
+
"step": 23000
|
| 735 |
+
},
|
| 736 |
+
{
|
| 737 |
+
"epoch": 0.14,
|
| 738 |
+
"eval_bleu": 38.1234,
|
| 739 |
+
"eval_gen_len": 24.1049,
|
| 740 |
+
"eval_loss": 1.7679733037948608,
|
| 741 |
+
"eval_runtime": 1113.7045,
|
| 742 |
+
"eval_samples_per_second": 2.705,
|
| 743 |
+
"eval_steps_per_second": 0.677,
|
| 744 |
+
"step": 23000
|
| 745 |
+
},
|
| 746 |
+
{
|
| 747 |
+
"epoch": 0.15,
|
| 748 |
+
"learning_rate": 4.264627651252011e-05,
|
| 749 |
+
"loss": 1.2851,
|
| 750 |
+
"step": 23500
|
| 751 |
+
},
|
| 752 |
+
{
|
| 753 |
+
"epoch": 0.15,
|
| 754 |
+
"eval_bleu": 38.1328,
|
| 755 |
+
"eval_gen_len": 24.1069,
|
| 756 |
+
"eval_loss": 1.801455020904541,
|
| 757 |
+
"eval_runtime": 1136.226,
|
| 758 |
+
"eval_samples_per_second": 2.652,
|
| 759 |
+
"eval_steps_per_second": 0.664,
|
| 760 |
+
"step": 23500
|
| 761 |
+
},
|
| 762 |
+
{
|
| 763 |
+
"epoch": 0.15,
|
| 764 |
+
"learning_rate": 4.248981431065883e-05,
|
| 765 |
+
"loss": 1.2732,
|
| 766 |
+
"step": 24000
|
| 767 |
+
},
|
| 768 |
+
{
|
| 769 |
+
"epoch": 0.15,
|
| 770 |
+
"eval_bleu": 37.8569,
|
| 771 |
+
"eval_gen_len": 23.8579,
|
| 772 |
+
"eval_loss": 1.8218289613723755,
|
| 773 |
+
"eval_runtime": 1124.2084,
|
| 774 |
+
"eval_samples_per_second": 2.68,
|
| 775 |
+
"eval_steps_per_second": 0.671,
|
| 776 |
+
"step": 24000
|
| 777 |
+
},
|
| 778 |
+
{
|
| 779 |
+
"epoch": 0.15,
|
| 780 |
+
"learning_rate": 4.233335210879756e-05,
|
| 781 |
+
"loss": 1.2666,
|
| 782 |
+
"step": 24500
|
| 783 |
+
},
|
| 784 |
+
{
|
| 785 |
+
"epoch": 0.15,
|
| 786 |
+
"eval_bleu": 37.3143,
|
| 787 |
+
"eval_gen_len": 24.2327,
|
| 788 |
+
"eval_loss": 1.8298362493515015,
|
| 789 |
+
"eval_runtime": 1169.4328,
|
| 790 |
+
"eval_samples_per_second": 2.576,
|
| 791 |
+
"eval_steps_per_second": 0.645,
|
| 792 |
+
"step": 24500
|
| 793 |
+
},
|
| 794 |
+
{
|
| 795 |
+
"epoch": 0.16,
|
| 796 |
+
"learning_rate": 4.2176889906936285e-05,
|
| 797 |
+
"loss": 1.248,
|
| 798 |
+
"step": 25000
|
| 799 |
+
},
|
| 800 |
+
{
|
| 801 |
+
"epoch": 0.16,
|
| 802 |
+
"eval_bleu": 37.4488,
|
| 803 |
+
"eval_gen_len": 24.0846,
|
| 804 |
+
"eval_loss": 1.7926667928695679,
|
| 805 |
+
"eval_runtime": 1136.492,
|
| 806 |
+
"eval_samples_per_second": 2.651,
|
| 807 |
+
"eval_steps_per_second": 0.663,
|
| 808 |
+
"step": 25000
|
| 809 |
+
},
|
| 810 |
+
{
|
| 811 |
+
"epoch": 0.16,
|
| 812 |
+
"learning_rate": 4.202042770507501e-05,
|
| 813 |
+
"loss": 1.2586,
|
| 814 |
+
"step": 25500
|
| 815 |
+
},
|
| 816 |
+
{
|
| 817 |
+
"epoch": 0.16,
|
| 818 |
+
"eval_bleu": 36.9098,
|
| 819 |
+
"eval_gen_len": 23.5061,
|
| 820 |
+
"eval_loss": 1.8259001970291138,
|
| 821 |
+
"eval_runtime": 1094.3744,
|
| 822 |
+
"eval_samples_per_second": 2.753,
|
| 823 |
+
"eval_steps_per_second": 0.689,
|
| 824 |
+
"step": 25500
|
| 825 |
+
},
|
| 826 |
+
{
|
| 827 |
+
"epoch": 0.16,
|
| 828 |
+
"learning_rate": 4.186396550321374e-05,
|
| 829 |
+
"loss": 1.2305,
|
| 830 |
+
"step": 26000
|
| 831 |
+
},
|
| 832 |
+
{
|
| 833 |
+
"epoch": 0.16,
|
| 834 |
+
"eval_bleu": 37.6963,
|
| 835 |
+
"eval_gen_len": 23.9618,
|
| 836 |
+
"eval_loss": 1.7810018062591553,
|
| 837 |
+
"eval_runtime": 1112.8793,
|
| 838 |
+
"eval_samples_per_second": 2.707,
|
| 839 |
+
"eval_steps_per_second": 0.678,
|
| 840 |
+
"step": 26000
|
| 841 |
+
},
|
| 842 |
+
{
|
| 843 |
+
"epoch": 0.17,
|
| 844 |
+
"learning_rate": 4.1707503301352464e-05,
|
| 845 |
+
"loss": 1.2435,
|
| 846 |
+
"step": 26500
|
| 847 |
+
},
|
| 848 |
+
{
|
| 849 |
+
"epoch": 0.17,
|
| 850 |
+
"eval_bleu": 38.219,
|
| 851 |
+
"eval_gen_len": 24.2695,
|
| 852 |
+
"eval_loss": 1.8444660902023315,
|
| 853 |
+
"eval_runtime": 1136.4697,
|
| 854 |
+
"eval_samples_per_second": 2.651,
|
| 855 |
+
"eval_steps_per_second": 0.663,
|
| 856 |
+
"step": 26500
|
| 857 |
+
},
|
| 858 |
+
{
|
| 859 |
+
"epoch": 0.17,
|
| 860 |
+
"learning_rate": 4.155104109949119e-05,
|
| 861 |
+
"loss": 1.2681,
|
| 862 |
+
"step": 27000
|
| 863 |
+
},
|
| 864 |
+
{
|
| 865 |
+
"epoch": 0.17,
|
| 866 |
+
"eval_bleu": 38.0339,
|
| 867 |
+
"eval_gen_len": 23.8065,
|
| 868 |
+
"eval_loss": 1.8042678833007812,
|
| 869 |
+
"eval_runtime": 1110.1648,
|
| 870 |
+
"eval_samples_per_second": 2.714,
|
| 871 |
+
"eval_steps_per_second": 0.679,
|
| 872 |
+
"step": 27000
|
| 873 |
+
},
|
| 874 |
+
{
|
| 875 |
+
"epoch": 0.17,
|
| 876 |
+
"learning_rate": 4.139457889762991e-05,
|
| 877 |
+
"loss": 1.2581,
|
| 878 |
+
"step": 27500
|
| 879 |
+
},
|
| 880 |
+
{
|
| 881 |
+
"epoch": 0.17,
|
| 882 |
+
"eval_bleu": 36.7337,
|
| 883 |
+
"eval_gen_len": 23.6279,
|
| 884 |
+
"eval_loss": 1.7898603677749634,
|
| 885 |
+
"eval_runtime": 1101.1993,
|
| 886 |
+
"eval_samples_per_second": 2.736,
|
| 887 |
+
"eval_steps_per_second": 0.685,
|
| 888 |
+
"step": 27500
|
| 889 |
+
},
|
| 890 |
+
{
|
| 891 |
+
"epoch": 0.18,
|
| 892 |
+
"learning_rate": 4.1238116695768636e-05,
|
| 893 |
+
"loss": 1.2476,
|
| 894 |
+
"step": 28000
|
| 895 |
+
},
|
| 896 |
+
{
|
| 897 |
+
"epoch": 0.18,
|
| 898 |
+
"eval_bleu": 37.5418,
|
| 899 |
+
"eval_gen_len": 23.7527,
|
| 900 |
+
"eval_loss": 1.851403832435608,
|
| 901 |
+
"eval_runtime": 1100.3622,
|
| 902 |
+
"eval_samples_per_second": 2.738,
|
| 903 |
+
"eval_steps_per_second": 0.685,
|
| 904 |
+
"step": 28000
|
| 905 |
+
},
|
| 906 |
+
{
|
| 907 |
+
"epoch": 0.18,
|
| 908 |
+
"learning_rate": 4.108165449390736e-05,
|
| 909 |
+
"loss": 1.2778,
|
| 910 |
+
"step": 28500
|
| 911 |
+
},
|
| 912 |
+
{
|
| 913 |
+
"epoch": 0.18,
|
| 914 |
+
"eval_bleu": 36.9893,
|
| 915 |
+
"eval_gen_len": 23.8487,
|
| 916 |
+
"eval_loss": 1.7635945081710815,
|
| 917 |
+
"eval_runtime": 1113.9066,
|
| 918 |
+
"eval_samples_per_second": 2.705,
|
| 919 |
+
"eval_steps_per_second": 0.677,
|
| 920 |
+
"step": 28500
|
| 921 |
+
},
|
| 922 |
+
{
|
| 923 |
+
"epoch": 0.18,
|
| 924 |
+
"learning_rate": 4.092519229204609e-05,
|
| 925 |
+
"loss": 1.2335,
|
| 926 |
+
"step": 29000
|
| 927 |
+
},
|
| 928 |
+
{
|
| 929 |
+
"epoch": 0.18,
|
| 930 |
+
"eval_bleu": 38.4148,
|
| 931 |
+
"eval_gen_len": 24.0743,
|
| 932 |
+
"eval_loss": 1.7612364292144775,
|
| 933 |
+
"eval_runtime": 1118.162,
|
| 934 |
+
"eval_samples_per_second": 2.695,
|
| 935 |
+
"eval_steps_per_second": 0.674,
|
| 936 |
+
"step": 29000
|
| 937 |
+
},
|
| 938 |
+
{
|
| 939 |
+
"epoch": 0.18,
|
| 940 |
+
"learning_rate": 4.0768730090184814e-05,
|
| 941 |
+
"loss": 1.2229,
|
| 942 |
+
"step": 29500
|
| 943 |
+
},
|
| 944 |
+
{
|
| 945 |
+
"epoch": 0.18,
|
| 946 |
+
"eval_bleu": 38.387,
|
| 947 |
+
"eval_gen_len": 24.1324,
|
| 948 |
+
"eval_loss": 1.724638819694519,
|
| 949 |
+
"eval_runtime": 1133.5566,
|
| 950 |
+
"eval_samples_per_second": 2.658,
|
| 951 |
+
"eval_steps_per_second": 0.665,
|
| 952 |
+
"step": 29500
|
| 953 |
+
},
|
| 954 |
+
{
|
| 955 |
+
"epoch": 0.19,
|
| 956 |
+
"learning_rate": 4.061226788832354e-05,
|
| 957 |
+
"loss": 1.198,
|
| 958 |
+
"step": 30000
|
| 959 |
+
},
|
| 960 |
+
{
|
| 961 |
+
"epoch": 0.19,
|
| 962 |
+
"eval_bleu": 38.2916,
|
| 963 |
+
"eval_gen_len": 23.918,
|
| 964 |
+
"eval_loss": 1.7657166719436646,
|
| 965 |
+
"eval_runtime": 1258.8163,
|
| 966 |
+
"eval_samples_per_second": 2.394,
|
| 967 |
+
"eval_steps_per_second": 0.599,
|
| 968 |
+
"step": 30000
|
| 969 |
+
},
|
| 970 |
+
{
|
| 971 |
+
"epoch": 0.19,
|
| 972 |
+
"learning_rate": 4.045580568646227e-05,
|
| 973 |
+
"loss": 1.281,
|
| 974 |
+
"step": 30500
|
| 975 |
+
},
|
| 976 |
+
{
|
| 977 |
+
"epoch": 0.19,
|
| 978 |
+
"eval_bleu": 36.5976,
|
| 979 |
+
"eval_gen_len": 23.24,
|
| 980 |
+
"eval_loss": 1.758962631225586,
|
| 981 |
+
"eval_runtime": 1321.212,
|
| 982 |
+
"eval_samples_per_second": 2.28,
|
| 983 |
+
"eval_steps_per_second": 0.571,
|
| 984 |
+
"step": 30500
|
| 985 |
+
},
|
| 986 |
+
{
|
| 987 |
+
"epoch": 0.19,
|
| 988 |
+
"learning_rate": 4.029934348460099e-05,
|
| 989 |
+
"loss": 1.231,
|
| 990 |
+
"step": 31000
|
| 991 |
+
},
|
| 992 |
+
{
|
| 993 |
+
"epoch": 0.19,
|
| 994 |
+
"eval_bleu": 37.8396,
|
| 995 |
+
"eval_gen_len": 23.7385,
|
| 996 |
+
"eval_loss": 1.733566164970398,
|
| 997 |
+
"eval_runtime": 1419.0837,
|
| 998 |
+
"eval_samples_per_second": 2.123,
|
| 999 |
+
"eval_steps_per_second": 0.531,
|
| 1000 |
+
"step": 31000
|
| 1001 |
+
},
|
| 1002 |
+
{
|
| 1003 |
+
"epoch": 0.2,
|
| 1004 |
+
"learning_rate": 4.014288128273972e-05,
|
| 1005 |
+
"loss": 1.2014,
|
| 1006 |
+
"step": 31500
|
| 1007 |
+
},
|
| 1008 |
+
{
|
| 1009 |
+
"epoch": 0.2,
|
| 1010 |
+
"eval_bleu": 37.8945,
|
| 1011 |
+
"eval_gen_len": 23.8025,
|
| 1012 |
+
"eval_loss": 1.7839374542236328,
|
| 1013 |
+
"eval_runtime": 1450.5502,
|
| 1014 |
+
"eval_samples_per_second": 2.077,
|
| 1015 |
+
"eval_steps_per_second": 0.52,
|
| 1016 |
+
"step": 31500
|
| 1017 |
+
},
|
| 1018 |
+
{
|
| 1019 |
+
"epoch": 0.2,
|
| 1020 |
+
"learning_rate": 3.9986419080878445e-05,
|
| 1021 |
+
"loss": 1.2295,
|
| 1022 |
+
"step": 32000
|
| 1023 |
+
},
|
| 1024 |
+
{
|
| 1025 |
+
"epoch": 0.2,
|
| 1026 |
+
"eval_bleu": 37.9001,
|
| 1027 |
+
"eval_gen_len": 23.92,
|
| 1028 |
+
"eval_loss": 1.757741928100586,
|
| 1029 |
+
"eval_runtime": 1476.1175,
|
| 1030 |
+
"eval_samples_per_second": 2.041,
|
| 1031 |
+
"eval_steps_per_second": 0.511,
|
| 1032 |
+
"step": 32000
|
| 1033 |
+
},
|
| 1034 |
+
{
|
| 1035 |
+
"epoch": 0.2,
|
| 1036 |
+
"learning_rate": 3.9829956879017165e-05,
|
| 1037 |
+
"loss": 1.2154,
|
| 1038 |
+
"step": 32500
|
| 1039 |
+
},
|
| 1040 |
+
{
|
| 1041 |
+
"epoch": 0.2,
|
| 1042 |
+
"eval_bleu": 37.5283,
|
| 1043 |
+
"eval_gen_len": 23.537,
|
| 1044 |
+
"eval_loss": 1.80224609375,
|
| 1045 |
+
"eval_runtime": 1473.4024,
|
| 1046 |
+
"eval_samples_per_second": 2.045,
|
| 1047 |
+
"eval_steps_per_second": 0.512,
|
| 1048 |
+
"step": 32500
|
| 1049 |
+
},
|
| 1050 |
+
{
|
| 1051 |
+
"epoch": 0.21,
|
| 1052 |
+
"learning_rate": 3.967349467715589e-05,
|
| 1053 |
+
"loss": 1.1956,
|
| 1054 |
+
"step": 33000
|
| 1055 |
+
},
|
| 1056 |
+
{
|
| 1057 |
+
"epoch": 0.21,
|
| 1058 |
+
"eval_bleu": 38.5709,
|
| 1059 |
+
"eval_gen_len": 24.0189,
|
| 1060 |
+
"eval_loss": 1.7815015316009521,
|
| 1061 |
+
"eval_runtime": 1398.0131,
|
| 1062 |
+
"eval_samples_per_second": 2.155,
|
| 1063 |
+
"eval_steps_per_second": 0.539,
|
| 1064 |
+
"step": 33000
|
| 1065 |
+
},
|
| 1066 |
+
{
|
| 1067 |
+
"epoch": 0.21,
|
| 1068 |
+
"learning_rate": 3.951703247529462e-05,
|
| 1069 |
+
"loss": 1.2045,
|
| 1070 |
+
"step": 33500
|
| 1071 |
+
},
|
| 1072 |
+
{
|
| 1073 |
+
"epoch": 0.21,
|
| 1074 |
+
"eval_bleu": 37.2706,
|
| 1075 |
+
"eval_gen_len": 23.2522,
|
| 1076 |
+
"eval_loss": 1.7592459917068481,
|
| 1077 |
+
"eval_runtime": 1445.8551,
|
| 1078 |
+
"eval_samples_per_second": 2.084,
|
| 1079 |
+
"eval_steps_per_second": 0.521,
|
| 1080 |
+
"step": 33500
|
| 1081 |
+
},
|
| 1082 |
+
{
|
| 1083 |
+
"epoch": 0.21,
|
| 1084 |
+
"learning_rate": 3.9360570273433344e-05,
|
| 1085 |
+
"loss": 1.1796,
|
| 1086 |
+
"step": 34000
|
| 1087 |
+
},
|
| 1088 |
+
{
|
| 1089 |
+
"epoch": 0.21,
|
| 1090 |
+
"eval_bleu": 38.5198,
|
| 1091 |
+
"eval_gen_len": 24.078,
|
| 1092 |
+
"eval_loss": 1.8075897693634033,
|
| 1093 |
+
"eval_runtime": 1528.5032,
|
| 1094 |
+
"eval_samples_per_second": 1.971,
|
| 1095 |
+
"eval_steps_per_second": 0.493,
|
| 1096 |
+
"step": 34000
|
| 1097 |
+
},
|
| 1098 |
+
{
|
| 1099 |
+
"epoch": 0.22,
|
| 1100 |
+
"learning_rate": 3.920410807157207e-05,
|
| 1101 |
+
"loss": 1.226,
|
| 1102 |
+
"step": 34500
|
| 1103 |
+
},
|
| 1104 |
+
{
|
| 1105 |
+
"epoch": 0.22,
|
| 1106 |
+
"eval_bleu": 38.1729,
|
| 1107 |
+
"eval_gen_len": 23.9038,
|
| 1108 |
+
"eval_loss": 1.731508493423462,
|
| 1109 |
+
"eval_runtime": 1501.9126,
|
| 1110 |
+
"eval_samples_per_second": 2.006,
|
| 1111 |
+
"eval_steps_per_second": 0.502,
|
| 1112 |
+
"step": 34500
|
| 1113 |
+
},
|
| 1114 |
+
{
|
| 1115 |
+
"epoch": 0.22,
|
| 1116 |
+
"learning_rate": 3.9047645869710796e-05,
|
| 1117 |
+
"loss": 1.2304,
|
| 1118 |
+
"step": 35000
|
| 1119 |
+
},
|
| 1120 |
+
{
|
| 1121 |
+
"epoch": 0.22,
|
| 1122 |
+
"eval_bleu": 37.8875,
|
| 1123 |
+
"eval_gen_len": 23.8493,
|
| 1124 |
+
"eval_loss": 1.7690067291259766,
|
| 1125 |
+
"eval_runtime": 1584.8349,
|
| 1126 |
+
"eval_samples_per_second": 1.901,
|
| 1127 |
+
"eval_steps_per_second": 0.476,
|
| 1128 |
+
"step": 35000
|
| 1129 |
+
},
|
| 1130 |
+
{
|
| 1131 |
+
"epoch": 0.22,
|
| 1132 |
+
"learning_rate": 3.889118366784952e-05,
|
| 1133 |
+
"loss": 1.197,
|
| 1134 |
+
"step": 35500
|
| 1135 |
+
},
|
| 1136 |
+
{
|
| 1137 |
+
"epoch": 0.22,
|
| 1138 |
+
"eval_bleu": 37.7553,
|
| 1139 |
+
"eval_gen_len": 23.9353,
|
| 1140 |
+
"eval_loss": 1.7940903902053833,
|
| 1141 |
+
"eval_runtime": 1636.4218,
|
| 1142 |
+
"eval_samples_per_second": 1.841,
|
| 1143 |
+
"eval_steps_per_second": 0.461,
|
| 1144 |
+
"step": 35500
|
| 1145 |
+
},
|
| 1146 |
+
{
|
| 1147 |
+
"epoch": 0.23,
|
| 1148 |
+
"learning_rate": 3.873472146598825e-05,
|
| 1149 |
+
"loss": 1.1844,
|
| 1150 |
+
"step": 36000
|
| 1151 |
+
},
|
| 1152 |
+
{
|
| 1153 |
+
"epoch": 0.23,
|
| 1154 |
+
"eval_bleu": 38.3847,
|
| 1155 |
+
"eval_gen_len": 23.9668,
|
| 1156 |
+
"eval_loss": 1.7370097637176514,
|
| 1157 |
+
"eval_runtime": 1586.621,
|
| 1158 |
+
"eval_samples_per_second": 1.899,
|
| 1159 |
+
"eval_steps_per_second": 0.475,
|
| 1160 |
+
"step": 36000
|
| 1161 |
+
},
|
| 1162 |
+
{
|
| 1163 |
+
"epoch": 0.23,
|
| 1164 |
+
"learning_rate": 3.8578259264126975e-05,
|
| 1165 |
+
"loss": 1.2099,
|
| 1166 |
+
"step": 36500
|
| 1167 |
+
},
|
| 1168 |
+
{
|
| 1169 |
+
"epoch": 0.23,
|
| 1170 |
+
"eval_bleu": 38.4795,
|
| 1171 |
+
"eval_gen_len": 23.932,
|
| 1172 |
+
"eval_loss": 1.7586994171142578,
|
| 1173 |
+
"eval_runtime": 1530.6865,
|
| 1174 |
+
"eval_samples_per_second": 1.968,
|
| 1175 |
+
"eval_steps_per_second": 0.493,
|
| 1176 |
+
"step": 36500
|
| 1177 |
+
},
|
| 1178 |
+
{
|
| 1179 |
+
"epoch": 0.23,
|
| 1180 |
+
"learning_rate": 3.84217970622657e-05,
|
| 1181 |
+
"loss": 1.1798,
|
| 1182 |
+
"step": 37000
|
| 1183 |
+
},
|
| 1184 |
+
{
|
| 1185 |
+
"epoch": 0.23,
|
| 1186 |
+
"eval_bleu": 37.463,
|
| 1187 |
+
"eval_gen_len": 23.6664,
|
| 1188 |
+
"eval_loss": 1.752753496170044,
|
| 1189 |
+
"eval_runtime": 1408.4249,
|
| 1190 |
+
"eval_samples_per_second": 2.139,
|
| 1191 |
+
"eval_steps_per_second": 0.535,
|
| 1192 |
+
"step": 37000
|
| 1193 |
+
},
|
| 1194 |
+
{
|
| 1195 |
+
"epoch": 0.23,
|
| 1196 |
+
"learning_rate": 3.826533486040442e-05,
|
| 1197 |
+
"loss": 1.1959,
|
| 1198 |
+
"step": 37500
|
| 1199 |
+
},
|
| 1200 |
+
{
|
| 1201 |
+
"epoch": 0.23,
|
| 1202 |
+
"eval_bleu": 38.1287,
|
| 1203 |
+
"eval_gen_len": 23.7693,
|
| 1204 |
+
"eval_loss": 1.7170414924621582,
|
| 1205 |
+
"eval_runtime": 1482.0293,
|
| 1206 |
+
"eval_samples_per_second": 2.033,
|
| 1207 |
+
"eval_steps_per_second": 0.509,
|
| 1208 |
+
"step": 37500
|
| 1209 |
+
},
|
| 1210 |
+
{
|
| 1211 |
+
"epoch": 0.24,
|
| 1212 |
+
"learning_rate": 3.8108872658543146e-05,
|
| 1213 |
+
"loss": 1.2061,
|
| 1214 |
+
"step": 38000
|
| 1215 |
+
},
|
| 1216 |
+
{
|
| 1217 |
+
"epoch": 0.24,
|
| 1218 |
+
"eval_bleu": 38.8039,
|
| 1219 |
+
"eval_gen_len": 23.9973,
|
| 1220 |
+
"eval_loss": 1.731540322303772,
|
| 1221 |
+
"eval_runtime": 1484.0478,
|
| 1222 |
+
"eval_samples_per_second": 2.03,
|
| 1223 |
+
"eval_steps_per_second": 0.508,
|
| 1224 |
+
"step": 38000
|
| 1225 |
+
},
|
| 1226 |
+
{
|
| 1227 |
+
"epoch": 0.24,
|
| 1228 |
+
"learning_rate": 3.795241045668187e-05,
|
| 1229 |
+
"loss": 1.1606,
|
| 1230 |
+
"step": 38500
|
| 1231 |
+
},
|
| 1232 |
+
{
|
| 1233 |
+
"epoch": 0.24,
|
| 1234 |
+
"eval_bleu": 37.93,
|
| 1235 |
+
"eval_gen_len": 23.8347,
|
| 1236 |
+
"eval_loss": 1.7204748392105103,
|
| 1237 |
+
"eval_runtime": 1527.1349,
|
| 1238 |
+
"eval_samples_per_second": 1.973,
|
| 1239 |
+
"eval_steps_per_second": 0.494,
|
| 1240 |
+
"step": 38500
|
| 1241 |
+
},
|
| 1242 |
+
{
|
| 1243 |
+
"epoch": 0.24,
|
| 1244 |
+
"learning_rate": 3.77959482548206e-05,
|
| 1245 |
+
"loss": 1.1605,
|
| 1246 |
+
"step": 39000
|
| 1247 |
+
},
|
| 1248 |
+
{
|
| 1249 |
+
"epoch": 0.24,
|
| 1250 |
+
"eval_bleu": 38.2637,
|
| 1251 |
+
"eval_gen_len": 24.2688,
|
| 1252 |
+
"eval_loss": 1.7610867023468018,
|
| 1253 |
+
"eval_runtime": 1550.9436,
|
| 1254 |
+
"eval_samples_per_second": 1.943,
|
| 1255 |
+
"eval_steps_per_second": 0.486,
|
| 1256 |
+
"step": 39000
|
| 1257 |
+
},
|
| 1258 |
+
{
|
| 1259 |
+
"epoch": 0.25,
|
| 1260 |
+
"learning_rate": 3.7639486052959325e-05,
|
| 1261 |
+
"loss": 1.1792,
|
| 1262 |
+
"step": 39500
|
| 1263 |
+
},
|
| 1264 |
+
{
|
| 1265 |
+
"epoch": 0.25,
|
| 1266 |
+
"eval_bleu": 38.3329,
|
| 1267 |
+
"eval_gen_len": 24.088,
|
| 1268 |
+
"eval_loss": 1.728115200996399,
|
| 1269 |
+
"eval_runtime": 1598.5717,
|
| 1270 |
+
"eval_samples_per_second": 1.885,
|
| 1271 |
+
"eval_steps_per_second": 0.472,
|
| 1272 |
+
"step": 39500
|
| 1273 |
+
},
|
| 1274 |
+
{
|
| 1275 |
+
"epoch": 0.25,
|
| 1276 |
+
"learning_rate": 3.748302385109805e-05,
|
| 1277 |
+
"loss": 1.1613,
|
| 1278 |
+
"step": 40000
|
| 1279 |
+
},
|
| 1280 |
+
{
|
| 1281 |
+
"epoch": 0.25,
|
| 1282 |
+
"eval_bleu": 38.7357,
|
| 1283 |
+
"eval_gen_len": 24.1019,
|
| 1284 |
+
"eval_loss": 1.7274950742721558,
|
| 1285 |
+
"eval_runtime": 1552.6289,
|
| 1286 |
+
"eval_samples_per_second": 1.941,
|
| 1287 |
+
"eval_steps_per_second": 0.486,
|
| 1288 |
+
"step": 40000
|
| 1289 |
+
}
|
| 1290 |
+
],
|
| 1291 |
+
"max_steps": 159783,
|
| 1292 |
+
"num_train_epochs": 1,
|
| 1293 |
+
"total_flos": 6683673699483648.0,
|
| 1294 |
+
"trial_name": null,
|
| 1295 |
+
"trial_params": null
|
| 1296 |
+
}
|