Commit
·
69b978b
1
Parent(s):
d448748
End of training
Browse files- all_results.json +10 -10
- eval_results.json +6 -6
- train_results.json +5 -5
- trainer_state.json +956 -20
all_results.json
CHANGED
|
@@ -1,14 +1,14 @@
|
|
| 1 |
{
|
| 2 |
-
"epoch":
|
| 3 |
-
"eval_loss": 1.
|
| 4 |
-
"eval_runtime":
|
| 5 |
"eval_samples": 10388,
|
| 6 |
-
"eval_samples_per_second":
|
| 7 |
-
"eval_steps_per_second":
|
| 8 |
-
"eval_wer": 0.
|
| 9 |
-
"train_loss":
|
| 10 |
-
"train_runtime":
|
| 11 |
"train_samples": 38209,
|
| 12 |
-
"train_samples_per_second":
|
| 13 |
-
"train_steps_per_second": 0.
|
| 14 |
}
|
|
|
|
| 1 |
{
|
| 2 |
+
"epoch": 30.0,
|
| 3 |
+
"eval_loss": 1.137310266494751,
|
| 4 |
+
"eval_runtime": 417.5245,
|
| 5 |
"eval_samples": 10388,
|
| 6 |
+
"eval_samples_per_second": 24.88,
|
| 7 |
+
"eval_steps_per_second": 3.111,
|
| 8 |
+
"eval_wer": 0.860665593725142,
|
| 9 |
+
"train_loss": 1.4937853462266097,
|
| 10 |
+
"train_runtime": 52137.1608,
|
| 11 |
"train_samples": 38209,
|
| 12 |
+
"train_samples_per_second": 21.986,
|
| 13 |
+
"train_steps_per_second": 0.344
|
| 14 |
}
|
eval_results.json
CHANGED
|
@@ -1,9 +1,9 @@
|
|
| 1 |
{
|
| 2 |
-
"epoch":
|
| 3 |
-
"eval_loss": 1.
|
| 4 |
-
"eval_runtime":
|
| 5 |
"eval_samples": 10388,
|
| 6 |
-
"eval_samples_per_second":
|
| 7 |
-
"eval_steps_per_second":
|
| 8 |
-
"eval_wer": 0.
|
| 9 |
}
|
|
|
|
| 1 |
{
|
| 2 |
+
"epoch": 30.0,
|
| 3 |
+
"eval_loss": 1.137310266494751,
|
| 4 |
+
"eval_runtime": 417.5245,
|
| 5 |
"eval_samples": 10388,
|
| 6 |
+
"eval_samples_per_second": 24.88,
|
| 7 |
+
"eval_steps_per_second": 3.111,
|
| 8 |
+
"eval_wer": 0.860665593725142
|
| 9 |
}
|
train_results.json
CHANGED
|
@@ -1,8 +1,8 @@
|
|
| 1 |
{
|
| 2 |
-
"epoch":
|
| 3 |
-
"train_loss":
|
| 4 |
-
"train_runtime":
|
| 5 |
"train_samples": 38209,
|
| 6 |
-
"train_samples_per_second":
|
| 7 |
-
"train_steps_per_second": 0.
|
| 8 |
}
|
|
|
|
| 1 |
{
|
| 2 |
+
"epoch": 30.0,
|
| 3 |
+
"train_loss": 1.4937853462266097,
|
| 4 |
+
"train_runtime": 52137.1608,
|
| 5 |
"train_samples": 38209,
|
| 6 |
+
"train_samples_per_second": 21.986,
|
| 7 |
+
"train_steps_per_second": 0.344
|
| 8 |
}
|
trainer_state.json
CHANGED
|
@@ -1,8 +1,8 @@
|
|
| 1 |
{
|
| 2 |
"best_metric": null,
|
| 3 |
"best_model_checkpoint": null,
|
| 4 |
-
"epoch":
|
| 5 |
-
"global_step":
|
| 6 |
"is_hyper_param_search": false,
|
| 7 |
"is_local_process_zero": true,
|
| 8 |
"is_world_process_zero": true,
|
|
@@ -438,41 +438,977 @@
|
|
| 438 |
},
|
| 439 |
{
|
| 440 |
"epoch": 9.38,
|
| 441 |
-
"learning_rate":
|
| 442 |
-
"loss": 2.
|
| 443 |
"step": 5600
|
| 444 |
},
|
| 445 |
{
|
| 446 |
"epoch": 9.55,
|
| 447 |
-
"learning_rate":
|
| 448 |
-
"loss": 2.
|
| 449 |
"step": 5700
|
| 450 |
},
|
| 451 |
{
|
| 452 |
"epoch": 9.71,
|
| 453 |
-
"learning_rate":
|
| 454 |
-
"loss": 2.
|
| 455 |
"step": 5800
|
| 456 |
},
|
| 457 |
{
|
| 458 |
"epoch": 9.88,
|
| 459 |
-
"learning_rate":
|
| 460 |
-
"loss": 2.
|
| 461 |
"step": 5900
|
| 462 |
},
|
| 463 |
{
|
| 464 |
-
"epoch": 10.
|
| 465 |
-
"
|
| 466 |
-
"
|
| 467 |
-
"
|
| 468 |
-
|
| 469 |
-
|
| 470 |
-
"
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 471 |
}
|
| 472 |
],
|
| 473 |
-
"max_steps":
|
| 474 |
-
"num_train_epochs":
|
| 475 |
-
"total_flos":
|
| 476 |
"trial_name": null,
|
| 477 |
"trial_params": null
|
| 478 |
}
|
|
|
|
| 1 |
{
|
| 2 |
"best_metric": null,
|
| 3 |
"best_model_checkpoint": null,
|
| 4 |
+
"epoch": 29.999581414817914,
|
| 5 |
+
"global_step": 17910,
|
| 6 |
"is_hyper_param_search": false,
|
| 7 |
"is_local_process_zero": true,
|
| 8 |
"is_world_process_zero": true,
|
|
|
|
| 438 |
},
|
| 439 |
{
|
| 440 |
"epoch": 9.38,
|
| 441 |
+
"learning_rate": 0.000773852922690132,
|
| 442 |
+
"loss": 2.3213,
|
| 443 |
"step": 5600
|
| 444 |
},
|
| 445 |
{
|
| 446 |
"epoch": 9.55,
|
| 447 |
+
"learning_rate": 0.0007675675675675676,
|
| 448 |
+
"loss": 2.3412,
|
| 449 |
"step": 5700
|
| 450 |
},
|
| 451 |
{
|
| 452 |
"epoch": 9.71,
|
| 453 |
+
"learning_rate": 0.0007612822124450032,
|
| 454 |
+
"loss": 2.3508,
|
| 455 |
"step": 5800
|
| 456 |
},
|
| 457 |
{
|
| 458 |
"epoch": 9.88,
|
| 459 |
+
"learning_rate": 0.0007549968573224387,
|
| 460 |
+
"loss": 2.3767,
|
| 461 |
"step": 5900
|
| 462 |
},
|
| 463 |
{
|
| 464 |
+
"epoch": 10.05,
|
| 465 |
+
"learning_rate": 0.0007487115021998742,
|
| 466 |
+
"loss": 2.3768,
|
| 467 |
+
"step": 6000
|
| 468 |
+
},
|
| 469 |
+
{
|
| 470 |
+
"epoch": 10.05,
|
| 471 |
+
"eval_loss": 1.6662662029266357,
|
| 472 |
+
"eval_runtime": 424.8561,
|
| 473 |
+
"eval_samples_per_second": 24.451,
|
| 474 |
+
"eval_steps_per_second": 3.058,
|
| 475 |
+
"eval_wer": 0.9664798268494367,
|
| 476 |
+
"step": 6000
|
| 477 |
+
},
|
| 478 |
+
{
|
| 479 |
+
"epoch": 10.22,
|
| 480 |
+
"learning_rate": 0.0007424261470773099,
|
| 481 |
+
"loss": 2.3681,
|
| 482 |
+
"step": 6100
|
| 483 |
+
},
|
| 484 |
+
{
|
| 485 |
+
"epoch": 10.39,
|
| 486 |
+
"learning_rate": 0.0007361407919547455,
|
| 487 |
+
"loss": 2.389,
|
| 488 |
+
"step": 6200
|
| 489 |
+
},
|
| 490 |
+
{
|
| 491 |
+
"epoch": 10.55,
|
| 492 |
+
"learning_rate": 0.0007299182903834066,
|
| 493 |
+
"loss": 2.3671,
|
| 494 |
+
"step": 6300
|
| 495 |
+
},
|
| 496 |
+
{
|
| 497 |
+
"epoch": 10.72,
|
| 498 |
+
"learning_rate": 0.0007236329352608423,
|
| 499 |
+
"loss": 2.3784,
|
| 500 |
+
"step": 6400
|
| 501 |
+
},
|
| 502 |
+
{
|
| 503 |
+
"epoch": 10.89,
|
| 504 |
+
"learning_rate": 0.0007173475801382778,
|
| 505 |
+
"loss": 2.3804,
|
| 506 |
+
"step": 6500
|
| 507 |
+
},
|
| 508 |
+
{
|
| 509 |
+
"epoch": 10.89,
|
| 510 |
+
"eval_loss": 1.6570764780044556,
|
| 511 |
+
"eval_runtime": 413.7975,
|
| 512 |
+
"eval_samples_per_second": 25.104,
|
| 513 |
+
"eval_steps_per_second": 3.139,
|
| 514 |
+
"eval_wer": 0.9719740274155059,
|
| 515 |
+
"step": 6500
|
| 516 |
+
},
|
| 517 |
+
{
|
| 518 |
+
"epoch": 11.06,
|
| 519 |
+
"learning_rate": 0.0007110622250157134,
|
| 520 |
+
"loss": 2.3757,
|
| 521 |
+
"step": 6600
|
| 522 |
+
},
|
| 523 |
+
{
|
| 524 |
+
"epoch": 11.22,
|
| 525 |
+
"learning_rate": 0.0007047768698931489,
|
| 526 |
+
"loss": 2.3487,
|
| 527 |
+
"step": 6700
|
| 528 |
+
},
|
| 529 |
+
{
|
| 530 |
+
"epoch": 11.39,
|
| 531 |
+
"learning_rate": 0.0006984915147705846,
|
| 532 |
+
"loss": 2.3473,
|
| 533 |
+
"step": 6800
|
| 534 |
+
},
|
| 535 |
+
{
|
| 536 |
+
"epoch": 11.56,
|
| 537 |
+
"learning_rate": 0.0006922061596480202,
|
| 538 |
+
"loss": 2.3481,
|
| 539 |
+
"step": 6900
|
| 540 |
+
},
|
| 541 |
+
{
|
| 542 |
+
"epoch": 11.72,
|
| 543 |
+
"learning_rate": 0.0006859208045254557,
|
| 544 |
+
"loss": 2.3237,
|
| 545 |
+
"step": 7000
|
| 546 |
+
},
|
| 547 |
+
{
|
| 548 |
+
"epoch": 11.72,
|
| 549 |
+
"eval_loss": 1.604884147644043,
|
| 550 |
+
"eval_runtime": 412.0866,
|
| 551 |
+
"eval_samples_per_second": 25.208,
|
| 552 |
+
"eval_steps_per_second": 3.152,
|
| 553 |
+
"eval_wer": 0.9637049780786947,
|
| 554 |
+
"step": 7000
|
| 555 |
+
},
|
| 556 |
+
{
|
| 557 |
+
"epoch": 11.89,
|
| 558 |
+
"learning_rate": 0.0006796354494028913,
|
| 559 |
+
"loss": 2.3379,
|
| 560 |
+
"step": 7100
|
| 561 |
+
},
|
| 562 |
+
{
|
| 563 |
+
"epoch": 12.06,
|
| 564 |
+
"learning_rate": 0.0006733500942803269,
|
| 565 |
+
"loss": 2.3362,
|
| 566 |
+
"step": 7200
|
| 567 |
+
},
|
| 568 |
+
{
|
| 569 |
+
"epoch": 12.23,
|
| 570 |
+
"learning_rate": 0.0006670647391577624,
|
| 571 |
+
"loss": 2.3148,
|
| 572 |
+
"step": 7300
|
| 573 |
+
},
|
| 574 |
+
{
|
| 575 |
+
"epoch": 12.4,
|
| 576 |
+
"learning_rate": 0.000660779384035198,
|
| 577 |
+
"loss": 2.3242,
|
| 578 |
+
"step": 7400
|
| 579 |
+
},
|
| 580 |
+
{
|
| 581 |
+
"epoch": 12.56,
|
| 582 |
+
"learning_rate": 0.0006544940289126335,
|
| 583 |
+
"loss": 2.317,
|
| 584 |
+
"step": 7500
|
| 585 |
+
},
|
| 586 |
+
{
|
| 587 |
+
"epoch": 12.56,
|
| 588 |
+
"eval_loss": 1.5874534845352173,
|
| 589 |
+
"eval_runtime": 412.3759,
|
| 590 |
+
"eval_samples_per_second": 25.191,
|
| 591 |
+
"eval_steps_per_second": 3.15,
|
| 592 |
+
"eval_wer": 0.9655363782673845,
|
| 593 |
+
"step": 7500
|
| 594 |
+
},
|
| 595 |
+
{
|
| 596 |
+
"epoch": 12.73,
|
| 597 |
+
"learning_rate": 0.0006482086737900693,
|
| 598 |
+
"loss": 2.3342,
|
| 599 |
+
"step": 7600
|
| 600 |
+
},
|
| 601 |
+
{
|
| 602 |
+
"epoch": 12.9,
|
| 603 |
+
"learning_rate": 0.0006419233186675048,
|
| 604 |
+
"loss": 2.3229,
|
| 605 |
+
"step": 7700
|
| 606 |
+
},
|
| 607 |
+
{
|
| 608 |
+
"epoch": 13.07,
|
| 609 |
+
"learning_rate": 0.0006356379635449403,
|
| 610 |
+
"loss": 2.3128,
|
| 611 |
+
"step": 7800
|
| 612 |
+
},
|
| 613 |
+
{
|
| 614 |
+
"epoch": 13.23,
|
| 615 |
+
"learning_rate": 0.0006293526084223759,
|
| 616 |
+
"loss": 2.2992,
|
| 617 |
+
"step": 7900
|
| 618 |
+
},
|
| 619 |
+
{
|
| 620 |
+
"epoch": 13.4,
|
| 621 |
+
"learning_rate": 0.0006230672532998114,
|
| 622 |
+
"loss": 2.2988,
|
| 623 |
+
"step": 8000
|
| 624 |
+
},
|
| 625 |
+
{
|
| 626 |
+
"epoch": 13.4,
|
| 627 |
+
"eval_loss": 1.5357071161270142,
|
| 628 |
+
"eval_runtime": 411.1707,
|
| 629 |
+
"eval_samples_per_second": 25.264,
|
| 630 |
+
"eval_steps_per_second": 3.159,
|
| 631 |
+
"eval_wer": 0.9603381615701945,
|
| 632 |
+
"step": 8000
|
| 633 |
+
},
|
| 634 |
+
{
|
| 635 |
+
"epoch": 13.57,
|
| 636 |
+
"learning_rate": 0.000616781898177247,
|
| 637 |
+
"loss": 2.3041,
|
| 638 |
+
"step": 8100
|
| 639 |
+
},
|
| 640 |
+
{
|
| 641 |
+
"epoch": 13.74,
|
| 642 |
+
"learning_rate": 0.0006104965430546826,
|
| 643 |
+
"loss": 2.2905,
|
| 644 |
+
"step": 8200
|
| 645 |
+
},
|
| 646 |
+
{
|
| 647 |
+
"epoch": 13.9,
|
| 648 |
+
"learning_rate": 0.0006043368950345695,
|
| 649 |
+
"loss": 2.2946,
|
| 650 |
+
"step": 8300
|
| 651 |
+
},
|
| 652 |
+
{
|
| 653 |
+
"epoch": 14.07,
|
| 654 |
+
"learning_rate": 0.000598051539912005,
|
| 655 |
+
"loss": 2.3022,
|
| 656 |
+
"step": 8400
|
| 657 |
+
},
|
| 658 |
+
{
|
| 659 |
+
"epoch": 14.24,
|
| 660 |
+
"learning_rate": 0.0005917661847894407,
|
| 661 |
+
"loss": 2.2906,
|
| 662 |
+
"step": 8500
|
| 663 |
+
},
|
| 664 |
+
{
|
| 665 |
+
"epoch": 14.24,
|
| 666 |
+
"eval_loss": 1.5637153387069702,
|
| 667 |
+
"eval_runtime": 411.7098,
|
| 668 |
+
"eval_samples_per_second": 25.231,
|
| 669 |
+
"eval_steps_per_second": 3.155,
|
| 670 |
+
"eval_wer": 0.9592097230700927,
|
| 671 |
+
"step": 8500
|
| 672 |
+
},
|
| 673 |
+
{
|
| 674 |
+
"epoch": 14.41,
|
| 675 |
+
"learning_rate": 0.0005854808296668762,
|
| 676 |
+
"loss": 2.2918,
|
| 677 |
+
"step": 8600
|
| 678 |
+
},
|
| 679 |
+
{
|
| 680 |
+
"epoch": 14.57,
|
| 681 |
+
"learning_rate": 0.0005791954745443117,
|
| 682 |
+
"loss": 2.2805,
|
| 683 |
+
"step": 8700
|
| 684 |
+
},
|
| 685 |
+
{
|
| 686 |
+
"epoch": 14.74,
|
| 687 |
+
"learning_rate": 0.0005729101194217473,
|
| 688 |
+
"loss": 2.2951,
|
| 689 |
+
"step": 8800
|
| 690 |
+
},
|
| 691 |
+
{
|
| 692 |
+
"epoch": 14.91,
|
| 693 |
+
"learning_rate": 0.0005666247642991829,
|
| 694 |
+
"loss": 2.2876,
|
| 695 |
+
"step": 8900
|
| 696 |
+
},
|
| 697 |
+
{
|
| 698 |
+
"epoch": 15.08,
|
| 699 |
+
"learning_rate": 0.0005603394091766186,
|
| 700 |
+
"loss": 2.2848,
|
| 701 |
+
"step": 9000
|
| 702 |
+
},
|
| 703 |
+
{
|
| 704 |
+
"epoch": 15.08,
|
| 705 |
+
"eval_loss": 1.5325744152069092,
|
| 706 |
+
"eval_runtime": 411.6076,
|
| 707 |
+
"eval_samples_per_second": 25.238,
|
| 708 |
+
"eval_steps_per_second": 3.156,
|
| 709 |
+
"eval_wer": 0.9537340214958285,
|
| 710 |
+
"step": 9000
|
| 711 |
+
},
|
| 712 |
+
{
|
| 713 |
+
"epoch": 15.24,
|
| 714 |
+
"learning_rate": 0.0005540540540540541,
|
| 715 |
+
"loss": 2.2537,
|
| 716 |
+
"step": 9100
|
| 717 |
+
},
|
| 718 |
+
{
|
| 719 |
+
"epoch": 15.41,
|
| 720 |
+
"learning_rate": 0.0005477686989314896,
|
| 721 |
+
"loss": 2.2504,
|
| 722 |
+
"step": 9200
|
| 723 |
+
},
|
| 724 |
+
{
|
| 725 |
+
"epoch": 15.58,
|
| 726 |
+
"learning_rate": 0.0005414833438089252,
|
| 727 |
+
"loss": 2.2542,
|
| 728 |
+
"step": 9300
|
| 729 |
+
},
|
| 730 |
+
{
|
| 731 |
+
"epoch": 15.75,
|
| 732 |
+
"learning_rate": 0.0005351979886863608,
|
| 733 |
+
"loss": 2.2455,
|
| 734 |
+
"step": 9400
|
| 735 |
+
},
|
| 736 |
+
{
|
| 737 |
+
"epoch": 15.91,
|
| 738 |
+
"learning_rate": 0.0005289126335637963,
|
| 739 |
+
"loss": 2.2381,
|
| 740 |
+
"step": 9500
|
| 741 |
+
},
|
| 742 |
+
{
|
| 743 |
+
"epoch": 15.91,
|
| 744 |
+
"eval_loss": 1.563069462776184,
|
| 745 |
+
"eval_runtime": 410.1326,
|
| 746 |
+
"eval_samples_per_second": 25.328,
|
| 747 |
+
"eval_steps_per_second": 3.167,
|
| 748 |
+
"eval_wer": 0.9508296797824518,
|
| 749 |
+
"step": 9500
|
| 750 |
+
},
|
| 751 |
+
{
|
| 752 |
+
"epoch": 16.08,
|
| 753 |
+
"learning_rate": 0.0005226272784412319,
|
| 754 |
+
"loss": 2.2406,
|
| 755 |
+
"step": 9600
|
| 756 |
+
},
|
| 757 |
+
{
|
| 758 |
+
"epoch": 16.25,
|
| 759 |
+
"learning_rate": 0.0005163419233186674,
|
| 760 |
+
"loss": 2.2265,
|
| 761 |
+
"step": 9700
|
| 762 |
+
},
|
| 763 |
+
{
|
| 764 |
+
"epoch": 16.42,
|
| 765 |
+
"learning_rate": 0.0005100565681961032,
|
| 766 |
+
"loss": 2.2221,
|
| 767 |
+
"step": 9800
|
| 768 |
+
},
|
| 769 |
+
{
|
| 770 |
+
"epoch": 16.58,
|
| 771 |
+
"learning_rate": 0.0005037712130735387,
|
| 772 |
+
"loss": 2.2122,
|
| 773 |
+
"step": 9900
|
| 774 |
+
},
|
| 775 |
+
{
|
| 776 |
+
"epoch": 16.75,
|
| 777 |
+
"learning_rate": 0.0004974858579509742,
|
| 778 |
+
"loss": 2.2072,
|
| 779 |
+
"step": 10000
|
| 780 |
+
},
|
| 781 |
+
{
|
| 782 |
+
"epoch": 16.75,
|
| 783 |
+
"eval_loss": 1.4565062522888184,
|
| 784 |
+
"eval_runtime": 409.2254,
|
| 785 |
+
"eval_samples_per_second": 25.385,
|
| 786 |
+
"eval_steps_per_second": 3.174,
|
| 787 |
+
"eval_wer": 0.9395452947814344,
|
| 788 |
+
"step": 10000
|
| 789 |
+
},
|
| 790 |
+
{
|
| 791 |
+
"epoch": 16.92,
|
| 792 |
+
"learning_rate": 0.0004912005028284098,
|
| 793 |
+
"loss": 2.1876,
|
| 794 |
+
"step": 10100
|
| 795 |
+
},
|
| 796 |
+
{
|
| 797 |
+
"epoch": 17.09,
|
| 798 |
+
"learning_rate": 0.00048491514770584537,
|
| 799 |
+
"loss": 2.2144,
|
| 800 |
+
"step": 10200
|
| 801 |
+
},
|
| 802 |
+
{
|
| 803 |
+
"epoch": 17.25,
|
| 804 |
+
"learning_rate": 0.00047862979258328096,
|
| 805 |
+
"loss": 2.1943,
|
| 806 |
+
"step": 10300
|
| 807 |
+
},
|
| 808 |
+
{
|
| 809 |
+
"epoch": 17.42,
|
| 810 |
+
"learning_rate": 0.00047234443746071655,
|
| 811 |
+
"loss": 2.1901,
|
| 812 |
+
"step": 10400
|
| 813 |
+
},
|
| 814 |
+
{
|
| 815 |
+
"epoch": 17.59,
|
| 816 |
+
"learning_rate": 0.00046605908233815214,
|
| 817 |
+
"loss": 2.197,
|
| 818 |
+
"step": 10500
|
| 819 |
+
},
|
| 820 |
+
{
|
| 821 |
+
"epoch": 17.59,
|
| 822 |
+
"eval_loss": 1.430406093597412,
|
| 823 |
+
"eval_runtime": 410.1605,
|
| 824 |
+
"eval_samples_per_second": 25.327,
|
| 825 |
+
"eval_steps_per_second": 3.167,
|
| 826 |
+
"eval_wer": 0.9405997373143163,
|
| 827 |
+
"step": 10500
|
| 828 |
+
},
|
| 829 |
+
{
|
| 830 |
+
"epoch": 17.76,
|
| 831 |
+
"learning_rate": 0.0004597737272155877,
|
| 832 |
+
"loss": 2.1872,
|
| 833 |
+
"step": 10600
|
| 834 |
+
},
|
| 835 |
+
{
|
| 836 |
+
"epoch": 17.92,
|
| 837 |
+
"learning_rate": 0.0004534883720930232,
|
| 838 |
+
"loss": 2.2033,
|
| 839 |
+
"step": 10700
|
| 840 |
+
},
|
| 841 |
+
{
|
| 842 |
+
"epoch": 18.09,
|
| 843 |
+
"learning_rate": 0.00044720301697045886,
|
| 844 |
+
"loss": 2.1865,
|
| 845 |
+
"step": 10800
|
| 846 |
+
},
|
| 847 |
+
{
|
| 848 |
+
"epoch": 18.26,
|
| 849 |
+
"learning_rate": 0.0004409176618478944,
|
| 850 |
+
"loss": 2.194,
|
| 851 |
+
"step": 10900
|
| 852 |
+
},
|
| 853 |
+
{
|
| 854 |
+
"epoch": 18.43,
|
| 855 |
+
"learning_rate": 0.00043463230672533,
|
| 856 |
+
"loss": 2.198,
|
| 857 |
+
"step": 11000
|
| 858 |
+
},
|
| 859 |
+
{
|
| 860 |
+
"epoch": 18.43,
|
| 861 |
+
"eval_loss": 1.423040747642517,
|
| 862 |
+
"eval_runtime": 411.9246,
|
| 863 |
+
"eval_samples_per_second": 25.218,
|
| 864 |
+
"eval_steps_per_second": 3.153,
|
| 865 |
+
"eval_wer": 0.9382318663632832,
|
| 866 |
+
"step": 11000
|
| 867 |
+
},
|
| 868 |
+
{
|
| 869 |
+
"epoch": 18.59,
|
| 870 |
+
"learning_rate": 0.0004283469516027655,
|
| 871 |
+
"loss": 2.1784,
|
| 872 |
+
"step": 11100
|
| 873 |
+
},
|
| 874 |
+
{
|
| 875 |
+
"epoch": 18.76,
|
| 876 |
+
"learning_rate": 0.00042206159648020117,
|
| 877 |
+
"loss": 2.1739,
|
| 878 |
+
"step": 11200
|
| 879 |
+
},
|
| 880 |
+
{
|
| 881 |
+
"epoch": 18.93,
|
| 882 |
+
"learning_rate": 0.0004157762413576367,
|
| 883 |
+
"loss": 2.1686,
|
| 884 |
+
"step": 11300
|
| 885 |
+
},
|
| 886 |
+
{
|
| 887 |
+
"epoch": 19.1,
|
| 888 |
+
"learning_rate": 0.0004094908862350723,
|
| 889 |
+
"loss": 2.1639,
|
| 890 |
+
"step": 11400
|
| 891 |
+
},
|
| 892 |
+
{
|
| 893 |
+
"epoch": 19.26,
|
| 894 |
+
"learning_rate": 0.00040320553111250783,
|
| 895 |
+
"loss": 2.1668,
|
| 896 |
+
"step": 11500
|
| 897 |
+
},
|
| 898 |
+
{
|
| 899 |
+
"epoch": 19.26,
|
| 900 |
+
"eval_loss": 1.3998422622680664,
|
| 901 |
+
"eval_runtime": 412.8679,
|
| 902 |
+
"eval_samples_per_second": 25.161,
|
| 903 |
+
"eval_steps_per_second": 3.146,
|
| 904 |
+
"eval_wer": 0.9314982333462827,
|
| 905 |
+
"step": 11500
|
| 906 |
+
},
|
| 907 |
+
{
|
| 908 |
+
"epoch": 19.43,
|
| 909 |
+
"learning_rate": 0.00039692017598994347,
|
| 910 |
+
"loss": 2.1694,
|
| 911 |
+
"step": 11600
|
| 912 |
+
},
|
| 913 |
+
{
|
| 914 |
+
"epoch": 19.6,
|
| 915 |
+
"learning_rate": 0.000390634820867379,
|
| 916 |
+
"loss": 2.1492,
|
| 917 |
+
"step": 11700
|
| 918 |
+
},
|
| 919 |
+
{
|
| 920 |
+
"epoch": 19.77,
|
| 921 |
+
"learning_rate": 0.0003843494657448146,
|
| 922 |
+
"loss": 2.1465,
|
| 923 |
+
"step": 11800
|
| 924 |
+
},
|
| 925 |
+
{
|
| 926 |
+
"epoch": 19.93,
|
| 927 |
+
"learning_rate": 0.00037806411062225013,
|
| 928 |
+
"loss": 2.1484,
|
| 929 |
+
"step": 11900
|
| 930 |
+
},
|
| 931 |
+
{
|
| 932 |
+
"epoch": 20.1,
|
| 933 |
+
"learning_rate": 0.0003718416090509114,
|
| 934 |
+
"loss": 2.1498,
|
| 935 |
+
"step": 12000
|
| 936 |
+
},
|
| 937 |
+
{
|
| 938 |
+
"epoch": 20.1,
|
| 939 |
+
"eval_loss": 1.3919602632522583,
|
| 940 |
+
"eval_runtime": 412.6773,
|
| 941 |
+
"eval_samples_per_second": 25.172,
|
| 942 |
+
"eval_steps_per_second": 3.148,
|
| 943 |
+
"eval_wer": 0.9257635458867491,
|
| 944 |
+
"step": 12000
|
| 945 |
+
},
|
| 946 |
+
{
|
| 947 |
+
"epoch": 20.27,
|
| 948 |
+
"learning_rate": 0.00036555625392834694,
|
| 949 |
+
"loss": 2.1295,
|
| 950 |
+
"step": 12100
|
| 951 |
+
},
|
| 952 |
+
{
|
| 953 |
+
"epoch": 20.44,
|
| 954 |
+
"learning_rate": 0.00035927089880578253,
|
| 955 |
+
"loss": 2.1346,
|
| 956 |
+
"step": 12200
|
| 957 |
+
},
|
| 958 |
+
{
|
| 959 |
+
"epoch": 20.6,
|
| 960 |
+
"learning_rate": 0.0003529855436832181,
|
| 961 |
+
"loss": 2.1227,
|
| 962 |
+
"step": 12300
|
| 963 |
+
},
|
| 964 |
+
{
|
| 965 |
+
"epoch": 20.77,
|
| 966 |
+
"learning_rate": 0.00034670018856065366,
|
| 967 |
+
"loss": 2.1205,
|
| 968 |
+
"step": 12400
|
| 969 |
+
},
|
| 970 |
+
{
|
| 971 |
+
"epoch": 20.94,
|
| 972 |
+
"learning_rate": 0.00034041483343808925,
|
| 973 |
+
"loss": 2.1244,
|
| 974 |
+
"step": 12500
|
| 975 |
+
},
|
| 976 |
+
{
|
| 977 |
+
"epoch": 20.94,
|
| 978 |
+
"eval_loss": 1.3584457635879517,
|
| 979 |
+
"eval_runtime": 410.923,
|
| 980 |
+
"eval_samples_per_second": 25.28,
|
| 981 |
+
"eval_steps_per_second": 3.161,
|
| 982 |
+
"eval_wer": 0.9152561185415394,
|
| 983 |
+
"step": 12500
|
| 984 |
+
},
|
| 985 |
+
{
|
| 986 |
+
"epoch": 21.11,
|
| 987 |
+
"learning_rate": 0.00033412947831552484,
|
| 988 |
+
"loss": 2.1163,
|
| 989 |
+
"step": 12600
|
| 990 |
+
},
|
| 991 |
+
{
|
| 992 |
+
"epoch": 21.27,
|
| 993 |
+
"learning_rate": 0.00032784412319296043,
|
| 994 |
+
"loss": 2.1141,
|
| 995 |
+
"step": 12700
|
| 996 |
+
},
|
| 997 |
+
{
|
| 998 |
+
"epoch": 21.44,
|
| 999 |
+
"learning_rate": 0.00032155876807039597,
|
| 1000 |
+
"loss": 2.1122,
|
| 1001 |
+
"step": 12800
|
| 1002 |
+
},
|
| 1003 |
+
{
|
| 1004 |
+
"epoch": 21.61,
|
| 1005 |
+
"learning_rate": 0.00031527341294783156,
|
| 1006 |
+
"loss": 2.0937,
|
| 1007 |
+
"step": 12900
|
| 1008 |
+
},
|
| 1009 |
+
{
|
| 1010 |
+
"epoch": 21.78,
|
| 1011 |
+
"learning_rate": 0.00030898805782526715,
|
| 1012 |
+
"loss": 2.0953,
|
| 1013 |
+
"step": 13000
|
| 1014 |
+
},
|
| 1015 |
+
{
|
| 1016 |
+
"epoch": 21.78,
|
| 1017 |
+
"eval_loss": 1.327351450920105,
|
| 1018 |
+
"eval_runtime": 411.8656,
|
| 1019 |
+
"eval_samples_per_second": 25.222,
|
| 1020 |
+
"eval_steps_per_second": 3.154,
|
| 1021 |
+
"eval_wer": 0.905377656917698,
|
| 1022 |
+
"step": 13000
|
| 1023 |
+
},
|
| 1024 |
+
{
|
| 1025 |
+
"epoch": 21.94,
|
| 1026 |
+
"learning_rate": 0.00030270270270270274,
|
| 1027 |
+
"loss": 2.096,
|
| 1028 |
+
"step": 13100
|
| 1029 |
+
},
|
| 1030 |
+
{
|
| 1031 |
+
"epoch": 22.11,
|
| 1032 |
+
"learning_rate": 0.0002964173475801383,
|
| 1033 |
+
"loss": 2.1102,
|
| 1034 |
+
"step": 13200
|
| 1035 |
+
},
|
| 1036 |
+
{
|
| 1037 |
+
"epoch": 22.28,
|
| 1038 |
+
"learning_rate": 0.00029013199245757386,
|
| 1039 |
+
"loss": 2.0892,
|
| 1040 |
+
"step": 13300
|
| 1041 |
+
},
|
| 1042 |
+
{
|
| 1043 |
+
"epoch": 22.45,
|
| 1044 |
+
"learning_rate": 0.00028384663733500945,
|
| 1045 |
+
"loss": 2.0805,
|
| 1046 |
+
"step": 13400
|
| 1047 |
+
},
|
| 1048 |
+
{
|
| 1049 |
+
"epoch": 22.61,
|
| 1050 |
+
"learning_rate": 0.00027756128221244504,
|
| 1051 |
+
"loss": 2.0762,
|
| 1052 |
+
"step": 13500
|
| 1053 |
+
},
|
| 1054 |
+
{
|
| 1055 |
+
"epoch": 22.61,
|
| 1056 |
+
"eval_loss": 1.2932939529418945,
|
| 1057 |
+
"eval_runtime": 410.2802,
|
| 1058 |
+
"eval_samples_per_second": 25.319,
|
| 1059 |
+
"eval_steps_per_second": 3.166,
|
| 1060 |
+
"eval_wer": 0.9073015520654124,
|
| 1061 |
+
"step": 13500
|
| 1062 |
+
},
|
| 1063 |
+
{
|
| 1064 |
+
"epoch": 22.78,
|
| 1065 |
+
"learning_rate": 0.0002712759270898806,
|
| 1066 |
+
"loss": 2.0867,
|
| 1067 |
+
"step": 13600
|
| 1068 |
+
},
|
| 1069 |
+
{
|
| 1070 |
+
"epoch": 22.95,
|
| 1071 |
+
"learning_rate": 0.00026499057196731617,
|
| 1072 |
+
"loss": 2.0757,
|
| 1073 |
+
"step": 13700
|
| 1074 |
+
},
|
| 1075 |
+
{
|
| 1076 |
+
"epoch": 23.12,
|
| 1077 |
+
"learning_rate": 0.00025870521684475176,
|
| 1078 |
+
"loss": 2.0883,
|
| 1079 |
+
"step": 13800
|
| 1080 |
+
},
|
| 1081 |
+
{
|
| 1082 |
+
"epoch": 23.28,
|
| 1083 |
+
"learning_rate": 0.0002524198617221873,
|
| 1084 |
+
"loss": 2.0696,
|
| 1085 |
+
"step": 13900
|
| 1086 |
+
},
|
| 1087 |
+
{
|
| 1088 |
+
"epoch": 23.45,
|
| 1089 |
+
"learning_rate": 0.0002461345065996229,
|
| 1090 |
+
"loss": 2.0587,
|
| 1091 |
+
"step": 14000
|
| 1092 |
+
},
|
| 1093 |
+
{
|
| 1094 |
+
"epoch": 23.45,
|
| 1095 |
+
"eval_loss": 1.2515921592712402,
|
| 1096 |
+
"eval_runtime": 410.7551,
|
| 1097 |
+
"eval_samples_per_second": 25.29,
|
| 1098 |
+
"eval_steps_per_second": 3.162,
|
| 1099 |
+
"eval_wer": 0.8944447527609746,
|
| 1100 |
+
"step": 14000
|
| 1101 |
+
},
|
| 1102 |
+
{
|
| 1103 |
+
"epoch": 23.62,
|
| 1104 |
+
"learning_rate": 0.00023984915147705848,
|
| 1105 |
+
"loss": 2.0661,
|
| 1106 |
+
"step": 14100
|
| 1107 |
+
},
|
| 1108 |
+
{
|
| 1109 |
+
"epoch": 23.79,
|
| 1110 |
+
"learning_rate": 0.00023356379635449404,
|
| 1111 |
+
"loss": 2.0529,
|
| 1112 |
+
"step": 14200
|
| 1113 |
+
},
|
| 1114 |
+
{
|
| 1115 |
+
"epoch": 23.95,
|
| 1116 |
+
"learning_rate": 0.00022727844123192963,
|
| 1117 |
+
"loss": 2.0509,
|
| 1118 |
+
"step": 14300
|
| 1119 |
+
},
|
| 1120 |
+
{
|
| 1121 |
+
"epoch": 24.12,
|
| 1122 |
+
"learning_rate": 0.0002209930861093652,
|
| 1123 |
+
"loss": 2.0481,
|
| 1124 |
+
"step": 14400
|
| 1125 |
+
},
|
| 1126 |
+
{
|
| 1127 |
+
"epoch": 24.29,
|
| 1128 |
+
"learning_rate": 0.00021470773098680078,
|
| 1129 |
+
"loss": 2.0363,
|
| 1130 |
+
"step": 14500
|
| 1131 |
+
},
|
| 1132 |
+
{
|
| 1133 |
+
"epoch": 24.29,
|
| 1134 |
+
"eval_loss": 1.2214268445968628,
|
| 1135 |
+
"eval_runtime": 412.9112,
|
| 1136 |
+
"eval_samples_per_second": 25.158,
|
| 1137 |
+
"eval_steps_per_second": 3.146,
|
| 1138 |
+
"eval_wer": 0.8901529866622269,
|
| 1139 |
+
"step": 14500
|
| 1140 |
+
},
|
| 1141 |
+
{
|
| 1142 |
+
"epoch": 24.46,
|
| 1143 |
+
"learning_rate": 0.00020842237586423635,
|
| 1144 |
+
"loss": 2.0412,
|
| 1145 |
+
"step": 14600
|
| 1146 |
+
},
|
| 1147 |
+
{
|
| 1148 |
+
"epoch": 24.62,
|
| 1149 |
+
"learning_rate": 0.00020213702074167188,
|
| 1150 |
+
"loss": 2.0264,
|
| 1151 |
+
"step": 14700
|
| 1152 |
+
},
|
| 1153 |
+
{
|
| 1154 |
+
"epoch": 24.79,
|
| 1155 |
+
"learning_rate": 0.00019585166561910747,
|
| 1156 |
+
"loss": 2.0373,
|
| 1157 |
+
"step": 14800
|
| 1158 |
+
},
|
| 1159 |
+
{
|
| 1160 |
+
"epoch": 24.96,
|
| 1161 |
+
"learning_rate": 0.00018956631049654304,
|
| 1162 |
+
"loss": 2.0373,
|
| 1163 |
+
"step": 14900
|
| 1164 |
+
},
|
| 1165 |
+
{
|
| 1166 |
+
"epoch": 25.13,
|
| 1167 |
+
"learning_rate": 0.00018328095537397863,
|
| 1168 |
+
"loss": 2.0302,
|
| 1169 |
+
"step": 15000
|
| 1170 |
+
},
|
| 1171 |
+
{
|
| 1172 |
+
"epoch": 25.13,
|
| 1173 |
+
"eval_loss": 1.2087428569793701,
|
| 1174 |
+
"eval_runtime": 412.179,
|
| 1175 |
+
"eval_samples_per_second": 25.203,
|
| 1176 |
+
"eval_steps_per_second": 3.152,
|
| 1177 |
+
"eval_wer": 0.8871191520062157,
|
| 1178 |
+
"step": 15000
|
| 1179 |
+
},
|
| 1180 |
+
{
|
| 1181 |
+
"epoch": 25.29,
|
| 1182 |
+
"learning_rate": 0.0001769956002514142,
|
| 1183 |
+
"loss": 2.0109,
|
| 1184 |
+
"step": 15100
|
| 1185 |
+
},
|
| 1186 |
+
{
|
| 1187 |
+
"epoch": 25.46,
|
| 1188 |
+
"learning_rate": 0.00017071024512884978,
|
| 1189 |
+
"loss": 2.0215,
|
| 1190 |
+
"step": 15200
|
| 1191 |
+
},
|
| 1192 |
+
{
|
| 1193 |
+
"epoch": 25.63,
|
| 1194 |
+
"learning_rate": 0.00016442489000628534,
|
| 1195 |
+
"loss": 2.0137,
|
| 1196 |
+
"step": 15300
|
| 1197 |
+
},
|
| 1198 |
+
{
|
| 1199 |
+
"epoch": 25.8,
|
| 1200 |
+
"learning_rate": 0.00015813953488372093,
|
| 1201 |
+
"loss": 2.0084,
|
| 1202 |
+
"step": 15400
|
| 1203 |
+
},
|
| 1204 |
+
{
|
| 1205 |
+
"epoch": 25.96,
|
| 1206 |
+
"learning_rate": 0.0001518541797611565,
|
| 1207 |
+
"loss": 2.0071,
|
| 1208 |
+
"step": 15500
|
| 1209 |
+
},
|
| 1210 |
+
{
|
| 1211 |
+
"epoch": 25.96,
|
| 1212 |
+
"eval_loss": 1.1953096389770508,
|
| 1213 |
+
"eval_runtime": 413.1745,
|
| 1214 |
+
"eval_samples_per_second": 25.142,
|
| 1215 |
+
"eval_steps_per_second": 3.144,
|
| 1216 |
+
"eval_wer": 0.8785726177923303,
|
| 1217 |
+
"step": 15500
|
| 1218 |
+
},
|
| 1219 |
+
{
|
| 1220 |
+
"epoch": 26.13,
|
| 1221 |
+
"learning_rate": 0.00014556882463859208,
|
| 1222 |
+
"loss": 2.0112,
|
| 1223 |
+
"step": 15600
|
| 1224 |
+
},
|
| 1225 |
+
{
|
| 1226 |
+
"epoch": 26.3,
|
| 1227 |
+
"learning_rate": 0.00013928346951602765,
|
| 1228 |
+
"loss": 2.0077,
|
| 1229 |
+
"step": 15700
|
| 1230 |
+
},
|
| 1231 |
+
{
|
| 1232 |
+
"epoch": 26.47,
|
| 1233 |
+
"learning_rate": 0.00013299811439346324,
|
| 1234 |
+
"loss": 2.0052,
|
| 1235 |
+
"step": 15800
|
| 1236 |
+
},
|
| 1237 |
+
{
|
| 1238 |
+
"epoch": 26.63,
|
| 1239 |
+
"learning_rate": 0.0001267127592708988,
|
| 1240 |
+
"loss": 1.9947,
|
| 1241 |
+
"step": 15900
|
| 1242 |
+
},
|
| 1243 |
+
{
|
| 1244 |
+
"epoch": 26.8,
|
| 1245 |
+
"learning_rate": 0.00012042740414833438,
|
| 1246 |
+
"loss": 1.9882,
|
| 1247 |
+
"step": 16000
|
| 1248 |
+
},
|
| 1249 |
+
{
|
| 1250 |
+
"epoch": 26.8,
|
| 1251 |
+
"eval_loss": 1.1737616062164307,
|
| 1252 |
+
"eval_runtime": 414.4584,
|
| 1253 |
+
"eval_samples_per_second": 25.064,
|
| 1254 |
+
"eval_steps_per_second": 3.134,
|
| 1255 |
+
"eval_wer": 0.8711730210703517,
|
| 1256 |
+
"step": 16000
|
| 1257 |
+
},
|
| 1258 |
+
{
|
| 1259 |
+
"epoch": 26.97,
|
| 1260 |
+
"learning_rate": 0.0001142049025769956,
|
| 1261 |
+
"loss": 1.9926,
|
| 1262 |
+
"step": 16100
|
| 1263 |
+
},
|
| 1264 |
+
{
|
| 1265 |
+
"epoch": 27.14,
|
| 1266 |
+
"learning_rate": 0.00010791954745443117,
|
| 1267 |
+
"loss": 2.0032,
|
| 1268 |
+
"step": 16200
|
| 1269 |
+
},
|
| 1270 |
+
{
|
| 1271 |
+
"epoch": 27.3,
|
| 1272 |
+
"learning_rate": 0.00010163419233186675,
|
| 1273 |
+
"loss": 1.996,
|
| 1274 |
+
"step": 16300
|
| 1275 |
+
},
|
| 1276 |
+
{
|
| 1277 |
+
"epoch": 27.47,
|
| 1278 |
+
"learning_rate": 9.534883720930233e-05,
|
| 1279 |
+
"loss": 1.9746,
|
| 1280 |
+
"step": 16400
|
| 1281 |
+
},
|
| 1282 |
+
{
|
| 1283 |
+
"epoch": 27.64,
|
| 1284 |
+
"learning_rate": 8.906348208673789e-05,
|
| 1285 |
+
"loss": 1.9772,
|
| 1286 |
+
"step": 16500
|
| 1287 |
+
},
|
| 1288 |
+
{
|
| 1289 |
+
"epoch": 27.64,
|
| 1290 |
+
"eval_loss": 1.164720892906189,
|
| 1291 |
+
"eval_runtime": 425.7197,
|
| 1292 |
+
"eval_samples_per_second": 24.401,
|
| 1293 |
+
"eval_steps_per_second": 3.051,
|
| 1294 |
+
"eval_wer": 0.867214236824093,
|
| 1295 |
+
"step": 16500
|
| 1296 |
+
},
|
| 1297 |
+
{
|
| 1298 |
+
"epoch": 27.81,
|
| 1299 |
+
"learning_rate": 8.277812696417347e-05,
|
| 1300 |
+
"loss": 1.9759,
|
| 1301 |
+
"step": 16600
|
| 1302 |
+
},
|
| 1303 |
+
{
|
| 1304 |
+
"epoch": 27.97,
|
| 1305 |
+
"learning_rate": 7.649277184160904e-05,
|
| 1306 |
+
"loss": 1.9657,
|
| 1307 |
+
"step": 16700
|
| 1308 |
+
},
|
| 1309 |
+
{
|
| 1310 |
+
"epoch": 28.14,
|
| 1311 |
+
"learning_rate": 7.020741671904462e-05,
|
| 1312 |
+
"loss": 1.9806,
|
| 1313 |
+
"step": 16800
|
| 1314 |
+
},
|
| 1315 |
+
{
|
| 1316 |
+
"epoch": 28.31,
|
| 1317 |
+
"learning_rate": 6.39220615964802e-05,
|
| 1318 |
+
"loss": 1.9802,
|
| 1319 |
+
"step": 16900
|
| 1320 |
+
},
|
| 1321 |
+
{
|
| 1322 |
+
"epoch": 28.48,
|
| 1323 |
+
"learning_rate": 5.763670647391578e-05,
|
| 1324 |
+
"loss": 1.9585,
|
| 1325 |
+
"step": 17000
|
| 1326 |
+
},
|
| 1327 |
+
{
|
| 1328 |
+
"epoch": 28.48,
|
| 1329 |
+
"eval_loss": 1.1459153890609741,
|
| 1330 |
+
"eval_runtime": 417.3472,
|
| 1331 |
+
"eval_samples_per_second": 24.891,
|
| 1332 |
+
"eval_steps_per_second": 3.113,
|
| 1333 |
+
"eval_wer": 0.8634774404794938,
|
| 1334 |
+
"step": 17000
|
| 1335 |
+
},
|
| 1336 |
+
{
|
| 1337 |
+
"epoch": 28.64,
|
| 1338 |
+
"learning_rate": 5.135135135135136e-05,
|
| 1339 |
+
"loss": 1.9573,
|
| 1340 |
+
"step": 17100
|
| 1341 |
+
},
|
| 1342 |
+
{
|
| 1343 |
+
"epoch": 28.81,
|
| 1344 |
+
"learning_rate": 4.506599622878693e-05,
|
| 1345 |
+
"loss": 1.962,
|
| 1346 |
+
"step": 17200
|
| 1347 |
+
},
|
| 1348 |
+
{
|
| 1349 |
+
"epoch": 28.98,
|
| 1350 |
+
"learning_rate": 3.8780641106222504e-05,
|
| 1351 |
+
"loss": 1.9653,
|
| 1352 |
+
"step": 17300
|
| 1353 |
+
},
|
| 1354 |
+
{
|
| 1355 |
+
"epoch": 29.15,
|
| 1356 |
+
"learning_rate": 3.249528598365808e-05,
|
| 1357 |
+
"loss": 1.9631,
|
| 1358 |
+
"step": 17400
|
| 1359 |
+
},
|
| 1360 |
+
{
|
| 1361 |
+
"epoch": 29.31,
|
| 1362 |
+
"learning_rate": 2.620993086109365e-05,
|
| 1363 |
+
"loss": 1.944,
|
| 1364 |
+
"step": 17500
|
| 1365 |
+
},
|
| 1366 |
+
{
|
| 1367 |
+
"epoch": 29.31,
|
| 1368 |
+
"eval_loss": 1.1414194107055664,
|
| 1369 |
+
"eval_runtime": 414.6007,
|
| 1370 |
+
"eval_samples_per_second": 25.055,
|
| 1371 |
+
"eval_steps_per_second": 3.133,
|
| 1372 |
+
"eval_wer": 0.8616275412989992,
|
| 1373 |
+
"step": 17500
|
| 1374 |
+
},
|
| 1375 |
+
{
|
| 1376 |
+
"epoch": 29.48,
|
| 1377 |
+
"learning_rate": 1.9924575738529227e-05,
|
| 1378 |
+
"loss": 1.9547,
|
| 1379 |
+
"step": 17600
|
| 1380 |
+
},
|
| 1381 |
+
{
|
| 1382 |
+
"epoch": 29.65,
|
| 1383 |
+
"learning_rate": 1.3639220615964803e-05,
|
| 1384 |
+
"loss": 1.9557,
|
| 1385 |
+
"step": 17700
|
| 1386 |
+
},
|
| 1387 |
+
{
|
| 1388 |
+
"epoch": 29.82,
|
| 1389 |
+
"learning_rate": 7.353865493400377e-06,
|
| 1390 |
+
"loss": 1.9464,
|
| 1391 |
+
"step": 17800
|
| 1392 |
+
},
|
| 1393 |
+
{
|
| 1394 |
+
"epoch": 29.98,
|
| 1395 |
+
"learning_rate": 1.0685103708359522e-06,
|
| 1396 |
+
"loss": 1.9654,
|
| 1397 |
+
"step": 17900
|
| 1398 |
+
},
|
| 1399 |
+
{
|
| 1400 |
+
"epoch": 30.0,
|
| 1401 |
+
"step": 17910,
|
| 1402 |
+
"total_flos": 3.015307099908152e+20,
|
| 1403 |
+
"train_loss": 1.4937853462266097,
|
| 1404 |
+
"train_runtime": 52137.1608,
|
| 1405 |
+
"train_samples_per_second": 21.986,
|
| 1406 |
+
"train_steps_per_second": 0.344
|
| 1407 |
}
|
| 1408 |
],
|
| 1409 |
+
"max_steps": 17910,
|
| 1410 |
+
"num_train_epochs": 30,
|
| 1411 |
+
"total_flos": 3.015307099908152e+20,
|
| 1412 |
"trial_name": null,
|
| 1413 |
"trial_params": null
|
| 1414 |
}
|