Commit
•
23b1976
1
Parent(s):
9467ba6
End of training
Browse files- all_results.json +14 -0
- eval_results.json +9 -0
- pytorch_model.bin +1 -1
- train_results.json +8 -0
- trainer_state.json +1003 -0
all_results.json
ADDED
|
@@ -0,0 +1,14 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"epoch": 50.0,
|
| 3 |
+
"eval_loss": 0.25618529319763184,
|
| 4 |
+
"eval_runtime": 109.3859,
|
| 5 |
+
"eval_samples": 3427,
|
| 6 |
+
"eval_samples_per_second": 31.329,
|
| 7 |
+
"eval_steps_per_second": 0.987,
|
| 8 |
+
"eval_wer": 0.31124942562345964,
|
| 9 |
+
"train_loss": 1.318804485697142,
|
| 10 |
+
"train_runtime": 22379.119,
|
| 11 |
+
"train_samples": 9072,
|
| 12 |
+
"train_samples_per_second": 20.269,
|
| 13 |
+
"train_steps_per_second": 0.635
|
| 14 |
+
}
|
eval_results.json
ADDED
|
@@ -0,0 +1,9 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"epoch": 50.0,
|
| 3 |
+
"eval_loss": 0.25618529319763184,
|
| 4 |
+
"eval_runtime": 109.3859,
|
| 5 |
+
"eval_samples": 3427,
|
| 6 |
+
"eval_samples_per_second": 31.329,
|
| 7 |
+
"eval_steps_per_second": 0.987,
|
| 8 |
+
"eval_wer": 0.31124942562345964
|
| 9 |
+
}
|
pytorch_model.bin
CHANGED
|
@@ -1,3 +1,3 @@
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
-
oid sha256:
|
| 3 |
size 1262087665
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:421a4dfcb32cc9a27b984a6f5358add60c34009843e11f84647e11205f12ed12
|
| 3 |
size 1262087665
|
train_results.json
ADDED
|
@@ -0,0 +1,8 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"epoch": 50.0,
|
| 3 |
+
"train_loss": 1.318804485697142,
|
| 4 |
+
"train_runtime": 22379.119,
|
| 5 |
+
"train_samples": 9072,
|
| 6 |
+
"train_samples_per_second": 20.269,
|
| 7 |
+
"train_steps_per_second": 0.635
|
| 8 |
+
}
|
trainer_state.json
ADDED
|
@@ -0,0 +1,1003 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"best_metric": null,
|
| 3 |
+
"best_model_checkpoint": null,
|
| 4 |
+
"epoch": 50.0,
|
| 5 |
+
"global_step": 14200,
|
| 6 |
+
"is_hyper_param_search": false,
|
| 7 |
+
"is_local_process_zero": true,
|
| 8 |
+
"is_world_process_zero": true,
|
| 9 |
+
"log_history": [
|
| 10 |
+
{
|
| 11 |
+
"epoch": 0.35,
|
| 12 |
+
"learning_rate": 6.789999999999999e-06,
|
| 13 |
+
"loss": 13.5868,
|
| 14 |
+
"step": 100
|
| 15 |
+
},
|
| 16 |
+
{
|
| 17 |
+
"epoch": 0.7,
|
| 18 |
+
"learning_rate": 1.379e-05,
|
| 19 |
+
"loss": 5.6931,
|
| 20 |
+
"step": 200
|
| 21 |
+
},
|
| 22 |
+
{
|
| 23 |
+
"epoch": 1.06,
|
| 24 |
+
"learning_rate": 2.0789999999999996e-05,
|
| 25 |
+
"loss": 4.1485,
|
| 26 |
+
"step": 300
|
| 27 |
+
},
|
| 28 |
+
{
|
| 29 |
+
"epoch": 1.41,
|
| 30 |
+
"learning_rate": 2.779e-05,
|
| 31 |
+
"loss": 3.5861,
|
| 32 |
+
"step": 400
|
| 33 |
+
},
|
| 34 |
+
{
|
| 35 |
+
"epoch": 1.76,
|
| 36 |
+
"learning_rate": 3.479e-05,
|
| 37 |
+
"loss": 3.2885,
|
| 38 |
+
"step": 500
|
| 39 |
+
},
|
| 40 |
+
{
|
| 41 |
+
"epoch": 2.11,
|
| 42 |
+
"learning_rate": 4.178999999999999e-05,
|
| 43 |
+
"loss": 3.2023,
|
| 44 |
+
"step": 600
|
| 45 |
+
},
|
| 46 |
+
{
|
| 47 |
+
"epoch": 2.46,
|
| 48 |
+
"learning_rate": 4.878999999999999e-05,
|
| 49 |
+
"loss": 3.1447,
|
| 50 |
+
"step": 700
|
| 51 |
+
},
|
| 52 |
+
{
|
| 53 |
+
"epoch": 2.82,
|
| 54 |
+
"learning_rate": 5.579e-05,
|
| 55 |
+
"loss": 3.0982,
|
| 56 |
+
"step": 800
|
| 57 |
+
},
|
| 58 |
+
{
|
| 59 |
+
"epoch": 3.17,
|
| 60 |
+
"learning_rate": 6.279e-05,
|
| 61 |
+
"loss": 2.9573,
|
| 62 |
+
"step": 900
|
| 63 |
+
},
|
| 64 |
+
{
|
| 65 |
+
"epoch": 3.52,
|
| 66 |
+
"learning_rate": 6.979e-05,
|
| 67 |
+
"loss": 2.3964,
|
| 68 |
+
"step": 1000
|
| 69 |
+
},
|
| 70 |
+
{
|
| 71 |
+
"epoch": 3.52,
|
| 72 |
+
"eval_loss": 1.225058913230896,
|
| 73 |
+
"eval_runtime": 113.8433,
|
| 74 |
+
"eval_samples_per_second": 30.103,
|
| 75 |
+
"eval_steps_per_second": 0.949,
|
| 76 |
+
"eval_wer": 0.878148627762229,
|
| 77 |
+
"step": 1000
|
| 78 |
+
},
|
| 79 |
+
{
|
| 80 |
+
"epoch": 3.87,
|
| 81 |
+
"learning_rate": 6.948560606060606e-05,
|
| 82 |
+
"loss": 1.9563,
|
| 83 |
+
"step": 1100
|
| 84 |
+
},
|
| 85 |
+
{
|
| 86 |
+
"epoch": 4.23,
|
| 87 |
+
"learning_rate": 6.895530303030303e-05,
|
| 88 |
+
"loss": 1.7124,
|
| 89 |
+
"step": 1200
|
| 90 |
+
},
|
| 91 |
+
{
|
| 92 |
+
"epoch": 4.58,
|
| 93 |
+
"learning_rate": 6.8425e-05,
|
| 94 |
+
"loss": 1.5888,
|
| 95 |
+
"step": 1300
|
| 96 |
+
},
|
| 97 |
+
{
|
| 98 |
+
"epoch": 4.93,
|
| 99 |
+
"learning_rate": 6.789469696969696e-05,
|
| 100 |
+
"loss": 1.514,
|
| 101 |
+
"step": 1400
|
| 102 |
+
},
|
| 103 |
+
{
|
| 104 |
+
"epoch": 5.28,
|
| 105 |
+
"learning_rate": 6.736439393939394e-05,
|
| 106 |
+
"loss": 1.4347,
|
| 107 |
+
"step": 1500
|
| 108 |
+
},
|
| 109 |
+
{
|
| 110 |
+
"epoch": 5.63,
|
| 111 |
+
"learning_rate": 6.68340909090909e-05,
|
| 112 |
+
"loss": 1.4113,
|
| 113 |
+
"step": 1600
|
| 114 |
+
},
|
| 115 |
+
{
|
| 116 |
+
"epoch": 5.99,
|
| 117 |
+
"learning_rate": 6.630378787878786e-05,
|
| 118 |
+
"loss": 1.4,
|
| 119 |
+
"step": 1700
|
| 120 |
+
},
|
| 121 |
+
{
|
| 122 |
+
"epoch": 6.34,
|
| 123 |
+
"learning_rate": 6.577348484848484e-05,
|
| 124 |
+
"loss": 1.3486,
|
| 125 |
+
"step": 1800
|
| 126 |
+
},
|
| 127 |
+
{
|
| 128 |
+
"epoch": 6.69,
|
| 129 |
+
"learning_rate": 6.52431818181818e-05,
|
| 130 |
+
"loss": 1.3427,
|
| 131 |
+
"step": 1900
|
| 132 |
+
},
|
| 133 |
+
{
|
| 134 |
+
"epoch": 7.04,
|
| 135 |
+
"learning_rate": 6.471287878787878e-05,
|
| 136 |
+
"loss": 1.3176,
|
| 137 |
+
"step": 2000
|
| 138 |
+
},
|
| 139 |
+
{
|
| 140 |
+
"epoch": 7.04,
|
| 141 |
+
"eval_loss": 0.3872202932834625,
|
| 142 |
+
"eval_runtime": 113.3065,
|
| 143 |
+
"eval_samples_per_second": 30.245,
|
| 144 |
+
"eval_steps_per_second": 0.953,
|
| 145 |
+
"eval_wer": 0.4461756965620953,
|
| 146 |
+
"step": 2000
|
| 147 |
+
},
|
| 148 |
+
{
|
| 149 |
+
"epoch": 7.39,
|
| 150 |
+
"learning_rate": 6.418257575757575e-05,
|
| 151 |
+
"loss": 1.2983,
|
| 152 |
+
"step": 2100
|
| 153 |
+
},
|
| 154 |
+
{
|
| 155 |
+
"epoch": 7.75,
|
| 156 |
+
"learning_rate": 6.365227272727272e-05,
|
| 157 |
+
"loss": 1.2845,
|
| 158 |
+
"step": 2200
|
| 159 |
+
},
|
| 160 |
+
{
|
| 161 |
+
"epoch": 8.1,
|
| 162 |
+
"learning_rate": 6.312196969696969e-05,
|
| 163 |
+
"loss": 1.2803,
|
| 164 |
+
"step": 2300
|
| 165 |
+
},
|
| 166 |
+
{
|
| 167 |
+
"epoch": 8.45,
|
| 168 |
+
"learning_rate": 6.259166666666667e-05,
|
| 169 |
+
"loss": 1.2482,
|
| 170 |
+
"step": 2400
|
| 171 |
+
},
|
| 172 |
+
{
|
| 173 |
+
"epoch": 8.8,
|
| 174 |
+
"learning_rate": 6.206136363636363e-05,
|
| 175 |
+
"loss": 1.2734,
|
| 176 |
+
"step": 2500
|
| 177 |
+
},
|
| 178 |
+
{
|
| 179 |
+
"epoch": 9.15,
|
| 180 |
+
"learning_rate": 6.153106060606061e-05,
|
| 181 |
+
"loss": 1.2382,
|
| 182 |
+
"step": 2600
|
| 183 |
+
},
|
| 184 |
+
{
|
| 185 |
+
"epoch": 9.51,
|
| 186 |
+
"learning_rate": 6.100075757575757e-05,
|
| 187 |
+
"loss": 1.2278,
|
| 188 |
+
"step": 2700
|
| 189 |
+
},
|
| 190 |
+
{
|
| 191 |
+
"epoch": 9.86,
|
| 192 |
+
"learning_rate": 6.047045454545454e-05,
|
| 193 |
+
"loss": 1.2254,
|
| 194 |
+
"step": 2800
|
| 195 |
+
},
|
| 196 |
+
{
|
| 197 |
+
"epoch": 10.21,
|
| 198 |
+
"learning_rate": 5.994015151515151e-05,
|
| 199 |
+
"loss": 1.2084,
|
| 200 |
+
"step": 2900
|
| 201 |
+
},
|
| 202 |
+
{
|
| 203 |
+
"epoch": 10.56,
|
| 204 |
+
"learning_rate": 5.940984848484848e-05,
|
| 205 |
+
"loss": 1.1999,
|
| 206 |
+
"step": 3000
|
| 207 |
+
},
|
| 208 |
+
{
|
| 209 |
+
"epoch": 10.56,
|
| 210 |
+
"eval_loss": 0.32438814640045166,
|
| 211 |
+
"eval_runtime": 116.219,
|
| 212 |
+
"eval_samples_per_second": 29.487,
|
| 213 |
+
"eval_steps_per_second": 0.929,
|
| 214 |
+
"eval_wer": 0.39216341534734117,
|
| 215 |
+
"step": 3000
|
| 216 |
+
},
|
| 217 |
+
{
|
| 218 |
+
"epoch": 10.92,
|
| 219 |
+
"learning_rate": 5.887954545454545e-05,
|
| 220 |
+
"loss": 1.2041,
|
| 221 |
+
"step": 3100
|
| 222 |
+
},
|
| 223 |
+
{
|
| 224 |
+
"epoch": 11.27,
|
| 225 |
+
"learning_rate": 5.8349242424242423e-05,
|
| 226 |
+
"loss": 1.1884,
|
| 227 |
+
"step": 3200
|
| 228 |
+
},
|
| 229 |
+
{
|
| 230 |
+
"epoch": 11.62,
|
| 231 |
+
"learning_rate": 5.781893939393939e-05,
|
| 232 |
+
"loss": 1.1938,
|
| 233 |
+
"step": 3300
|
| 234 |
+
},
|
| 235 |
+
{
|
| 236 |
+
"epoch": 11.97,
|
| 237 |
+
"learning_rate": 5.728863636363636e-05,
|
| 238 |
+
"loss": 1.1871,
|
| 239 |
+
"step": 3400
|
| 240 |
+
},
|
| 241 |
+
{
|
| 242 |
+
"epoch": 12.32,
|
| 243 |
+
"learning_rate": 5.675833333333333e-05,
|
| 244 |
+
"loss": 1.1859,
|
| 245 |
+
"step": 3500
|
| 246 |
+
},
|
| 247 |
+
{
|
| 248 |
+
"epoch": 12.68,
|
| 249 |
+
"learning_rate": 5.62280303030303e-05,
|
| 250 |
+
"loss": 1.1824,
|
| 251 |
+
"step": 3600
|
| 252 |
+
},
|
| 253 |
+
{
|
| 254 |
+
"epoch": 13.03,
|
| 255 |
+
"learning_rate": 5.569772727272727e-05,
|
| 256 |
+
"loss": 1.1627,
|
| 257 |
+
"step": 3700
|
| 258 |
+
},
|
| 259 |
+
{
|
| 260 |
+
"epoch": 13.38,
|
| 261 |
+
"learning_rate": 5.516742424242423e-05,
|
| 262 |
+
"loss": 1.1557,
|
| 263 |
+
"step": 3800
|
| 264 |
+
},
|
| 265 |
+
{
|
| 266 |
+
"epoch": 13.73,
|
| 267 |
+
"learning_rate": 5.4637121212121204e-05,
|
| 268 |
+
"loss": 1.1679,
|
| 269 |
+
"step": 3900
|
| 270 |
+
},
|
| 271 |
+
{
|
| 272 |
+
"epoch": 14.08,
|
| 273 |
+
"learning_rate": 5.4106818181818174e-05,
|
| 274 |
+
"loss": 1.1633,
|
| 275 |
+
"step": 4000
|
| 276 |
+
},
|
| 277 |
+
{
|
| 278 |
+
"epoch": 14.08,
|
| 279 |
+
"eval_loss": 0.3014352321624756,
|
| 280 |
+
"eval_runtime": 112.4096,
|
| 281 |
+
"eval_samples_per_second": 30.487,
|
| 282 |
+
"eval_steps_per_second": 0.961,
|
| 283 |
+
"eval_wer": 0.3703997660720999,
|
| 284 |
+
"step": 4000
|
| 285 |
+
},
|
| 286 |
+
{
|
| 287 |
+
"epoch": 14.44,
|
| 288 |
+
"learning_rate": 5.3576515151515145e-05,
|
| 289 |
+
"loss": 1.1435,
|
| 290 |
+
"step": 4100
|
| 291 |
+
},
|
| 292 |
+
{
|
| 293 |
+
"epoch": 14.79,
|
| 294 |
+
"learning_rate": 5.3046212121212116e-05,
|
| 295 |
+
"loss": 1.14,
|
| 296 |
+
"step": 4200
|
| 297 |
+
},
|
| 298 |
+
{
|
| 299 |
+
"epoch": 15.14,
|
| 300 |
+
"learning_rate": 5.252121212121211e-05,
|
| 301 |
+
"loss": 1.1508,
|
| 302 |
+
"step": 4300
|
| 303 |
+
},
|
| 304 |
+
{
|
| 305 |
+
"epoch": 15.49,
|
| 306 |
+
"learning_rate": 5.199090909090908e-05,
|
| 307 |
+
"loss": 1.1158,
|
| 308 |
+
"step": 4400
|
| 309 |
+
},
|
| 310 |
+
{
|
| 311 |
+
"epoch": 15.85,
|
| 312 |
+
"learning_rate": 5.146060606060605e-05,
|
| 313 |
+
"loss": 1.1181,
|
| 314 |
+
"step": 4500
|
| 315 |
+
},
|
| 316 |
+
{
|
| 317 |
+
"epoch": 16.2,
|
| 318 |
+
"learning_rate": 5.093030303030302e-05,
|
| 319 |
+
"loss": 1.1156,
|
| 320 |
+
"step": 4600
|
| 321 |
+
},
|
| 322 |
+
{
|
| 323 |
+
"epoch": 16.55,
|
| 324 |
+
"learning_rate": 5.039999999999999e-05,
|
| 325 |
+
"loss": 1.1205,
|
| 326 |
+
"step": 4700
|
| 327 |
+
},
|
| 328 |
+
{
|
| 329 |
+
"epoch": 16.9,
|
| 330 |
+
"learning_rate": 4.986969696969696e-05,
|
| 331 |
+
"loss": 1.1364,
|
| 332 |
+
"step": 4800
|
| 333 |
+
},
|
| 334 |
+
{
|
| 335 |
+
"epoch": 17.25,
|
| 336 |
+
"learning_rate": 4.933939393939393e-05,
|
| 337 |
+
"loss": 1.1098,
|
| 338 |
+
"step": 4900
|
| 339 |
+
},
|
| 340 |
+
{
|
| 341 |
+
"epoch": 17.61,
|
| 342 |
+
"learning_rate": 4.8809090909090903e-05,
|
| 343 |
+
"loss": 1.1132,
|
| 344 |
+
"step": 5000
|
| 345 |
+
},
|
| 346 |
+
{
|
| 347 |
+
"epoch": 17.61,
|
| 348 |
+
"eval_loss": 0.2912500500679016,
|
| 349 |
+
"eval_runtime": 113.7465,
|
| 350 |
+
"eval_samples_per_second": 30.128,
|
| 351 |
+
"eval_steps_per_second": 0.949,
|
| 352 |
+
"eval_wer": 0.3622540624086219,
|
| 353 |
+
"step": 5000
|
| 354 |
+
},
|
| 355 |
+
{
|
| 356 |
+
"epoch": 17.96,
|
| 357 |
+
"learning_rate": 4.8278787878787874e-05,
|
| 358 |
+
"loss": 1.1102,
|
| 359 |
+
"step": 5100
|
| 360 |
+
},
|
| 361 |
+
{
|
| 362 |
+
"epoch": 18.31,
|
| 363 |
+
"learning_rate": 4.7748484848484845e-05,
|
| 364 |
+
"loss": 1.12,
|
| 365 |
+
"step": 5200
|
| 366 |
+
},
|
| 367 |
+
{
|
| 368 |
+
"epoch": 18.66,
|
| 369 |
+
"learning_rate": 4.7218181818181815e-05,
|
| 370 |
+
"loss": 1.1068,
|
| 371 |
+
"step": 5300
|
| 372 |
+
},
|
| 373 |
+
{
|
| 374 |
+
"epoch": 19.01,
|
| 375 |
+
"learning_rate": 4.6687878787878786e-05,
|
| 376 |
+
"loss": 1.1125,
|
| 377 |
+
"step": 5400
|
| 378 |
+
},
|
| 379 |
+
{
|
| 380 |
+
"epoch": 19.37,
|
| 381 |
+
"learning_rate": 4.6157575757575756e-05,
|
| 382 |
+
"loss": 1.0941,
|
| 383 |
+
"step": 5500
|
| 384 |
+
},
|
| 385 |
+
{
|
| 386 |
+
"epoch": 19.72,
|
| 387 |
+
"learning_rate": 4.562727272727273e-05,
|
| 388 |
+
"loss": 1.0938,
|
| 389 |
+
"step": 5600
|
| 390 |
+
},
|
| 391 |
+
{
|
| 392 |
+
"epoch": 20.07,
|
| 393 |
+
"learning_rate": 4.50969696969697e-05,
|
| 394 |
+
"loss": 1.071,
|
| 395 |
+
"step": 5700
|
| 396 |
+
},
|
| 397 |
+
{
|
| 398 |
+
"epoch": 20.42,
|
| 399 |
+
"learning_rate": 4.456666666666667e-05,
|
| 400 |
+
"loss": 1.0878,
|
| 401 |
+
"step": 5800
|
| 402 |
+
},
|
| 403 |
+
{
|
| 404 |
+
"epoch": 20.77,
|
| 405 |
+
"learning_rate": 4.403636363636364e-05,
|
| 406 |
+
"loss": 1.0815,
|
| 407 |
+
"step": 5900
|
| 408 |
+
},
|
| 409 |
+
{
|
| 410 |
+
"epoch": 21.13,
|
| 411 |
+
"learning_rate": 4.3506060606060596e-05,
|
| 412 |
+
"loss": 1.0888,
|
| 413 |
+
"step": 6000
|
| 414 |
+
},
|
| 415 |
+
{
|
| 416 |
+
"epoch": 21.13,
|
| 417 |
+
"eval_loss": 0.28635460138320923,
|
| 418 |
+
"eval_runtime": 111.8452,
|
| 419 |
+
"eval_samples_per_second": 30.641,
|
| 420 |
+
"eval_steps_per_second": 0.966,
|
| 421 |
+
"eval_wer": 0.3498057562972555,
|
| 422 |
+
"step": 6000
|
| 423 |
+
},
|
| 424 |
+
{
|
| 425 |
+
"epoch": 21.48,
|
| 426 |
+
"learning_rate": 4.2975757575757566e-05,
|
| 427 |
+
"loss": 1.0875,
|
| 428 |
+
"step": 6100
|
| 429 |
+
},
|
| 430 |
+
{
|
| 431 |
+
"epoch": 21.83,
|
| 432 |
+
"learning_rate": 4.244545454545454e-05,
|
| 433 |
+
"loss": 1.0609,
|
| 434 |
+
"step": 6200
|
| 435 |
+
},
|
| 436 |
+
{
|
| 437 |
+
"epoch": 22.18,
|
| 438 |
+
"learning_rate": 4.191515151515151e-05,
|
| 439 |
+
"loss": 1.0796,
|
| 440 |
+
"step": 6300
|
| 441 |
+
},
|
| 442 |
+
{
|
| 443 |
+
"epoch": 22.54,
|
| 444 |
+
"learning_rate": 4.1390151515151515e-05,
|
| 445 |
+
"loss": 1.0701,
|
| 446 |
+
"step": 6400
|
| 447 |
+
},
|
| 448 |
+
{
|
| 449 |
+
"epoch": 22.89,
|
| 450 |
+
"learning_rate": 4.0859848484848485e-05,
|
| 451 |
+
"loss": 1.0633,
|
| 452 |
+
"step": 6500
|
| 453 |
+
},
|
| 454 |
+
{
|
| 455 |
+
"epoch": 23.24,
|
| 456 |
+
"learning_rate": 4.0329545454545456e-05,
|
| 457 |
+
"loss": 1.0491,
|
| 458 |
+
"step": 6600
|
| 459 |
+
},
|
| 460 |
+
{
|
| 461 |
+
"epoch": 23.59,
|
| 462 |
+
"learning_rate": 3.9799242424242427e-05,
|
| 463 |
+
"loss": 1.0673,
|
| 464 |
+
"step": 6700
|
| 465 |
+
},
|
| 466 |
+
{
|
| 467 |
+
"epoch": 23.94,
|
| 468 |
+
"learning_rate": 3.927424242424242e-05,
|
| 469 |
+
"loss": 1.0574,
|
| 470 |
+
"step": 6800
|
| 471 |
+
},
|
| 472 |
+
{
|
| 473 |
+
"epoch": 24.3,
|
| 474 |
+
"learning_rate": 3.874393939393939e-05,
|
| 475 |
+
"loss": 1.0371,
|
| 476 |
+
"step": 6900
|
| 477 |
+
},
|
| 478 |
+
{
|
| 479 |
+
"epoch": 24.65,
|
| 480 |
+
"learning_rate": 3.821363636363636e-05,
|
| 481 |
+
"loss": 1.0487,
|
| 482 |
+
"step": 7000
|
| 483 |
+
},
|
| 484 |
+
{
|
| 485 |
+
"epoch": 24.65,
|
| 486 |
+
"eval_loss": 0.28213828802108765,
|
| 487 |
+
"eval_runtime": 109.6819,
|
| 488 |
+
"eval_samples_per_second": 31.245,
|
| 489 |
+
"eval_steps_per_second": 0.985,
|
| 490 |
+
"eval_wer": 0.3434562847236727,
|
| 491 |
+
"step": 7000
|
| 492 |
+
},
|
| 493 |
+
{
|
| 494 |
+
"epoch": 25.0,
|
| 495 |
+
"learning_rate": 3.768333333333333e-05,
|
| 496 |
+
"loss": 1.0749,
|
| 497 |
+
"step": 7100
|
| 498 |
+
},
|
| 499 |
+
{
|
| 500 |
+
"epoch": 25.35,
|
| 501 |
+
"learning_rate": 3.71530303030303e-05,
|
| 502 |
+
"loss": 1.0492,
|
| 503 |
+
"step": 7200
|
| 504 |
+
},
|
| 505 |
+
{
|
| 506 |
+
"epoch": 25.7,
|
| 507 |
+
"learning_rate": 3.662272727272727e-05,
|
| 508 |
+
"loss": 1.0189,
|
| 509 |
+
"step": 7300
|
| 510 |
+
},
|
| 511 |
+
{
|
| 512 |
+
"epoch": 26.06,
|
| 513 |
+
"learning_rate": 3.609242424242424e-05,
|
| 514 |
+
"loss": 1.0428,
|
| 515 |
+
"step": 7400
|
| 516 |
+
},
|
| 517 |
+
{
|
| 518 |
+
"epoch": 26.41,
|
| 519 |
+
"learning_rate": 3.556212121212121e-05,
|
| 520 |
+
"loss": 1.0372,
|
| 521 |
+
"step": 7500
|
| 522 |
+
},
|
| 523 |
+
{
|
| 524 |
+
"epoch": 26.76,
|
| 525 |
+
"learning_rate": 3.503181818181818e-05,
|
| 526 |
+
"loss": 1.0324,
|
| 527 |
+
"step": 7600
|
| 528 |
+
},
|
| 529 |
+
{
|
| 530 |
+
"epoch": 27.11,
|
| 531 |
+
"learning_rate": 3.450151515151515e-05,
|
| 532 |
+
"loss": 1.0244,
|
| 533 |
+
"step": 7700
|
| 534 |
+
},
|
| 535 |
+
{
|
| 536 |
+
"epoch": 27.46,
|
| 537 |
+
"learning_rate": 3.397121212121212e-05,
|
| 538 |
+
"loss": 1.0217,
|
| 539 |
+
"step": 7800
|
| 540 |
+
},
|
| 541 |
+
{
|
| 542 |
+
"epoch": 27.82,
|
| 543 |
+
"learning_rate": 3.344090909090909e-05,
|
| 544 |
+
"loss": 1.024,
|
| 545 |
+
"step": 7900
|
| 546 |
+
},
|
| 547 |
+
{
|
| 548 |
+
"epoch": 28.17,
|
| 549 |
+
"learning_rate": 3.291060606060606e-05,
|
| 550 |
+
"loss": 1.0431,
|
| 551 |
+
"step": 8000
|
| 552 |
+
},
|
| 553 |
+
{
|
| 554 |
+
"epoch": 28.17,
|
| 555 |
+
"eval_loss": 0.2738790810108185,
|
| 556 |
+
"eval_runtime": 112.4344,
|
| 557 |
+
"eval_samples_per_second": 30.48,
|
| 558 |
+
"eval_steps_per_second": 0.961,
|
| 559 |
+
"eval_wer": 0.33075734157650694,
|
| 560 |
+
"step": 8000
|
| 561 |
+
},
|
| 562 |
+
{
|
| 563 |
+
"epoch": 28.52,
|
| 564 |
+
"learning_rate": 3.238030303030303e-05,
|
| 565 |
+
"loss": 1.0062,
|
| 566 |
+
"step": 8100
|
| 567 |
+
},
|
| 568 |
+
{
|
| 569 |
+
"epoch": 28.87,
|
| 570 |
+
"learning_rate": 3.185e-05,
|
| 571 |
+
"loss": 1.0167,
|
| 572 |
+
"step": 8200
|
| 573 |
+
},
|
| 574 |
+
{
|
| 575 |
+
"epoch": 29.23,
|
| 576 |
+
"learning_rate": 3.1319696969696965e-05,
|
| 577 |
+
"loss": 1.0051,
|
| 578 |
+
"step": 8300
|
| 579 |
+
},
|
| 580 |
+
{
|
| 581 |
+
"epoch": 29.58,
|
| 582 |
+
"learning_rate": 3.0789393939393936e-05,
|
| 583 |
+
"loss": 1.015,
|
| 584 |
+
"step": 8400
|
| 585 |
+
},
|
| 586 |
+
{
|
| 587 |
+
"epoch": 29.93,
|
| 588 |
+
"learning_rate": 3.0259090909090907e-05,
|
| 589 |
+
"loss": 1.0221,
|
| 590 |
+
"step": 8500
|
| 591 |
+
},
|
| 592 |
+
{
|
| 593 |
+
"epoch": 30.28,
|
| 594 |
+
"learning_rate": 2.9728787878787874e-05,
|
| 595 |
+
"loss": 0.9987,
|
| 596 |
+
"step": 8600
|
| 597 |
+
},
|
| 598 |
+
{
|
| 599 |
+
"epoch": 30.63,
|
| 600 |
+
"learning_rate": 2.9198484848484844e-05,
|
| 601 |
+
"loss": 1.0008,
|
| 602 |
+
"step": 8700
|
| 603 |
+
},
|
| 604 |
+
{
|
| 605 |
+
"epoch": 30.99,
|
| 606 |
+
"learning_rate": 2.8668181818181815e-05,
|
| 607 |
+
"loss": 0.9992,
|
| 608 |
+
"step": 8800
|
| 609 |
+
},
|
| 610 |
+
{
|
| 611 |
+
"epoch": 31.34,
|
| 612 |
+
"learning_rate": 2.8143181818181816e-05,
|
| 613 |
+
"loss": 1.0122,
|
| 614 |
+
"step": 8900
|
| 615 |
+
},
|
| 616 |
+
{
|
| 617 |
+
"epoch": 31.69,
|
| 618 |
+
"learning_rate": 2.7612878787878786e-05,
|
| 619 |
+
"loss": 0.9896,
|
| 620 |
+
"step": 9000
|
| 621 |
+
},
|
| 622 |
+
{
|
| 623 |
+
"epoch": 31.69,
|
| 624 |
+
"eval_loss": 0.2629273533821106,
|
| 625 |
+
"eval_runtime": 110.3673,
|
| 626 |
+
"eval_samples_per_second": 31.051,
|
| 627 |
+
"eval_steps_per_second": 0.979,
|
| 628 |
+
"eval_wer": 0.32432432432432434,
|
| 629 |
+
"step": 9000
|
| 630 |
+
},
|
| 631 |
+
{
|
| 632 |
+
"epoch": 32.04,
|
| 633 |
+
"learning_rate": 2.7082575757575757e-05,
|
| 634 |
+
"loss": 0.9994,
|
| 635 |
+
"step": 9100
|
| 636 |
+
},
|
| 637 |
+
{
|
| 638 |
+
"epoch": 32.39,
|
| 639 |
+
"learning_rate": 2.6552272727272727e-05,
|
| 640 |
+
"loss": 0.9966,
|
| 641 |
+
"step": 9200
|
| 642 |
+
},
|
| 643 |
+
{
|
| 644 |
+
"epoch": 32.75,
|
| 645 |
+
"learning_rate": 2.6021969696969695e-05,
|
| 646 |
+
"loss": 0.9905,
|
| 647 |
+
"step": 9300
|
| 648 |
+
},
|
| 649 |
+
{
|
| 650 |
+
"epoch": 33.1,
|
| 651 |
+
"learning_rate": 2.5491666666666665e-05,
|
| 652 |
+
"loss": 0.9819,
|
| 653 |
+
"step": 9400
|
| 654 |
+
},
|
| 655 |
+
{
|
| 656 |
+
"epoch": 33.45,
|
| 657 |
+
"learning_rate": 2.4966666666666666e-05,
|
| 658 |
+
"loss": 1.0018,
|
| 659 |
+
"step": 9500
|
| 660 |
+
},
|
| 661 |
+
{
|
| 662 |
+
"epoch": 33.8,
|
| 663 |
+
"learning_rate": 2.4436363636363636e-05,
|
| 664 |
+
"loss": 0.9881,
|
| 665 |
+
"step": 9600
|
| 666 |
+
},
|
| 667 |
+
{
|
| 668 |
+
"epoch": 34.15,
|
| 669 |
+
"learning_rate": 2.3906060606060604e-05,
|
| 670 |
+
"loss": 0.9939,
|
| 671 |
+
"step": 9700
|
| 672 |
+
},
|
| 673 |
+
{
|
| 674 |
+
"epoch": 34.51,
|
| 675 |
+
"learning_rate": 2.3375757575757574e-05,
|
| 676 |
+
"loss": 0.9874,
|
| 677 |
+
"step": 9800
|
| 678 |
+
},
|
| 679 |
+
{
|
| 680 |
+
"epoch": 34.86,
|
| 681 |
+
"learning_rate": 2.284545454545454e-05,
|
| 682 |
+
"loss": 0.9617,
|
| 683 |
+
"step": 9900
|
| 684 |
+
},
|
| 685 |
+
{
|
| 686 |
+
"epoch": 35.21,
|
| 687 |
+
"learning_rate": 2.2315151515151512e-05,
|
| 688 |
+
"loss": 0.9839,
|
| 689 |
+
"step": 10000
|
| 690 |
+
},
|
| 691 |
+
{
|
| 692 |
+
"epoch": 35.21,
|
| 693 |
+
"eval_loss": 0.28058063983917236,
|
| 694 |
+
"eval_runtime": 111.4323,
|
| 695 |
+
"eval_samples_per_second": 30.754,
|
| 696 |
+
"eval_steps_per_second": 0.969,
|
| 697 |
+
"eval_wer": 0.33079911441580684,
|
| 698 |
+
"step": 10000
|
| 699 |
+
},
|
| 700 |
+
{
|
| 701 |
+
"epoch": 35.56,
|
| 702 |
+
"learning_rate": 2.1784848484848483e-05,
|
| 703 |
+
"loss": 0.9761,
|
| 704 |
+
"step": 10100
|
| 705 |
+
},
|
| 706 |
+
{
|
| 707 |
+
"epoch": 35.92,
|
| 708 |
+
"learning_rate": 2.1254545454545453e-05,
|
| 709 |
+
"loss": 0.9892,
|
| 710 |
+
"step": 10200
|
| 711 |
+
},
|
| 712 |
+
{
|
| 713 |
+
"epoch": 36.27,
|
| 714 |
+
"learning_rate": 2.0724242424242424e-05,
|
| 715 |
+
"loss": 0.9712,
|
| 716 |
+
"step": 10300
|
| 717 |
+
},
|
| 718 |
+
{
|
| 719 |
+
"epoch": 36.62,
|
| 720 |
+
"learning_rate": 2.0193939393939394e-05,
|
| 721 |
+
"loss": 0.9725,
|
| 722 |
+
"step": 10400
|
| 723 |
+
},
|
| 724 |
+
{
|
| 725 |
+
"epoch": 36.97,
|
| 726 |
+
"learning_rate": 1.966363636363636e-05,
|
| 727 |
+
"loss": 0.9738,
|
| 728 |
+
"step": 10500
|
| 729 |
+
},
|
| 730 |
+
{
|
| 731 |
+
"epoch": 37.32,
|
| 732 |
+
"learning_rate": 1.9133333333333332e-05,
|
| 733 |
+
"loss": 0.9622,
|
| 734 |
+
"step": 10600
|
| 735 |
+
},
|
| 736 |
+
{
|
| 737 |
+
"epoch": 37.68,
|
| 738 |
+
"learning_rate": 1.86030303030303e-05,
|
| 739 |
+
"loss": 0.9612,
|
| 740 |
+
"step": 10700
|
| 741 |
+
},
|
| 742 |
+
{
|
| 743 |
+
"epoch": 38.03,
|
| 744 |
+
"learning_rate": 1.807272727272727e-05,
|
| 745 |
+
"loss": 0.9645,
|
| 746 |
+
"step": 10800
|
| 747 |
+
},
|
| 748 |
+
{
|
| 749 |
+
"epoch": 38.38,
|
| 750 |
+
"learning_rate": 1.754242424242424e-05,
|
| 751 |
+
"loss": 0.9703,
|
| 752 |
+
"step": 10900
|
| 753 |
+
},
|
| 754 |
+
{
|
| 755 |
+
"epoch": 38.73,
|
| 756 |
+
"learning_rate": 1.701212121212121e-05,
|
| 757 |
+
"loss": 0.9586,
|
| 758 |
+
"step": 11000
|
| 759 |
+
},
|
| 760 |
+
{
|
| 761 |
+
"epoch": 38.73,
|
| 762 |
+
"eval_loss": 0.26495224237442017,
|
| 763 |
+
"eval_runtime": 115.232,
|
| 764 |
+
"eval_samples_per_second": 29.74,
|
| 765 |
+
"eval_steps_per_second": 0.937,
|
| 766 |
+
"eval_wer": 0.32353064037762647,
|
| 767 |
+
"step": 11000
|
| 768 |
+
},
|
| 769 |
+
{
|
| 770 |
+
"epoch": 39.08,
|
| 771 |
+
"learning_rate": 1.6481818181818178e-05,
|
| 772 |
+
"loss": 0.9597,
|
| 773 |
+
"step": 11100
|
| 774 |
+
},
|
| 775 |
+
{
|
| 776 |
+
"epoch": 39.44,
|
| 777 |
+
"learning_rate": 1.595151515151515e-05,
|
| 778 |
+
"loss": 0.9617,
|
| 779 |
+
"step": 11200
|
| 780 |
+
},
|
| 781 |
+
{
|
| 782 |
+
"epoch": 39.79,
|
| 783 |
+
"learning_rate": 1.542121212121212e-05,
|
| 784 |
+
"loss": 0.9615,
|
| 785 |
+
"step": 11300
|
| 786 |
+
},
|
| 787 |
+
{
|
| 788 |
+
"epoch": 40.14,
|
| 789 |
+
"learning_rate": 1.489090909090909e-05,
|
| 790 |
+
"loss": 0.9476,
|
| 791 |
+
"step": 11400
|
| 792 |
+
},
|
| 793 |
+
{
|
| 794 |
+
"epoch": 40.49,
|
| 795 |
+
"learning_rate": 1.4360606060606059e-05,
|
| 796 |
+
"loss": 0.958,
|
| 797 |
+
"step": 11500
|
| 798 |
+
},
|
| 799 |
+
{
|
| 800 |
+
"epoch": 40.85,
|
| 801 |
+
"learning_rate": 1.3830303030303029e-05,
|
| 802 |
+
"loss": 0.9515,
|
| 803 |
+
"step": 11600
|
| 804 |
+
},
|
| 805 |
+
{
|
| 806 |
+
"epoch": 41.2,
|
| 807 |
+
"learning_rate": 1.33e-05,
|
| 808 |
+
"loss": 0.9525,
|
| 809 |
+
"step": 11700
|
| 810 |
+
},
|
| 811 |
+
{
|
| 812 |
+
"epoch": 41.55,
|
| 813 |
+
"learning_rate": 1.2769696969696969e-05,
|
| 814 |
+
"loss": 0.9484,
|
| 815 |
+
"step": 11800
|
| 816 |
+
},
|
| 817 |
+
{
|
| 818 |
+
"epoch": 41.9,
|
| 819 |
+
"learning_rate": 1.2239393939393937e-05,
|
| 820 |
+
"loss": 0.9326,
|
| 821 |
+
"step": 11900
|
| 822 |
+
},
|
| 823 |
+
{
|
| 824 |
+
"epoch": 42.25,
|
| 825 |
+
"learning_rate": 1.1719696969696968e-05,
|
| 826 |
+
"loss": 0.9501,
|
| 827 |
+
"step": 12000
|
| 828 |
+
},
|
| 829 |
+
{
|
| 830 |
+
"epoch": 42.25,
|
| 831 |
+
"eval_loss": 0.25854310393333435,
|
| 832 |
+
"eval_runtime": 111.5579,
|
| 833 |
+
"eval_samples_per_second": 30.719,
|
| 834 |
+
"eval_steps_per_second": 0.968,
|
| 835 |
+
"eval_wer": 0.3173064873219433,
|
| 836 |
+
"step": 12000
|
| 837 |
+
},
|
| 838 |
+
{
|
| 839 |
+
"epoch": 42.61,
|
| 840 |
+
"learning_rate": 1.1189393939393939e-05,
|
| 841 |
+
"loss": 0.9395,
|
| 842 |
+
"step": 12100
|
| 843 |
+
},
|
| 844 |
+
{
|
| 845 |
+
"epoch": 42.96,
|
| 846 |
+
"learning_rate": 1.065909090909091e-05,
|
| 847 |
+
"loss": 0.9519,
|
| 848 |
+
"step": 12200
|
| 849 |
+
},
|
| 850 |
+
{
|
| 851 |
+
"epoch": 43.31,
|
| 852 |
+
"learning_rate": 1.0128787878787877e-05,
|
| 853 |
+
"loss": 0.9445,
|
| 854 |
+
"step": 12300
|
| 855 |
+
},
|
| 856 |
+
{
|
| 857 |
+
"epoch": 43.66,
|
| 858 |
+
"learning_rate": 9.598484848484847e-06,
|
| 859 |
+
"loss": 0.9293,
|
| 860 |
+
"step": 12400
|
| 861 |
+
},
|
| 862 |
+
{
|
| 863 |
+
"epoch": 44.01,
|
| 864 |
+
"learning_rate": 9.068181818181818e-06,
|
| 865 |
+
"loss": 0.9421,
|
| 866 |
+
"step": 12500
|
| 867 |
+
},
|
| 868 |
+
{
|
| 869 |
+
"epoch": 44.37,
|
| 870 |
+
"learning_rate": 8.537878787878787e-06,
|
| 871 |
+
"loss": 0.9305,
|
| 872 |
+
"step": 12600
|
| 873 |
+
},
|
| 874 |
+
{
|
| 875 |
+
"epoch": 44.72,
|
| 876 |
+
"learning_rate": 8.007575757575757e-06,
|
| 877 |
+
"loss": 0.9305,
|
| 878 |
+
"step": 12700
|
| 879 |
+
},
|
| 880 |
+
{
|
| 881 |
+
"epoch": 45.07,
|
| 882 |
+
"learning_rate": 7.477272727272726e-06,
|
| 883 |
+
"loss": 0.9238,
|
| 884 |
+
"step": 12800
|
| 885 |
+
},
|
| 886 |
+
{
|
| 887 |
+
"epoch": 45.42,
|
| 888 |
+
"learning_rate": 6.946969696969697e-06,
|
| 889 |
+
"loss": 0.9291,
|
| 890 |
+
"step": 12900
|
| 891 |
+
},
|
| 892 |
+
{
|
| 893 |
+
"epoch": 45.77,
|
| 894 |
+
"learning_rate": 6.416666666666666e-06,
|
| 895 |
+
"loss": 0.938,
|
| 896 |
+
"step": 13000
|
| 897 |
+
},
|
| 898 |
+
{
|
| 899 |
+
"epoch": 45.77,
|
| 900 |
+
"eval_loss": 0.2560683786869049,
|
| 901 |
+
"eval_runtime": 110.8197,
|
| 902 |
+
"eval_samples_per_second": 30.924,
|
| 903 |
+
"eval_steps_per_second": 0.975,
|
| 904 |
+
"eval_wer": 0.3117089268557584,
|
| 905 |
+
"step": 13000
|
| 906 |
+
},
|
| 907 |
+
{
|
| 908 |
+
"epoch": 46.13,
|
| 909 |
+
"learning_rate": 5.886363636363636e-06,
|
| 910 |
+
"loss": 0.9387,
|
| 911 |
+
"step": 13100
|
| 912 |
+
},
|
| 913 |
+
{
|
| 914 |
+
"epoch": 46.48,
|
| 915 |
+
"learning_rate": 5.356060606060605e-06,
|
| 916 |
+
"loss": 0.9103,
|
| 917 |
+
"step": 13200
|
| 918 |
+
},
|
| 919 |
+
{
|
| 920 |
+
"epoch": 46.83,
|
| 921 |
+
"learning_rate": 4.825757575757576e-06,
|
| 922 |
+
"loss": 0.9283,
|
| 923 |
+
"step": 13300
|
| 924 |
+
},
|
| 925 |
+
{
|
| 926 |
+
"epoch": 47.18,
|
| 927 |
+
"learning_rate": 4.295454545454545e-06,
|
| 928 |
+
"loss": 0.9282,
|
| 929 |
+
"step": 13400
|
| 930 |
+
},
|
| 931 |
+
{
|
| 932 |
+
"epoch": 47.54,
|
| 933 |
+
"learning_rate": 3.7651515151515147e-06,
|
| 934 |
+
"loss": 0.9168,
|
| 935 |
+
"step": 13500
|
| 936 |
+
},
|
| 937 |
+
{
|
| 938 |
+
"epoch": 47.89,
|
| 939 |
+
"learning_rate": 3.2348484848484844e-06,
|
| 940 |
+
"loss": 0.925,
|
| 941 |
+
"step": 13600
|
| 942 |
+
},
|
| 943 |
+
{
|
| 944 |
+
"epoch": 48.24,
|
| 945 |
+
"learning_rate": 2.704545454545454e-06,
|
| 946 |
+
"loss": 0.928,
|
| 947 |
+
"step": 13700
|
| 948 |
+
},
|
| 949 |
+
{
|
| 950 |
+
"epoch": 48.59,
|
| 951 |
+
"learning_rate": 2.174242424242424e-06,
|
| 952 |
+
"loss": 0.9301,
|
| 953 |
+
"step": 13800
|
| 954 |
+
},
|
| 955 |
+
{
|
| 956 |
+
"epoch": 48.94,
|
| 957 |
+
"learning_rate": 1.6439393939393937e-06,
|
| 958 |
+
"loss": 0.9404,
|
| 959 |
+
"step": 13900
|
| 960 |
+
},
|
| 961 |
+
{
|
| 962 |
+
"epoch": 49.3,
|
| 963 |
+
"learning_rate": 1.1136363636363635e-06,
|
| 964 |
+
"loss": 0.921,
|
| 965 |
+
"step": 14000
|
| 966 |
+
},
|
| 967 |
+
{
|
| 968 |
+
"epoch": 49.3,
|
| 969 |
+
"eval_loss": 0.2559361755847931,
|
| 970 |
+
"eval_runtime": 110.0973,
|
| 971 |
+
"eval_samples_per_second": 31.127,
|
| 972 |
+
"eval_steps_per_second": 0.981,
|
| 973 |
+
"eval_wer": 0.31154183549855885,
|
| 974 |
+
"step": 14000
|
| 975 |
+
},
|
| 976 |
+
{
|
| 977 |
+
"epoch": 49.65,
|
| 978 |
+
"learning_rate": 5.833333333333333e-07,
|
| 979 |
+
"loss": 0.9243,
|
| 980 |
+
"step": 14100
|
| 981 |
+
},
|
| 982 |
+
{
|
| 983 |
+
"epoch": 50.0,
|
| 984 |
+
"learning_rate": 5.303030303030303e-08,
|
| 985 |
+
"loss": 0.9223,
|
| 986 |
+
"step": 14200
|
| 987 |
+
},
|
| 988 |
+
{
|
| 989 |
+
"epoch": 50.0,
|
| 990 |
+
"step": 14200,
|
| 991 |
+
"total_flos": 5.595667068571877e+19,
|
| 992 |
+
"train_loss": 1.318804485697142,
|
| 993 |
+
"train_runtime": 22379.119,
|
| 994 |
+
"train_samples_per_second": 20.269,
|
| 995 |
+
"train_steps_per_second": 0.635
|
| 996 |
+
}
|
| 997 |
+
],
|
| 998 |
+
"max_steps": 14200,
|
| 999 |
+
"num_train_epochs": 50,
|
| 1000 |
+
"total_flos": 5.595667068571877e+19,
|
| 1001 |
+
"trial_name": null,
|
| 1002 |
+
"trial_params": null
|
| 1003 |
+
}
|