End of training
Browse files- README.md +5 -3
- all_results.json +15 -0
- eval_results.json +9 -0
- train_results.json +9 -0
- trainer_state.json +3330 -0
README.md
CHANGED
|
@@ -3,6 +3,8 @@ library_name: transformers
|
|
| 3 |
license: apache-2.0
|
| 4 |
base_model: facebook/wav2vec2-xls-r-300m
|
| 5 |
tags:
|
|
|
|
|
|
|
| 6 |
- generated_from_trainer
|
| 7 |
metrics:
|
| 8 |
- wer
|
|
@@ -16,10 +18,10 @@ should probably proofread and complete it, then remove this comment. -->
|
|
| 16 |
|
| 17 |
# Wav2vec2-fula
|
| 18 |
|
| 19 |
-
This model is a fine-tuned version of [facebook/wav2vec2-xls-r-300m](https://huggingface.co/facebook/wav2vec2-xls-r-300m) on
|
| 20 |
It achieves the following results on the evaluation set:
|
| 21 |
-
- Loss: 0.
|
| 22 |
-
- Wer: 0.
|
| 23 |
|
| 24 |
## Model description
|
| 25 |
|
|
|
|
| 3 |
license: apache-2.0
|
| 4 |
base_model: facebook/wav2vec2-xls-r-300m
|
| 5 |
tags:
|
| 6 |
+
- automatic-speech-recognition
|
| 7 |
+
- LAfricaMobile/fulfulde
|
| 8 |
- generated_from_trainer
|
| 9 |
metrics:
|
| 10 |
- wer
|
|
|
|
| 18 |
|
| 19 |
# Wav2vec2-fula
|
| 20 |
|
| 21 |
+
This model is a fine-tuned version of [facebook/wav2vec2-xls-r-300m](https://huggingface.co/facebook/wav2vec2-xls-r-300m) on the LAFRICAMOBILE/FULFULDE - DEFAULT dataset.
|
| 22 |
It achieves the following results on the evaluation set:
|
| 23 |
+
- Loss: 0.3143
|
| 24 |
+
- Wer: 0.5455
|
| 25 |
|
| 26 |
## Model description
|
| 27 |
|
all_results.json
ADDED
|
@@ -0,0 +1,15 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"epoch": 7.565045540470871,
|
| 3 |
+
"eval_loss": 0.3142752945423126,
|
| 4 |
+
"eval_runtime": 150.4483,
|
| 5 |
+
"eval_samples": 1225,
|
| 6 |
+
"eval_samples_per_second": 8.142,
|
| 7 |
+
"eval_steps_per_second": 1.024,
|
| 8 |
+
"eval_wer": 0.5454641797371768,
|
| 9 |
+
"total_flos": 8.071230701734987e+19,
|
| 10 |
+
"train_loss": 0.7137202631343494,
|
| 11 |
+
"train_runtime": 97593.296,
|
| 12 |
+
"train_samples": 46550,
|
| 13 |
+
"train_samples_per_second": 28.619,
|
| 14 |
+
"train_steps_per_second": 0.894
|
| 15 |
+
}
|
eval_results.json
ADDED
|
@@ -0,0 +1,9 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"epoch": 7.565045540470871,
|
| 3 |
+
"eval_loss": 0.3142752945423126,
|
| 4 |
+
"eval_runtime": 150.4483,
|
| 5 |
+
"eval_samples": 1225,
|
| 6 |
+
"eval_samples_per_second": 8.142,
|
| 7 |
+
"eval_steps_per_second": 1.024,
|
| 8 |
+
"eval_wer": 0.5454641797371768
|
| 9 |
+
}
|
train_results.json
ADDED
|
@@ -0,0 +1,9 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"epoch": 7.565045540470871,
|
| 3 |
+
"total_flos": 8.071230701734987e+19,
|
| 4 |
+
"train_loss": 0.7137202631343494,
|
| 5 |
+
"train_runtime": 97593.296,
|
| 6 |
+
"train_samples": 46550,
|
| 7 |
+
"train_samples_per_second": 28.619,
|
| 8 |
+
"train_steps_per_second": 0.894
|
| 9 |
+
}
|
trainer_state.json
ADDED
|
@@ -0,0 +1,3330 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"best_global_step": 8500,
|
| 3 |
+
"best_metric": 0.3142754137516022,
|
| 4 |
+
"best_model_checkpoint": "./Wav2vec2-fula/checkpoint-8500",
|
| 5 |
+
"epoch": 7.565045540470871,
|
| 6 |
+
"eval_steps": 500,
|
| 7 |
+
"global_step": 11000,
|
| 8 |
+
"is_hyper_param_search": false,
|
| 9 |
+
"is_local_process_zero": true,
|
| 10 |
+
"is_world_process_zero": true,
|
| 11 |
+
"log_history": [
|
| 12 |
+
{
|
| 13 |
+
"epoch": 0.017185083347654236,
|
| 14 |
+
"grad_norm": 10.79065990447998,
|
| 15 |
+
"learning_rate": 1.2000000000000002e-06,
|
| 16 |
+
"loss": 14.779,
|
| 17 |
+
"step": 25
|
| 18 |
+
},
|
| 19 |
+
{
|
| 20 |
+
"epoch": 0.03437016669530847,
|
| 21 |
+
"grad_norm": 11.638880729675293,
|
| 22 |
+
"learning_rate": 2.4000000000000003e-06,
|
| 23 |
+
"loss": 13.4402,
|
| 24 |
+
"step": 50
|
| 25 |
+
},
|
| 26 |
+
{
|
| 27 |
+
"epoch": 0.05155525004296271,
|
| 28 |
+
"grad_norm": 12.49268913269043,
|
| 29 |
+
"learning_rate": 3.6499999999999998e-06,
|
| 30 |
+
"loss": 14.1815,
|
| 31 |
+
"step": 75
|
| 32 |
+
},
|
| 33 |
+
{
|
| 34 |
+
"epoch": 0.06874033339061694,
|
| 35 |
+
"grad_norm": 13.672736167907715,
|
| 36 |
+
"learning_rate": 4.9000000000000005e-06,
|
| 37 |
+
"loss": 12.3835,
|
| 38 |
+
"step": 100
|
| 39 |
+
},
|
| 40 |
+
{
|
| 41 |
+
"epoch": 0.08592541673827118,
|
| 42 |
+
"grad_norm": 13.192852973937988,
|
| 43 |
+
"learning_rate": 6.15e-06,
|
| 44 |
+
"loss": 11.3532,
|
| 45 |
+
"step": 125
|
| 46 |
+
},
|
| 47 |
+
{
|
| 48 |
+
"epoch": 0.10311050008592541,
|
| 49 |
+
"grad_norm": 13.134611129760742,
|
| 50 |
+
"learning_rate": 7.4e-06,
|
| 51 |
+
"loss": 7.4199,
|
| 52 |
+
"step": 150
|
| 53 |
+
},
|
| 54 |
+
{
|
| 55 |
+
"epoch": 0.12029558343357966,
|
| 56 |
+
"grad_norm": 10.945942878723145,
|
| 57 |
+
"learning_rate": 8.65e-06,
|
| 58 |
+
"loss": 6.3555,
|
| 59 |
+
"step": 175
|
| 60 |
+
},
|
| 61 |
+
{
|
| 62 |
+
"epoch": 0.13748066678123388,
|
| 63 |
+
"grad_norm": 10.094327926635742,
|
| 64 |
+
"learning_rate": 9.900000000000002e-06,
|
| 65 |
+
"loss": 5.2388,
|
| 66 |
+
"step": 200
|
| 67 |
+
},
|
| 68 |
+
{
|
| 69 |
+
"epoch": 0.1546657501288881,
|
| 70 |
+
"grad_norm": 9.148414611816406,
|
| 71 |
+
"learning_rate": 1.115e-05,
|
| 72 |
+
"loss": 5.0771,
|
| 73 |
+
"step": 225
|
| 74 |
+
},
|
| 75 |
+
{
|
| 76 |
+
"epoch": 0.17185083347654237,
|
| 77 |
+
"grad_norm": 7.876718997955322,
|
| 78 |
+
"learning_rate": 1.24e-05,
|
| 79 |
+
"loss": 4.6181,
|
| 80 |
+
"step": 250
|
| 81 |
+
},
|
| 82 |
+
{
|
| 83 |
+
"epoch": 0.1890359168241966,
|
| 84 |
+
"grad_norm": 7.483435153961182,
|
| 85 |
+
"learning_rate": 1.3650000000000001e-05,
|
| 86 |
+
"loss": 4.5182,
|
| 87 |
+
"step": 275
|
| 88 |
+
},
|
| 89 |
+
{
|
| 90 |
+
"epoch": 0.20622100017185083,
|
| 91 |
+
"grad_norm": 6.882397174835205,
|
| 92 |
+
"learning_rate": 1.49e-05,
|
| 93 |
+
"loss": 4.1879,
|
| 94 |
+
"step": 300
|
| 95 |
+
},
|
| 96 |
+
{
|
| 97 |
+
"epoch": 0.22340608351950508,
|
| 98 |
+
"grad_norm": 6.444328308105469,
|
| 99 |
+
"learning_rate": 1.6150000000000003e-05,
|
| 100 |
+
"loss": 4.1217,
|
| 101 |
+
"step": 325
|
| 102 |
+
},
|
| 103 |
+
{
|
| 104 |
+
"epoch": 0.2405911668671593,
|
| 105 |
+
"grad_norm": 4.407646179199219,
|
| 106 |
+
"learning_rate": 1.74e-05,
|
| 107 |
+
"loss": 3.8535,
|
| 108 |
+
"step": 350
|
| 109 |
+
},
|
| 110 |
+
{
|
| 111 |
+
"epoch": 0.25777625021481354,
|
| 112 |
+
"grad_norm": 4.332294940948486,
|
| 113 |
+
"learning_rate": 1.865e-05,
|
| 114 |
+
"loss": 3.7725,
|
| 115 |
+
"step": 375
|
| 116 |
+
},
|
| 117 |
+
{
|
| 118 |
+
"epoch": 0.27496133356246777,
|
| 119 |
+
"grad_norm": 3.5310115814208984,
|
| 120 |
+
"learning_rate": 1.9900000000000003e-05,
|
| 121 |
+
"loss": 3.5538,
|
| 122 |
+
"step": 400
|
| 123 |
+
},
|
| 124 |
+
{
|
| 125 |
+
"epoch": 0.292146416910122,
|
| 126 |
+
"grad_norm": 2.884195566177368,
|
| 127 |
+
"learning_rate": 2.115e-05,
|
| 128 |
+
"loss": 3.4667,
|
| 129 |
+
"step": 425
|
| 130 |
+
},
|
| 131 |
+
{
|
| 132 |
+
"epoch": 0.3093315002577762,
|
| 133 |
+
"grad_norm": 2.221975564956665,
|
| 134 |
+
"learning_rate": 2.2400000000000002e-05,
|
| 135 |
+
"loss": 3.3157,
|
| 136 |
+
"step": 450
|
| 137 |
+
},
|
| 138 |
+
{
|
| 139 |
+
"epoch": 0.3265165836054305,
|
| 140 |
+
"grad_norm": 1.393004059791565,
|
| 141 |
+
"learning_rate": 2.365e-05,
|
| 142 |
+
"loss": 3.2574,
|
| 143 |
+
"step": 475
|
| 144 |
+
},
|
| 145 |
+
{
|
| 146 |
+
"epoch": 0.34370166695308474,
|
| 147 |
+
"grad_norm": 1.5566725730895996,
|
| 148 |
+
"learning_rate": 2.4900000000000002e-05,
|
| 149 |
+
"loss": 3.1344,
|
| 150 |
+
"step": 500
|
| 151 |
+
},
|
| 152 |
+
{
|
| 153 |
+
"epoch": 0.34370166695308474,
|
| 154 |
+
"eval_loss": 3.093510150909424,
|
| 155 |
+
"eval_runtime": 148.9649,
|
| 156 |
+
"eval_samples_per_second": 8.223,
|
| 157 |
+
"eval_steps_per_second": 1.034,
|
| 158 |
+
"eval_wer": 1.0,
|
| 159 |
+
"step": 500
|
| 160 |
+
},
|
| 161 |
+
{
|
| 162 |
+
"epoch": 0.36088675030073897,
|
| 163 |
+
"grad_norm": 1.1272865533828735,
|
| 164 |
+
"learning_rate": 2.6150000000000002e-05,
|
| 165 |
+
"loss": 3.0829,
|
| 166 |
+
"step": 525
|
| 167 |
+
},
|
| 168 |
+
{
|
| 169 |
+
"epoch": 0.3780718336483932,
|
| 170 |
+
"grad_norm": 0.8798616528511047,
|
| 171 |
+
"learning_rate": 2.7400000000000002e-05,
|
| 172 |
+
"loss": 3.0138,
|
| 173 |
+
"step": 550
|
| 174 |
+
},
|
| 175 |
+
{
|
| 176 |
+
"epoch": 0.3952569169960474,
|
| 177 |
+
"grad_norm": 1.233052134513855,
|
| 178 |
+
"learning_rate": 2.865e-05,
|
| 179 |
+
"loss": 2.9976,
|
| 180 |
+
"step": 575
|
| 181 |
+
},
|
| 182 |
+
{
|
| 183 |
+
"epoch": 0.41244200034370165,
|
| 184 |
+
"grad_norm": 0.4763319492340088,
|
| 185 |
+
"learning_rate": 2.9900000000000002e-05,
|
| 186 |
+
"loss": 2.9609,
|
| 187 |
+
"step": 600
|
| 188 |
+
},
|
| 189 |
+
{
|
| 190 |
+
"epoch": 0.4296270836913559,
|
| 191 |
+
"grad_norm": 0.35058021545410156,
|
| 192 |
+
"learning_rate": 3.115e-05,
|
| 193 |
+
"loss": 2.9375,
|
| 194 |
+
"step": 625
|
| 195 |
+
},
|
| 196 |
+
{
|
| 197 |
+
"epoch": 0.44681216703901017,
|
| 198 |
+
"grad_norm": 0.9661968946456909,
|
| 199 |
+
"learning_rate": 3.24e-05,
|
| 200 |
+
"loss": 2.8961,
|
| 201 |
+
"step": 650
|
| 202 |
+
},
|
| 203 |
+
{
|
| 204 |
+
"epoch": 0.4639972503866644,
|
| 205 |
+
"grad_norm": 1.0027278661727905,
|
| 206 |
+
"learning_rate": 3.3650000000000005e-05,
|
| 207 |
+
"loss": 2.8169,
|
| 208 |
+
"step": 675
|
| 209 |
+
},
|
| 210 |
+
{
|
| 211 |
+
"epoch": 0.4811823337343186,
|
| 212 |
+
"grad_norm": 0.6780478358268738,
|
| 213 |
+
"learning_rate": 3.49e-05,
|
| 214 |
+
"loss": 2.674,
|
| 215 |
+
"step": 700
|
| 216 |
+
},
|
| 217 |
+
{
|
| 218 |
+
"epoch": 0.49836741708197285,
|
| 219 |
+
"grad_norm": 0.7937625646591187,
|
| 220 |
+
"learning_rate": 3.615e-05,
|
| 221 |
+
"loss": 2.4667,
|
| 222 |
+
"step": 725
|
| 223 |
+
},
|
| 224 |
+
{
|
| 225 |
+
"epoch": 0.5155525004296271,
|
| 226 |
+
"grad_norm": 1.0229036808013916,
|
| 227 |
+
"learning_rate": 3.74e-05,
|
| 228 |
+
"loss": 2.2186,
|
| 229 |
+
"step": 750
|
| 230 |
+
},
|
| 231 |
+
{
|
| 232 |
+
"epoch": 0.5327375837772813,
|
| 233 |
+
"grad_norm": 1.1023578643798828,
|
| 234 |
+
"learning_rate": 3.8650000000000004e-05,
|
| 235 |
+
"loss": 1.9797,
|
| 236 |
+
"step": 775
|
| 237 |
+
},
|
| 238 |
+
{
|
| 239 |
+
"epoch": 0.5499226671249355,
|
| 240 |
+
"grad_norm": 1.084370493888855,
|
| 241 |
+
"learning_rate": 3.99e-05,
|
| 242 |
+
"loss": 1.6958,
|
| 243 |
+
"step": 800
|
| 244 |
+
},
|
| 245 |
+
{
|
| 246 |
+
"epoch": 0.5671077504725898,
|
| 247 |
+
"grad_norm": 0.8571304082870483,
|
| 248 |
+
"learning_rate": 4.115e-05,
|
| 249 |
+
"loss": 1.3521,
|
| 250 |
+
"step": 825
|
| 251 |
+
},
|
| 252 |
+
{
|
| 253 |
+
"epoch": 0.584292833820244,
|
| 254 |
+
"grad_norm": 1.0564861297607422,
|
| 255 |
+
"learning_rate": 4.24e-05,
|
| 256 |
+
"loss": 1.1168,
|
| 257 |
+
"step": 850
|
| 258 |
+
},
|
| 259 |
+
{
|
| 260 |
+
"epoch": 0.6014779171678982,
|
| 261 |
+
"grad_norm": 0.8157157301902771,
|
| 262 |
+
"learning_rate": 4.3650000000000004e-05,
|
| 263 |
+
"loss": 0.9829,
|
| 264 |
+
"step": 875
|
| 265 |
+
},
|
| 266 |
+
{
|
| 267 |
+
"epoch": 0.6186630005155525,
|
| 268 |
+
"grad_norm": 1.098561406135559,
|
| 269 |
+
"learning_rate": 4.49e-05,
|
| 270 |
+
"loss": 0.904,
|
| 271 |
+
"step": 900
|
| 272 |
+
},
|
| 273 |
+
{
|
| 274 |
+
"epoch": 0.6358480838632067,
|
| 275 |
+
"grad_norm": 0.6373503804206848,
|
| 276 |
+
"learning_rate": 4.6150000000000004e-05,
|
| 277 |
+
"loss": 0.8686,
|
| 278 |
+
"step": 925
|
| 279 |
+
},
|
| 280 |
+
{
|
| 281 |
+
"epoch": 0.653033167210861,
|
| 282 |
+
"grad_norm": 1.2277697324752808,
|
| 283 |
+
"learning_rate": 4.74e-05,
|
| 284 |
+
"loss": 0.7969,
|
| 285 |
+
"step": 950
|
| 286 |
+
},
|
| 287 |
+
{
|
| 288 |
+
"epoch": 0.6702182505585152,
|
| 289 |
+
"grad_norm": 0.5907439589500427,
|
| 290 |
+
"learning_rate": 4.8650000000000003e-05,
|
| 291 |
+
"loss": 0.7524,
|
| 292 |
+
"step": 975
|
| 293 |
+
},
|
| 294 |
+
{
|
| 295 |
+
"epoch": 0.6874033339061695,
|
| 296 |
+
"grad_norm": 0.7881184220314026,
|
| 297 |
+
"learning_rate": 4.99e-05,
|
| 298 |
+
"loss": 0.7323,
|
| 299 |
+
"step": 1000
|
| 300 |
+
},
|
| 301 |
+
{
|
| 302 |
+
"epoch": 0.6874033339061695,
|
| 303 |
+
"eval_loss": 0.630436360836029,
|
| 304 |
+
"eval_runtime": 155.009,
|
| 305 |
+
"eval_samples_per_second": 7.903,
|
| 306 |
+
"eval_steps_per_second": 0.993,
|
| 307 |
+
"eval_wer": 0.711954217888936,
|
| 308 |
+
"step": 1000
|
| 309 |
+
},
|
| 310 |
+
{
|
| 311 |
+
"epoch": 0.7045884172538237,
|
| 312 |
+
"grad_norm": 0.7166395783424377,
|
| 313 |
+
"learning_rate": 4.99866651205937e-05,
|
| 314 |
+
"loss": 0.705,
|
| 315 |
+
"step": 1025
|
| 316 |
+
},
|
| 317 |
+
{
|
| 318 |
+
"epoch": 0.7217735006014779,
|
| 319 |
+
"grad_norm": 0.7382345199584961,
|
| 320 |
+
"learning_rate": 4.9972170686456406e-05,
|
| 321 |
+
"loss": 0.7207,
|
| 322 |
+
"step": 1050
|
| 323 |
+
},
|
| 324 |
+
{
|
| 325 |
+
"epoch": 0.7389585839491322,
|
| 326 |
+
"grad_norm": 0.6300435066223145,
|
| 327 |
+
"learning_rate": 4.995767625231911e-05,
|
| 328 |
+
"loss": 0.7097,
|
| 329 |
+
"step": 1075
|
| 330 |
+
},
|
| 331 |
+
{
|
| 332 |
+
"epoch": 0.7561436672967864,
|
| 333 |
+
"grad_norm": 0.7419637441635132,
|
| 334 |
+
"learning_rate": 4.994318181818182e-05,
|
| 335 |
+
"loss": 0.6865,
|
| 336 |
+
"step": 1100
|
| 337 |
+
},
|
| 338 |
+
{
|
| 339 |
+
"epoch": 0.7733287506444406,
|
| 340 |
+
"grad_norm": 0.5746726989746094,
|
| 341 |
+
"learning_rate": 4.9928687384044535e-05,
|
| 342 |
+
"loss": 0.6694,
|
| 343 |
+
"step": 1125
|
| 344 |
+
},
|
| 345 |
+
{
|
| 346 |
+
"epoch": 0.7905138339920948,
|
| 347 |
+
"grad_norm": 0.7999011874198914,
|
| 348 |
+
"learning_rate": 4.991419294990724e-05,
|
| 349 |
+
"loss": 0.6729,
|
| 350 |
+
"step": 1150
|
| 351 |
+
},
|
| 352 |
+
{
|
| 353 |
+
"epoch": 0.8076989173397491,
|
| 354 |
+
"grad_norm": 0.5744255781173706,
|
| 355 |
+
"learning_rate": 4.989969851576995e-05,
|
| 356 |
+
"loss": 0.6253,
|
| 357 |
+
"step": 1175
|
| 358 |
+
},
|
| 359 |
+
{
|
| 360 |
+
"epoch": 0.8248840006874033,
|
| 361 |
+
"grad_norm": 0.8337430357933044,
|
| 362 |
+
"learning_rate": 4.988520408163265e-05,
|
| 363 |
+
"loss": 0.6616,
|
| 364 |
+
"step": 1200
|
| 365 |
+
},
|
| 366 |
+
{
|
| 367 |
+
"epoch": 0.8420690840350575,
|
| 368 |
+
"grad_norm": 0.5431417226791382,
|
| 369 |
+
"learning_rate": 4.9870709647495364e-05,
|
| 370 |
+
"loss": 0.6129,
|
| 371 |
+
"step": 1225
|
| 372 |
+
},
|
| 373 |
+
{
|
| 374 |
+
"epoch": 0.8592541673827118,
|
| 375 |
+
"grad_norm": 0.8542420864105225,
|
| 376 |
+
"learning_rate": 4.985621521335807e-05,
|
| 377 |
+
"loss": 0.6252,
|
| 378 |
+
"step": 1250
|
| 379 |
+
},
|
| 380 |
+
{
|
| 381 |
+
"epoch": 0.876439250730366,
|
| 382 |
+
"grad_norm": 0.5819442272186279,
|
| 383 |
+
"learning_rate": 4.984172077922078e-05,
|
| 384 |
+
"loss": 0.5723,
|
| 385 |
+
"step": 1275
|
| 386 |
+
},
|
| 387 |
+
{
|
| 388 |
+
"epoch": 0.8936243340780203,
|
| 389 |
+
"grad_norm": 0.9629399180412292,
|
| 390 |
+
"learning_rate": 4.982722634508349e-05,
|
| 391 |
+
"loss": 0.582,
|
| 392 |
+
"step": 1300
|
| 393 |
+
},
|
| 394 |
+
{
|
| 395 |
+
"epoch": 0.9108094174256746,
|
| 396 |
+
"grad_norm": 0.5698046088218689,
|
| 397 |
+
"learning_rate": 4.98127319109462e-05,
|
| 398 |
+
"loss": 0.5745,
|
| 399 |
+
"step": 1325
|
| 400 |
+
},
|
| 401 |
+
{
|
| 402 |
+
"epoch": 0.9279945007733288,
|
| 403 |
+
"grad_norm": 0.893267035484314,
|
| 404 |
+
"learning_rate": 4.979823747680891e-05,
|
| 405 |
+
"loss": 0.5887,
|
| 406 |
+
"step": 1350
|
| 407 |
+
},
|
| 408 |
+
{
|
| 409 |
+
"epoch": 0.945179584120983,
|
| 410 |
+
"grad_norm": 0.4911533296108246,
|
| 411 |
+
"learning_rate": 4.9783743042671616e-05,
|
| 412 |
+
"loss": 0.5657,
|
| 413 |
+
"step": 1375
|
| 414 |
+
},
|
| 415 |
+
{
|
| 416 |
+
"epoch": 0.9623646674686372,
|
| 417 |
+
"grad_norm": 0.9226717948913574,
|
| 418 |
+
"learning_rate": 4.976924860853432e-05,
|
| 419 |
+
"loss": 0.5615,
|
| 420 |
+
"step": 1400
|
| 421 |
+
},
|
| 422 |
+
{
|
| 423 |
+
"epoch": 0.9795497508162915,
|
| 424 |
+
"grad_norm": 0.5409913659095764,
|
| 425 |
+
"learning_rate": 4.975475417439703e-05,
|
| 426 |
+
"loss": 0.5562,
|
| 427 |
+
"step": 1425
|
| 428 |
+
},
|
| 429 |
+
{
|
| 430 |
+
"epoch": 0.9967348341639457,
|
| 431 |
+
"grad_norm": 0.7855440974235535,
|
| 432 |
+
"learning_rate": 4.9740259740259745e-05,
|
| 433 |
+
"loss": 0.5641,
|
| 434 |
+
"step": 1450
|
| 435 |
+
},
|
| 436 |
+
{
|
| 437 |
+
"epoch": 1.0144354700120295,
|
| 438 |
+
"grad_norm": 0.48342418670654297,
|
| 439 |
+
"learning_rate": 4.972576530612245e-05,
|
| 440 |
+
"loss": 0.5216,
|
| 441 |
+
"step": 1475
|
| 442 |
+
},
|
| 443 |
+
{
|
| 444 |
+
"epoch": 1.0316205533596838,
|
| 445 |
+
"grad_norm": 0.7128458023071289,
|
| 446 |
+
"learning_rate": 4.971127087198516e-05,
|
| 447 |
+
"loss": 0.5416,
|
| 448 |
+
"step": 1500
|
| 449 |
+
},
|
| 450 |
+
{
|
| 451 |
+
"epoch": 1.0316205533596838,
|
| 452 |
+
"eval_loss": 0.47852450609207153,
|
| 453 |
+
"eval_runtime": 156.3389,
|
| 454 |
+
"eval_samples_per_second": 7.836,
|
| 455 |
+
"eval_steps_per_second": 0.985,
|
| 456 |
+
"eval_wer": 0.6490568037303942,
|
| 457 |
+
"step": 1500
|
| 458 |
+
},
|
| 459 |
+
{
|
| 460 |
+
"epoch": 1.048805636707338,
|
| 461 |
+
"grad_norm": 0.4331744909286499,
|
| 462 |
+
"learning_rate": 4.969677643784787e-05,
|
| 463 |
+
"loss": 0.5337,
|
| 464 |
+
"step": 1525
|
| 465 |
+
},
|
| 466 |
+
{
|
| 467 |
+
"epoch": 1.0659907200549923,
|
| 468 |
+
"grad_norm": 0.7347924113273621,
|
| 469 |
+
"learning_rate": 4.968228200371058e-05,
|
| 470 |
+
"loss": 0.5115,
|
| 471 |
+
"step": 1550
|
| 472 |
+
},
|
| 473 |
+
{
|
| 474 |
+
"epoch": 1.0831758034026464,
|
| 475 |
+
"grad_norm": 0.47687920928001404,
|
| 476 |
+
"learning_rate": 4.966778756957329e-05,
|
| 477 |
+
"loss": 0.5525,
|
| 478 |
+
"step": 1575
|
| 479 |
+
},
|
| 480 |
+
{
|
| 481 |
+
"epoch": 1.1003608867503007,
|
| 482 |
+
"grad_norm": 0.6743185520172119,
|
| 483 |
+
"learning_rate": 4.9653293135435996e-05,
|
| 484 |
+
"loss": 0.4931,
|
| 485 |
+
"step": 1600
|
| 486 |
+
},
|
| 487 |
+
{
|
| 488 |
+
"epoch": 1.117545970097955,
|
| 489 |
+
"grad_norm": 0.5449199676513672,
|
| 490 |
+
"learning_rate": 4.9638798701298704e-05,
|
| 491 |
+
"loss": 0.4773,
|
| 492 |
+
"step": 1625
|
| 493 |
+
},
|
| 494 |
+
{
|
| 495 |
+
"epoch": 1.1347310534456092,
|
| 496 |
+
"grad_norm": 0.6370054483413696,
|
| 497 |
+
"learning_rate": 4.962430426716141e-05,
|
| 498 |
+
"loss": 0.4937,
|
| 499 |
+
"step": 1650
|
| 500 |
+
},
|
| 501 |
+
{
|
| 502 |
+
"epoch": 1.1519161367932633,
|
| 503 |
+
"grad_norm": 0.5630868077278137,
|
| 504 |
+
"learning_rate": 4.9609809833024125e-05,
|
| 505 |
+
"loss": 0.541,
|
| 506 |
+
"step": 1675
|
| 507 |
+
},
|
| 508 |
+
{
|
| 509 |
+
"epoch": 1.1691012201409177,
|
| 510 |
+
"grad_norm": 0.6498411297798157,
|
| 511 |
+
"learning_rate": 4.959531539888683e-05,
|
| 512 |
+
"loss": 0.5057,
|
| 513 |
+
"step": 1700
|
| 514 |
+
},
|
| 515 |
+
{
|
| 516 |
+
"epoch": 1.186286303488572,
|
| 517 |
+
"grad_norm": 0.468143105506897,
|
| 518 |
+
"learning_rate": 4.9580820964749533e-05,
|
| 519 |
+
"loss": 0.5155,
|
| 520 |
+
"step": 1725
|
| 521 |
+
},
|
| 522 |
+
{
|
| 523 |
+
"epoch": 1.2034713868362261,
|
| 524 |
+
"grad_norm": 0.7214887738227844,
|
| 525 |
+
"learning_rate": 4.956632653061225e-05,
|
| 526 |
+
"loss": 0.5132,
|
| 527 |
+
"step": 1750
|
| 528 |
+
},
|
| 529 |
+
{
|
| 530 |
+
"epoch": 1.2206564701838805,
|
| 531 |
+
"grad_norm": 0.4510962963104248,
|
| 532 |
+
"learning_rate": 4.9551832096474955e-05,
|
| 533 |
+
"loss": 0.4643,
|
| 534 |
+
"step": 1775
|
| 535 |
+
},
|
| 536 |
+
{
|
| 537 |
+
"epoch": 1.2378415535315346,
|
| 538 |
+
"grad_norm": 0.6587559580802917,
|
| 539 |
+
"learning_rate": 4.953733766233766e-05,
|
| 540 |
+
"loss": 0.4701,
|
| 541 |
+
"step": 1800
|
| 542 |
+
},
|
| 543 |
+
{
|
| 544 |
+
"epoch": 1.255026636879189,
|
| 545 |
+
"grad_norm": 0.5024493336677551,
|
| 546 |
+
"learning_rate": 4.952284322820037e-05,
|
| 547 |
+
"loss": 0.5198,
|
| 548 |
+
"step": 1825
|
| 549 |
+
},
|
| 550 |
+
{
|
| 551 |
+
"epoch": 1.272211720226843,
|
| 552 |
+
"grad_norm": 0.7172912359237671,
|
| 553 |
+
"learning_rate": 4.950834879406308e-05,
|
| 554 |
+
"loss": 0.4555,
|
| 555 |
+
"step": 1850
|
| 556 |
+
},
|
| 557 |
+
{
|
| 558 |
+
"epoch": 1.2893968035744974,
|
| 559 |
+
"grad_norm": 0.5447876453399658,
|
| 560 |
+
"learning_rate": 4.949385435992579e-05,
|
| 561 |
+
"loss": 0.5041,
|
| 562 |
+
"step": 1875
|
| 563 |
+
},
|
| 564 |
+
{
|
| 565 |
+
"epoch": 1.3065818869221515,
|
| 566 |
+
"grad_norm": 0.6178200244903564,
|
| 567 |
+
"learning_rate": 4.94793599257885e-05,
|
| 568 |
+
"loss": 0.4686,
|
| 569 |
+
"step": 1900
|
| 570 |
+
},
|
| 571 |
+
{
|
| 572 |
+
"epoch": 1.3237669702698058,
|
| 573 |
+
"grad_norm": 0.49054816365242004,
|
| 574 |
+
"learning_rate": 4.9464865491651207e-05,
|
| 575 |
+
"loss": 0.4803,
|
| 576 |
+
"step": 1925
|
| 577 |
+
},
|
| 578 |
+
{
|
| 579 |
+
"epoch": 1.34095205361746,
|
| 580 |
+
"grad_norm": 0.6802988648414612,
|
| 581 |
+
"learning_rate": 4.9450371057513914e-05,
|
| 582 |
+
"loss": 0.4644,
|
| 583 |
+
"step": 1950
|
| 584 |
+
},
|
| 585 |
+
{
|
| 586 |
+
"epoch": 1.3581371369651143,
|
| 587 |
+
"grad_norm": 0.41138285398483276,
|
| 588 |
+
"learning_rate": 4.943587662337663e-05,
|
| 589 |
+
"loss": 0.513,
|
| 590 |
+
"step": 1975
|
| 591 |
+
},
|
| 592 |
+
{
|
| 593 |
+
"epoch": 1.3753222203127686,
|
| 594 |
+
"grad_norm": 0.8374213576316833,
|
| 595 |
+
"learning_rate": 4.9421382189239336e-05,
|
| 596 |
+
"loss": 0.4479,
|
| 597 |
+
"step": 2000
|
| 598 |
+
},
|
| 599 |
+
{
|
| 600 |
+
"epoch": 1.3753222203127686,
|
| 601 |
+
"eval_loss": 0.42023956775665283,
|
| 602 |
+
"eval_runtime": 156.1936,
|
| 603 |
+
"eval_samples_per_second": 7.843,
|
| 604 |
+
"eval_steps_per_second": 0.986,
|
| 605 |
+
"eval_wer": 0.6206549385332768,
|
| 606 |
+
"step": 2000
|
| 607 |
+
},
|
| 608 |
+
{
|
| 609 |
+
"epoch": 1.3925073036604227,
|
| 610 |
+
"grad_norm": 0.7108762860298157,
|
| 611 |
+
"learning_rate": 4.940688775510204e-05,
|
| 612 |
+
"loss": 0.465,
|
| 613 |
+
"step": 2025
|
| 614 |
+
},
|
| 615 |
+
{
|
| 616 |
+
"epoch": 1.4096923870080769,
|
| 617 |
+
"grad_norm": 0.6862374544143677,
|
| 618 |
+
"learning_rate": 4.939239332096475e-05,
|
| 619 |
+
"loss": 0.4513,
|
| 620 |
+
"step": 2050
|
| 621 |
+
},
|
| 622 |
+
{
|
| 623 |
+
"epoch": 1.4268774703557312,
|
| 624 |
+
"grad_norm": 0.5275014638900757,
|
| 625 |
+
"learning_rate": 4.9377898886827465e-05,
|
| 626 |
+
"loss": 0.5135,
|
| 627 |
+
"step": 2075
|
| 628 |
+
},
|
| 629 |
+
{
|
| 630 |
+
"epoch": 1.4440625537033855,
|
| 631 |
+
"grad_norm": 0.7356075644493103,
|
| 632 |
+
"learning_rate": 4.936340445269017e-05,
|
| 633 |
+
"loss": 0.4843,
|
| 634 |
+
"step": 2100
|
| 635 |
+
},
|
| 636 |
+
{
|
| 637 |
+
"epoch": 1.4612476370510397,
|
| 638 |
+
"grad_norm": 0.46985840797424316,
|
| 639 |
+
"learning_rate": 4.934891001855288e-05,
|
| 640 |
+
"loss": 0.544,
|
| 641 |
+
"step": 2125
|
| 642 |
+
},
|
| 643 |
+
{
|
| 644 |
+
"epoch": 1.478432720398694,
|
| 645 |
+
"grad_norm": 0.8410568237304688,
|
| 646 |
+
"learning_rate": 4.933441558441559e-05,
|
| 647 |
+
"loss": 0.453,
|
| 648 |
+
"step": 2150
|
| 649 |
+
},
|
| 650 |
+
{
|
| 651 |
+
"epoch": 1.4956178037463481,
|
| 652 |
+
"grad_norm": 0.5153818130493164,
|
| 653 |
+
"learning_rate": 4.9319921150278294e-05,
|
| 654 |
+
"loss": 0.5145,
|
| 655 |
+
"step": 2175
|
| 656 |
+
},
|
| 657 |
+
{
|
| 658 |
+
"epoch": 1.5128028870940025,
|
| 659 |
+
"grad_norm": 0.8318967223167419,
|
| 660 |
+
"learning_rate": 4.930542671614101e-05,
|
| 661 |
+
"loss": 0.4583,
|
| 662 |
+
"step": 2200
|
| 663 |
+
},
|
| 664 |
+
{
|
| 665 |
+
"epoch": 1.5299879704416566,
|
| 666 |
+
"grad_norm": 0.5471023321151733,
|
| 667 |
+
"learning_rate": 4.9290932282003716e-05,
|
| 668 |
+
"loss": 0.4622,
|
| 669 |
+
"step": 2225
|
| 670 |
+
},
|
| 671 |
+
{
|
| 672 |
+
"epoch": 1.547173053789311,
|
| 673 |
+
"grad_norm": 0.6953750848770142,
|
| 674 |
+
"learning_rate": 4.927643784786642e-05,
|
| 675 |
+
"loss": 0.4712,
|
| 676 |
+
"step": 2250
|
| 677 |
+
},
|
| 678 |
+
{
|
| 679 |
+
"epoch": 1.5643581371369653,
|
| 680 |
+
"grad_norm": 0.5455946326255798,
|
| 681 |
+
"learning_rate": 4.926194341372913e-05,
|
| 682 |
+
"loss": 0.4702,
|
| 683 |
+
"step": 2275
|
| 684 |
+
},
|
| 685 |
+
{
|
| 686 |
+
"epoch": 1.5815432204846194,
|
| 687 |
+
"grad_norm": 0.6755653023719788,
|
| 688 |
+
"learning_rate": 4.924744897959184e-05,
|
| 689 |
+
"loss": 0.4356,
|
| 690 |
+
"step": 2300
|
| 691 |
+
},
|
| 692 |
+
{
|
| 693 |
+
"epoch": 1.5987283038322735,
|
| 694 |
+
"grad_norm": 0.47134584188461304,
|
| 695 |
+
"learning_rate": 4.9232954545454546e-05,
|
| 696 |
+
"loss": 0.4333,
|
| 697 |
+
"step": 2325
|
| 698 |
+
},
|
| 699 |
+
{
|
| 700 |
+
"epoch": 1.6159133871799278,
|
| 701 |
+
"grad_norm": 0.6077346205711365,
|
| 702 |
+
"learning_rate": 4.921846011131725e-05,
|
| 703 |
+
"loss": 0.4432,
|
| 704 |
+
"step": 2350
|
| 705 |
+
},
|
| 706 |
+
{
|
| 707 |
+
"epoch": 1.6330984705275822,
|
| 708 |
+
"grad_norm": 0.5316899418830872,
|
| 709 |
+
"learning_rate": 4.920396567717996e-05,
|
| 710 |
+
"loss": 0.45,
|
| 711 |
+
"step": 2375
|
| 712 |
+
},
|
| 713 |
+
{
|
| 714 |
+
"epoch": 1.6502835538752363,
|
| 715 |
+
"grad_norm": 0.79489666223526,
|
| 716 |
+
"learning_rate": 4.9189471243042675e-05,
|
| 717 |
+
"loss": 0.4637,
|
| 718 |
+
"step": 2400
|
| 719 |
+
},
|
| 720 |
+
{
|
| 721 |
+
"epoch": 1.6674686372228904,
|
| 722 |
+
"grad_norm": 0.5166532397270203,
|
| 723 |
+
"learning_rate": 4.917497680890538e-05,
|
| 724 |
+
"loss": 0.4375,
|
| 725 |
+
"step": 2425
|
| 726 |
+
},
|
| 727 |
+
{
|
| 728 |
+
"epoch": 1.6846537205705447,
|
| 729 |
+
"grad_norm": 0.6109660863876343,
|
| 730 |
+
"learning_rate": 4.916048237476809e-05,
|
| 731 |
+
"loss": 0.418,
|
| 732 |
+
"step": 2450
|
| 733 |
+
},
|
| 734 |
+
{
|
| 735 |
+
"epoch": 1.701838803918199,
|
| 736 |
+
"grad_norm": 0.6634232401847839,
|
| 737 |
+
"learning_rate": 4.91459879406308e-05,
|
| 738 |
+
"loss": 0.4315,
|
| 739 |
+
"step": 2475
|
| 740 |
+
},
|
| 741 |
+
{
|
| 742 |
+
"epoch": 1.7190238872658532,
|
| 743 |
+
"grad_norm": 0.5985594987869263,
|
| 744 |
+
"learning_rate": 4.913149350649351e-05,
|
| 745 |
+
"loss": 0.4541,
|
| 746 |
+
"step": 2500
|
| 747 |
+
},
|
| 748 |
+
{
|
| 749 |
+
"epoch": 1.7190238872658532,
|
| 750 |
+
"eval_loss": 0.3850683569908142,
|
| 751 |
+
"eval_runtime": 155.7689,
|
| 752 |
+
"eval_samples_per_second": 7.864,
|
| 753 |
+
"eval_steps_per_second": 0.989,
|
| 754 |
+
"eval_wer": 0.6006252649427724,
|
| 755 |
+
"step": 2500
|
| 756 |
+
},
|
| 757 |
+
{
|
| 758 |
+
"epoch": 1.7362089706135073,
|
| 759 |
+
"grad_norm": 0.5531134605407715,
|
| 760 |
+
"learning_rate": 4.911699907235622e-05,
|
| 761 |
+
"loss": 0.4558,
|
| 762 |
+
"step": 2525
|
| 763 |
+
},
|
| 764 |
+
{
|
| 765 |
+
"epoch": 1.7533940539611617,
|
| 766 |
+
"grad_norm": 0.594007670879364,
|
| 767 |
+
"learning_rate": 4.9102504638218926e-05,
|
| 768 |
+
"loss": 0.4422,
|
| 769 |
+
"step": 2550
|
| 770 |
+
},
|
| 771 |
+
{
|
| 772 |
+
"epoch": 1.770579137308816,
|
| 773 |
+
"grad_norm": 0.4865548312664032,
|
| 774 |
+
"learning_rate": 4.9088010204081634e-05,
|
| 775 |
+
"loss": 0.4216,
|
| 776 |
+
"step": 2575
|
| 777 |
+
},
|
| 778 |
+
{
|
| 779 |
+
"epoch": 1.7877642206564701,
|
| 780 |
+
"grad_norm": 0.5752180814743042,
|
| 781 |
+
"learning_rate": 4.907351576994435e-05,
|
| 782 |
+
"loss": 0.429,
|
| 783 |
+
"step": 2600
|
| 784 |
+
},
|
| 785 |
+
{
|
| 786 |
+
"epoch": 1.8049493040041245,
|
| 787 |
+
"grad_norm": 0.4255332350730896,
|
| 788 |
+
"learning_rate": 4.9059021335807055e-05,
|
| 789 |
+
"loss": 0.4074,
|
| 790 |
+
"step": 2625
|
| 791 |
+
},
|
| 792 |
+
{
|
| 793 |
+
"epoch": 1.8221343873517788,
|
| 794 |
+
"grad_norm": 0.6160002946853638,
|
| 795 |
+
"learning_rate": 4.904452690166976e-05,
|
| 796 |
+
"loss": 0.4388,
|
| 797 |
+
"step": 2650
|
| 798 |
+
},
|
| 799 |
+
{
|
| 800 |
+
"epoch": 1.839319470699433,
|
| 801 |
+
"grad_norm": 0.49583593010902405,
|
| 802 |
+
"learning_rate": 4.903003246753247e-05,
|
| 803 |
+
"loss": 0.4211,
|
| 804 |
+
"step": 2675
|
| 805 |
+
},
|
| 806 |
+
{
|
| 807 |
+
"epoch": 1.856504554047087,
|
| 808 |
+
"grad_norm": 0.7490856647491455,
|
| 809 |
+
"learning_rate": 4.901553803339518e-05,
|
| 810 |
+
"loss": 0.4559,
|
| 811 |
+
"step": 2700
|
| 812 |
+
},
|
| 813 |
+
{
|
| 814 |
+
"epoch": 1.8736896373947414,
|
| 815 |
+
"grad_norm": 0.4634897708892822,
|
| 816 |
+
"learning_rate": 4.900104359925789e-05,
|
| 817 |
+
"loss": 0.4323,
|
| 818 |
+
"step": 2725
|
| 819 |
+
},
|
| 820 |
+
{
|
| 821 |
+
"epoch": 1.8908747207423957,
|
| 822 |
+
"grad_norm": 1.0682307481765747,
|
| 823 |
+
"learning_rate": 4.89865491651206e-05,
|
| 824 |
+
"loss": 0.4361,
|
| 825 |
+
"step": 2750
|
| 826 |
+
},
|
| 827 |
+
{
|
| 828 |
+
"epoch": 1.9080598040900498,
|
| 829 |
+
"grad_norm": 0.41067609190940857,
|
| 830 |
+
"learning_rate": 4.89720547309833e-05,
|
| 831 |
+
"loss": 0.4627,
|
| 832 |
+
"step": 2775
|
| 833 |
+
},
|
| 834 |
+
{
|
| 835 |
+
"epoch": 1.925244887437704,
|
| 836 |
+
"grad_norm": 0.674846887588501,
|
| 837 |
+
"learning_rate": 4.8957560296846014e-05,
|
| 838 |
+
"loss": 0.4357,
|
| 839 |
+
"step": 2800
|
| 840 |
+
},
|
| 841 |
+
{
|
| 842 |
+
"epoch": 1.9424299707853583,
|
| 843 |
+
"grad_norm": 0.4671356678009033,
|
| 844 |
+
"learning_rate": 4.894306586270872e-05,
|
| 845 |
+
"loss": 0.429,
|
| 846 |
+
"step": 2825
|
| 847 |
+
},
|
| 848 |
+
{
|
| 849 |
+
"epoch": 1.9596150541330126,
|
| 850 |
+
"grad_norm": 0.6439480781555176,
|
| 851 |
+
"learning_rate": 4.892857142857143e-05,
|
| 852 |
+
"loss": 0.3965,
|
| 853 |
+
"step": 2850
|
| 854 |
+
},
|
| 855 |
+
{
|
| 856 |
+
"epoch": 1.9768001374806667,
|
| 857 |
+
"grad_norm": 0.5034067034721375,
|
| 858 |
+
"learning_rate": 4.8914076994434137e-05,
|
| 859 |
+
"loss": 0.4131,
|
| 860 |
+
"step": 2875
|
| 861 |
+
},
|
| 862 |
+
{
|
| 863 |
+
"epoch": 1.9939852208283209,
|
| 864 |
+
"grad_norm": 0.67892986536026,
|
| 865 |
+
"learning_rate": 4.8899582560296844e-05,
|
| 866 |
+
"loss": 0.4306,
|
| 867 |
+
"step": 2900
|
| 868 |
+
},
|
| 869 |
+
{
|
| 870 |
+
"epoch": 2.011685856676405,
|
| 871 |
+
"grad_norm": 0.4826701879501343,
|
| 872 |
+
"learning_rate": 4.888508812615956e-05,
|
| 873 |
+
"loss": 0.4437,
|
| 874 |
+
"step": 2925
|
| 875 |
+
},
|
| 876 |
+
{
|
| 877 |
+
"epoch": 2.028870940024059,
|
| 878 |
+
"grad_norm": 0.5658535361289978,
|
| 879 |
+
"learning_rate": 4.8870593692022266e-05,
|
| 880 |
+
"loss": 0.3842,
|
| 881 |
+
"step": 2950
|
| 882 |
+
},
|
| 883 |
+
{
|
| 884 |
+
"epoch": 2.0460560233717136,
|
| 885 |
+
"grad_norm": 0.5490546226501465,
|
| 886 |
+
"learning_rate": 4.885609925788497e-05,
|
| 887 |
+
"loss": 0.3872,
|
| 888 |
+
"step": 2975
|
| 889 |
+
},
|
| 890 |
+
{
|
| 891 |
+
"epoch": 2.0632411067193677,
|
| 892 |
+
"grad_norm": 0.59776371717453,
|
| 893 |
+
"learning_rate": 4.884160482374768e-05,
|
| 894 |
+
"loss": 0.365,
|
| 895 |
+
"step": 3000
|
| 896 |
+
},
|
| 897 |
+
{
|
| 898 |
+
"epoch": 2.0632411067193677,
|
| 899 |
+
"eval_loss": 0.37011492252349854,
|
| 900 |
+
"eval_runtime": 150.8939,
|
| 901 |
+
"eval_samples_per_second": 8.118,
|
| 902 |
+
"eval_steps_per_second": 1.021,
|
| 903 |
+
"eval_wer": 0.588543874523103,
|
| 904 |
+
"step": 3000
|
| 905 |
+
},
|
| 906 |
+
{
|
| 907 |
+
"epoch": 2.080426190067022,
|
| 908 |
+
"grad_norm": 0.4145926237106323,
|
| 909 |
+
"learning_rate": 4.8827110389610395e-05,
|
| 910 |
+
"loss": 0.4335,
|
| 911 |
+
"step": 3025
|
| 912 |
+
},
|
| 913 |
+
{
|
| 914 |
+
"epoch": 2.097611273414676,
|
| 915 |
+
"grad_norm": 0.6833218336105347,
|
| 916 |
+
"learning_rate": 4.88126159554731e-05,
|
| 917 |
+
"loss": 0.3801,
|
| 918 |
+
"step": 3050
|
| 919 |
+
},
|
| 920 |
+
{
|
| 921 |
+
"epoch": 2.1147963567623305,
|
| 922 |
+
"grad_norm": 0.5245340466499329,
|
| 923 |
+
"learning_rate": 4.879812152133581e-05,
|
| 924 |
+
"loss": 0.4167,
|
| 925 |
+
"step": 3075
|
| 926 |
+
},
|
| 927 |
+
{
|
| 928 |
+
"epoch": 2.1319814401099846,
|
| 929 |
+
"grad_norm": 0.5739388465881348,
|
| 930 |
+
"learning_rate": 4.878362708719852e-05,
|
| 931 |
+
"loss": 0.3793,
|
| 932 |
+
"step": 3100
|
| 933 |
+
},
|
| 934 |
+
{
|
| 935 |
+
"epoch": 2.1491665234576387,
|
| 936 |
+
"grad_norm": 0.502185046672821,
|
| 937 |
+
"learning_rate": 4.876913265306123e-05,
|
| 938 |
+
"loss": 0.4444,
|
| 939 |
+
"step": 3125
|
| 940 |
+
},
|
| 941 |
+
{
|
| 942 |
+
"epoch": 2.166351606805293,
|
| 943 |
+
"grad_norm": 0.635421633720398,
|
| 944 |
+
"learning_rate": 4.875463821892394e-05,
|
| 945 |
+
"loss": 0.3668,
|
| 946 |
+
"step": 3150
|
| 947 |
+
},
|
| 948 |
+
{
|
| 949 |
+
"epoch": 2.1835366901529474,
|
| 950 |
+
"grad_norm": 0.4521035850048065,
|
| 951 |
+
"learning_rate": 4.8740143784786646e-05,
|
| 952 |
+
"loss": 0.3772,
|
| 953 |
+
"step": 3175
|
| 954 |
+
},
|
| 955 |
+
{
|
| 956 |
+
"epoch": 2.2007217735006015,
|
| 957 |
+
"grad_norm": 0.7126047015190125,
|
| 958 |
+
"learning_rate": 4.8725649350649354e-05,
|
| 959 |
+
"loss": 0.374,
|
| 960 |
+
"step": 3200
|
| 961 |
+
},
|
| 962 |
+
{
|
| 963 |
+
"epoch": 2.2179068568482556,
|
| 964 |
+
"grad_norm": 0.44621542096138,
|
| 965 |
+
"learning_rate": 4.871115491651206e-05,
|
| 966 |
+
"loss": 0.448,
|
| 967 |
+
"step": 3225
|
| 968 |
+
},
|
| 969 |
+
{
|
| 970 |
+
"epoch": 2.23509194019591,
|
| 971 |
+
"grad_norm": 0.6418918967247009,
|
| 972 |
+
"learning_rate": 4.8696660482374775e-05,
|
| 973 |
+
"loss": 0.4099,
|
| 974 |
+
"step": 3250
|
| 975 |
+
},
|
| 976 |
+
{
|
| 977 |
+
"epoch": 2.2522770235435643,
|
| 978 |
+
"grad_norm": 0.6630382537841797,
|
| 979 |
+
"learning_rate": 4.8682166048237476e-05,
|
| 980 |
+
"loss": 0.4028,
|
| 981 |
+
"step": 3275
|
| 982 |
+
},
|
| 983 |
+
{
|
| 984 |
+
"epoch": 2.2694621068912184,
|
| 985 |
+
"grad_norm": 0.6084064841270447,
|
| 986 |
+
"learning_rate": 4.866767161410018e-05,
|
| 987 |
+
"loss": 0.3737,
|
| 988 |
+
"step": 3300
|
| 989 |
+
},
|
| 990 |
+
{
|
| 991 |
+
"epoch": 2.2866471902388725,
|
| 992 |
+
"grad_norm": 0.49679034948349,
|
| 993 |
+
"learning_rate": 4.86531771799629e-05,
|
| 994 |
+
"loss": 0.4054,
|
| 995 |
+
"step": 3325
|
| 996 |
+
},
|
| 997 |
+
{
|
| 998 |
+
"epoch": 2.3038322735865266,
|
| 999 |
+
"grad_norm": 0.5616837739944458,
|
| 1000 |
+
"learning_rate": 4.8638682745825605e-05,
|
| 1001 |
+
"loss": 0.3562,
|
| 1002 |
+
"step": 3350
|
| 1003 |
+
},
|
| 1004 |
+
{
|
| 1005 |
+
"epoch": 2.321017356934181,
|
| 1006 |
+
"grad_norm": 0.501042366027832,
|
| 1007 |
+
"learning_rate": 4.862418831168831e-05,
|
| 1008 |
+
"loss": 0.3613,
|
| 1009 |
+
"step": 3375
|
| 1010 |
+
},
|
| 1011 |
+
{
|
| 1012 |
+
"epoch": 2.3382024402818353,
|
| 1013 |
+
"grad_norm": 0.7408941984176636,
|
| 1014 |
+
"learning_rate": 4.860969387755102e-05,
|
| 1015 |
+
"loss": 0.3774,
|
| 1016 |
+
"step": 3400
|
| 1017 |
+
},
|
| 1018 |
+
{
|
| 1019 |
+
"epoch": 2.3553875236294894,
|
| 1020 |
+
"grad_norm": 0.4748440086841583,
|
| 1021 |
+
"learning_rate": 4.859519944341373e-05,
|
| 1022 |
+
"loss": 0.4066,
|
| 1023 |
+
"step": 3425
|
| 1024 |
+
},
|
| 1025 |
+
{
|
| 1026 |
+
"epoch": 2.372572606977144,
|
| 1027 |
+
"grad_norm": 0.6289178729057312,
|
| 1028 |
+
"learning_rate": 4.858070500927644e-05,
|
| 1029 |
+
"loss": 0.3748,
|
| 1030 |
+
"step": 3450
|
| 1031 |
+
},
|
| 1032 |
+
{
|
| 1033 |
+
"epoch": 2.389757690324798,
|
| 1034 |
+
"grad_norm": 0.4348820149898529,
|
| 1035 |
+
"learning_rate": 4.856621057513915e-05,
|
| 1036 |
+
"loss": 0.4605,
|
| 1037 |
+
"step": 3475
|
| 1038 |
+
},
|
| 1039 |
+
{
|
| 1040 |
+
"epoch": 2.4069427736724522,
|
| 1041 |
+
"grad_norm": 0.5051092505455017,
|
| 1042 |
+
"learning_rate": 4.8551716141001856e-05,
|
| 1043 |
+
"loss": 0.3433,
|
| 1044 |
+
"step": 3500
|
| 1045 |
+
},
|
| 1046 |
+
{
|
| 1047 |
+
"epoch": 2.4069427736724522,
|
| 1048 |
+
"eval_loss": 0.364750474691391,
|
| 1049 |
+
"eval_runtime": 151.4529,
|
| 1050 |
+
"eval_samples_per_second": 8.088,
|
| 1051 |
+
"eval_steps_per_second": 1.017,
|
| 1052 |
+
"eval_wer": 0.5796947859262399,
|
| 1053 |
+
"step": 3500
|
| 1054 |
+
},
|
| 1055 |
+
{
|
| 1056 |
+
"epoch": 2.4241278570201064,
|
| 1057 |
+
"grad_norm": 0.5361665487289429,
|
| 1058 |
+
"learning_rate": 4.8537221706864564e-05,
|
| 1059 |
+
"loss": 0.4373,
|
| 1060 |
+
"step": 3525
|
| 1061 |
+
},
|
| 1062 |
+
{
|
| 1063 |
+
"epoch": 2.441312940367761,
|
| 1064 |
+
"grad_norm": 0.5831518769264221,
|
| 1065 |
+
"learning_rate": 4.852272727272728e-05,
|
| 1066 |
+
"loss": 0.4052,
|
| 1067 |
+
"step": 3550
|
| 1068 |
+
},
|
| 1069 |
+
{
|
| 1070 |
+
"epoch": 2.458498023715415,
|
| 1071 |
+
"grad_norm": 0.42250296473503113,
|
| 1072 |
+
"learning_rate": 4.8508232838589985e-05,
|
| 1073 |
+
"loss": 0.3786,
|
| 1074 |
+
"step": 3575
|
| 1075 |
+
},
|
| 1076 |
+
{
|
| 1077 |
+
"epoch": 2.475683107063069,
|
| 1078 |
+
"grad_norm": 0.6438080072402954,
|
| 1079 |
+
"learning_rate": 4.849373840445269e-05,
|
| 1080 |
+
"loss": 0.3529,
|
| 1081 |
+
"step": 3600
|
| 1082 |
+
},
|
| 1083 |
+
{
|
| 1084 |
+
"epoch": 2.4928681904107233,
|
| 1085 |
+
"grad_norm": 0.41823628544807434,
|
| 1086 |
+
"learning_rate": 4.84792439703154e-05,
|
| 1087 |
+
"loss": 0.4122,
|
| 1088 |
+
"step": 3625
|
| 1089 |
+
},
|
| 1090 |
+
{
|
| 1091 |
+
"epoch": 2.510053273758378,
|
| 1092 |
+
"grad_norm": 0.6400181651115417,
|
| 1093 |
+
"learning_rate": 4.8464749536178115e-05,
|
| 1094 |
+
"loss": 0.3602,
|
| 1095 |
+
"step": 3650
|
| 1096 |
+
},
|
| 1097 |
+
{
|
| 1098 |
+
"epoch": 2.527238357106032,
|
| 1099 |
+
"grad_norm": 0.40071290731430054,
|
| 1100 |
+
"learning_rate": 4.845025510204082e-05,
|
| 1101 |
+
"loss": 0.3968,
|
| 1102 |
+
"step": 3675
|
| 1103 |
+
},
|
| 1104 |
+
{
|
| 1105 |
+
"epoch": 2.544423440453686,
|
| 1106 |
+
"grad_norm": 1.0052437782287598,
|
| 1107 |
+
"learning_rate": 4.843576066790353e-05,
|
| 1108 |
+
"loss": 0.3927,
|
| 1109 |
+
"step": 3700
|
| 1110 |
+
},
|
| 1111 |
+
{
|
| 1112 |
+
"epoch": 2.5616085238013406,
|
| 1113 |
+
"grad_norm": 0.43629390001296997,
|
| 1114 |
+
"learning_rate": 4.842126623376624e-05,
|
| 1115 |
+
"loss": 0.3858,
|
| 1116 |
+
"step": 3725
|
| 1117 |
+
},
|
| 1118 |
+
{
|
| 1119 |
+
"epoch": 2.5787936071489947,
|
| 1120 |
+
"grad_norm": 0.8912670016288757,
|
| 1121 |
+
"learning_rate": 4.8406771799628944e-05,
|
| 1122 |
+
"loss": 0.3688,
|
| 1123 |
+
"step": 3750
|
| 1124 |
+
},
|
| 1125 |
+
{
|
| 1126 |
+
"epoch": 2.595978690496649,
|
| 1127 |
+
"grad_norm": 0.4484070837497711,
|
| 1128 |
+
"learning_rate": 4.839227736549166e-05,
|
| 1129 |
+
"loss": 0.4143,
|
| 1130 |
+
"step": 3775
|
| 1131 |
+
},
|
| 1132 |
+
{
|
| 1133 |
+
"epoch": 2.613163773844303,
|
| 1134 |
+
"grad_norm": 0.6882378458976746,
|
| 1135 |
+
"learning_rate": 4.837778293135436e-05,
|
| 1136 |
+
"loss": 0.3575,
|
| 1137 |
+
"step": 3800
|
| 1138 |
+
},
|
| 1139 |
+
{
|
| 1140 |
+
"epoch": 2.6303488571919575,
|
| 1141 |
+
"grad_norm": 0.4403606653213501,
|
| 1142 |
+
"learning_rate": 4.8363288497217067e-05,
|
| 1143 |
+
"loss": 0.4099,
|
| 1144 |
+
"step": 3825
|
| 1145 |
+
},
|
| 1146 |
+
{
|
| 1147 |
+
"epoch": 2.6475339405396117,
|
| 1148 |
+
"grad_norm": 0.7600814700126648,
|
| 1149 |
+
"learning_rate": 4.834879406307978e-05,
|
| 1150 |
+
"loss": 0.3732,
|
| 1151 |
+
"step": 3850
|
| 1152 |
+
},
|
| 1153 |
+
{
|
| 1154 |
+
"epoch": 2.664719023887266,
|
| 1155 |
+
"grad_norm": 0.3970819115638733,
|
| 1156 |
+
"learning_rate": 4.833429962894249e-05,
|
| 1157 |
+
"loss": 0.3994,
|
| 1158 |
+
"step": 3875
|
| 1159 |
+
},
|
| 1160 |
+
{
|
| 1161 |
+
"epoch": 2.68190410723492,
|
| 1162 |
+
"grad_norm": 0.8571271300315857,
|
| 1163 |
+
"learning_rate": 4.8319805194805196e-05,
|
| 1164 |
+
"loss": 0.3354,
|
| 1165 |
+
"step": 3900
|
| 1166 |
+
},
|
| 1167 |
+
{
|
| 1168 |
+
"epoch": 2.6990891905825745,
|
| 1169 |
+
"grad_norm": 0.514602780342102,
|
| 1170 |
+
"learning_rate": 4.83053107606679e-05,
|
| 1171 |
+
"loss": 0.4109,
|
| 1172 |
+
"step": 3925
|
| 1173 |
+
},
|
| 1174 |
+
{
|
| 1175 |
+
"epoch": 2.7162742739302286,
|
| 1176 |
+
"grad_norm": 0.5113664269447327,
|
| 1177 |
+
"learning_rate": 4.829081632653061e-05,
|
| 1178 |
+
"loss": 0.3742,
|
| 1179 |
+
"step": 3950
|
| 1180 |
+
},
|
| 1181 |
+
{
|
| 1182 |
+
"epoch": 2.7334593572778827,
|
| 1183 |
+
"grad_norm": 0.44583624601364136,
|
| 1184 |
+
"learning_rate": 4.8276321892393325e-05,
|
| 1185 |
+
"loss": 0.3896,
|
| 1186 |
+
"step": 3975
|
| 1187 |
+
},
|
| 1188 |
+
{
|
| 1189 |
+
"epoch": 2.7506444406255373,
|
| 1190 |
+
"grad_norm": 0.5685079097747803,
|
| 1191 |
+
"learning_rate": 4.826182745825603e-05,
|
| 1192 |
+
"loss": 0.3561,
|
| 1193 |
+
"step": 4000
|
| 1194 |
+
},
|
| 1195 |
+
{
|
| 1196 |
+
"epoch": 2.7506444406255373,
|
| 1197 |
+
"eval_loss": 0.3437730669975281,
|
| 1198 |
+
"eval_runtime": 152.7484,
|
| 1199 |
+
"eval_samples_per_second": 8.02,
|
| 1200 |
+
"eval_steps_per_second": 1.008,
|
| 1201 |
+
"eval_wer": 0.5716405256464604,
|
| 1202 |
+
"step": 4000
|
| 1203 |
+
},
|
| 1204 |
+
{
|
| 1205 |
+
"epoch": 2.7678295239731914,
|
| 1206 |
+
"grad_norm": 0.48729953169822693,
|
| 1207 |
+
"learning_rate": 4.824733302411874e-05,
|
| 1208 |
+
"loss": 0.3752,
|
| 1209 |
+
"step": 4025
|
| 1210 |
+
},
|
| 1211 |
+
{
|
| 1212 |
+
"epoch": 2.7850146073208455,
|
| 1213 |
+
"grad_norm": 0.64439457654953,
|
| 1214 |
+
"learning_rate": 4.823283858998145e-05,
|
| 1215 |
+
"loss": 0.3513,
|
| 1216 |
+
"step": 4050
|
| 1217 |
+
},
|
| 1218 |
+
{
|
| 1219 |
+
"epoch": 2.8021996906684996,
|
| 1220 |
+
"grad_norm": 0.4153486490249634,
|
| 1221 |
+
"learning_rate": 4.821834415584416e-05,
|
| 1222 |
+
"loss": 0.3979,
|
| 1223 |
+
"step": 4075
|
| 1224 |
+
},
|
| 1225 |
+
{
|
| 1226 |
+
"epoch": 2.8193847740161537,
|
| 1227 |
+
"grad_norm": 0.5988856554031372,
|
| 1228 |
+
"learning_rate": 4.820384972170687e-05,
|
| 1229 |
+
"loss": 0.3825,
|
| 1230 |
+
"step": 4100
|
| 1231 |
+
},
|
| 1232 |
+
{
|
| 1233 |
+
"epoch": 2.8365698573638083,
|
| 1234 |
+
"grad_norm": 0.8136705160140991,
|
| 1235 |
+
"learning_rate": 4.8189355287569576e-05,
|
| 1236 |
+
"loss": 0.4321,
|
| 1237 |
+
"step": 4125
|
| 1238 |
+
},
|
| 1239 |
+
{
|
| 1240 |
+
"epoch": 2.8537549407114624,
|
| 1241 |
+
"grad_norm": 0.6228049993515015,
|
| 1242 |
+
"learning_rate": 4.8174860853432284e-05,
|
| 1243 |
+
"loss": 0.3569,
|
| 1244 |
+
"step": 4150
|
| 1245 |
+
},
|
| 1246 |
+
{
|
| 1247 |
+
"epoch": 2.8709400240591165,
|
| 1248 |
+
"grad_norm": 0.5307970643043518,
|
| 1249 |
+
"learning_rate": 4.816036641929499e-05,
|
| 1250 |
+
"loss": 0.3775,
|
| 1251 |
+
"step": 4175
|
| 1252 |
+
},
|
| 1253 |
+
{
|
| 1254 |
+
"epoch": 2.888125107406771,
|
| 1255 |
+
"grad_norm": 0.5696175694465637,
|
| 1256 |
+
"learning_rate": 4.8145871985157705e-05,
|
| 1257 |
+
"loss": 0.3306,
|
| 1258 |
+
"step": 4200
|
| 1259 |
+
},
|
| 1260 |
+
{
|
| 1261 |
+
"epoch": 2.905310190754425,
|
| 1262 |
+
"grad_norm": 0.46450933814048767,
|
| 1263 |
+
"learning_rate": 4.813137755102041e-05,
|
| 1264 |
+
"loss": 0.4307,
|
| 1265 |
+
"step": 4225
|
| 1266 |
+
},
|
| 1267 |
+
{
|
| 1268 |
+
"epoch": 2.9224952741020793,
|
| 1269 |
+
"grad_norm": 0.6127625703811646,
|
| 1270 |
+
"learning_rate": 4.811688311688312e-05,
|
| 1271 |
+
"loss": 0.3604,
|
| 1272 |
+
"step": 4250
|
| 1273 |
+
},
|
| 1274 |
+
{
|
| 1275 |
+
"epoch": 2.939680357449734,
|
| 1276 |
+
"grad_norm": 0.5017271637916565,
|
| 1277 |
+
"learning_rate": 4.810238868274583e-05,
|
| 1278 |
+
"loss": 0.3721,
|
| 1279 |
+
"step": 4275
|
| 1280 |
+
},
|
| 1281 |
+
{
|
| 1282 |
+
"epoch": 2.956865440797388,
|
| 1283 |
+
"grad_norm": 0.6709030866622925,
|
| 1284 |
+
"learning_rate": 4.808789424860854e-05,
|
| 1285 |
+
"loss": 0.3603,
|
| 1286 |
+
"step": 4300
|
| 1287 |
+
},
|
| 1288 |
+
{
|
| 1289 |
+
"epoch": 2.974050524145042,
|
| 1290 |
+
"grad_norm": 0.40175503492355347,
|
| 1291 |
+
"learning_rate": 4.807339981447124e-05,
|
| 1292 |
+
"loss": 0.376,
|
| 1293 |
+
"step": 4325
|
| 1294 |
+
},
|
| 1295 |
+
{
|
| 1296 |
+
"epoch": 2.9912356074926962,
|
| 1297 |
+
"grad_norm": 0.6254987716674805,
|
| 1298 |
+
"learning_rate": 4.805890538033395e-05,
|
| 1299 |
+
"loss": 0.3316,
|
| 1300 |
+
"step": 4350
|
| 1301 |
+
},
|
| 1302 |
+
{
|
| 1303 |
+
"epoch": 3.0089362433407802,
|
| 1304 |
+
"grad_norm": 0.4055463373661041,
|
| 1305 |
+
"learning_rate": 4.804441094619666e-05,
|
| 1306 |
+
"loss": 0.4126,
|
| 1307 |
+
"step": 4375
|
| 1308 |
+
},
|
| 1309 |
+
{
|
| 1310 |
+
"epoch": 3.0261213266884344,
|
| 1311 |
+
"grad_norm": 0.5347069501876831,
|
| 1312 |
+
"learning_rate": 4.802991651205937e-05,
|
| 1313 |
+
"loss": 0.3477,
|
| 1314 |
+
"step": 4400
|
| 1315 |
+
},
|
| 1316 |
+
{
|
| 1317 |
+
"epoch": 3.0433064100360885,
|
| 1318 |
+
"grad_norm": 0.4559672474861145,
|
| 1319 |
+
"learning_rate": 4.801542207792208e-05,
|
| 1320 |
+
"loss": 0.3458,
|
| 1321 |
+
"step": 4425
|
| 1322 |
+
},
|
| 1323 |
+
{
|
| 1324 |
+
"epoch": 3.060491493383743,
|
| 1325 |
+
"grad_norm": 0.46173930168151855,
|
| 1326 |
+
"learning_rate": 4.8000927643784786e-05,
|
| 1327 |
+
"loss": 0.3316,
|
| 1328 |
+
"step": 4450
|
| 1329 |
+
},
|
| 1330 |
+
{
|
| 1331 |
+
"epoch": 3.077676576731397,
|
| 1332 |
+
"grad_norm": 0.414719820022583,
|
| 1333 |
+
"learning_rate": 4.7986433209647494e-05,
|
| 1334 |
+
"loss": 0.3719,
|
| 1335 |
+
"step": 4475
|
| 1336 |
+
},
|
| 1337 |
+
{
|
| 1338 |
+
"epoch": 3.0948616600790513,
|
| 1339 |
+
"grad_norm": 0.7085908055305481,
|
| 1340 |
+
"learning_rate": 4.797193877551021e-05,
|
| 1341 |
+
"loss": 0.3237,
|
| 1342 |
+
"step": 4500
|
| 1343 |
+
},
|
| 1344 |
+
{
|
| 1345 |
+
"epoch": 3.0948616600790513,
|
| 1346 |
+
"eval_loss": 0.3646816313266754,
|
| 1347 |
+
"eval_runtime": 153.9661,
|
| 1348 |
+
"eval_samples_per_second": 7.956,
|
| 1349 |
+
"eval_steps_per_second": 1.0,
|
| 1350 |
+
"eval_wer": 0.5677193726155151,
|
| 1351 |
+
"step": 4500
|
| 1352 |
+
},
|
| 1353 |
+
{
|
| 1354 |
+
"epoch": 3.1120467434267054,
|
| 1355 |
+
"grad_norm": 0.4840669631958008,
|
| 1356 |
+
"learning_rate": 4.7957444341372916e-05,
|
| 1357 |
+
"loss": 0.3515,
|
| 1358 |
+
"step": 4525
|
| 1359 |
+
},
|
| 1360 |
+
{
|
| 1361 |
+
"epoch": 3.12923182677436,
|
| 1362 |
+
"grad_norm": 0.6030757427215576,
|
| 1363 |
+
"learning_rate": 4.794294990723562e-05,
|
| 1364 |
+
"loss": 0.3263,
|
| 1365 |
+
"step": 4550
|
| 1366 |
+
},
|
| 1367 |
+
{
|
| 1368 |
+
"epoch": 3.146416910122014,
|
| 1369 |
+
"grad_norm": 0.5091059803962708,
|
| 1370 |
+
"learning_rate": 4.792845547309833e-05,
|
| 1371 |
+
"loss": 0.3315,
|
| 1372 |
+
"step": 4575
|
| 1373 |
+
},
|
| 1374 |
+
{
|
| 1375 |
+
"epoch": 3.163601993469668,
|
| 1376 |
+
"grad_norm": 0.7523996829986572,
|
| 1377 |
+
"learning_rate": 4.7913961038961045e-05,
|
| 1378 |
+
"loss": 0.3368,
|
| 1379 |
+
"step": 4600
|
| 1380 |
+
},
|
| 1381 |
+
{
|
| 1382 |
+
"epoch": 3.1807870768173228,
|
| 1383 |
+
"grad_norm": 1.0035797357559204,
|
| 1384 |
+
"learning_rate": 4.789946660482375e-05,
|
| 1385 |
+
"loss": 0.3931,
|
| 1386 |
+
"step": 4625
|
| 1387 |
+
},
|
| 1388 |
+
{
|
| 1389 |
+
"epoch": 3.197972160164977,
|
| 1390 |
+
"grad_norm": 0.5936137437820435,
|
| 1391 |
+
"learning_rate": 4.788497217068646e-05,
|
| 1392 |
+
"loss": 0.2896,
|
| 1393 |
+
"step": 4650
|
| 1394 |
+
},
|
| 1395 |
+
{
|
| 1396 |
+
"epoch": 3.215157243512631,
|
| 1397 |
+
"grad_norm": 0.5628079771995544,
|
| 1398 |
+
"learning_rate": 4.787047773654917e-05,
|
| 1399 |
+
"loss": 0.3632,
|
| 1400 |
+
"step": 4675
|
| 1401 |
+
},
|
| 1402 |
+
{
|
| 1403 |
+
"epoch": 3.232342326860285,
|
| 1404 |
+
"grad_norm": 0.5175526738166809,
|
| 1405 |
+
"learning_rate": 4.7855983302411874e-05,
|
| 1406 |
+
"loss": 0.33,
|
| 1407 |
+
"step": 4700
|
| 1408 |
+
},
|
| 1409 |
+
{
|
| 1410 |
+
"epoch": 3.2495274102079397,
|
| 1411 |
+
"grad_norm": 0.36029067635536194,
|
| 1412 |
+
"learning_rate": 4.784148886827459e-05,
|
| 1413 |
+
"loss": 0.3487,
|
| 1414 |
+
"step": 4725
|
| 1415 |
+
},
|
| 1416 |
+
{
|
| 1417 |
+
"epoch": 3.266712493555594,
|
| 1418 |
+
"grad_norm": 0.6295140981674194,
|
| 1419 |
+
"learning_rate": 4.7826994434137296e-05,
|
| 1420 |
+
"loss": 0.3391,
|
| 1421 |
+
"step": 4750
|
| 1422 |
+
},
|
| 1423 |
+
{
|
| 1424 |
+
"epoch": 3.283897576903248,
|
| 1425 |
+
"grad_norm": 0.535555362701416,
|
| 1426 |
+
"learning_rate": 4.7812500000000003e-05,
|
| 1427 |
+
"loss": 0.3804,
|
| 1428 |
+
"step": 4775
|
| 1429 |
+
},
|
| 1430 |
+
{
|
| 1431 |
+
"epoch": 3.301082660250902,
|
| 1432 |
+
"grad_norm": 1.0075314044952393,
|
| 1433 |
+
"learning_rate": 4.779800556586271e-05,
|
| 1434 |
+
"loss": 0.3338,
|
| 1435 |
+
"step": 4800
|
| 1436 |
+
},
|
| 1437 |
+
{
|
| 1438 |
+
"epoch": 3.3182677435985566,
|
| 1439 |
+
"grad_norm": 0.3681296110153198,
|
| 1440 |
+
"learning_rate": 4.7783511131725425e-05,
|
| 1441 |
+
"loss": 0.3727,
|
| 1442 |
+
"step": 4825
|
| 1443 |
+
},
|
| 1444 |
+
{
|
| 1445 |
+
"epoch": 3.3354528269462107,
|
| 1446 |
+
"grad_norm": 0.5938307642936707,
|
| 1447 |
+
"learning_rate": 4.7769016697588126e-05,
|
| 1448 |
+
"loss": 0.3565,
|
| 1449 |
+
"step": 4850
|
| 1450 |
+
},
|
| 1451 |
+
{
|
| 1452 |
+
"epoch": 3.352637910293865,
|
| 1453 |
+
"grad_norm": 0.4364496171474457,
|
| 1454 |
+
"learning_rate": 4.775452226345083e-05,
|
| 1455 |
+
"loss": 0.3459,
|
| 1456 |
+
"step": 4875
|
| 1457 |
+
},
|
| 1458 |
+
{
|
| 1459 |
+
"epoch": 3.3698229936415194,
|
| 1460 |
+
"grad_norm": 0.5793933272361755,
|
| 1461 |
+
"learning_rate": 4.774002782931354e-05,
|
| 1462 |
+
"loss": 0.3653,
|
| 1463 |
+
"step": 4900
|
| 1464 |
+
},
|
| 1465 |
+
{
|
| 1466 |
+
"epoch": 3.3870080769891735,
|
| 1467 |
+
"grad_norm": 0.41033703088760376,
|
| 1468 |
+
"learning_rate": 4.7725533395176255e-05,
|
| 1469 |
+
"loss": 0.3629,
|
| 1470 |
+
"step": 4925
|
| 1471 |
+
},
|
| 1472 |
+
{
|
| 1473 |
+
"epoch": 3.4041931603368276,
|
| 1474 |
+
"grad_norm": 0.6783180236816406,
|
| 1475 |
+
"learning_rate": 4.771103896103896e-05,
|
| 1476 |
+
"loss": 0.3258,
|
| 1477 |
+
"step": 4950
|
| 1478 |
+
},
|
| 1479 |
+
{
|
| 1480 |
+
"epoch": 3.4213782436844817,
|
| 1481 |
+
"grad_norm": 0.4356047511100769,
|
| 1482 |
+
"learning_rate": 4.769654452690167e-05,
|
| 1483 |
+
"loss": 0.3423,
|
| 1484 |
+
"step": 4975
|
| 1485 |
+
},
|
| 1486 |
+
{
|
| 1487 |
+
"epoch": 3.4385633270321363,
|
| 1488 |
+
"grad_norm": 0.645196795463562,
|
| 1489 |
+
"learning_rate": 4.768205009276438e-05,
|
| 1490 |
+
"loss": 0.322,
|
| 1491 |
+
"step": 5000
|
| 1492 |
+
},
|
| 1493 |
+
{
|
| 1494 |
+
"epoch": 3.4385633270321363,
|
| 1495 |
+
"eval_loss": 0.3426838219165802,
|
| 1496 |
+
"eval_runtime": 154.5652,
|
| 1497 |
+
"eval_samples_per_second": 7.925,
|
| 1498 |
+
"eval_steps_per_second": 0.996,
|
| 1499 |
+
"eval_wer": 0.5637982195845698,
|
| 1500 |
+
"step": 5000
|
| 1501 |
+
},
|
| 1502 |
+
{
|
| 1503 |
+
"epoch": 3.4557484103797904,
|
| 1504 |
+
"grad_norm": 0.7585852742195129,
|
| 1505 |
+
"learning_rate": 4.766755565862709e-05,
|
| 1506 |
+
"loss": 0.3615,
|
| 1507 |
+
"step": 5025
|
| 1508 |
+
},
|
| 1509 |
+
{
|
| 1510 |
+
"epoch": 3.4729334937274445,
|
| 1511 |
+
"grad_norm": 0.6891105771064758,
|
| 1512 |
+
"learning_rate": 4.76530612244898e-05,
|
| 1513 |
+
"loss": 0.3016,
|
| 1514 |
+
"step": 5050
|
| 1515 |
+
},
|
| 1516 |
+
{
|
| 1517 |
+
"epoch": 3.4901185770750986,
|
| 1518 |
+
"grad_norm": 0.40171509981155396,
|
| 1519 |
+
"learning_rate": 4.7638566790352506e-05,
|
| 1520 |
+
"loss": 0.3389,
|
| 1521 |
+
"step": 5075
|
| 1522 |
+
},
|
| 1523 |
+
{
|
| 1524 |
+
"epoch": 3.507303660422753,
|
| 1525 |
+
"grad_norm": 0.6688668131828308,
|
| 1526 |
+
"learning_rate": 4.7624072356215214e-05,
|
| 1527 |
+
"loss": 0.3196,
|
| 1528 |
+
"step": 5100
|
| 1529 |
+
},
|
| 1530 |
+
{
|
| 1531 |
+
"epoch": 3.5244887437704073,
|
| 1532 |
+
"grad_norm": 0.7812600135803223,
|
| 1533 |
+
"learning_rate": 4.760957792207793e-05,
|
| 1534 |
+
"loss": 0.3795,
|
| 1535 |
+
"step": 5125
|
| 1536 |
+
},
|
| 1537 |
+
{
|
| 1538 |
+
"epoch": 3.5416738271180614,
|
| 1539 |
+
"grad_norm": 0.6127042174339294,
|
| 1540 |
+
"learning_rate": 4.7595083487940635e-05,
|
| 1541 |
+
"loss": 0.3422,
|
| 1542 |
+
"step": 5150
|
| 1543 |
+
},
|
| 1544 |
+
{
|
| 1545 |
+
"epoch": 3.558858910465716,
|
| 1546 |
+
"grad_norm": 0.4893074333667755,
|
| 1547 |
+
"learning_rate": 4.758058905380334e-05,
|
| 1548 |
+
"loss": 0.3759,
|
| 1549 |
+
"step": 5175
|
| 1550 |
+
},
|
| 1551 |
+
{
|
| 1552 |
+
"epoch": 3.57604399381337,
|
| 1553 |
+
"grad_norm": 0.7298163175582886,
|
| 1554 |
+
"learning_rate": 4.756609461966605e-05,
|
| 1555 |
+
"loss": 0.3388,
|
| 1556 |
+
"step": 5200
|
| 1557 |
+
},
|
| 1558 |
+
{
|
| 1559 |
+
"epoch": 3.5932290771610242,
|
| 1560 |
+
"grad_norm": 0.5726421475410461,
|
| 1561 |
+
"learning_rate": 4.755160018552876e-05,
|
| 1562 |
+
"loss": 0.4172,
|
| 1563 |
+
"step": 5225
|
| 1564 |
+
},
|
| 1565 |
+
{
|
| 1566 |
+
"epoch": 3.6104141605086784,
|
| 1567 |
+
"grad_norm": 0.5771546959877014,
|
| 1568 |
+
"learning_rate": 4.753710575139147e-05,
|
| 1569 |
+
"loss": 0.3217,
|
| 1570 |
+
"step": 5250
|
| 1571 |
+
},
|
| 1572 |
+
{
|
| 1573 |
+
"epoch": 3.6275992438563325,
|
| 1574 |
+
"grad_norm": 0.48307299613952637,
|
| 1575 |
+
"learning_rate": 4.752261131725418e-05,
|
| 1576 |
+
"loss": 0.3864,
|
| 1577 |
+
"step": 5275
|
| 1578 |
+
},
|
| 1579 |
+
{
|
| 1580 |
+
"epoch": 3.644784327203987,
|
| 1581 |
+
"grad_norm": 0.5440219044685364,
|
| 1582 |
+
"learning_rate": 4.750811688311689e-05,
|
| 1583 |
+
"loss": 0.3288,
|
| 1584 |
+
"step": 5300
|
| 1585 |
+
},
|
| 1586 |
+
{
|
| 1587 |
+
"epoch": 3.661969410551641,
|
| 1588 |
+
"grad_norm": 0.4851985275745392,
|
| 1589 |
+
"learning_rate": 4.7493622448979594e-05,
|
| 1590 |
+
"loss": 0.4105,
|
| 1591 |
+
"step": 5325
|
| 1592 |
+
},
|
| 1593 |
+
{
|
| 1594 |
+
"epoch": 3.6791544938992953,
|
| 1595 |
+
"grad_norm": 0.537399172782898,
|
| 1596 |
+
"learning_rate": 4.74791280148423e-05,
|
| 1597 |
+
"loss": 0.3309,
|
| 1598 |
+
"step": 5350
|
| 1599 |
+
},
|
| 1600 |
+
{
|
| 1601 |
+
"epoch": 3.69633957724695,
|
| 1602 |
+
"grad_norm": 0.4729978144168854,
|
| 1603 |
+
"learning_rate": 4.746463358070501e-05,
|
| 1604 |
+
"loss": 0.3604,
|
| 1605 |
+
"step": 5375
|
| 1606 |
+
},
|
| 1607 |
+
{
|
| 1608 |
+
"epoch": 3.713524660594604,
|
| 1609 |
+
"grad_norm": 0.8028717041015625,
|
| 1610 |
+
"learning_rate": 4.7450139146567716e-05,
|
| 1611 |
+
"loss": 0.2908,
|
| 1612 |
+
"step": 5400
|
| 1613 |
+
},
|
| 1614 |
+
{
|
| 1615 |
+
"epoch": 3.730709743942258,
|
| 1616 |
+
"grad_norm": 0.40672922134399414,
|
| 1617 |
+
"learning_rate": 4.7435644712430424e-05,
|
| 1618 |
+
"loss": 0.3871,
|
| 1619 |
+
"step": 5425
|
| 1620 |
+
},
|
| 1621 |
+
{
|
| 1622 |
+
"epoch": 3.747894827289912,
|
| 1623 |
+
"grad_norm": 0.6138872504234314,
|
| 1624 |
+
"learning_rate": 4.742115027829314e-05,
|
| 1625 |
+
"loss": 0.3166,
|
| 1626 |
+
"step": 5450
|
| 1627 |
+
},
|
| 1628 |
+
{
|
| 1629 |
+
"epoch": 3.7650799106375668,
|
| 1630 |
+
"grad_norm": 0.4270385801792145,
|
| 1631 |
+
"learning_rate": 4.7406655844155846e-05,
|
| 1632 |
+
"loss": 0.3449,
|
| 1633 |
+
"step": 5475
|
| 1634 |
+
},
|
| 1635 |
+
{
|
| 1636 |
+
"epoch": 3.782264993985221,
|
| 1637 |
+
"grad_norm": 0.535121500492096,
|
| 1638 |
+
"learning_rate": 4.739216141001855e-05,
|
| 1639 |
+
"loss": 0.2921,
|
| 1640 |
+
"step": 5500
|
| 1641 |
+
},
|
| 1642 |
+
{
|
| 1643 |
+
"epoch": 3.782264993985221,
|
| 1644 |
+
"eval_loss": 0.3344533145427704,
|
| 1645 |
+
"eval_runtime": 150.3102,
|
| 1646 |
+
"eval_samples_per_second": 8.15,
|
| 1647 |
+
"eval_steps_per_second": 1.025,
|
| 1648 |
+
"eval_wer": 0.5604069520983468,
|
| 1649 |
+
"step": 5500
|
| 1650 |
+
},
|
| 1651 |
+
{
|
| 1652 |
+
"epoch": 3.799450077332875,
|
| 1653 |
+
"grad_norm": 0.3871770203113556,
|
| 1654 |
+
"learning_rate": 4.737766697588126e-05,
|
| 1655 |
+
"loss": 0.349,
|
| 1656 |
+
"step": 5525
|
| 1657 |
+
},
|
| 1658 |
+
{
|
| 1659 |
+
"epoch": 3.816635160680529,
|
| 1660 |
+
"grad_norm": 0.503182053565979,
|
| 1661 |
+
"learning_rate": 4.7363172541743975e-05,
|
| 1662 |
+
"loss": 0.3707,
|
| 1663 |
+
"step": 5550
|
| 1664 |
+
},
|
| 1665 |
+
{
|
| 1666 |
+
"epoch": 3.8338202440281837,
|
| 1667 |
+
"grad_norm": 0.4528012275695801,
|
| 1668 |
+
"learning_rate": 4.734867810760668e-05,
|
| 1669 |
+
"loss": 0.359,
|
| 1670 |
+
"step": 5575
|
| 1671 |
+
},
|
| 1672 |
+
{
|
| 1673 |
+
"epoch": 3.851005327375838,
|
| 1674 |
+
"grad_norm": 0.630174458026886,
|
| 1675 |
+
"learning_rate": 4.733418367346939e-05,
|
| 1676 |
+
"loss": 0.3558,
|
| 1677 |
+
"step": 5600
|
| 1678 |
+
},
|
| 1679 |
+
{
|
| 1680 |
+
"epoch": 3.868190410723492,
|
| 1681 |
+
"grad_norm": 0.4319029450416565,
|
| 1682 |
+
"learning_rate": 4.73196892393321e-05,
|
| 1683 |
+
"loss": 0.3812,
|
| 1684 |
+
"step": 5625
|
| 1685 |
+
},
|
| 1686 |
+
{
|
| 1687 |
+
"epoch": 3.8853754940711465,
|
| 1688 |
+
"grad_norm": 0.5308706760406494,
|
| 1689 |
+
"learning_rate": 4.730519480519481e-05,
|
| 1690 |
+
"loss": 0.2885,
|
| 1691 |
+
"step": 5650
|
| 1692 |
+
},
|
| 1693 |
+
{
|
| 1694 |
+
"epoch": 3.9025605774188006,
|
| 1695 |
+
"grad_norm": 0.4054734408855438,
|
| 1696 |
+
"learning_rate": 4.729070037105752e-05,
|
| 1697 |
+
"loss": 0.363,
|
| 1698 |
+
"step": 5675
|
| 1699 |
+
},
|
| 1700 |
+
{
|
| 1701 |
+
"epoch": 3.9197456607664547,
|
| 1702 |
+
"grad_norm": 0.8012121319770813,
|
| 1703 |
+
"learning_rate": 4.7276205936920226e-05,
|
| 1704 |
+
"loss": 0.3398,
|
| 1705 |
+
"step": 5700
|
| 1706 |
+
},
|
| 1707 |
+
{
|
| 1708 |
+
"epoch": 3.936930744114109,
|
| 1709 |
+
"grad_norm": 0.4499848783016205,
|
| 1710 |
+
"learning_rate": 4.7261711502782933e-05,
|
| 1711 |
+
"loss": 0.3485,
|
| 1712 |
+
"step": 5725
|
| 1713 |
+
},
|
| 1714 |
+
{
|
| 1715 |
+
"epoch": 3.954115827461763,
|
| 1716 |
+
"grad_norm": 0.5845701098442078,
|
| 1717 |
+
"learning_rate": 4.724721706864564e-05,
|
| 1718 |
+
"loss": 0.3188,
|
| 1719 |
+
"step": 5750
|
| 1720 |
+
},
|
| 1721 |
+
{
|
| 1722 |
+
"epoch": 3.9713009108094175,
|
| 1723 |
+
"grad_norm": 0.4100358486175537,
|
| 1724 |
+
"learning_rate": 4.7232722634508355e-05,
|
| 1725 |
+
"loss": 0.3479,
|
| 1726 |
+
"step": 5775
|
| 1727 |
+
},
|
| 1728 |
+
{
|
| 1729 |
+
"epoch": 3.9884859941570716,
|
| 1730 |
+
"grad_norm": 0.42875462770462036,
|
| 1731 |
+
"learning_rate": 4.721822820037106e-05,
|
| 1732 |
+
"loss": 0.3043,
|
| 1733 |
+
"step": 5800
|
| 1734 |
+
},
|
| 1735 |
+
{
|
| 1736 |
+
"epoch": 4.006186630005155,
|
| 1737 |
+
"grad_norm": 0.4040299355983734,
|
| 1738 |
+
"learning_rate": 4.720373376623377e-05,
|
| 1739 |
+
"loss": 0.4045,
|
| 1740 |
+
"step": 5825
|
| 1741 |
+
},
|
| 1742 |
+
{
|
| 1743 |
+
"epoch": 4.02337171335281,
|
| 1744 |
+
"grad_norm": 0.4601922035217285,
|
| 1745 |
+
"learning_rate": 4.718923933209648e-05,
|
| 1746 |
+
"loss": 0.3044,
|
| 1747 |
+
"step": 5850
|
| 1748 |
+
},
|
| 1749 |
+
{
|
| 1750 |
+
"epoch": 4.040556796700464,
|
| 1751 |
+
"grad_norm": 0.5125726461410522,
|
| 1752 |
+
"learning_rate": 4.7174744897959185e-05,
|
| 1753 |
+
"loss": 0.342,
|
| 1754 |
+
"step": 5875
|
| 1755 |
+
},
|
| 1756 |
+
{
|
| 1757 |
+
"epoch": 4.057741880048118,
|
| 1758 |
+
"grad_norm": 0.5830023884773254,
|
| 1759 |
+
"learning_rate": 4.716025046382189e-05,
|
| 1760 |
+
"loss": 0.2916,
|
| 1761 |
+
"step": 5900
|
| 1762 |
+
},
|
| 1763 |
+
{
|
| 1764 |
+
"epoch": 4.0749269633957725,
|
| 1765 |
+
"grad_norm": 0.3486079275608063,
|
| 1766 |
+
"learning_rate": 4.71457560296846e-05,
|
| 1767 |
+
"loss": 0.3215,
|
| 1768 |
+
"step": 5925
|
| 1769 |
+
},
|
| 1770 |
+
{
|
| 1771 |
+
"epoch": 4.092112046743427,
|
| 1772 |
+
"grad_norm": 0.5681314468383789,
|
| 1773 |
+
"learning_rate": 4.713126159554731e-05,
|
| 1774 |
+
"loss": 0.3049,
|
| 1775 |
+
"step": 5950
|
| 1776 |
+
},
|
| 1777 |
+
{
|
| 1778 |
+
"epoch": 4.109297130091081,
|
| 1779 |
+
"grad_norm": 0.3579752743244171,
|
| 1780 |
+
"learning_rate": 4.711676716141002e-05,
|
| 1781 |
+
"loss": 0.3167,
|
| 1782 |
+
"step": 5975
|
| 1783 |
+
},
|
| 1784 |
+
{
|
| 1785 |
+
"epoch": 4.126482213438735,
|
| 1786 |
+
"grad_norm": 0.8702667355537415,
|
| 1787 |
+
"learning_rate": 4.710227272727273e-05,
|
| 1788 |
+
"loss": 0.3037,
|
| 1789 |
+
"step": 6000
|
| 1790 |
+
},
|
| 1791 |
+
{
|
| 1792 |
+
"epoch": 4.126482213438735,
|
| 1793 |
+
"eval_loss": 0.335175484418869,
|
| 1794 |
+
"eval_runtime": 150.3279,
|
| 1795 |
+
"eval_samples_per_second": 8.149,
|
| 1796 |
+
"eval_steps_per_second": 1.024,
|
| 1797 |
+
"eval_wer": 0.5541013141161509,
|
| 1798 |
+
"step": 6000
|
| 1799 |
+
},
|
| 1800 |
+
{
|
| 1801 |
+
"epoch": 4.143667296786389,
|
| 1802 |
+
"grad_norm": 0.6025490760803223,
|
| 1803 |
+
"learning_rate": 4.7087778293135436e-05,
|
| 1804 |
+
"loss": 0.3527,
|
| 1805 |
+
"step": 6025
|
| 1806 |
+
},
|
| 1807 |
+
{
|
| 1808 |
+
"epoch": 4.160852380134044,
|
| 1809 |
+
"grad_norm": 0.46763402223587036,
|
| 1810 |
+
"learning_rate": 4.7073283858998144e-05,
|
| 1811 |
+
"loss": 0.3182,
|
| 1812 |
+
"step": 6050
|
| 1813 |
+
},
|
| 1814 |
+
{
|
| 1815 |
+
"epoch": 4.178037463481698,
|
| 1816 |
+
"grad_norm": 0.38680383563041687,
|
| 1817 |
+
"learning_rate": 4.705878942486086e-05,
|
| 1818 |
+
"loss": 0.3234,
|
| 1819 |
+
"step": 6075
|
| 1820 |
+
},
|
| 1821 |
+
{
|
| 1822 |
+
"epoch": 4.195222546829352,
|
| 1823 |
+
"grad_norm": 0.45606276392936707,
|
| 1824 |
+
"learning_rate": 4.7044294990723565e-05,
|
| 1825 |
+
"loss": 0.3418,
|
| 1826 |
+
"step": 6100
|
| 1827 |
+
},
|
| 1828 |
+
{
|
| 1829 |
+
"epoch": 4.212407630177006,
|
| 1830 |
+
"grad_norm": 0.5080279111862183,
|
| 1831 |
+
"learning_rate": 4.702980055658627e-05,
|
| 1832 |
+
"loss": 0.3402,
|
| 1833 |
+
"step": 6125
|
| 1834 |
+
},
|
| 1835 |
+
{
|
| 1836 |
+
"epoch": 4.229592713524661,
|
| 1837 |
+
"grad_norm": 0.5734138488769531,
|
| 1838 |
+
"learning_rate": 4.701530612244898e-05,
|
| 1839 |
+
"loss": 0.3026,
|
| 1840 |
+
"step": 6150
|
| 1841 |
+
},
|
| 1842 |
+
{
|
| 1843 |
+
"epoch": 4.246777796872315,
|
| 1844 |
+
"grad_norm": 0.34839344024658203,
|
| 1845 |
+
"learning_rate": 4.7000811688311694e-05,
|
| 1846 |
+
"loss": 0.3422,
|
| 1847 |
+
"step": 6175
|
| 1848 |
+
},
|
| 1849 |
+
{
|
| 1850 |
+
"epoch": 4.263962880219969,
|
| 1851 |
+
"grad_norm": 0.5648381114006042,
|
| 1852 |
+
"learning_rate": 4.69863172541744e-05,
|
| 1853 |
+
"loss": 0.3075,
|
| 1854 |
+
"step": 6200
|
| 1855 |
+
},
|
| 1856 |
+
{
|
| 1857 |
+
"epoch": 4.281147963567623,
|
| 1858 |
+
"grad_norm": 0.5454294681549072,
|
| 1859 |
+
"learning_rate": 4.697182282003711e-05,
|
| 1860 |
+
"loss": 0.3528,
|
| 1861 |
+
"step": 6225
|
| 1862 |
+
},
|
| 1863 |
+
{
|
| 1864 |
+
"epoch": 4.298333046915277,
|
| 1865 |
+
"grad_norm": 0.5028226375579834,
|
| 1866 |
+
"learning_rate": 4.695732838589982e-05,
|
| 1867 |
+
"loss": 0.2952,
|
| 1868 |
+
"step": 6250
|
| 1869 |
+
},
|
| 1870 |
+
{
|
| 1871 |
+
"epoch": 4.315518130262932,
|
| 1872 |
+
"grad_norm": 0.45058056712150574,
|
| 1873 |
+
"learning_rate": 4.6942833951762524e-05,
|
| 1874 |
+
"loss": 0.3535,
|
| 1875 |
+
"step": 6275
|
| 1876 |
+
},
|
| 1877 |
+
{
|
| 1878 |
+
"epoch": 4.332703213610586,
|
| 1879 |
+
"grad_norm": 0.6654832363128662,
|
| 1880 |
+
"learning_rate": 4.692833951762524e-05,
|
| 1881 |
+
"loss": 0.3127,
|
| 1882 |
+
"step": 6300
|
| 1883 |
+
},
|
| 1884 |
+
{
|
| 1885 |
+
"epoch": 4.34988829695824,
|
| 1886 |
+
"grad_norm": 0.49009886384010315,
|
| 1887 |
+
"learning_rate": 4.6913845083487946e-05,
|
| 1888 |
+
"loss": 0.3419,
|
| 1889 |
+
"step": 6325
|
| 1890 |
+
},
|
| 1891 |
+
{
|
| 1892 |
+
"epoch": 4.367073380305895,
|
| 1893 |
+
"grad_norm": 0.5751463770866394,
|
| 1894 |
+
"learning_rate": 4.689935064935065e-05,
|
| 1895 |
+
"loss": 0.3069,
|
| 1896 |
+
"step": 6350
|
| 1897 |
+
},
|
| 1898 |
+
{
|
| 1899 |
+
"epoch": 4.384258463653548,
|
| 1900 |
+
"grad_norm": 0.7767444849014282,
|
| 1901 |
+
"learning_rate": 4.688485621521336e-05,
|
| 1902 |
+
"loss": 0.3666,
|
| 1903 |
+
"step": 6375
|
| 1904 |
+
},
|
| 1905 |
+
{
|
| 1906 |
+
"epoch": 4.401443547001203,
|
| 1907 |
+
"grad_norm": 0.5131709575653076,
|
| 1908 |
+
"learning_rate": 4.687036178107607e-05,
|
| 1909 |
+
"loss": 0.329,
|
| 1910 |
+
"step": 6400
|
| 1911 |
+
},
|
| 1912 |
+
{
|
| 1913 |
+
"epoch": 4.4186286303488576,
|
| 1914 |
+
"grad_norm": 0.4997400939464569,
|
| 1915 |
+
"learning_rate": 4.6855867346938776e-05,
|
| 1916 |
+
"loss": 0.365,
|
| 1917 |
+
"step": 6425
|
| 1918 |
+
},
|
| 1919 |
+
{
|
| 1920 |
+
"epoch": 4.435813713696511,
|
| 1921 |
+
"grad_norm": 0.5275589227676392,
|
| 1922 |
+
"learning_rate": 4.684137291280148e-05,
|
| 1923 |
+
"loss": 0.31,
|
| 1924 |
+
"step": 6450
|
| 1925 |
+
},
|
| 1926 |
+
{
|
| 1927 |
+
"epoch": 4.452998797044166,
|
| 1928 |
+
"grad_norm": 0.43651729822158813,
|
| 1929 |
+
"learning_rate": 4.682687847866419e-05,
|
| 1930 |
+
"loss": 0.3253,
|
| 1931 |
+
"step": 6475
|
| 1932 |
+
},
|
| 1933 |
+
{
|
| 1934 |
+
"epoch": 4.47018388039182,
|
| 1935 |
+
"grad_norm": 0.49254560470581055,
|
| 1936 |
+
"learning_rate": 4.6812384044526905e-05,
|
| 1937 |
+
"loss": 0.2695,
|
| 1938 |
+
"step": 6500
|
| 1939 |
+
},
|
| 1940 |
+
{
|
| 1941 |
+
"epoch": 4.47018388039182,
|
| 1942 |
+
"eval_loss": 0.32015639543533325,
|
| 1943 |
+
"eval_runtime": 154.5447,
|
| 1944 |
+
"eval_samples_per_second": 7.927,
|
| 1945 |
+
"eval_steps_per_second": 0.996,
|
| 1946 |
+
"eval_wer": 0.5515048749470114,
|
| 1947 |
+
"step": 6500
|
| 1948 |
+
},
|
| 1949 |
+
{
|
| 1950 |
+
"epoch": 4.487368963739474,
|
| 1951 |
+
"grad_norm": 0.3705599904060364,
|
| 1952 |
+
"learning_rate": 4.679788961038961e-05,
|
| 1953 |
+
"loss": 0.3453,
|
| 1954 |
+
"step": 6525
|
| 1955 |
+
},
|
| 1956 |
+
{
|
| 1957 |
+
"epoch": 4.504554047087129,
|
| 1958 |
+
"grad_norm": 0.5254660844802856,
|
| 1959 |
+
"learning_rate": 4.678339517625232e-05,
|
| 1960 |
+
"loss": 0.282,
|
| 1961 |
+
"step": 6550
|
| 1962 |
+
},
|
| 1963 |
+
{
|
| 1964 |
+
"epoch": 4.521739130434782,
|
| 1965 |
+
"grad_norm": 0.37494751811027527,
|
| 1966 |
+
"learning_rate": 4.676890074211503e-05,
|
| 1967 |
+
"loss": 0.3491,
|
| 1968 |
+
"step": 6575
|
| 1969 |
+
},
|
| 1970 |
+
{
|
| 1971 |
+
"epoch": 4.538924213782437,
|
| 1972 |
+
"grad_norm": 0.5620461702346802,
|
| 1973 |
+
"learning_rate": 4.675440630797774e-05,
|
| 1974 |
+
"loss": 0.3074,
|
| 1975 |
+
"step": 6600
|
| 1976 |
+
},
|
| 1977 |
+
{
|
| 1978 |
+
"epoch": 4.556109297130091,
|
| 1979 |
+
"grad_norm": 0.8100690245628357,
|
| 1980 |
+
"learning_rate": 4.673991187384045e-05,
|
| 1981 |
+
"loss": 0.3366,
|
| 1982 |
+
"step": 6625
|
| 1983 |
+
},
|
| 1984 |
+
{
|
| 1985 |
+
"epoch": 4.573294380477745,
|
| 1986 |
+
"grad_norm": 0.7091922760009766,
|
| 1987 |
+
"learning_rate": 4.6725417439703156e-05,
|
| 1988 |
+
"loss": 0.2864,
|
| 1989 |
+
"step": 6650
|
| 1990 |
+
},
|
| 1991 |
+
{
|
| 1992 |
+
"epoch": 4.5904794638254,
|
| 1993 |
+
"grad_norm": 0.38283970952033997,
|
| 1994 |
+
"learning_rate": 4.6710923005565864e-05,
|
| 1995 |
+
"loss": 0.3512,
|
| 1996 |
+
"step": 6675
|
| 1997 |
+
},
|
| 1998 |
+
{
|
| 1999 |
+
"epoch": 4.607664547173053,
|
| 2000 |
+
"grad_norm": 0.5631033182144165,
|
| 2001 |
+
"learning_rate": 4.669642857142857e-05,
|
| 2002 |
+
"loss": 0.2884,
|
| 2003 |
+
"step": 6700
|
| 2004 |
+
},
|
| 2005 |
+
{
|
| 2006 |
+
"epoch": 4.624849630520708,
|
| 2007 |
+
"grad_norm": 0.3868861794471741,
|
| 2008 |
+
"learning_rate": 4.6681934137291285e-05,
|
| 2009 |
+
"loss": 0.3413,
|
| 2010 |
+
"step": 6725
|
| 2011 |
+
},
|
| 2012 |
+
{
|
| 2013 |
+
"epoch": 4.642034713868362,
|
| 2014 |
+
"grad_norm": 0.6378294229507446,
|
| 2015 |
+
"learning_rate": 4.666743970315399e-05,
|
| 2016 |
+
"loss": 0.2949,
|
| 2017 |
+
"step": 6750
|
| 2018 |
+
},
|
| 2019 |
+
{
|
| 2020 |
+
"epoch": 4.659219797216016,
|
| 2021 |
+
"grad_norm": 0.5135634541511536,
|
| 2022 |
+
"learning_rate": 4.66529452690167e-05,
|
| 2023 |
+
"loss": 0.3463,
|
| 2024 |
+
"step": 6775
|
| 2025 |
+
},
|
| 2026 |
+
{
|
| 2027 |
+
"epoch": 4.676404880563671,
|
| 2028 |
+
"grad_norm": 0.4989064633846283,
|
| 2029 |
+
"learning_rate": 4.663845083487941e-05,
|
| 2030 |
+
"loss": 0.3027,
|
| 2031 |
+
"step": 6800
|
| 2032 |
+
},
|
| 2033 |
+
{
|
| 2034 |
+
"epoch": 4.693589963911325,
|
| 2035 |
+
"grad_norm": 0.5633465647697449,
|
| 2036 |
+
"learning_rate": 4.662395640074212e-05,
|
| 2037 |
+
"loss": 0.3626,
|
| 2038 |
+
"step": 6825
|
| 2039 |
+
},
|
| 2040 |
+
{
|
| 2041 |
+
"epoch": 4.710775047258979,
|
| 2042 |
+
"grad_norm": 0.6484938859939575,
|
| 2043 |
+
"learning_rate": 4.660946196660483e-05,
|
| 2044 |
+
"loss": 0.2881,
|
| 2045 |
+
"step": 6850
|
| 2046 |
+
},
|
| 2047 |
+
{
|
| 2048 |
+
"epoch": 4.7279601306066334,
|
| 2049 |
+
"grad_norm": 0.4283730983734131,
|
| 2050 |
+
"learning_rate": 4.6594967532467537e-05,
|
| 2051 |
+
"loss": 0.3452,
|
| 2052 |
+
"step": 6875
|
| 2053 |
+
},
|
| 2054 |
+
{
|
| 2055 |
+
"epoch": 4.745145213954288,
|
| 2056 |
+
"grad_norm": 0.5287323594093323,
|
| 2057 |
+
"learning_rate": 4.6580473098330244e-05,
|
| 2058 |
+
"loss": 0.3015,
|
| 2059 |
+
"step": 6900
|
| 2060 |
+
},
|
| 2061 |
+
{
|
| 2062 |
+
"epoch": 4.762330297301942,
|
| 2063 |
+
"grad_norm": 0.41760608553886414,
|
| 2064 |
+
"learning_rate": 4.656597866419295e-05,
|
| 2065 |
+
"loss": 0.3032,
|
| 2066 |
+
"step": 6925
|
| 2067 |
+
},
|
| 2068 |
+
{
|
| 2069 |
+
"epoch": 4.779515380649596,
|
| 2070 |
+
"grad_norm": 0.6855202317237854,
|
| 2071 |
+
"learning_rate": 4.655148423005566e-05,
|
| 2072 |
+
"loss": 0.3006,
|
| 2073 |
+
"step": 6950
|
| 2074 |
+
},
|
| 2075 |
+
{
|
| 2076 |
+
"epoch": 4.796700463997251,
|
| 2077 |
+
"grad_norm": 0.45387232303619385,
|
| 2078 |
+
"learning_rate": 4.6536989795918366e-05,
|
| 2079 |
+
"loss": 0.3428,
|
| 2080 |
+
"step": 6975
|
| 2081 |
+
},
|
| 2082 |
+
{
|
| 2083 |
+
"epoch": 4.8138855473449045,
|
| 2084 |
+
"grad_norm": 0.8281689286231995,
|
| 2085 |
+
"learning_rate": 4.6522495361781074e-05,
|
| 2086 |
+
"loss": 0.2804,
|
| 2087 |
+
"step": 7000
|
| 2088 |
+
},
|
| 2089 |
+
{
|
| 2090 |
+
"epoch": 4.8138855473449045,
|
| 2091 |
+
"eval_loss": 0.3352712392807007,
|
| 2092 |
+
"eval_runtime": 156.292,
|
| 2093 |
+
"eval_samples_per_second": 7.838,
|
| 2094 |
+
"eval_steps_per_second": 0.985,
|
| 2095 |
+
"eval_wer": 0.5525116574819839,
|
| 2096 |
+
"step": 7000
|
| 2097 |
+
},
|
| 2098 |
+
{
|
| 2099 |
+
"epoch": 4.831070630692559,
|
| 2100 |
+
"grad_norm": 0.4252523183822632,
|
| 2101 |
+
"learning_rate": 4.650800092764379e-05,
|
| 2102 |
+
"loss": 0.3456,
|
| 2103 |
+
"step": 7025
|
| 2104 |
+
},
|
| 2105 |
+
{
|
| 2106 |
+
"epoch": 4.848255714040213,
|
| 2107 |
+
"grad_norm": 0.536359429359436,
|
| 2108 |
+
"learning_rate": 4.6493506493506495e-05,
|
| 2109 |
+
"loss": 0.3016,
|
| 2110 |
+
"step": 7050
|
| 2111 |
+
},
|
| 2112 |
+
{
|
| 2113 |
+
"epoch": 4.865440797387867,
|
| 2114 |
+
"grad_norm": 0.48082077503204346,
|
| 2115 |
+
"learning_rate": 4.64790120593692e-05,
|
| 2116 |
+
"loss": 0.3323,
|
| 2117 |
+
"step": 7075
|
| 2118 |
+
},
|
| 2119 |
+
{
|
| 2120 |
+
"epoch": 4.882625880735522,
|
| 2121 |
+
"grad_norm": 0.7152004837989807,
|
| 2122 |
+
"learning_rate": 4.646451762523191e-05,
|
| 2123 |
+
"loss": 0.3102,
|
| 2124 |
+
"step": 7100
|
| 2125 |
+
},
|
| 2126 |
+
{
|
| 2127 |
+
"epoch": 4.8998109640831755,
|
| 2128 |
+
"grad_norm": 0.43397244811058044,
|
| 2129 |
+
"learning_rate": 4.6450023191094625e-05,
|
| 2130 |
+
"loss": 0.3671,
|
| 2131 |
+
"step": 7125
|
| 2132 |
+
},
|
| 2133 |
+
{
|
| 2134 |
+
"epoch": 4.91699604743083,
|
| 2135 |
+
"grad_norm": 1.093762755393982,
|
| 2136 |
+
"learning_rate": 4.643552875695733e-05,
|
| 2137 |
+
"loss": 0.299,
|
| 2138 |
+
"step": 7150
|
| 2139 |
+
},
|
| 2140 |
+
{
|
| 2141 |
+
"epoch": 4.934181130778485,
|
| 2142 |
+
"grad_norm": 0.3550453782081604,
|
| 2143 |
+
"learning_rate": 4.642103432282004e-05,
|
| 2144 |
+
"loss": 0.3351,
|
| 2145 |
+
"step": 7175
|
| 2146 |
+
},
|
| 2147 |
+
{
|
| 2148 |
+
"epoch": 4.951366214126138,
|
| 2149 |
+
"grad_norm": 0.6337935924530029,
|
| 2150 |
+
"learning_rate": 4.640653988868275e-05,
|
| 2151 |
+
"loss": 0.2952,
|
| 2152 |
+
"step": 7200
|
| 2153 |
+
},
|
| 2154 |
+
{
|
| 2155 |
+
"epoch": 4.968551297473793,
|
| 2156 |
+
"grad_norm": 0.5126771926879883,
|
| 2157 |
+
"learning_rate": 4.6392045454545454e-05,
|
| 2158 |
+
"loss": 0.3431,
|
| 2159 |
+
"step": 7225
|
| 2160 |
+
},
|
| 2161 |
+
{
|
| 2162 |
+
"epoch": 4.9857363808214465,
|
| 2163 |
+
"grad_norm": 0.4208792448043823,
|
| 2164 |
+
"learning_rate": 4.637755102040817e-05,
|
| 2165 |
+
"loss": 0.3025,
|
| 2166 |
+
"step": 7250
|
| 2167 |
+
},
|
| 2168 |
+
{
|
| 2169 |
+
"epoch": 5.003437016669531,
|
| 2170 |
+
"grad_norm": 0.5154265761375427,
|
| 2171 |
+
"learning_rate": 4.6363056586270876e-05,
|
| 2172 |
+
"loss": 0.3215,
|
| 2173 |
+
"step": 7275
|
| 2174 |
+
},
|
| 2175 |
+
{
|
| 2176 |
+
"epoch": 5.020622100017185,
|
| 2177 |
+
"grad_norm": 0.4937199652194977,
|
| 2178 |
+
"learning_rate": 4.634856215213358e-05,
|
| 2179 |
+
"loss": 0.2898,
|
| 2180 |
+
"step": 7300
|
| 2181 |
+
},
|
| 2182 |
+
{
|
| 2183 |
+
"epoch": 5.03780718336484,
|
| 2184 |
+
"grad_norm": 0.4737917482852936,
|
| 2185 |
+
"learning_rate": 4.633406771799629e-05,
|
| 2186 |
+
"loss": 0.3186,
|
| 2187 |
+
"step": 7325
|
| 2188 |
+
},
|
| 2189 |
+
{
|
| 2190 |
+
"epoch": 5.054992266712493,
|
| 2191 |
+
"grad_norm": 0.39069080352783203,
|
| 2192 |
+
"learning_rate": 4.6319573283859005e-05,
|
| 2193 |
+
"loss": 0.299,
|
| 2194 |
+
"step": 7350
|
| 2195 |
+
},
|
| 2196 |
+
{
|
| 2197 |
+
"epoch": 5.072177350060148,
|
| 2198 |
+
"grad_norm": 0.41207846999168396,
|
| 2199 |
+
"learning_rate": 4.630507884972171e-05,
|
| 2200 |
+
"loss": 0.3261,
|
| 2201 |
+
"step": 7375
|
| 2202 |
+
},
|
| 2203 |
+
{
|
| 2204 |
+
"epoch": 5.089362433407802,
|
| 2205 |
+
"grad_norm": 0.5971049070358276,
|
| 2206 |
+
"learning_rate": 4.629058441558442e-05,
|
| 2207 |
+
"loss": 0.279,
|
| 2208 |
+
"step": 7400
|
| 2209 |
+
},
|
| 2210 |
+
{
|
| 2211 |
+
"epoch": 5.106547516755456,
|
| 2212 |
+
"grad_norm": 0.41475459933280945,
|
| 2213 |
+
"learning_rate": 4.627608998144712e-05,
|
| 2214 |
+
"loss": 0.3126,
|
| 2215 |
+
"step": 7425
|
| 2216 |
+
},
|
| 2217 |
+
{
|
| 2218 |
+
"epoch": 5.123732600103111,
|
| 2219 |
+
"grad_norm": 0.5062717795372009,
|
| 2220 |
+
"learning_rate": 4.6261595547309835e-05,
|
| 2221 |
+
"loss": 0.2741,
|
| 2222 |
+
"step": 7450
|
| 2223 |
+
},
|
| 2224 |
+
{
|
| 2225 |
+
"epoch": 5.140917683450764,
|
| 2226 |
+
"grad_norm": 0.5244805812835693,
|
| 2227 |
+
"learning_rate": 4.624710111317254e-05,
|
| 2228 |
+
"loss": 0.2913,
|
| 2229 |
+
"step": 7475
|
| 2230 |
+
},
|
| 2231 |
+
{
|
| 2232 |
+
"epoch": 5.158102766798419,
|
| 2233 |
+
"grad_norm": 0.7847909927368164,
|
| 2234 |
+
"learning_rate": 4.623260667903525e-05,
|
| 2235 |
+
"loss": 0.2908,
|
| 2236 |
+
"step": 7500
|
| 2237 |
+
},
|
| 2238 |
+
{
|
| 2239 |
+
"epoch": 5.158102766798419,
|
| 2240 |
+
"eval_loss": 0.3383817970752716,
|
| 2241 |
+
"eval_runtime": 159.5151,
|
| 2242 |
+
"eval_samples_per_second": 7.68,
|
| 2243 |
+
"eval_steps_per_second": 0.965,
|
| 2244 |
+
"eval_wer": 0.5484845273420941,
|
| 2245 |
+
"step": 7500
|
| 2246 |
+
},
|
| 2247 |
+
{
|
| 2248 |
+
"epoch": 5.1752878501460735,
|
| 2249 |
+
"grad_norm": 0.44492971897125244,
|
| 2250 |
+
"learning_rate": 4.621811224489796e-05,
|
| 2251 |
+
"loss": 0.3359,
|
| 2252 |
+
"step": 7525
|
| 2253 |
+
},
|
| 2254 |
+
{
|
| 2255 |
+
"epoch": 5.192472933493727,
|
| 2256 |
+
"grad_norm": 0.8884369730949402,
|
| 2257 |
+
"learning_rate": 4.620361781076067e-05,
|
| 2258 |
+
"loss": 0.286,
|
| 2259 |
+
"step": 7550
|
| 2260 |
+
},
|
| 2261 |
+
{
|
| 2262 |
+
"epoch": 5.209658016841382,
|
| 2263 |
+
"grad_norm": 0.4650115966796875,
|
| 2264 |
+
"learning_rate": 4.618912337662338e-05,
|
| 2265 |
+
"loss": 0.3538,
|
| 2266 |
+
"step": 7575
|
| 2267 |
+
},
|
| 2268 |
+
{
|
| 2269 |
+
"epoch": 5.226843100189036,
|
| 2270 |
+
"grad_norm": 0.46531254053115845,
|
| 2271 |
+
"learning_rate": 4.6174628942486086e-05,
|
| 2272 |
+
"loss": 0.2828,
|
| 2273 |
+
"step": 7600
|
| 2274 |
+
},
|
| 2275 |
+
{
|
| 2276 |
+
"epoch": 5.24402818353669,
|
| 2277 |
+
"grad_norm": 0.5248584747314453,
|
| 2278 |
+
"learning_rate": 4.6160134508348794e-05,
|
| 2279 |
+
"loss": 0.3166,
|
| 2280 |
+
"step": 7625
|
| 2281 |
+
},
|
| 2282 |
+
{
|
| 2283 |
+
"epoch": 5.2612132668843445,
|
| 2284 |
+
"grad_norm": 0.676996648311615,
|
| 2285 |
+
"learning_rate": 4.614564007421151e-05,
|
| 2286 |
+
"loss": 0.3141,
|
| 2287 |
+
"step": 7650
|
| 2288 |
+
},
|
| 2289 |
+
{
|
| 2290 |
+
"epoch": 5.278398350231998,
|
| 2291 |
+
"grad_norm": 0.40085482597351074,
|
| 2292 |
+
"learning_rate": 4.6131145640074215e-05,
|
| 2293 |
+
"loss": 0.3099,
|
| 2294 |
+
"step": 7675
|
| 2295 |
+
},
|
| 2296 |
+
{
|
| 2297 |
+
"epoch": 5.295583433579653,
|
| 2298 |
+
"grad_norm": 0.5248492956161499,
|
| 2299 |
+
"learning_rate": 4.611665120593692e-05,
|
| 2300 |
+
"loss": 0.2905,
|
| 2301 |
+
"step": 7700
|
| 2302 |
+
},
|
| 2303 |
+
{
|
| 2304 |
+
"epoch": 5.312768516927307,
|
| 2305 |
+
"grad_norm": 0.3710981607437134,
|
| 2306 |
+
"learning_rate": 4.610215677179963e-05,
|
| 2307 |
+
"loss": 0.3114,
|
| 2308 |
+
"step": 7725
|
| 2309 |
+
},
|
| 2310 |
+
{
|
| 2311 |
+
"epoch": 5.329953600274961,
|
| 2312 |
+
"grad_norm": 0.607997477054596,
|
| 2313 |
+
"learning_rate": 4.608766233766234e-05,
|
| 2314 |
+
"loss": 0.2814,
|
| 2315 |
+
"step": 7750
|
| 2316 |
+
},
|
| 2317 |
+
{
|
| 2318 |
+
"epoch": 5.347138683622616,
|
| 2319 |
+
"grad_norm": 0.41846323013305664,
|
| 2320 |
+
"learning_rate": 4.607316790352505e-05,
|
| 2321 |
+
"loss": 0.3442,
|
| 2322 |
+
"step": 7775
|
| 2323 |
+
},
|
| 2324 |
+
{
|
| 2325 |
+
"epoch": 5.36432376697027,
|
| 2326 |
+
"grad_norm": 0.7187564373016357,
|
| 2327 |
+
"learning_rate": 4.605867346938776e-05,
|
| 2328 |
+
"loss": 0.3089,
|
| 2329 |
+
"step": 7800
|
| 2330 |
+
},
|
| 2331 |
+
{
|
| 2332 |
+
"epoch": 5.381508850317924,
|
| 2333 |
+
"grad_norm": 0.6370894312858582,
|
| 2334 |
+
"learning_rate": 4.604417903525047e-05,
|
| 2335 |
+
"loss": 0.2802,
|
| 2336 |
+
"step": 7825
|
| 2337 |
+
},
|
| 2338 |
+
{
|
| 2339 |
+
"epoch": 5.398693933665578,
|
| 2340 |
+
"grad_norm": 0.4946443736553192,
|
| 2341 |
+
"learning_rate": 4.6029684601113174e-05,
|
| 2342 |
+
"loss": 0.2752,
|
| 2343 |
+
"step": 7850
|
| 2344 |
+
},
|
| 2345 |
+
{
|
| 2346 |
+
"epoch": 5.415879017013232,
|
| 2347 |
+
"grad_norm": 0.5713298916816711,
|
| 2348 |
+
"learning_rate": 4.601519016697589e-05,
|
| 2349 |
+
"loss": 0.3335,
|
| 2350 |
+
"step": 7875
|
| 2351 |
+
},
|
| 2352 |
+
{
|
| 2353 |
+
"epoch": 5.433064100360887,
|
| 2354 |
+
"grad_norm": 0.5185278058052063,
|
| 2355 |
+
"learning_rate": 4.6000695732838596e-05,
|
| 2356 |
+
"loss": 0.2779,
|
| 2357 |
+
"step": 7900
|
| 2358 |
+
},
|
| 2359 |
+
{
|
| 2360 |
+
"epoch": 5.450249183708541,
|
| 2361 |
+
"grad_norm": 0.6777128577232361,
|
| 2362 |
+
"learning_rate": 4.59862012987013e-05,
|
| 2363 |
+
"loss": 0.2907,
|
| 2364 |
+
"step": 7925
|
| 2365 |
+
},
|
| 2366 |
+
{
|
| 2367 |
+
"epoch": 5.467434267056195,
|
| 2368 |
+
"grad_norm": 0.4836239814758301,
|
| 2369 |
+
"learning_rate": 4.5971706864564004e-05,
|
| 2370 |
+
"loss": 0.2726,
|
| 2371 |
+
"step": 7950
|
| 2372 |
+
},
|
| 2373 |
+
{
|
| 2374 |
+
"epoch": 5.484619350403849,
|
| 2375 |
+
"grad_norm": 0.4827396273612976,
|
| 2376 |
+
"learning_rate": 4.595721243042672e-05,
|
| 2377 |
+
"loss": 0.2956,
|
| 2378 |
+
"step": 7975
|
| 2379 |
+
},
|
| 2380 |
+
{
|
| 2381 |
+
"epoch": 5.501804433751504,
|
| 2382 |
+
"grad_norm": 0.4124370813369751,
|
| 2383 |
+
"learning_rate": 4.5942717996289425e-05,
|
| 2384 |
+
"loss": 0.2646,
|
| 2385 |
+
"step": 8000
|
| 2386 |
+
},
|
| 2387 |
+
{
|
| 2388 |
+
"epoch": 5.501804433751504,
|
| 2389 |
+
"eval_loss": 0.31637853384017944,
|
| 2390 |
+
"eval_runtime": 154.147,
|
| 2391 |
+
"eval_samples_per_second": 7.947,
|
| 2392 |
+
"eval_steps_per_second": 0.999,
|
| 2393 |
+
"eval_wer": 0.5461530309453159,
|
| 2394 |
+
"step": 8000
|
| 2395 |
+
},
|
| 2396 |
+
{
|
| 2397 |
+
"epoch": 5.518989517099158,
|
| 2398 |
+
"grad_norm": 0.4147077202796936,
|
| 2399 |
+
"learning_rate": 4.592822356215213e-05,
|
| 2400 |
+
"loss": 0.3084,
|
| 2401 |
+
"step": 8025
|
| 2402 |
+
},
|
| 2403 |
+
{
|
| 2404 |
+
"epoch": 5.536174600446812,
|
| 2405 |
+
"grad_norm": 0.5999482274055481,
|
| 2406 |
+
"learning_rate": 4.591372912801484e-05,
|
| 2407 |
+
"loss": 0.2824,
|
| 2408 |
+
"step": 8050
|
| 2409 |
+
},
|
| 2410 |
+
{
|
| 2411 |
+
"epoch": 5.553359683794467,
|
| 2412 |
+
"grad_norm": 0.4082586169242859,
|
| 2413 |
+
"learning_rate": 4.5899234693877555e-05,
|
| 2414 |
+
"loss": 0.3095,
|
| 2415 |
+
"step": 8075
|
| 2416 |
+
},
|
| 2417 |
+
{
|
| 2418 |
+
"epoch": 5.57054476714212,
|
| 2419 |
+
"grad_norm": 0.45266950130462646,
|
| 2420 |
+
"learning_rate": 4.588474025974026e-05,
|
| 2421 |
+
"loss": 0.2672,
|
| 2422 |
+
"step": 8100
|
| 2423 |
+
},
|
| 2424 |
+
{
|
| 2425 |
+
"epoch": 5.587729850489775,
|
| 2426 |
+
"grad_norm": 0.4024001955986023,
|
| 2427 |
+
"learning_rate": 4.587024582560297e-05,
|
| 2428 |
+
"loss": 0.3147,
|
| 2429 |
+
"step": 8125
|
| 2430 |
+
},
|
| 2431 |
+
{
|
| 2432 |
+
"epoch": 5.60491493383743,
|
| 2433 |
+
"grad_norm": 0.61323481798172,
|
| 2434 |
+
"learning_rate": 4.585575139146568e-05,
|
| 2435 |
+
"loss": 0.2461,
|
| 2436 |
+
"step": 8150
|
| 2437 |
+
},
|
| 2438 |
+
{
|
| 2439 |
+
"epoch": 5.622100017185083,
|
| 2440 |
+
"grad_norm": 0.7370169758796692,
|
| 2441 |
+
"learning_rate": 4.584125695732839e-05,
|
| 2442 |
+
"loss": 0.3364,
|
| 2443 |
+
"step": 8175
|
| 2444 |
+
},
|
| 2445 |
+
{
|
| 2446 |
+
"epoch": 5.639285100532738,
|
| 2447 |
+
"grad_norm": 0.5105010867118835,
|
| 2448 |
+
"learning_rate": 4.58267625231911e-05,
|
| 2449 |
+
"loss": 0.2749,
|
| 2450 |
+
"step": 8200
|
| 2451 |
+
},
|
| 2452 |
+
{
|
| 2453 |
+
"epoch": 5.6564701838803915,
|
| 2454 |
+
"grad_norm": 0.4862951636314392,
|
| 2455 |
+
"learning_rate": 4.5812268089053806e-05,
|
| 2456 |
+
"loss": 0.318,
|
| 2457 |
+
"step": 8225
|
| 2458 |
+
},
|
| 2459 |
+
{
|
| 2460 |
+
"epoch": 5.673655267228046,
|
| 2461 |
+
"grad_norm": 0.5929895639419556,
|
| 2462 |
+
"learning_rate": 4.5797773654916513e-05,
|
| 2463 |
+
"loss": 0.3215,
|
| 2464 |
+
"step": 8250
|
| 2465 |
+
},
|
| 2466 |
+
{
|
| 2467 |
+
"epoch": 5.690840350575701,
|
| 2468 |
+
"grad_norm": 0.5355464220046997,
|
| 2469 |
+
"learning_rate": 4.578327922077922e-05,
|
| 2470 |
+
"loss": 0.343,
|
| 2471 |
+
"step": 8275
|
| 2472 |
+
},
|
| 2473 |
+
{
|
| 2474 |
+
"epoch": 5.708025433923354,
|
| 2475 |
+
"grad_norm": 0.6044451594352722,
|
| 2476 |
+
"learning_rate": 4.5768784786641935e-05,
|
| 2477 |
+
"loss": 0.2761,
|
| 2478 |
+
"step": 8300
|
| 2479 |
+
},
|
| 2480 |
+
{
|
| 2481 |
+
"epoch": 5.725210517271009,
|
| 2482 |
+
"grad_norm": 0.5010135769844055,
|
| 2483 |
+
"learning_rate": 4.575429035250464e-05,
|
| 2484 |
+
"loss": 0.3525,
|
| 2485 |
+
"step": 8325
|
| 2486 |
+
},
|
| 2487 |
+
{
|
| 2488 |
+
"epoch": 5.742395600618663,
|
| 2489 |
+
"grad_norm": 0.592808723449707,
|
| 2490 |
+
"learning_rate": 4.573979591836735e-05,
|
| 2491 |
+
"loss": 0.2749,
|
| 2492 |
+
"step": 8350
|
| 2493 |
+
},
|
| 2494 |
+
{
|
| 2495 |
+
"epoch": 5.759580683966317,
|
| 2496 |
+
"grad_norm": 0.5672963261604309,
|
| 2497 |
+
"learning_rate": 4.572530148423006e-05,
|
| 2498 |
+
"loss": 0.3711,
|
| 2499 |
+
"step": 8375
|
| 2500 |
+
},
|
| 2501 |
+
{
|
| 2502 |
+
"epoch": 5.776765767313972,
|
| 2503 |
+
"grad_norm": 0.8346742391586304,
|
| 2504 |
+
"learning_rate": 4.571080705009277e-05,
|
| 2505 |
+
"loss": 0.272,
|
| 2506 |
+
"step": 8400
|
| 2507 |
+
},
|
| 2508 |
+
{
|
| 2509 |
+
"epoch": 5.793950850661625,
|
| 2510 |
+
"grad_norm": 0.5474342107772827,
|
| 2511 |
+
"learning_rate": 4.569631261595548e-05,
|
| 2512 |
+
"loss": 0.3022,
|
| 2513 |
+
"step": 8425
|
| 2514 |
+
},
|
| 2515 |
+
{
|
| 2516 |
+
"epoch": 5.81113593400928,
|
| 2517 |
+
"grad_norm": 0.5067320466041565,
|
| 2518 |
+
"learning_rate": 4.5681818181818186e-05,
|
| 2519 |
+
"loss": 0.2632,
|
| 2520 |
+
"step": 8450
|
| 2521 |
+
},
|
| 2522 |
+
{
|
| 2523 |
+
"epoch": 5.828321017356934,
|
| 2524 |
+
"grad_norm": 0.4363684356212616,
|
| 2525 |
+
"learning_rate": 4.566732374768089e-05,
|
| 2526 |
+
"loss": 0.2961,
|
| 2527 |
+
"step": 8475
|
| 2528 |
+
},
|
| 2529 |
+
{
|
| 2530 |
+
"epoch": 5.845506100704588,
|
| 2531 |
+
"grad_norm": 0.6637271642684937,
|
| 2532 |
+
"learning_rate": 4.56528293135436e-05,
|
| 2533 |
+
"loss": 0.2982,
|
| 2534 |
+
"step": 8500
|
| 2535 |
+
},
|
| 2536 |
+
{
|
| 2537 |
+
"epoch": 5.845506100704588,
|
| 2538 |
+
"eval_loss": 0.3142754137516022,
|
| 2539 |
+
"eval_runtime": 153.2422,
|
| 2540 |
+
"eval_samples_per_second": 7.994,
|
| 2541 |
+
"eval_steps_per_second": 1.005,
|
| 2542 |
+
"eval_wer": 0.5454641797371768,
|
| 2543 |
+
"step": 8500
|
| 2544 |
+
},
|
| 2545 |
+
{
|
| 2546 |
+
"epoch": 5.862691184052243,
|
| 2547 |
+
"grad_norm": 0.5597206950187683,
|
| 2548 |
+
"learning_rate": 4.563833487940631e-05,
|
| 2549 |
+
"loss": 0.3282,
|
| 2550 |
+
"step": 8525
|
| 2551 |
+
},
|
| 2552 |
+
{
|
| 2553 |
+
"epoch": 5.879876267399897,
|
| 2554 |
+
"grad_norm": 0.5069429278373718,
|
| 2555 |
+
"learning_rate": 4.5623840445269016e-05,
|
| 2556 |
+
"loss": 0.2819,
|
| 2557 |
+
"step": 8550
|
| 2558 |
+
},
|
| 2559 |
+
{
|
| 2560 |
+
"epoch": 5.897061350747551,
|
| 2561 |
+
"grad_norm": 0.4272073805332184,
|
| 2562 |
+
"learning_rate": 4.5609346011131724e-05,
|
| 2563 |
+
"loss": 0.3032,
|
| 2564 |
+
"step": 8575
|
| 2565 |
+
},
|
| 2566 |
+
{
|
| 2567 |
+
"epoch": 5.9142464340952055,
|
| 2568 |
+
"grad_norm": 0.46952158212661743,
|
| 2569 |
+
"learning_rate": 4.559485157699444e-05,
|
| 2570 |
+
"loss": 0.2979,
|
| 2571 |
+
"step": 8600
|
| 2572 |
+
},
|
| 2573 |
+
{
|
| 2574 |
+
"epoch": 5.93143151744286,
|
| 2575 |
+
"grad_norm": 0.46362563967704773,
|
| 2576 |
+
"learning_rate": 4.5580357142857145e-05,
|
| 2577 |
+
"loss": 0.3042,
|
| 2578 |
+
"step": 8625
|
| 2579 |
+
},
|
| 2580 |
+
{
|
| 2581 |
+
"epoch": 5.948616600790514,
|
| 2582 |
+
"grad_norm": 0.6578242778778076,
|
| 2583 |
+
"learning_rate": 4.556586270871985e-05,
|
| 2584 |
+
"loss": 0.2558,
|
| 2585 |
+
"step": 8650
|
| 2586 |
+
},
|
| 2587 |
+
{
|
| 2588 |
+
"epoch": 5.965801684138168,
|
| 2589 |
+
"grad_norm": 0.5568517446517944,
|
| 2590 |
+
"learning_rate": 4.555136827458256e-05,
|
| 2591 |
+
"loss": 0.3127,
|
| 2592 |
+
"step": 8675
|
| 2593 |
+
},
|
| 2594 |
+
{
|
| 2595 |
+
"epoch": 5.982986767485822,
|
| 2596 |
+
"grad_norm": 0.5202658772468567,
|
| 2597 |
+
"learning_rate": 4.5536873840445274e-05,
|
| 2598 |
+
"loss": 0.2956,
|
| 2599 |
+
"step": 8700
|
| 2600 |
+
},
|
| 2601 |
+
{
|
| 2602 |
+
"epoch": 6.000687403333906,
|
| 2603 |
+
"grad_norm": 0.5851805210113525,
|
| 2604 |
+
"learning_rate": 4.552237940630798e-05,
|
| 2605 |
+
"loss": 0.3197,
|
| 2606 |
+
"step": 8725
|
| 2607 |
+
},
|
| 2608 |
+
{
|
| 2609 |
+
"epoch": 6.0178724866815605,
|
| 2610 |
+
"grad_norm": 0.4245930314064026,
|
| 2611 |
+
"learning_rate": 4.550788497217069e-05,
|
| 2612 |
+
"loss": 0.2537,
|
| 2613 |
+
"step": 8750
|
| 2614 |
+
},
|
| 2615 |
+
{
|
| 2616 |
+
"epoch": 6.035057570029215,
|
| 2617 |
+
"grad_norm": 0.3843390643596649,
|
| 2618 |
+
"learning_rate": 4.54933905380334e-05,
|
| 2619 |
+
"loss": 0.2552,
|
| 2620 |
+
"step": 8775
|
| 2621 |
+
},
|
| 2622 |
+
{
|
| 2623 |
+
"epoch": 6.052242653376869,
|
| 2624 |
+
"grad_norm": 0.4880934953689575,
|
| 2625 |
+
"learning_rate": 4.5478896103896104e-05,
|
| 2626 |
+
"loss": 0.2669,
|
| 2627 |
+
"step": 8800
|
| 2628 |
+
},
|
| 2629 |
+
{
|
| 2630 |
+
"epoch": 6.069427736724523,
|
| 2631 |
+
"grad_norm": 0.44671300053596497,
|
| 2632 |
+
"learning_rate": 4.546440166975882e-05,
|
| 2633 |
+
"loss": 0.2894,
|
| 2634 |
+
"step": 8825
|
| 2635 |
+
},
|
| 2636 |
+
{
|
| 2637 |
+
"epoch": 6.086612820072177,
|
| 2638 |
+
"grad_norm": 0.6307169795036316,
|
| 2639 |
+
"learning_rate": 4.5449907235621526e-05,
|
| 2640 |
+
"loss": 0.3032,
|
| 2641 |
+
"step": 8850
|
| 2642 |
+
},
|
| 2643 |
+
{
|
| 2644 |
+
"epoch": 6.1037979034198315,
|
| 2645 |
+
"grad_norm": 0.4117954969406128,
|
| 2646 |
+
"learning_rate": 4.543541280148423e-05,
|
| 2647 |
+
"loss": 0.2863,
|
| 2648 |
+
"step": 8875
|
| 2649 |
+
},
|
| 2650 |
+
{
|
| 2651 |
+
"epoch": 6.120982986767486,
|
| 2652 |
+
"grad_norm": 0.41599756479263306,
|
| 2653 |
+
"learning_rate": 4.542091836734694e-05,
|
| 2654 |
+
"loss": 0.2804,
|
| 2655 |
+
"step": 8900
|
| 2656 |
+
},
|
| 2657 |
+
{
|
| 2658 |
+
"epoch": 6.13816807011514,
|
| 2659 |
+
"grad_norm": 0.5033993124961853,
|
| 2660 |
+
"learning_rate": 4.5406423933209655e-05,
|
| 2661 |
+
"loss": 0.2971,
|
| 2662 |
+
"step": 8925
|
| 2663 |
+
},
|
| 2664 |
+
{
|
| 2665 |
+
"epoch": 6.155353153462794,
|
| 2666 |
+
"grad_norm": 0.39483895897865295,
|
| 2667 |
+
"learning_rate": 4.539192949907236e-05,
|
| 2668 |
+
"loss": 0.2485,
|
| 2669 |
+
"step": 8950
|
| 2670 |
+
},
|
| 2671 |
+
{
|
| 2672 |
+
"epoch": 6.172538236810449,
|
| 2673 |
+
"grad_norm": 0.48659563064575195,
|
| 2674 |
+
"learning_rate": 4.537743506493506e-05,
|
| 2675 |
+
"loss": 0.2847,
|
| 2676 |
+
"step": 8975
|
| 2677 |
+
},
|
| 2678 |
+
{
|
| 2679 |
+
"epoch": 6.189723320158103,
|
| 2680 |
+
"grad_norm": 0.7809969186782837,
|
| 2681 |
+
"learning_rate": 4.536294063079777e-05,
|
| 2682 |
+
"loss": 0.2978,
|
| 2683 |
+
"step": 9000
|
| 2684 |
+
},
|
| 2685 |
+
{
|
| 2686 |
+
"epoch": 6.189723320158103,
|
| 2687 |
+
"eval_loss": 0.3218280076980591,
|
| 2688 |
+
"eval_runtime": 151.9921,
|
| 2689 |
+
"eval_samples_per_second": 8.06,
|
| 2690 |
+
"eval_steps_per_second": 1.013,
|
| 2691 |
+
"eval_wer": 0.5423908435777872,
|
| 2692 |
+
"step": 9000
|
| 2693 |
+
},
|
| 2694 |
+
{
|
| 2695 |
+
"epoch": 6.206908403505757,
|
| 2696 |
+
"grad_norm": 0.6053128242492676,
|
| 2697 |
+
"learning_rate": 4.5348446196660485e-05,
|
| 2698 |
+
"loss": 0.275,
|
| 2699 |
+
"step": 9025
|
| 2700 |
+
},
|
| 2701 |
+
{
|
| 2702 |
+
"epoch": 6.224093486853411,
|
| 2703 |
+
"grad_norm": 0.4554101526737213,
|
| 2704 |
+
"learning_rate": 4.533395176252319e-05,
|
| 2705 |
+
"loss": 0.2926,
|
| 2706 |
+
"step": 9050
|
| 2707 |
+
},
|
| 2708 |
+
{
|
| 2709 |
+
"epoch": 6.241278570201065,
|
| 2710 |
+
"grad_norm": 0.4731072187423706,
|
| 2711 |
+
"learning_rate": 4.53194573283859e-05,
|
| 2712 |
+
"loss": 0.2921,
|
| 2713 |
+
"step": 9075
|
| 2714 |
+
},
|
| 2715 |
+
{
|
| 2716 |
+
"epoch": 6.25846365354872,
|
| 2717 |
+
"grad_norm": 0.4384573996067047,
|
| 2718 |
+
"learning_rate": 4.530496289424861e-05,
|
| 2719 |
+
"loss": 0.2827,
|
| 2720 |
+
"step": 9100
|
| 2721 |
+
},
|
| 2722 |
+
{
|
| 2723 |
+
"epoch": 6.275648736896374,
|
| 2724 |
+
"grad_norm": 0.5838501453399658,
|
| 2725 |
+
"learning_rate": 4.529046846011132e-05,
|
| 2726 |
+
"loss": 0.2924,
|
| 2727 |
+
"step": 9125
|
| 2728 |
+
},
|
| 2729 |
+
{
|
| 2730 |
+
"epoch": 6.292833820244028,
|
| 2731 |
+
"grad_norm": 0.5555030703544617,
|
| 2732 |
+
"learning_rate": 4.527597402597403e-05,
|
| 2733 |
+
"loss": 0.2839,
|
| 2734 |
+
"step": 9150
|
| 2735 |
+
},
|
| 2736 |
+
{
|
| 2737 |
+
"epoch": 6.310018903591683,
|
| 2738 |
+
"grad_norm": 0.7978671193122864,
|
| 2739 |
+
"learning_rate": 4.5261479591836736e-05,
|
| 2740 |
+
"loss": 0.2808,
|
| 2741 |
+
"step": 9175
|
| 2742 |
+
},
|
| 2743 |
+
{
|
| 2744 |
+
"epoch": 6.327203986939336,
|
| 2745 |
+
"grad_norm": 0.4151977002620697,
|
| 2746 |
+
"learning_rate": 4.5246985157699443e-05,
|
| 2747 |
+
"loss": 0.2767,
|
| 2748 |
+
"step": 9200
|
| 2749 |
+
},
|
| 2750 |
+
{
|
| 2751 |
+
"epoch": 6.344389070286991,
|
| 2752 |
+
"grad_norm": 0.5371110439300537,
|
| 2753 |
+
"learning_rate": 4.523249072356215e-05,
|
| 2754 |
+
"loss": 0.3124,
|
| 2755 |
+
"step": 9225
|
| 2756 |
+
},
|
| 2757 |
+
{
|
| 2758 |
+
"epoch": 6.3615741536346455,
|
| 2759 |
+
"grad_norm": 0.4500742554664612,
|
| 2760 |
+
"learning_rate": 4.5217996289424865e-05,
|
| 2761 |
+
"loss": 0.2731,
|
| 2762 |
+
"step": 9250
|
| 2763 |
+
},
|
| 2764 |
+
{
|
| 2765 |
+
"epoch": 6.378759236982299,
|
| 2766 |
+
"grad_norm": 0.5459848642349243,
|
| 2767 |
+
"learning_rate": 4.520350185528757e-05,
|
| 2768 |
+
"loss": 0.3118,
|
| 2769 |
+
"step": 9275
|
| 2770 |
+
},
|
| 2771 |
+
{
|
| 2772 |
+
"epoch": 6.395944320329954,
|
| 2773 |
+
"grad_norm": 0.7165120244026184,
|
| 2774 |
+
"learning_rate": 4.518900742115028e-05,
|
| 2775 |
+
"loss": 0.2781,
|
| 2776 |
+
"step": 9300
|
| 2777 |
+
},
|
| 2778 |
+
{
|
| 2779 |
+
"epoch": 6.413129403677607,
|
| 2780 |
+
"grad_norm": 0.4492073655128479,
|
| 2781 |
+
"learning_rate": 4.517451298701299e-05,
|
| 2782 |
+
"loss": 0.2652,
|
| 2783 |
+
"step": 9325
|
| 2784 |
+
},
|
| 2785 |
+
{
|
| 2786 |
+
"epoch": 6.430314487025262,
|
| 2787 |
+
"grad_norm": 0.8615822196006775,
|
| 2788 |
+
"learning_rate": 4.51600185528757e-05,
|
| 2789 |
+
"loss": 0.265,
|
| 2790 |
+
"step": 9350
|
| 2791 |
+
},
|
| 2792 |
+
{
|
| 2793 |
+
"epoch": 6.4474995703729165,
|
| 2794 |
+
"grad_norm": 0.37190139293670654,
|
| 2795 |
+
"learning_rate": 4.514552411873841e-05,
|
| 2796 |
+
"loss": 0.2839,
|
| 2797 |
+
"step": 9375
|
| 2798 |
+
},
|
| 2799 |
+
{
|
| 2800 |
+
"epoch": 6.46468465372057,
|
| 2801 |
+
"grad_norm": 0.5044118762016296,
|
| 2802 |
+
"learning_rate": 4.5131029684601116e-05,
|
| 2803 |
+
"loss": 0.2737,
|
| 2804 |
+
"step": 9400
|
| 2805 |
+
},
|
| 2806 |
+
{
|
| 2807 |
+
"epoch": 6.481869737068225,
|
| 2808 |
+
"grad_norm": 0.8805606961250305,
|
| 2809 |
+
"learning_rate": 4.5116535250463824e-05,
|
| 2810 |
+
"loss": 0.2658,
|
| 2811 |
+
"step": 9425
|
| 2812 |
+
},
|
| 2813 |
+
{
|
| 2814 |
+
"epoch": 6.499054820415879,
|
| 2815 |
+
"grad_norm": 0.52882981300354,
|
| 2816 |
+
"learning_rate": 4.510204081632654e-05,
|
| 2817 |
+
"loss": 0.2875,
|
| 2818 |
+
"step": 9450
|
| 2819 |
+
},
|
| 2820 |
+
{
|
| 2821 |
+
"epoch": 6.516239903763533,
|
| 2822 |
+
"grad_norm": 0.8483859896659851,
|
| 2823 |
+
"learning_rate": 4.5087546382189246e-05,
|
| 2824 |
+
"loss": 0.3182,
|
| 2825 |
+
"step": 9475
|
| 2826 |
+
},
|
| 2827 |
+
{
|
| 2828 |
+
"epoch": 6.533424987111188,
|
| 2829 |
+
"grad_norm": 0.5920891165733337,
|
| 2830 |
+
"learning_rate": 4.5073051948051946e-05,
|
| 2831 |
+
"loss": 0.288,
|
| 2832 |
+
"step": 9500
|
| 2833 |
+
},
|
| 2834 |
+
{
|
| 2835 |
+
"epoch": 6.533424987111188,
|
| 2836 |
+
"eval_loss": 0.3151616156101227,
|
| 2837 |
+
"eval_runtime": 152.9497,
|
| 2838 |
+
"eval_samples_per_second": 8.009,
|
| 2839 |
+
"eval_steps_per_second": 1.007,
|
| 2840 |
+
"eval_wer": 0.5417549809241204,
|
| 2841 |
+
"step": 9500
|
| 2842 |
+
},
|
| 2843 |
+
{
|
| 2844 |
+
"epoch": 6.550610070458841,
|
| 2845 |
+
"grad_norm": 0.5297147035598755,
|
| 2846 |
+
"learning_rate": 4.5059137291280155e-05,
|
| 2847 |
+
"loss": 0.2972,
|
| 2848 |
+
"step": 9525
|
| 2849 |
+
},
|
| 2850 |
+
{
|
| 2851 |
+
"epoch": 6.567795153806496,
|
| 2852 |
+
"grad_norm": 0.43849292397499084,
|
| 2853 |
+
"learning_rate": 4.504464285714286e-05,
|
| 2854 |
+
"loss": 0.2753,
|
| 2855 |
+
"step": 9550
|
| 2856 |
+
},
|
| 2857 |
+
{
|
| 2858 |
+
"epoch": 6.58498023715415,
|
| 2859 |
+
"grad_norm": 0.5231007933616638,
|
| 2860 |
+
"learning_rate": 4.503014842300557e-05,
|
| 2861 |
+
"loss": 0.2916,
|
| 2862 |
+
"step": 9575
|
| 2863 |
+
},
|
| 2864 |
+
{
|
| 2865 |
+
"epoch": 6.602165320501804,
|
| 2866 |
+
"grad_norm": 0.3840227425098419,
|
| 2867 |
+
"learning_rate": 4.501565398886828e-05,
|
| 2868 |
+
"loss": 0.2569,
|
| 2869 |
+
"step": 9600
|
| 2870 |
+
},
|
| 2871 |
+
{
|
| 2872 |
+
"epoch": 6.619350403849459,
|
| 2873 |
+
"grad_norm": 0.4950826168060303,
|
| 2874 |
+
"learning_rate": 4.5001159554730984e-05,
|
| 2875 |
+
"loss": 0.3198,
|
| 2876 |
+
"step": 9625
|
| 2877 |
+
},
|
| 2878 |
+
{
|
| 2879 |
+
"epoch": 6.636535487197113,
|
| 2880 |
+
"grad_norm": 0.4441792368888855,
|
| 2881 |
+
"learning_rate": 4.49866651205937e-05,
|
| 2882 |
+
"loss": 0.2796,
|
| 2883 |
+
"step": 9650
|
| 2884 |
+
},
|
| 2885 |
+
{
|
| 2886 |
+
"epoch": 6.653720570544767,
|
| 2887 |
+
"grad_norm": 0.4779716730117798,
|
| 2888 |
+
"learning_rate": 4.4972170686456406e-05,
|
| 2889 |
+
"loss": 0.3165,
|
| 2890 |
+
"step": 9675
|
| 2891 |
+
},
|
| 2892 |
+
{
|
| 2893 |
+
"epoch": 6.670905653892421,
|
| 2894 |
+
"grad_norm": 0.6962916254997253,
|
| 2895 |
+
"learning_rate": 4.4957676252319113e-05,
|
| 2896 |
+
"loss": 0.3176,
|
| 2897 |
+
"step": 9700
|
| 2898 |
+
},
|
| 2899 |
+
{
|
| 2900 |
+
"epoch": 6.688090737240076,
|
| 2901 |
+
"grad_norm": 0.43603190779685974,
|
| 2902 |
+
"learning_rate": 4.494318181818182e-05,
|
| 2903 |
+
"loss": 0.3095,
|
| 2904 |
+
"step": 9725
|
| 2905 |
+
},
|
| 2906 |
+
{
|
| 2907 |
+
"epoch": 6.70527582058773,
|
| 2908 |
+
"grad_norm": 0.8133454918861389,
|
| 2909 |
+
"learning_rate": 4.492868738404453e-05,
|
| 2910 |
+
"loss": 0.2793,
|
| 2911 |
+
"step": 9750
|
| 2912 |
+
},
|
| 2913 |
+
{
|
| 2914 |
+
"epoch": 6.722460903935384,
|
| 2915 |
+
"grad_norm": 0.4596538245677948,
|
| 2916 |
+
"learning_rate": 4.4914192949907236e-05,
|
| 2917 |
+
"loss": 0.2626,
|
| 2918 |
+
"step": 9775
|
| 2919 |
+
},
|
| 2920 |
+
{
|
| 2921 |
+
"epoch": 6.739645987283039,
|
| 2922 |
+
"grad_norm": 0.6948014497756958,
|
| 2923 |
+
"learning_rate": 4.489969851576994e-05,
|
| 2924 |
+
"loss": 0.2546,
|
| 2925 |
+
"step": 9800
|
| 2926 |
+
},
|
| 2927 |
+
{
|
| 2928 |
+
"epoch": 6.756831070630692,
|
| 2929 |
+
"grad_norm": 0.40873172879219055,
|
| 2930 |
+
"learning_rate": 4.488520408163265e-05,
|
| 2931 |
+
"loss": 0.2867,
|
| 2932 |
+
"step": 9825
|
| 2933 |
+
},
|
| 2934 |
+
{
|
| 2935 |
+
"epoch": 6.774016153978347,
|
| 2936 |
+
"grad_norm": 0.36870279908180237,
|
| 2937 |
+
"learning_rate": 4.4870709647495365e-05,
|
| 2938 |
+
"loss": 0.3258,
|
| 2939 |
+
"step": 9850
|
| 2940 |
+
},
|
| 2941 |
+
{
|
| 2942 |
+
"epoch": 6.791201237326001,
|
| 2943 |
+
"grad_norm": 0.4158894121646881,
|
| 2944 |
+
"learning_rate": 4.485621521335807e-05,
|
| 2945 |
+
"loss": 0.3015,
|
| 2946 |
+
"step": 9875
|
| 2947 |
+
},
|
| 2948 |
+
{
|
| 2949 |
+
"epoch": 6.808386320673655,
|
| 2950 |
+
"grad_norm": 0.4422719180583954,
|
| 2951 |
+
"learning_rate": 4.484172077922078e-05,
|
| 2952 |
+
"loss": 0.285,
|
| 2953 |
+
"step": 9900
|
| 2954 |
+
},
|
| 2955 |
+
{
|
| 2956 |
+
"epoch": 6.82557140402131,
|
| 2957 |
+
"grad_norm": 0.7379534840583801,
|
| 2958 |
+
"learning_rate": 4.482780612244898e-05,
|
| 2959 |
+
"loss": 0.3093,
|
| 2960 |
+
"step": 9925
|
| 2961 |
+
},
|
| 2962 |
+
{
|
| 2963 |
+
"epoch": 6.8427564873689635,
|
| 2964 |
+
"grad_norm": 0.5119884014129639,
|
| 2965 |
+
"learning_rate": 4.481331168831169e-05,
|
| 2966 |
+
"loss": 0.2794,
|
| 2967 |
+
"step": 9950
|
| 2968 |
+
},
|
| 2969 |
+
{
|
| 2970 |
+
"epoch": 6.859941570716618,
|
| 2971 |
+
"grad_norm": 0.46535733342170715,
|
| 2972 |
+
"learning_rate": 4.4798817254174396e-05,
|
| 2973 |
+
"loss": 0.2784,
|
| 2974 |
+
"step": 9975
|
| 2975 |
+
},
|
| 2976 |
+
{
|
| 2977 |
+
"epoch": 6.877126654064273,
|
| 2978 |
+
"grad_norm": 0.49434012174606323,
|
| 2979 |
+
"learning_rate": 4.4784322820037104e-05,
|
| 2980 |
+
"loss": 0.2706,
|
| 2981 |
+
"step": 10000
|
| 2982 |
+
},
|
| 2983 |
+
{
|
| 2984 |
+
"epoch": 6.877126654064273,
|
| 2985 |
+
"eval_loss": 0.321118026971817,
|
| 2986 |
+
"eval_runtime": 152.9572,
|
| 2987 |
+
"eval_samples_per_second": 8.009,
|
| 2988 |
+
"eval_steps_per_second": 1.007,
|
| 2989 |
+
"eval_wer": 0.53984739296312,
|
| 2990 |
+
"step": 10000
|
| 2991 |
+
},
|
| 2992 |
+
{
|
| 2993 |
+
"epoch": 6.894311737411926,
|
| 2994 |
+
"grad_norm": 0.4334565997123718,
|
| 2995 |
+
"learning_rate": 4.476982838589982e-05,
|
| 2996 |
+
"loss": 0.2762,
|
| 2997 |
+
"step": 10025
|
| 2998 |
+
},
|
| 2999 |
+
{
|
| 3000 |
+
"epoch": 6.911496820759581,
|
| 3001 |
+
"grad_norm": 0.564243733882904,
|
| 3002 |
+
"learning_rate": 4.4755333951762525e-05,
|
| 3003 |
+
"loss": 0.2429,
|
| 3004 |
+
"step": 10050
|
| 3005 |
+
},
|
| 3006 |
+
{
|
| 3007 |
+
"epoch": 6.9286819041072345,
|
| 3008 |
+
"grad_norm": 0.4657536447048187,
|
| 3009 |
+
"learning_rate": 4.474083951762523e-05,
|
| 3010 |
+
"loss": 0.2528,
|
| 3011 |
+
"step": 10075
|
| 3012 |
+
},
|
| 3013 |
+
{
|
| 3014 |
+
"epoch": 6.945866987454889,
|
| 3015 |
+
"grad_norm": 0.5497089624404907,
|
| 3016 |
+
"learning_rate": 4.472634508348794e-05,
|
| 3017 |
+
"loss": 0.2982,
|
| 3018 |
+
"step": 10100
|
| 3019 |
+
},
|
| 3020 |
+
{
|
| 3021 |
+
"epoch": 6.963052070802544,
|
| 3022 |
+
"grad_norm": 0.7017095685005188,
|
| 3023 |
+
"learning_rate": 4.471185064935065e-05,
|
| 3024 |
+
"loss": 0.2762,
|
| 3025 |
+
"step": 10125
|
| 3026 |
+
},
|
| 3027 |
+
{
|
| 3028 |
+
"epoch": 6.980237154150197,
|
| 3029 |
+
"grad_norm": 0.4462623596191406,
|
| 3030 |
+
"learning_rate": 4.469735621521336e-05,
|
| 3031 |
+
"loss": 0.2814,
|
| 3032 |
+
"step": 10150
|
| 3033 |
+
},
|
| 3034 |
+
{
|
| 3035 |
+
"epoch": 6.997422237497852,
|
| 3036 |
+
"grad_norm": 0.42071837186813354,
|
| 3037 |
+
"learning_rate": 4.468286178107607e-05,
|
| 3038 |
+
"loss": 0.2461,
|
| 3039 |
+
"step": 10175
|
| 3040 |
+
},
|
| 3041 |
+
{
|
| 3042 |
+
"epoch": 7.015122873345936,
|
| 3043 |
+
"grad_norm": 0.5645279288291931,
|
| 3044 |
+
"learning_rate": 4.466836734693878e-05,
|
| 3045 |
+
"loss": 0.2862,
|
| 3046 |
+
"step": 10200
|
| 3047 |
+
},
|
| 3048 |
+
{
|
| 3049 |
+
"epoch": 7.0323079566935895,
|
| 3050 |
+
"grad_norm": 1.0430643558502197,
|
| 3051 |
+
"learning_rate": 4.4653872912801484e-05,
|
| 3052 |
+
"loss": 0.2689,
|
| 3053 |
+
"step": 10225
|
| 3054 |
+
},
|
| 3055 |
+
{
|
| 3056 |
+
"epoch": 7.049493040041244,
|
| 3057 |
+
"grad_norm": 0.5359970331192017,
|
| 3058 |
+
"learning_rate": 4.46393784786642e-05,
|
| 3059 |
+
"loss": 0.2685,
|
| 3060 |
+
"step": 10250
|
| 3061 |
+
},
|
| 3062 |
+
{
|
| 3063 |
+
"epoch": 7.066678123388899,
|
| 3064 |
+
"grad_norm": 0.7322932481765747,
|
| 3065 |
+
"learning_rate": 4.4624884044526906e-05,
|
| 3066 |
+
"loss": 0.2806,
|
| 3067 |
+
"step": 10275
|
| 3068 |
+
},
|
| 3069 |
+
{
|
| 3070 |
+
"epoch": 7.083863206736552,
|
| 3071 |
+
"grad_norm": 0.4568728804588318,
|
| 3072 |
+
"learning_rate": 4.461038961038961e-05,
|
| 3073 |
+
"loss": 0.2726,
|
| 3074 |
+
"step": 10300
|
| 3075 |
+
},
|
| 3076 |
+
{
|
| 3077 |
+
"epoch": 7.101048290084207,
|
| 3078 |
+
"grad_norm": 1.341957926750183,
|
| 3079 |
+
"learning_rate": 4.459589517625232e-05,
|
| 3080 |
+
"loss": 0.2647,
|
| 3081 |
+
"step": 10325
|
| 3082 |
+
},
|
| 3083 |
+
{
|
| 3084 |
+
"epoch": 7.1182333734318615,
|
| 3085 |
+
"grad_norm": 0.48318567872047424,
|
| 3086 |
+
"learning_rate": 4.4581400742115035e-05,
|
| 3087 |
+
"loss": 0.28,
|
| 3088 |
+
"step": 10350
|
| 3089 |
+
},
|
| 3090 |
+
{
|
| 3091 |
+
"epoch": 7.135418456779515,
|
| 3092 |
+
"grad_norm": 0.7370210289955139,
|
| 3093 |
+
"learning_rate": 4.456690630797774e-05,
|
| 3094 |
+
"loss": 0.2886,
|
| 3095 |
+
"step": 10375
|
| 3096 |
+
},
|
| 3097 |
+
{
|
| 3098 |
+
"epoch": 7.15260354012717,
|
| 3099 |
+
"grad_norm": 0.703504741191864,
|
| 3100 |
+
"learning_rate": 4.455241187384045e-05,
|
| 3101 |
+
"loss": 0.276,
|
| 3102 |
+
"step": 10400
|
| 3103 |
+
},
|
| 3104 |
+
{
|
| 3105 |
+
"epoch": 7.169788623474824,
|
| 3106 |
+
"grad_norm": 0.5735402703285217,
|
| 3107 |
+
"learning_rate": 4.453791743970316e-05,
|
| 3108 |
+
"loss": 0.2786,
|
| 3109 |
+
"step": 10425
|
| 3110 |
+
},
|
| 3111 |
+
{
|
| 3112 |
+
"epoch": 7.186973706822478,
|
| 3113 |
+
"grad_norm": 0.36307334899902344,
|
| 3114 |
+
"learning_rate": 4.4523423005565865e-05,
|
| 3115 |
+
"loss": 0.2714,
|
| 3116 |
+
"step": 10450
|
| 3117 |
+
},
|
| 3118 |
+
{
|
| 3119 |
+
"epoch": 7.2041587901701325,
|
| 3120 |
+
"grad_norm": 0.8428148031234741,
|
| 3121 |
+
"learning_rate": 4.450892857142857e-05,
|
| 3122 |
+
"loss": 0.2997,
|
| 3123 |
+
"step": 10475
|
| 3124 |
+
},
|
| 3125 |
+
{
|
| 3126 |
+
"epoch": 7.221343873517786,
|
| 3127 |
+
"grad_norm": 0.8433165550231934,
|
| 3128 |
+
"learning_rate": 4.449443413729128e-05,
|
| 3129 |
+
"loss": 0.3008,
|
| 3130 |
+
"step": 10500
|
| 3131 |
+
},
|
| 3132 |
+
{
|
| 3133 |
+
"epoch": 7.221343873517786,
|
| 3134 |
+
"eval_loss": 0.3265960216522217,
|
| 3135 |
+
"eval_runtime": 151.3692,
|
| 3136 |
+
"eval_samples_per_second": 8.093,
|
| 3137 |
+
"eval_steps_per_second": 1.017,
|
| 3138 |
+
"eval_wer": 0.53984739296312,
|
| 3139 |
+
"step": 10500
|
| 3140 |
+
},
|
| 3141 |
+
{
|
| 3142 |
+
"epoch": 7.238528956865441,
|
| 3143 |
+
"grad_norm": 0.7506297826766968,
|
| 3144 |
+
"learning_rate": 4.447993970315399e-05,
|
| 3145 |
+
"loss": 0.2789,
|
| 3146 |
+
"step": 10525
|
| 3147 |
+
},
|
| 3148 |
+
{
|
| 3149 |
+
"epoch": 7.255714040213095,
|
| 3150 |
+
"grad_norm": 0.4726732075214386,
|
| 3151 |
+
"learning_rate": 4.44654452690167e-05,
|
| 3152 |
+
"loss": 0.2734,
|
| 3153 |
+
"step": 10550
|
| 3154 |
+
},
|
| 3155 |
+
{
|
| 3156 |
+
"epoch": 7.272899123560749,
|
| 3157 |
+
"grad_norm": 0.5341284275054932,
|
| 3158 |
+
"learning_rate": 4.445095083487941e-05,
|
| 3159 |
+
"loss": 0.2281,
|
| 3160 |
+
"step": 10575
|
| 3161 |
+
},
|
| 3162 |
+
{
|
| 3163 |
+
"epoch": 7.2900842069084035,
|
| 3164 |
+
"grad_norm": 1.0162677764892578,
|
| 3165 |
+
"learning_rate": 4.4436456400742116e-05,
|
| 3166 |
+
"loss": 0.2581,
|
| 3167 |
+
"step": 10600
|
| 3168 |
+
},
|
| 3169 |
+
{
|
| 3170 |
+
"epoch": 7.307269290256058,
|
| 3171 |
+
"grad_norm": 0.6543082594871521,
|
| 3172 |
+
"learning_rate": 4.4421961966604823e-05,
|
| 3173 |
+
"loss": 0.245,
|
| 3174 |
+
"step": 10625
|
| 3175 |
+
},
|
| 3176 |
+
{
|
| 3177 |
+
"epoch": 7.324454373603712,
|
| 3178 |
+
"grad_norm": 0.46228596568107605,
|
| 3179 |
+
"learning_rate": 4.440746753246753e-05,
|
| 3180 |
+
"loss": 0.2835,
|
| 3181 |
+
"step": 10650
|
| 3182 |
+
},
|
| 3183 |
+
{
|
| 3184 |
+
"epoch": 7.341639456951366,
|
| 3185 |
+
"grad_norm": 1.2381787300109863,
|
| 3186 |
+
"learning_rate": 4.4392973098330245e-05,
|
| 3187 |
+
"loss": 0.241,
|
| 3188 |
+
"step": 10675
|
| 3189 |
+
},
|
| 3190 |
+
{
|
| 3191 |
+
"epoch": 7.35882454029902,
|
| 3192 |
+
"grad_norm": 0.6333453059196472,
|
| 3193 |
+
"learning_rate": 4.437847866419295e-05,
|
| 3194 |
+
"loss": 0.2691,
|
| 3195 |
+
"step": 10700
|
| 3196 |
+
},
|
| 3197 |
+
{
|
| 3198 |
+
"epoch": 7.376009623646675,
|
| 3199 |
+
"grad_norm": 0.5812168121337891,
|
| 3200 |
+
"learning_rate": 4.436398423005566e-05,
|
| 3201 |
+
"loss": 0.2679,
|
| 3202 |
+
"step": 10725
|
| 3203 |
+
},
|
| 3204 |
+
{
|
| 3205 |
+
"epoch": 7.393194706994329,
|
| 3206 |
+
"grad_norm": 0.4156525433063507,
|
| 3207 |
+
"learning_rate": 4.434948979591837e-05,
|
| 3208 |
+
"loss": 0.2355,
|
| 3209 |
+
"step": 10750
|
| 3210 |
+
},
|
| 3211 |
+
{
|
| 3212 |
+
"epoch": 7.410379790341983,
|
| 3213 |
+
"grad_norm": 0.7613847255706787,
|
| 3214 |
+
"learning_rate": 4.433499536178108e-05,
|
| 3215 |
+
"loss": 0.2578,
|
| 3216 |
+
"step": 10775
|
| 3217 |
+
},
|
| 3218 |
+
{
|
| 3219 |
+
"epoch": 7.427564873689637,
|
| 3220 |
+
"grad_norm": 0.4807249903678894,
|
| 3221 |
+
"learning_rate": 4.432050092764379e-05,
|
| 3222 |
+
"loss": 0.2873,
|
| 3223 |
+
"step": 10800
|
| 3224 |
+
},
|
| 3225 |
+
{
|
| 3226 |
+
"epoch": 7.444749957037292,
|
| 3227 |
+
"grad_norm": 0.6777732372283936,
|
| 3228 |
+
"learning_rate": 4.4306006493506496e-05,
|
| 3229 |
+
"loss": 0.2319,
|
| 3230 |
+
"step": 10825
|
| 3231 |
+
},
|
| 3232 |
+
{
|
| 3233 |
+
"epoch": 7.461935040384946,
|
| 3234 |
+
"grad_norm": 0.5690301060676575,
|
| 3235 |
+
"learning_rate": 4.4291512059369204e-05,
|
| 3236 |
+
"loss": 0.2933,
|
| 3237 |
+
"step": 10850
|
| 3238 |
+
},
|
| 3239 |
+
{
|
| 3240 |
+
"epoch": 7.4791201237326,
|
| 3241 |
+
"grad_norm": 0.6938736438751221,
|
| 3242 |
+
"learning_rate": 4.427701762523191e-05,
|
| 3243 |
+
"loss": 0.25,
|
| 3244 |
+
"step": 10875
|
| 3245 |
+
},
|
| 3246 |
+
{
|
| 3247 |
+
"epoch": 7.496305207080255,
|
| 3248 |
+
"grad_norm": 1.2933138608932495,
|
| 3249 |
+
"learning_rate": 4.4262523191094626e-05,
|
| 3250 |
+
"loss": 0.2932,
|
| 3251 |
+
"step": 10900
|
| 3252 |
+
},
|
| 3253 |
+
{
|
| 3254 |
+
"epoch": 7.513490290427908,
|
| 3255 |
+
"grad_norm": 0.6283312439918518,
|
| 3256 |
+
"learning_rate": 4.424802875695733e-05,
|
| 3257 |
+
"loss": 0.2703,
|
| 3258 |
+
"step": 10925
|
| 3259 |
+
},
|
| 3260 |
+
{
|
| 3261 |
+
"epoch": 7.530675373775563,
|
| 3262 |
+
"grad_norm": 0.6064092516899109,
|
| 3263 |
+
"learning_rate": 4.423353432282004e-05,
|
| 3264 |
+
"loss": 0.2762,
|
| 3265 |
+
"step": 10950
|
| 3266 |
+
},
|
| 3267 |
+
{
|
| 3268 |
+
"epoch": 7.5478604571232175,
|
| 3269 |
+
"grad_norm": 0.6407192349433899,
|
| 3270 |
+
"learning_rate": 4.421903988868275e-05,
|
| 3271 |
+
"loss": 0.2477,
|
| 3272 |
+
"step": 10975
|
| 3273 |
+
},
|
| 3274 |
+
{
|
| 3275 |
+
"epoch": 7.565045540470871,
|
| 3276 |
+
"grad_norm": 0.40185797214508057,
|
| 3277 |
+
"learning_rate": 4.4204545454545455e-05,
|
| 3278 |
+
"loss": 0.2674,
|
| 3279 |
+
"step": 11000
|
| 3280 |
+
},
|
| 3281 |
+
{
|
| 3282 |
+
"epoch": 7.565045540470871,
|
| 3283 |
+
"eval_loss": 0.318492591381073,
|
| 3284 |
+
"eval_runtime": 150.7169,
|
| 3285 |
+
"eval_samples_per_second": 8.128,
|
| 3286 |
+
"eval_steps_per_second": 1.022,
|
| 3287 |
+
"eval_wer": 0.5379398050021196,
|
| 3288 |
+
"step": 11000
|
| 3289 |
+
},
|
| 3290 |
+
{
|
| 3291 |
+
"epoch": 7.565045540470871,
|
| 3292 |
+
"step": 11000,
|
| 3293 |
+
"total_flos": 8.071230701734987e+19,
|
| 3294 |
+
"train_loss": 0.7137202631343494,
|
| 3295 |
+
"train_runtime": 97593.296,
|
| 3296 |
+
"train_samples_per_second": 28.619,
|
| 3297 |
+
"train_steps_per_second": 0.894
|
| 3298 |
+
}
|
| 3299 |
+
],
|
| 3300 |
+
"logging_steps": 25,
|
| 3301 |
+
"max_steps": 87240,
|
| 3302 |
+
"num_input_tokens_seen": 0,
|
| 3303 |
+
"num_train_epochs": 60,
|
| 3304 |
+
"save_steps": 500,
|
| 3305 |
+
"stateful_callbacks": {
|
| 3306 |
+
"EarlyStoppingCallback": {
|
| 3307 |
+
"args": {
|
| 3308 |
+
"early_stopping_patience": 5,
|
| 3309 |
+
"early_stopping_threshold": 0.0
|
| 3310 |
+
},
|
| 3311 |
+
"attributes": {
|
| 3312 |
+
"early_stopping_patience_counter": 5
|
| 3313 |
+
}
|
| 3314 |
+
},
|
| 3315 |
+
"TrainerControl": {
|
| 3316 |
+
"args": {
|
| 3317 |
+
"should_epoch_stop": false,
|
| 3318 |
+
"should_evaluate": false,
|
| 3319 |
+
"should_log": false,
|
| 3320 |
+
"should_save": true,
|
| 3321 |
+
"should_training_stop": true
|
| 3322 |
+
},
|
| 3323 |
+
"attributes": {}
|
| 3324 |
+
}
|
| 3325 |
+
},
|
| 3326 |
+
"total_flos": 8.071230701734987e+19,
|
| 3327 |
+
"train_batch_size": 8,
|
| 3328 |
+
"trial_name": null,
|
| 3329 |
+
"trial_params": null
|
| 3330 |
+
}
|