End of training
Browse files- README.md +5 -2
- adapter.engs.safetensors +3 -0
- all_results.json +15 -0
- eval_results.json +9 -0
- train_results.json +9 -0
- trainer_state.json +3242 -0
README.md
CHANGED
|
@@ -3,6 +3,9 @@ library_name: transformers
|
|
| 3 |
license: cc-by-nc-4.0
|
| 4 |
base_model: facebook/mms-300m
|
| 5 |
tags:
|
|
|
|
|
|
|
|
|
|
| 6 |
- generated_from_trainer
|
| 7 |
metrics:
|
| 8 |
- wer
|
|
@@ -16,10 +19,10 @@ should probably proofread and complete it, then remove this comment. -->
|
|
| 16 |
|
| 17 |
# mms-300m-librispeech-adapter-model
|
| 18 |
|
| 19 |
-
This model is a fine-tuned version of [facebook/mms-300m](https://huggingface.co/facebook/mms-300m) on
|
| 20 |
It achieves the following results on the evaluation set:
|
| 21 |
- Loss: 0.1966
|
| 22 |
-
- Wer: 0.
|
| 23 |
|
| 24 |
## Model description
|
| 25 |
|
|
|
|
| 3 |
license: cc-by-nc-4.0
|
| 4 |
base_model: facebook/mms-300m
|
| 5 |
tags:
|
| 6 |
+
- automatic-speech-recognition
|
| 7 |
+
- libri10h
|
| 8 |
+
- mms
|
| 9 |
- generated_from_trainer
|
| 10 |
metrics:
|
| 11 |
- wer
|
|
|
|
| 19 |
|
| 20 |
# mms-300m-librispeech-adapter-model
|
| 21 |
|
| 22 |
+
This model is a fine-tuned version of [facebook/mms-300m](https://huggingface.co/facebook/mms-300m) on the LIBRI10H - ENGS dataset.
|
| 23 |
It achieves the following results on the evaluation set:
|
| 24 |
- Loss: 0.1966
|
| 25 |
+
- Wer: 0.1474
|
| 26 |
|
| 27 |
## Model description
|
| 28 |
|
adapter.engs.safetensors
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:4a3a2e1a2ed0561b8d19293f36a841dcb662612c0a51e5ebfeb44fd8863dcb0a
|
| 3 |
+
size 3586804
|
all_results.json
ADDED
|
@@ -0,0 +1,15 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"epoch": 57.971014492753625,
|
| 3 |
+
"eval_loss": 0.19662442803382874,
|
| 4 |
+
"eval_runtime": 150.3481,
|
| 5 |
+
"eval_samples": 2604,
|
| 6 |
+
"eval_samples_per_second": 17.32,
|
| 7 |
+
"eval_steps_per_second": 4.33,
|
| 8 |
+
"eval_wer": 0.1473525082547639,
|
| 9 |
+
"total_flos": 6.201114678692461e+19,
|
| 10 |
+
"train_loss": 0.3545982142448425,
|
| 11 |
+
"train_runtime": 69759.9962,
|
| 12 |
+
"train_samples": 2759,
|
| 13 |
+
"train_samples_per_second": 2.294,
|
| 14 |
+
"train_steps_per_second": 0.287
|
| 15 |
+
}
|
eval_results.json
ADDED
|
@@ -0,0 +1,9 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"epoch": 57.971014492753625,
|
| 3 |
+
"eval_loss": 0.19662442803382874,
|
| 4 |
+
"eval_runtime": 150.3481,
|
| 5 |
+
"eval_samples": 2604,
|
| 6 |
+
"eval_samples_per_second": 17.32,
|
| 7 |
+
"eval_steps_per_second": 4.33,
|
| 8 |
+
"eval_wer": 0.1473525082547639
|
| 9 |
+
}
|
train_results.json
ADDED
|
@@ -0,0 +1,9 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"epoch": 57.971014492753625,
|
| 3 |
+
"total_flos": 6.201114678692461e+19,
|
| 4 |
+
"train_loss": 0.3545982142448425,
|
| 5 |
+
"train_runtime": 69759.9962,
|
| 6 |
+
"train_samples": 2759,
|
| 7 |
+
"train_samples_per_second": 2.294,
|
| 8 |
+
"train_steps_per_second": 0.287
|
| 9 |
+
}
|
trainer_state.json
ADDED
|
@@ -0,0 +1,3242 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"best_metric": 0.18993638455867767,
|
| 3 |
+
"best_model_checkpoint": "/scratch/skscla001/speech/results/mms-300m-librispeech-adapter-model/checkpoint-14700",
|
| 4 |
+
"epoch": 57.971014492753625,
|
| 5 |
+
"eval_steps": 100,
|
| 6 |
+
"global_step": 20000,
|
| 7 |
+
"is_hyper_param_search": false,
|
| 8 |
+
"is_local_process_zero": true,
|
| 9 |
+
"is_world_process_zero": true,
|
| 10 |
+
"log_history": [
|
| 11 |
+
{
|
| 12 |
+
"epoch": 0.2898550724637681,
|
| 13 |
+
"grad_norm": 1.1440045833587646,
|
| 14 |
+
"learning_rate": 0.00029699999999999996,
|
| 15 |
+
"loss": 6.2783,
|
| 16 |
+
"step": 100
|
| 17 |
+
},
|
| 18 |
+
{
|
| 19 |
+
"epoch": 0.2898550724637681,
|
| 20 |
+
"eval_loss": 2.9219679832458496,
|
| 21 |
+
"eval_runtime": 143.5175,
|
| 22 |
+
"eval_samples_per_second": 18.144,
|
| 23 |
+
"eval_steps_per_second": 4.536,
|
| 24 |
+
"eval_wer": 1.0,
|
| 25 |
+
"step": 100
|
| 26 |
+
},
|
| 27 |
+
{
|
| 28 |
+
"epoch": 0.5797101449275363,
|
| 29 |
+
"grad_norm": 0.45640963315963745,
|
| 30 |
+
"learning_rate": 0.0002985075376884422,
|
| 31 |
+
"loss": 2.8558,
|
| 32 |
+
"step": 200
|
| 33 |
+
},
|
| 34 |
+
{
|
| 35 |
+
"epoch": 0.5797101449275363,
|
| 36 |
+
"eval_loss": 2.8894731998443604,
|
| 37 |
+
"eval_runtime": 142.4415,
|
| 38 |
+
"eval_samples_per_second": 18.281,
|
| 39 |
+
"eval_steps_per_second": 4.57,
|
| 40 |
+
"eval_wer": 1.0,
|
| 41 |
+
"step": 200
|
| 42 |
+
},
|
| 43 |
+
{
|
| 44 |
+
"epoch": 0.8695652173913043,
|
| 45 |
+
"grad_norm": 0.38338732719421387,
|
| 46 |
+
"learning_rate": 0.00029699999999999996,
|
| 47 |
+
"loss": 2.8526,
|
| 48 |
+
"step": 300
|
| 49 |
+
},
|
| 50 |
+
{
|
| 51 |
+
"epoch": 0.8695652173913043,
|
| 52 |
+
"eval_loss": 2.8732688426971436,
|
| 53 |
+
"eval_runtime": 142.7771,
|
| 54 |
+
"eval_samples_per_second": 18.238,
|
| 55 |
+
"eval_steps_per_second": 4.56,
|
| 56 |
+
"eval_wer": 1.0,
|
| 57 |
+
"step": 300
|
| 58 |
+
},
|
| 59 |
+
{
|
| 60 |
+
"epoch": 1.1594202898550725,
|
| 61 |
+
"grad_norm": 0.6874628067016602,
|
| 62 |
+
"learning_rate": 0.00029549246231155775,
|
| 63 |
+
"loss": 2.841,
|
| 64 |
+
"step": 400
|
| 65 |
+
},
|
| 66 |
+
{
|
| 67 |
+
"epoch": 1.1594202898550725,
|
| 68 |
+
"eval_loss": 2.9120943546295166,
|
| 69 |
+
"eval_runtime": 140.6955,
|
| 70 |
+
"eval_samples_per_second": 18.508,
|
| 71 |
+
"eval_steps_per_second": 4.627,
|
| 72 |
+
"eval_wer": 1.0,
|
| 73 |
+
"step": 400
|
| 74 |
+
},
|
| 75 |
+
{
|
| 76 |
+
"epoch": 1.4492753623188406,
|
| 77 |
+
"grad_norm": 0.5108693242073059,
|
| 78 |
+
"learning_rate": 0.00029398492462311555,
|
| 79 |
+
"loss": 2.831,
|
| 80 |
+
"step": 500
|
| 81 |
+
},
|
| 82 |
+
{
|
| 83 |
+
"epoch": 1.4492753623188406,
|
| 84 |
+
"eval_loss": 2.8824281692504883,
|
| 85 |
+
"eval_runtime": 141.8812,
|
| 86 |
+
"eval_samples_per_second": 18.353,
|
| 87 |
+
"eval_steps_per_second": 4.588,
|
| 88 |
+
"eval_wer": 1.0,
|
| 89 |
+
"step": 500
|
| 90 |
+
},
|
| 91 |
+
{
|
| 92 |
+
"epoch": 1.7391304347826086,
|
| 93 |
+
"grad_norm": 0.7823837995529175,
|
| 94 |
+
"learning_rate": 0.00029247738693467335,
|
| 95 |
+
"loss": 2.7114,
|
| 96 |
+
"step": 600
|
| 97 |
+
},
|
| 98 |
+
{
|
| 99 |
+
"epoch": 1.7391304347826086,
|
| 100 |
+
"eval_loss": 2.6127419471740723,
|
| 101 |
+
"eval_runtime": 143.0376,
|
| 102 |
+
"eval_samples_per_second": 18.205,
|
| 103 |
+
"eval_steps_per_second": 4.551,
|
| 104 |
+
"eval_wer": 1.0,
|
| 105 |
+
"step": 600
|
| 106 |
+
},
|
| 107 |
+
{
|
| 108 |
+
"epoch": 2.028985507246377,
|
| 109 |
+
"grad_norm": 0.5136599540710449,
|
| 110 |
+
"learning_rate": 0.00029096984924623114,
|
| 111 |
+
"loss": 2.316,
|
| 112 |
+
"step": 700
|
| 113 |
+
},
|
| 114 |
+
{
|
| 115 |
+
"epoch": 2.028985507246377,
|
| 116 |
+
"eval_loss": 1.937637209892273,
|
| 117 |
+
"eval_runtime": 143.9661,
|
| 118 |
+
"eval_samples_per_second": 18.088,
|
| 119 |
+
"eval_steps_per_second": 4.522,
|
| 120 |
+
"eval_wer": 0.9999005450133269,
|
| 121 |
+
"step": 700
|
| 122 |
+
},
|
| 123 |
+
{
|
| 124 |
+
"epoch": 2.318840579710145,
|
| 125 |
+
"grad_norm": 0.8716309666633606,
|
| 126 |
+
"learning_rate": 0.00028946231155778894,
|
| 127 |
+
"loss": 1.7366,
|
| 128 |
+
"step": 800
|
| 129 |
+
},
|
| 130 |
+
{
|
| 131 |
+
"epoch": 2.318840579710145,
|
| 132 |
+
"eval_loss": 1.22072434425354,
|
| 133 |
+
"eval_runtime": 144.0146,
|
| 134 |
+
"eval_samples_per_second": 18.082,
|
| 135 |
+
"eval_steps_per_second": 4.52,
|
| 136 |
+
"eval_wer": 0.8263515932688865,
|
| 137 |
+
"step": 800
|
| 138 |
+
},
|
| 139 |
+
{
|
| 140 |
+
"epoch": 2.608695652173913,
|
| 141 |
+
"grad_norm": 0.626215398311615,
|
| 142 |
+
"learning_rate": 0.0002879547738693467,
|
| 143 |
+
"loss": 1.1927,
|
| 144 |
+
"step": 900
|
| 145 |
+
},
|
| 146 |
+
{
|
| 147 |
+
"epoch": 2.608695652173913,
|
| 148 |
+
"eval_loss": 0.8307786583900452,
|
| 149 |
+
"eval_runtime": 144.3438,
|
| 150 |
+
"eval_samples_per_second": 18.04,
|
| 151 |
+
"eval_steps_per_second": 4.51,
|
| 152 |
+
"eval_wer": 0.6531407884791344,
|
| 153 |
+
"step": 900
|
| 154 |
+
},
|
| 155 |
+
{
|
| 156 |
+
"epoch": 2.898550724637681,
|
| 157 |
+
"grad_norm": 0.458715558052063,
|
| 158 |
+
"learning_rate": 0.0002864472361809045,
|
| 159 |
+
"loss": 0.923,
|
| 160 |
+
"step": 1000
|
| 161 |
+
},
|
| 162 |
+
{
|
| 163 |
+
"epoch": 2.898550724637681,
|
| 164 |
+
"eval_loss": 0.6707094311714172,
|
| 165 |
+
"eval_runtime": 144.3744,
|
| 166 |
+
"eval_samples_per_second": 18.036,
|
| 167 |
+
"eval_steps_per_second": 4.509,
|
| 168 |
+
"eval_wer": 0.5563511954489399,
|
| 169 |
+
"step": 1000
|
| 170 |
+
},
|
| 171 |
+
{
|
| 172 |
+
"epoch": 3.1884057971014492,
|
| 173 |
+
"grad_norm": 0.5768333673477173,
|
| 174 |
+
"learning_rate": 0.0002849396984924623,
|
| 175 |
+
"loss": 0.7974,
|
| 176 |
+
"step": 1100
|
| 177 |
+
},
|
| 178 |
+
{
|
| 179 |
+
"epoch": 3.1884057971014492,
|
| 180 |
+
"eval_loss": 0.59714674949646,
|
| 181 |
+
"eval_runtime": 144.5219,
|
| 182 |
+
"eval_samples_per_second": 18.018,
|
| 183 |
+
"eval_steps_per_second": 4.505,
|
| 184 |
+
"eval_wer": 0.5258980785296575,
|
| 185 |
+
"step": 1100
|
| 186 |
+
},
|
| 187 |
+
{
|
| 188 |
+
"epoch": 3.4782608695652173,
|
| 189 |
+
"grad_norm": 0.5361514687538147,
|
| 190 |
+
"learning_rate": 0.0002834321608040201,
|
| 191 |
+
"loss": 0.7119,
|
| 192 |
+
"step": 1200
|
| 193 |
+
},
|
| 194 |
+
{
|
| 195 |
+
"epoch": 3.4782608695652173,
|
| 196 |
+
"eval_loss": 0.5403450131416321,
|
| 197 |
+
"eval_runtime": 144.9473,
|
| 198 |
+
"eval_samples_per_second": 17.965,
|
| 199 |
+
"eval_steps_per_second": 4.491,
|
| 200 |
+
"eval_wer": 0.4906711222500696,
|
| 201 |
+
"step": 1200
|
| 202 |
+
},
|
| 203 |
+
{
|
| 204 |
+
"epoch": 3.7681159420289854,
|
| 205 |
+
"grad_norm": 0.44001927971839905,
|
| 206 |
+
"learning_rate": 0.0002819246231155779,
|
| 207 |
+
"loss": 0.6491,
|
| 208 |
+
"step": 1300
|
| 209 |
+
},
|
| 210 |
+
{
|
| 211 |
+
"epoch": 3.7681159420289854,
|
| 212 |
+
"eval_loss": 0.49191784858703613,
|
| 213 |
+
"eval_runtime": 145.6845,
|
| 214 |
+
"eval_samples_per_second": 17.874,
|
| 215 |
+
"eval_steps_per_second": 4.469,
|
| 216 |
+
"eval_wer": 0.4562199148665314,
|
| 217 |
+
"step": 1300
|
| 218 |
+
},
|
| 219 |
+
{
|
| 220 |
+
"epoch": 4.057971014492754,
|
| 221 |
+
"grad_norm": 0.5078291296958923,
|
| 222 |
+
"learning_rate": 0.00028041708542713567,
|
| 223 |
+
"loss": 0.6167,
|
| 224 |
+
"step": 1400
|
| 225 |
+
},
|
| 226 |
+
{
|
| 227 |
+
"epoch": 4.057971014492754,
|
| 228 |
+
"eval_loss": 0.47232988476753235,
|
| 229 |
+
"eval_runtime": 145.8228,
|
| 230 |
+
"eval_samples_per_second": 17.857,
|
| 231 |
+
"eval_steps_per_second": 4.464,
|
| 232 |
+
"eval_wer": 0.44744798504197003,
|
| 233 |
+
"step": 1400
|
| 234 |
+
},
|
| 235 |
+
{
|
| 236 |
+
"epoch": 4.3478260869565215,
|
| 237 |
+
"grad_norm": 0.45701169967651367,
|
| 238 |
+
"learning_rate": 0.0002789095477386934,
|
| 239 |
+
"loss": 0.5899,
|
| 240 |
+
"step": 1500
|
| 241 |
+
},
|
| 242 |
+
{
|
| 243 |
+
"epoch": 4.3478260869565215,
|
| 244 |
+
"eval_loss": 0.43916764855384827,
|
| 245 |
+
"eval_runtime": 146.0418,
|
| 246 |
+
"eval_samples_per_second": 17.831,
|
| 247 |
+
"eval_steps_per_second": 4.458,
|
| 248 |
+
"eval_wer": 0.42153001551497793,
|
| 249 |
+
"step": 1500
|
| 250 |
+
},
|
| 251 |
+
{
|
| 252 |
+
"epoch": 4.63768115942029,
|
| 253 |
+
"grad_norm": 0.4275153577327728,
|
| 254 |
+
"learning_rate": 0.00027740201005025127,
|
| 255 |
+
"loss": 0.5461,
|
| 256 |
+
"step": 1600
|
| 257 |
+
},
|
| 258 |
+
{
|
| 259 |
+
"epoch": 4.63768115942029,
|
| 260 |
+
"eval_loss": 0.4208216369152069,
|
| 261 |
+
"eval_runtime": 146.3826,
|
| 262 |
+
"eval_samples_per_second": 17.789,
|
| 263 |
+
"eval_steps_per_second": 4.447,
|
| 264 |
+
"eval_wer": 0.4089986871941759,
|
| 265 |
+
"step": 1600
|
| 266 |
+
},
|
| 267 |
+
{
|
| 268 |
+
"epoch": 4.927536231884058,
|
| 269 |
+
"grad_norm": 0.5200296640396118,
|
| 270 |
+
"learning_rate": 0.000275894472361809,
|
| 271 |
+
"loss": 0.5331,
|
| 272 |
+
"step": 1700
|
| 273 |
+
},
|
| 274 |
+
{
|
| 275 |
+
"epoch": 4.927536231884058,
|
| 276 |
+
"eval_loss": 0.40836045145988464,
|
| 277 |
+
"eval_runtime": 146.2881,
|
| 278 |
+
"eval_samples_per_second": 17.8,
|
| 279 |
+
"eval_steps_per_second": 4.45,
|
| 280 |
+
"eval_wer": 0.40305127899112864,
|
| 281 |
+
"step": 1700
|
| 282 |
+
},
|
| 283 |
+
{
|
| 284 |
+
"epoch": 5.217391304347826,
|
| 285 |
+
"grad_norm": 0.4017215371131897,
|
| 286 |
+
"learning_rate": 0.0002743869346733668,
|
| 287 |
+
"loss": 0.5134,
|
| 288 |
+
"step": 1800
|
| 289 |
+
},
|
| 290 |
+
{
|
| 291 |
+
"epoch": 5.217391304347826,
|
| 292 |
+
"eval_loss": 0.38590207695961,
|
| 293 |
+
"eval_runtime": 146.2278,
|
| 294 |
+
"eval_samples_per_second": 17.808,
|
| 295 |
+
"eval_steps_per_second": 4.452,
|
| 296 |
+
"eval_wer": 0.3848311254326292,
|
| 297 |
+
"step": 1800
|
| 298 |
+
},
|
| 299 |
+
{
|
| 300 |
+
"epoch": 5.507246376811594,
|
| 301 |
+
"grad_norm": 0.5074508190155029,
|
| 302 |
+
"learning_rate": 0.0002728793969849246,
|
| 303 |
+
"loss": 0.4967,
|
| 304 |
+
"step": 1900
|
| 305 |
+
},
|
| 306 |
+
{
|
| 307 |
+
"epoch": 5.507246376811594,
|
| 308 |
+
"eval_loss": 0.3674832284450531,
|
| 309 |
+
"eval_runtime": 145.8489,
|
| 310 |
+
"eval_samples_per_second": 17.854,
|
| 311 |
+
"eval_steps_per_second": 4.464,
|
| 312 |
+
"eval_wer": 0.3747463897839838,
|
| 313 |
+
"step": 1900
|
| 314 |
+
},
|
| 315 |
+
{
|
| 316 |
+
"epoch": 5.797101449275362,
|
| 317 |
+
"grad_norm": 0.44507330656051636,
|
| 318 |
+
"learning_rate": 0.0002713718592964824,
|
| 319 |
+
"loss": 0.4735,
|
| 320 |
+
"step": 2000
|
| 321 |
+
},
|
| 322 |
+
{
|
| 323 |
+
"epoch": 5.797101449275362,
|
| 324 |
+
"eval_loss": 0.3600199818611145,
|
| 325 |
+
"eval_runtime": 146.0583,
|
| 326 |
+
"eval_samples_per_second": 17.828,
|
| 327 |
+
"eval_steps_per_second": 4.457,
|
| 328 |
+
"eval_wer": 0.36408481521263475,
|
| 329 |
+
"step": 2000
|
| 330 |
+
},
|
| 331 |
+
{
|
| 332 |
+
"epoch": 6.086956521739131,
|
| 333 |
+
"grad_norm": 0.4315125644207001,
|
| 334 |
+
"learning_rate": 0.0002698643216080402,
|
| 335 |
+
"loss": 0.4589,
|
| 336 |
+
"step": 2100
|
| 337 |
+
},
|
| 338 |
+
{
|
| 339 |
+
"epoch": 6.086956521739131,
|
| 340 |
+
"eval_loss": 0.34947261214256287,
|
| 341 |
+
"eval_runtime": 146.0671,
|
| 342 |
+
"eval_samples_per_second": 17.827,
|
| 343 |
+
"eval_steps_per_second": 4.457,
|
| 344 |
+
"eval_wer": 0.358057843020249,
|
| 345 |
+
"step": 2100
|
| 346 |
+
},
|
| 347 |
+
{
|
| 348 |
+
"epoch": 6.3768115942028984,
|
| 349 |
+
"grad_norm": 0.3876771628856659,
|
| 350 |
+
"learning_rate": 0.000268356783919598,
|
| 351 |
+
"loss": 0.4395,
|
| 352 |
+
"step": 2200
|
| 353 |
+
},
|
| 354 |
+
{
|
| 355 |
+
"epoch": 6.3768115942028984,
|
| 356 |
+
"eval_loss": 0.33656537532806396,
|
| 357 |
+
"eval_runtime": 146.4679,
|
| 358 |
+
"eval_samples_per_second": 17.779,
|
| 359 |
+
"eval_steps_per_second": 4.445,
|
| 360 |
+
"eval_wer": 0.348410709312965,
|
| 361 |
+
"step": 2200
|
| 362 |
+
},
|
| 363 |
+
{
|
| 364 |
+
"epoch": 6.666666666666667,
|
| 365 |
+
"grad_norm": 0.5842604041099548,
|
| 366 |
+
"learning_rate": 0.00026684924623115574,
|
| 367 |
+
"loss": 0.4463,
|
| 368 |
+
"step": 2300
|
| 369 |
+
},
|
| 370 |
+
{
|
| 371 |
+
"epoch": 6.666666666666667,
|
| 372 |
+
"eval_loss": 0.3318806290626526,
|
| 373 |
+
"eval_runtime": 146.1921,
|
| 374 |
+
"eval_samples_per_second": 17.812,
|
| 375 |
+
"eval_steps_per_second": 4.453,
|
| 376 |
+
"eval_wer": 0.34687910251820026,
|
| 377 |
+
"step": 2300
|
| 378 |
+
},
|
| 379 |
+
{
|
| 380 |
+
"epoch": 6.956521739130435,
|
| 381 |
+
"grad_norm": 0.48257681727409363,
|
| 382 |
+
"learning_rate": 0.00026534170854271353,
|
| 383 |
+
"loss": 0.4367,
|
| 384 |
+
"step": 2400
|
| 385 |
+
},
|
| 386 |
+
{
|
| 387 |
+
"epoch": 6.956521739130435,
|
| 388 |
+
"eval_loss": 0.3226897716522217,
|
| 389 |
+
"eval_runtime": 145.9874,
|
| 390 |
+
"eval_samples_per_second": 17.837,
|
| 391 |
+
"eval_steps_per_second": 4.459,
|
| 392 |
+
"eval_wer": 0.3391017225603692,
|
| 393 |
+
"step": 2400
|
| 394 |
+
},
|
| 395 |
+
{
|
| 396 |
+
"epoch": 7.246376811594203,
|
| 397 |
+
"grad_norm": 0.5191691517829895,
|
| 398 |
+
"learning_rate": 0.00026383417085427133,
|
| 399 |
+
"loss": 0.4034,
|
| 400 |
+
"step": 2500
|
| 401 |
+
},
|
| 402 |
+
{
|
| 403 |
+
"epoch": 7.246376811594203,
|
| 404 |
+
"eval_loss": 0.3165632486343384,
|
| 405 |
+
"eval_runtime": 151.3525,
|
| 406 |
+
"eval_samples_per_second": 17.205,
|
| 407 |
+
"eval_steps_per_second": 4.301,
|
| 408 |
+
"eval_wer": 0.3279229820583204,
|
| 409 |
+
"step": 2500
|
| 410 |
+
},
|
| 411 |
+
{
|
| 412 |
+
"epoch": 7.536231884057971,
|
| 413 |
+
"grad_norm": 0.45448774099349976,
|
| 414 |
+
"learning_rate": 0.00026232663316582913,
|
| 415 |
+
"loss": 0.4088,
|
| 416 |
+
"step": 2600
|
| 417 |
+
},
|
| 418 |
+
{
|
| 419 |
+
"epoch": 7.536231884057971,
|
| 420 |
+
"eval_loss": 0.3038766086101532,
|
| 421 |
+
"eval_runtime": 146.2949,
|
| 422 |
+
"eval_samples_per_second": 17.8,
|
| 423 |
+
"eval_steps_per_second": 4.45,
|
| 424 |
+
"eval_wer": 0.31960854517245496,
|
| 425 |
+
"step": 2600
|
| 426 |
+
},
|
| 427 |
+
{
|
| 428 |
+
"epoch": 7.826086956521739,
|
| 429 |
+
"grad_norm": 0.5683527588844299,
|
| 430 |
+
"learning_rate": 0.0002608190954773869,
|
| 431 |
+
"loss": 0.4024,
|
| 432 |
+
"step": 2700
|
| 433 |
+
},
|
| 434 |
+
{
|
| 435 |
+
"epoch": 7.826086956521739,
|
| 436 |
+
"eval_loss": 0.30069929361343384,
|
| 437 |
+
"eval_runtime": 145.9069,
|
| 438 |
+
"eval_samples_per_second": 17.847,
|
| 439 |
+
"eval_steps_per_second": 4.462,
|
| 440 |
+
"eval_wer": 0.3201058201058201,
|
| 441 |
+
"step": 2700
|
| 442 |
+
},
|
| 443 |
+
{
|
| 444 |
+
"epoch": 8.115942028985508,
|
| 445 |
+
"grad_norm": 0.5986935496330261,
|
| 446 |
+
"learning_rate": 0.0002593115577889447,
|
| 447 |
+
"loss": 0.3994,
|
| 448 |
+
"step": 2800
|
| 449 |
+
},
|
| 450 |
+
{
|
| 451 |
+
"epoch": 8.115942028985508,
|
| 452 |
+
"eval_loss": 0.2905051112174988,
|
| 453 |
+
"eval_runtime": 146.8783,
|
| 454 |
+
"eval_samples_per_second": 17.729,
|
| 455 |
+
"eval_steps_per_second": 4.432,
|
| 456 |
+
"eval_wer": 0.30687830687830686,
|
| 457 |
+
"step": 2800
|
| 458 |
+
},
|
| 459 |
+
{
|
| 460 |
+
"epoch": 8.405797101449275,
|
| 461 |
+
"grad_norm": 0.542873740196228,
|
| 462 |
+
"learning_rate": 0.00025780402010050247,
|
| 463 |
+
"loss": 0.3789,
|
| 464 |
+
"step": 2900
|
| 465 |
+
},
|
| 466 |
+
{
|
| 467 |
+
"epoch": 8.405797101449275,
|
| 468 |
+
"eval_loss": 0.28533315658569336,
|
| 469 |
+
"eval_runtime": 147.0496,
|
| 470 |
+
"eval_samples_per_second": 17.708,
|
| 471 |
+
"eval_steps_per_second": 4.427,
|
| 472 |
+
"eval_wer": 0.29866332497911446,
|
| 473 |
+
"step": 2900
|
| 474 |
+
},
|
| 475 |
+
{
|
| 476 |
+
"epoch": 8.695652173913043,
|
| 477 |
+
"grad_norm": 0.5281730890274048,
|
| 478 |
+
"learning_rate": 0.00025629648241206026,
|
| 479 |
+
"loss": 0.3789,
|
| 480 |
+
"step": 3000
|
| 481 |
+
},
|
| 482 |
+
{
|
| 483 |
+
"epoch": 8.695652173913043,
|
| 484 |
+
"eval_loss": 0.282540500164032,
|
| 485 |
+
"eval_runtime": 147.0526,
|
| 486 |
+
"eval_samples_per_second": 17.708,
|
| 487 |
+
"eval_steps_per_second": 4.427,
|
| 488 |
+
"eval_wer": 0.2968930262163345,
|
| 489 |
+
"step": 3000
|
| 490 |
+
},
|
| 491 |
+
{
|
| 492 |
+
"epoch": 8.985507246376812,
|
| 493 |
+
"grad_norm": 0.7449385523796082,
|
| 494 |
+
"learning_rate": 0.00025478894472361806,
|
| 495 |
+
"loss": 0.371,
|
| 496 |
+
"step": 3100
|
| 497 |
+
},
|
| 498 |
+
{
|
| 499 |
+
"epoch": 8.985507246376812,
|
| 500 |
+
"eval_loss": 0.27829161286354065,
|
| 501 |
+
"eval_runtime": 146.3003,
|
| 502 |
+
"eval_samples_per_second": 17.799,
|
| 503 |
+
"eval_steps_per_second": 4.45,
|
| 504 |
+
"eval_wer": 0.2889366272824919,
|
| 505 |
+
"step": 3100
|
| 506 |
+
},
|
| 507 |
+
{
|
| 508 |
+
"epoch": 9.27536231884058,
|
| 509 |
+
"grad_norm": 0.8428851366043091,
|
| 510 |
+
"learning_rate": 0.00025328140703517586,
|
| 511 |
+
"loss": 0.356,
|
| 512 |
+
"step": 3200
|
| 513 |
+
},
|
| 514 |
+
{
|
| 515 |
+
"epoch": 9.27536231884058,
|
| 516 |
+
"eval_loss": 0.27369388937950134,
|
| 517 |
+
"eval_runtime": 146.3486,
|
| 518 |
+
"eval_samples_per_second": 17.793,
|
| 519 |
+
"eval_steps_per_second": 4.448,
|
| 520 |
+
"eval_wer": 0.2853363567649282,
|
| 521 |
+
"step": 3200
|
| 522 |
+
},
|
| 523 |
+
{
|
| 524 |
+
"epoch": 9.565217391304348,
|
| 525 |
+
"grad_norm": 0.6694580316543579,
|
| 526 |
+
"learning_rate": 0.00025177386934673366,
|
| 527 |
+
"loss": 0.3519,
|
| 528 |
+
"step": 3300
|
| 529 |
+
},
|
| 530 |
+
{
|
| 531 |
+
"epoch": 9.565217391304348,
|
| 532 |
+
"eval_loss": 0.27026933431625366,
|
| 533 |
+
"eval_runtime": 146.7069,
|
| 534 |
+
"eval_samples_per_second": 17.75,
|
| 535 |
+
"eval_steps_per_second": 4.437,
|
| 536 |
+
"eval_wer": 0.2834666030154752,
|
| 537 |
+
"step": 3300
|
| 538 |
+
},
|
| 539 |
+
{
|
| 540 |
+
"epoch": 9.855072463768115,
|
| 541 |
+
"grad_norm": 0.667295515537262,
|
| 542 |
+
"learning_rate": 0.00025026633165829145,
|
| 543 |
+
"loss": 0.3547,
|
| 544 |
+
"step": 3400
|
| 545 |
+
},
|
| 546 |
+
{
|
| 547 |
+
"epoch": 9.855072463768115,
|
| 548 |
+
"eval_loss": 0.2659197747707367,
|
| 549 |
+
"eval_runtime": 146.6151,
|
| 550 |
+
"eval_samples_per_second": 17.761,
|
| 551 |
+
"eval_steps_per_second": 4.44,
|
| 552 |
+
"eval_wer": 0.2799458964872499,
|
| 553 |
+
"step": 3400
|
| 554 |
+
},
|
| 555 |
+
{
|
| 556 |
+
"epoch": 10.144927536231885,
|
| 557 |
+
"grad_norm": 3.2425894737243652,
|
| 558 |
+
"learning_rate": 0.0002487587939698492,
|
| 559 |
+
"loss": 0.3474,
|
| 560 |
+
"step": 3500
|
| 561 |
+
},
|
| 562 |
+
{
|
| 563 |
+
"epoch": 10.144927536231885,
|
| 564 |
+
"eval_loss": 0.2704484164714813,
|
| 565 |
+
"eval_runtime": 147.9392,
|
| 566 |
+
"eval_samples_per_second": 17.602,
|
| 567 |
+
"eval_steps_per_second": 4.4,
|
| 568 |
+
"eval_wer": 0.2744758722202331,
|
| 569 |
+
"step": 3500
|
| 570 |
+
},
|
| 571 |
+
{
|
| 572 |
+
"epoch": 10.434782608695652,
|
| 573 |
+
"grad_norm": 1.8763881921768188,
|
| 574 |
+
"learning_rate": 0.00024725125628140705,
|
| 575 |
+
"loss": 0.3402,
|
| 576 |
+
"step": 3600
|
| 577 |
+
},
|
| 578 |
+
{
|
| 579 |
+
"epoch": 10.434782608695652,
|
| 580 |
+
"eval_loss": 0.25888901948928833,
|
| 581 |
+
"eval_runtime": 146.8065,
|
| 582 |
+
"eval_samples_per_second": 17.738,
|
| 583 |
+
"eval_steps_per_second": 4.434,
|
| 584 |
+
"eval_wer": 0.2672753311851056,
|
| 585 |
+
"step": 3600
|
| 586 |
+
},
|
| 587 |
+
{
|
| 588 |
+
"epoch": 10.72463768115942,
|
| 589 |
+
"grad_norm": 1.3940495252609253,
|
| 590 |
+
"learning_rate": 0.0002457437185929648,
|
| 591 |
+
"loss": 0.3305,
|
| 592 |
+
"step": 3700
|
| 593 |
+
},
|
| 594 |
+
{
|
| 595 |
+
"epoch": 10.72463768115942,
|
| 596 |
+
"eval_loss": 0.26088258624076843,
|
| 597 |
+
"eval_runtime": 146.7967,
|
| 598 |
+
"eval_samples_per_second": 17.739,
|
| 599 |
+
"eval_steps_per_second": 4.435,
|
| 600 |
+
"eval_wer": 0.2688069379798703,
|
| 601 |
+
"step": 3700
|
| 602 |
+
},
|
| 603 |
+
{
|
| 604 |
+
"epoch": 11.014492753623188,
|
| 605 |
+
"grad_norm": 0.4827100336551666,
|
| 606 |
+
"learning_rate": 0.0002442361809045226,
|
| 607 |
+
"loss": 0.3306,
|
| 608 |
+
"step": 3800
|
| 609 |
+
},
|
| 610 |
+
{
|
| 611 |
+
"epoch": 11.014492753623188,
|
| 612 |
+
"eval_loss": 0.25174590945243835,
|
| 613 |
+
"eval_runtime": 147.1647,
|
| 614 |
+
"eval_samples_per_second": 17.694,
|
| 615 |
+
"eval_steps_per_second": 4.424,
|
| 616 |
+
"eval_wer": 0.2602935911206588,
|
| 617 |
+
"step": 3800
|
| 618 |
+
},
|
| 619 |
+
{
|
| 620 |
+
"epoch": 11.304347826086957,
|
| 621 |
+
"grad_norm": 0.503418505191803,
|
| 622 |
+
"learning_rate": 0.00024272864321608038,
|
| 623 |
+
"loss": 0.315,
|
| 624 |
+
"step": 3900
|
| 625 |
+
},
|
| 626 |
+
{
|
| 627 |
+
"epoch": 11.304347826086957,
|
| 628 |
+
"eval_loss": 0.2613690495491028,
|
| 629 |
+
"eval_runtime": 146.9874,
|
| 630 |
+
"eval_samples_per_second": 17.716,
|
| 631 |
+
"eval_steps_per_second": 4.429,
|
| 632 |
+
"eval_wer": 0.2663802363050483,
|
| 633 |
+
"step": 3900
|
| 634 |
+
},
|
| 635 |
+
{
|
| 636 |
+
"epoch": 11.594202898550725,
|
| 637 |
+
"grad_norm": 0.5063391923904419,
|
| 638 |
+
"learning_rate": 0.00024122110552763816,
|
| 639 |
+
"loss": 0.319,
|
| 640 |
+
"step": 4000
|
| 641 |
+
},
|
| 642 |
+
{
|
| 643 |
+
"epoch": 11.594202898550725,
|
| 644 |
+
"eval_loss": 0.249360591173172,
|
| 645 |
+
"eval_runtime": 147.2708,
|
| 646 |
+
"eval_samples_per_second": 17.682,
|
| 647 |
+
"eval_steps_per_second": 4.42,
|
| 648 |
+
"eval_wer": 0.25822492739785974,
|
| 649 |
+
"step": 4000
|
| 650 |
+
},
|
| 651 |
+
{
|
| 652 |
+
"epoch": 11.884057971014492,
|
| 653 |
+
"grad_norm": 0.5601128339767456,
|
| 654 |
+
"learning_rate": 0.00023971356783919598,
|
| 655 |
+
"loss": 0.316,
|
| 656 |
+
"step": 4100
|
| 657 |
+
},
|
| 658 |
+
{
|
| 659 |
+
"epoch": 11.884057971014492,
|
| 660 |
+
"eval_loss": 0.24954313039779663,
|
| 661 |
+
"eval_runtime": 147.9605,
|
| 662 |
+
"eval_samples_per_second": 17.599,
|
| 663 |
+
"eval_steps_per_second": 4.4,
|
| 664 |
+
"eval_wer": 0.2546047658829614,
|
| 665 |
+
"step": 4100
|
| 666 |
+
},
|
| 667 |
+
{
|
| 668 |
+
"epoch": 12.173913043478262,
|
| 669 |
+
"grad_norm": 0.5574008226394653,
|
| 670 |
+
"learning_rate": 0.00023820603015075375,
|
| 671 |
+
"loss": 0.3097,
|
| 672 |
+
"step": 4200
|
| 673 |
+
},
|
| 674 |
+
{
|
| 675 |
+
"epoch": 12.173913043478262,
|
| 676 |
+
"eval_loss": 0.24174192547798157,
|
| 677 |
+
"eval_runtime": 147.0046,
|
| 678 |
+
"eval_samples_per_second": 17.714,
|
| 679 |
+
"eval_steps_per_second": 4.428,
|
| 680 |
+
"eval_wer": 0.2503083104586864,
|
| 681 |
+
"step": 4200
|
| 682 |
+
},
|
| 683 |
+
{
|
| 684 |
+
"epoch": 12.46376811594203,
|
| 685 |
+
"grad_norm": 0.44565197825431824,
|
| 686 |
+
"learning_rate": 0.00023669849246231155,
|
| 687 |
+
"loss": 0.3027,
|
| 688 |
+
"step": 4300
|
| 689 |
+
},
|
| 690 |
+
{
|
| 691 |
+
"epoch": 12.46376811594203,
|
| 692 |
+
"eval_loss": 0.2395372986793518,
|
| 693 |
+
"eval_runtime": 148.3347,
|
| 694 |
+
"eval_samples_per_second": 17.555,
|
| 695 |
+
"eval_steps_per_second": 4.389,
|
| 696 |
+
"eval_wer": 0.24398297330628158,
|
| 697 |
+
"step": 4300
|
| 698 |
+
},
|
| 699 |
+
{
|
| 700 |
+
"epoch": 12.753623188405797,
|
| 701 |
+
"grad_norm": 0.4654025733470917,
|
| 702 |
+
"learning_rate": 0.00023519095477386932,
|
| 703 |
+
"loss": 0.2978,
|
| 704 |
+
"step": 4400
|
| 705 |
+
},
|
| 706 |
+
{
|
| 707 |
+
"epoch": 12.753623188405797,
|
| 708 |
+
"eval_loss": 0.236846461892128,
|
| 709 |
+
"eval_runtime": 147.8716,
|
| 710 |
+
"eval_samples_per_second": 17.61,
|
| 711 |
+
"eval_steps_per_second": 4.402,
|
| 712 |
+
"eval_wer": 0.2450173051676811,
|
| 713 |
+
"step": 4400
|
| 714 |
+
},
|
| 715 |
+
{
|
| 716 |
+
"epoch": 13.043478260869565,
|
| 717 |
+
"grad_norm": 0.5224162936210632,
|
| 718 |
+
"learning_rate": 0.00023368341708542711,
|
| 719 |
+
"loss": 0.2891,
|
| 720 |
+
"step": 4500
|
| 721 |
+
},
|
| 722 |
+
{
|
| 723 |
+
"epoch": 13.043478260869565,
|
| 724 |
+
"eval_loss": 0.24193796515464783,
|
| 725 |
+
"eval_runtime": 148.0829,
|
| 726 |
+
"eval_samples_per_second": 17.585,
|
| 727 |
+
"eval_steps_per_second": 4.396,
|
| 728 |
+
"eval_wer": 0.2487170306719179,
|
| 729 |
+
"step": 4500
|
| 730 |
+
},
|
| 731 |
+
{
|
| 732 |
+
"epoch": 13.333333333333334,
|
| 733 |
+
"grad_norm": 0.5733482241630554,
|
| 734 |
+
"learning_rate": 0.0002321758793969849,
|
| 735 |
+
"loss": 0.2931,
|
| 736 |
+
"step": 4600
|
| 737 |
+
},
|
| 738 |
+
{
|
| 739 |
+
"epoch": 13.333333333333334,
|
| 740 |
+
"eval_loss": 0.23379352688789368,
|
| 741 |
+
"eval_runtime": 147.4582,
|
| 742 |
+
"eval_samples_per_second": 17.659,
|
| 743 |
+
"eval_steps_per_second": 4.415,
|
| 744 |
+
"eval_wer": 0.23485300552969726,
|
| 745 |
+
"step": 4600
|
| 746 |
+
},
|
| 747 |
+
{
|
| 748 |
+
"epoch": 13.623188405797102,
|
| 749 |
+
"grad_norm": 0.5568373203277588,
|
| 750 |
+
"learning_rate": 0.0002306683417085427,
|
| 751 |
+
"loss": 0.2845,
|
| 752 |
+
"step": 4700
|
| 753 |
+
},
|
| 754 |
+
{
|
| 755 |
+
"epoch": 13.623188405797102,
|
| 756 |
+
"eval_loss": 0.23074574768543243,
|
| 757 |
+
"eval_runtime": 147.9177,
|
| 758 |
+
"eval_samples_per_second": 17.604,
|
| 759 |
+
"eval_steps_per_second": 4.401,
|
| 760 |
+
"eval_wer": 0.2324859768468791,
|
| 761 |
+
"step": 4700
|
| 762 |
+
},
|
| 763 |
+
{
|
| 764 |
+
"epoch": 13.91304347826087,
|
| 765 |
+
"grad_norm": 0.5352379083633423,
|
| 766 |
+
"learning_rate": 0.00022916080402010048,
|
| 767 |
+
"loss": 0.286,
|
| 768 |
+
"step": 4800
|
| 769 |
+
},
|
| 770 |
+
{
|
| 771 |
+
"epoch": 13.91304347826087,
|
| 772 |
+
"eval_loss": 0.23180751502513885,
|
| 773 |
+
"eval_runtime": 148.3255,
|
| 774 |
+
"eval_samples_per_second": 17.556,
|
| 775 |
+
"eval_steps_per_second": 4.389,
|
| 776 |
+
"eval_wer": 0.23457453156701277,
|
| 777 |
+
"step": 4800
|
| 778 |
+
},
|
| 779 |
+
{
|
| 780 |
+
"epoch": 14.202898550724637,
|
| 781 |
+
"grad_norm": 0.4989466071128845,
|
| 782 |
+
"learning_rate": 0.00022765326633165828,
|
| 783 |
+
"loss": 0.2839,
|
| 784 |
+
"step": 4900
|
| 785 |
+
},
|
| 786 |
+
{
|
| 787 |
+
"epoch": 14.202898550724637,
|
| 788 |
+
"eval_loss": 0.22843925654888153,
|
| 789 |
+
"eval_runtime": 149.221,
|
| 790 |
+
"eval_samples_per_second": 17.451,
|
| 791 |
+
"eval_steps_per_second": 4.363,
|
| 792 |
+
"eval_wer": 0.2305963321000915,
|
| 793 |
+
"step": 4900
|
| 794 |
+
},
|
| 795 |
+
{
|
| 796 |
+
"epoch": 14.492753623188406,
|
| 797 |
+
"grad_norm": 0.48703616857528687,
|
| 798 |
+
"learning_rate": 0.00022614572864321605,
|
| 799 |
+
"loss": 0.2854,
|
| 800 |
+
"step": 5000
|
| 801 |
+
},
|
| 802 |
+
{
|
| 803 |
+
"epoch": 14.492753623188406,
|
| 804 |
+
"eval_loss": 0.22595585882663727,
|
| 805 |
+
"eval_runtime": 148.0737,
|
| 806 |
+
"eval_samples_per_second": 17.586,
|
| 807 |
+
"eval_steps_per_second": 4.396,
|
| 808 |
+
"eval_wer": 0.22604129371046663,
|
| 809 |
+
"step": 5000
|
| 810 |
+
},
|
| 811 |
+
{
|
| 812 |
+
"epoch": 14.782608695652174,
|
| 813 |
+
"grad_norm": 0.5562406182289124,
|
| 814 |
+
"learning_rate": 0.00022463819095477384,
|
| 815 |
+
"loss": 0.2755,
|
| 816 |
+
"step": 5100
|
| 817 |
+
},
|
| 818 |
+
{
|
| 819 |
+
"epoch": 14.782608695652174,
|
| 820 |
+
"eval_loss": 0.22389821708202362,
|
| 821 |
+
"eval_runtime": 147.9028,
|
| 822 |
+
"eval_samples_per_second": 17.606,
|
| 823 |
+
"eval_steps_per_second": 4.402,
|
| 824 |
+
"eval_wer": 0.22576281974778215,
|
| 825 |
+
"step": 5100
|
| 826 |
+
},
|
| 827 |
+
{
|
| 828 |
+
"epoch": 15.072463768115941,
|
| 829 |
+
"grad_norm": 0.5024857521057129,
|
| 830 |
+
"learning_rate": 0.00022313065326633164,
|
| 831 |
+
"loss": 0.2759,
|
| 832 |
+
"step": 5200
|
| 833 |
+
},
|
| 834 |
+
{
|
| 835 |
+
"epoch": 15.072463768115941,
|
| 836 |
+
"eval_loss": 0.2225174903869629,
|
| 837 |
+
"eval_runtime": 148.6759,
|
| 838 |
+
"eval_samples_per_second": 17.515,
|
| 839 |
+
"eval_steps_per_second": 4.379,
|
| 840 |
+
"eval_wer": 0.22478816087838643,
|
| 841 |
+
"step": 5200
|
| 842 |
+
},
|
| 843 |
+
{
|
| 844 |
+
"epoch": 15.36231884057971,
|
| 845 |
+
"grad_norm": 0.49712419509887695,
|
| 846 |
+
"learning_rate": 0.00022162311557788944,
|
| 847 |
+
"loss": 0.2717,
|
| 848 |
+
"step": 5300
|
| 849 |
+
},
|
| 850 |
+
{
|
| 851 |
+
"epoch": 15.36231884057971,
|
| 852 |
+
"eval_loss": 0.22049400210380554,
|
| 853 |
+
"eval_runtime": 148.6086,
|
| 854 |
+
"eval_samples_per_second": 17.523,
|
| 855 |
+
"eval_steps_per_second": 4.381,
|
| 856 |
+
"eval_wer": 0.2202331224887616,
|
| 857 |
+
"step": 5300
|
| 858 |
+
},
|
| 859 |
+
{
|
| 860 |
+
"epoch": 15.652173913043478,
|
| 861 |
+
"grad_norm": 0.5845891237258911,
|
| 862 |
+
"learning_rate": 0.0002201155778894472,
|
| 863 |
+
"loss": 0.2712,
|
| 864 |
+
"step": 5400
|
| 865 |
+
},
|
| 866 |
+
{
|
| 867 |
+
"epoch": 15.652173913043478,
|
| 868 |
+
"eval_loss": 0.21882875263690948,
|
| 869 |
+
"eval_runtime": 148.6863,
|
| 870 |
+
"eval_samples_per_second": 17.513,
|
| 871 |
+
"eval_steps_per_second": 4.378,
|
| 872 |
+
"eval_wer": 0.22198353025420695,
|
| 873 |
+
"step": 5400
|
| 874 |
+
},
|
| 875 |
+
{
|
| 876 |
+
"epoch": 15.942028985507246,
|
| 877 |
+
"grad_norm": 0.464219331741333,
|
| 878 |
+
"learning_rate": 0.000218608040201005,
|
| 879 |
+
"loss": 0.2665,
|
| 880 |
+
"step": 5500
|
| 881 |
+
},
|
| 882 |
+
{
|
| 883 |
+
"epoch": 15.942028985507246,
|
| 884 |
+
"eval_loss": 0.21793150901794434,
|
| 885 |
+
"eval_runtime": 148.1837,
|
| 886 |
+
"eval_samples_per_second": 17.573,
|
| 887 |
+
"eval_steps_per_second": 4.393,
|
| 888 |
+
"eval_wer": 0.22017344949675777,
|
| 889 |
+
"step": 5500
|
| 890 |
+
},
|
| 891 |
+
{
|
| 892 |
+
"epoch": 16.231884057971016,
|
| 893 |
+
"grad_norm": 0.541633665561676,
|
| 894 |
+
"learning_rate": 0.00021711557788944722,
|
| 895 |
+
"loss": 0.2554,
|
| 896 |
+
"step": 5600
|
| 897 |
+
},
|
| 898 |
+
{
|
| 899 |
+
"epoch": 16.231884057971016,
|
| 900 |
+
"eval_loss": 0.22023917734622955,
|
| 901 |
+
"eval_runtime": 148.2405,
|
| 902 |
+
"eval_samples_per_second": 17.566,
|
| 903 |
+
"eval_steps_per_second": 4.392,
|
| 904 |
+
"eval_wer": 0.22130723634483032,
|
| 905 |
+
"step": 5600
|
| 906 |
+
},
|
| 907 |
+
{
|
| 908 |
+
"epoch": 16.52173913043478,
|
| 909 |
+
"grad_norm": 0.5381140112876892,
|
| 910 |
+
"learning_rate": 0.000215608040201005,
|
| 911 |
+
"loss": 0.2558,
|
| 912 |
+
"step": 5700
|
| 913 |
+
},
|
| 914 |
+
{
|
| 915 |
+
"epoch": 16.52173913043478,
|
| 916 |
+
"eval_loss": 0.21436667442321777,
|
| 917 |
+
"eval_runtime": 152.8954,
|
| 918 |
+
"eval_samples_per_second": 17.031,
|
| 919 |
+
"eval_steps_per_second": 4.258,
|
| 920 |
+
"eval_wer": 0.2124159605362613,
|
| 921 |
+
"step": 5700
|
| 922 |
+
},
|
| 923 |
+
{
|
| 924 |
+
"epoch": 16.81159420289855,
|
| 925 |
+
"grad_norm": 0.5217423439025879,
|
| 926 |
+
"learning_rate": 0.0002141005025125628,
|
| 927 |
+
"loss": 0.2611,
|
| 928 |
+
"step": 5800
|
| 929 |
+
},
|
| 930 |
+
{
|
| 931 |
+
"epoch": 16.81159420289855,
|
| 932 |
+
"eval_loss": 0.21614325046539307,
|
| 933 |
+
"eval_runtime": 147.2916,
|
| 934 |
+
"eval_samples_per_second": 17.679,
|
| 935 |
+
"eval_steps_per_second": 4.42,
|
| 936 |
+
"eval_wer": 0.21643394199785176,
|
| 937 |
+
"step": 5800
|
| 938 |
+
},
|
| 939 |
+
{
|
| 940 |
+
"epoch": 17.10144927536232,
|
| 941 |
+
"grad_norm": 0.5291975140571594,
|
| 942 |
+
"learning_rate": 0.00021259296482412058,
|
| 943 |
+
"loss": 0.261,
|
| 944 |
+
"step": 5900
|
| 945 |
+
},
|
| 946 |
+
{
|
| 947 |
+
"epoch": 17.10144927536232,
|
| 948 |
+
"eval_loss": 0.21743394434452057,
|
| 949 |
+
"eval_runtime": 148.5519,
|
| 950 |
+
"eval_samples_per_second": 17.529,
|
| 951 |
+
"eval_steps_per_second": 4.382,
|
| 952 |
+
"eval_wer": 0.2099693678641047,
|
| 953 |
+
"step": 5900
|
| 954 |
+
},
|
| 955 |
+
{
|
| 956 |
+
"epoch": 17.391304347826086,
|
| 957 |
+
"grad_norm": 0.4251558780670166,
|
| 958 |
+
"learning_rate": 0.00021108542713567838,
|
| 959 |
+
"loss": 0.2503,
|
| 960 |
+
"step": 6000
|
| 961 |
+
},
|
| 962 |
+
{
|
| 963 |
+
"epoch": 17.391304347826086,
|
| 964 |
+
"eval_loss": 0.2164338082075119,
|
| 965 |
+
"eval_runtime": 148.7374,
|
| 966 |
+
"eval_samples_per_second": 17.507,
|
| 967 |
+
"eval_steps_per_second": 4.377,
|
| 968 |
+
"eval_wer": 0.2117993396188885,
|
| 969 |
+
"step": 6000
|
| 970 |
+
},
|
| 971 |
+
{
|
| 972 |
+
"epoch": 17.681159420289855,
|
| 973 |
+
"grad_norm": 0.48946651816368103,
|
| 974 |
+
"learning_rate": 0.00020957788944723615,
|
| 975 |
+
"loss": 0.2533,
|
| 976 |
+
"step": 6100
|
| 977 |
+
},
|
| 978 |
+
{
|
| 979 |
+
"epoch": 17.681159420289855,
|
| 980 |
+
"eval_loss": 0.2096243053674698,
|
| 981 |
+
"eval_runtime": 148.6793,
|
| 982 |
+
"eval_samples_per_second": 17.514,
|
| 983 |
+
"eval_steps_per_second": 4.379,
|
| 984 |
+
"eval_wer": 0.20632931535187174,
|
| 985 |
+
"step": 6100
|
| 986 |
+
},
|
| 987 |
+
{
|
| 988 |
+
"epoch": 17.971014492753625,
|
| 989 |
+
"grad_norm": 1.0431879758834839,
|
| 990 |
+
"learning_rate": 0.00020807035175879395,
|
| 991 |
+
"loss": 0.2491,
|
| 992 |
+
"step": 6200
|
| 993 |
+
},
|
| 994 |
+
{
|
| 995 |
+
"epoch": 17.971014492753625,
|
| 996 |
+
"eval_loss": 0.21442431211471558,
|
| 997 |
+
"eval_runtime": 148.4753,
|
| 998 |
+
"eval_samples_per_second": 17.538,
|
| 999 |
+
"eval_steps_per_second": 4.385,
|
| 1000 |
+
"eval_wer": 0.2036838127063691,
|
| 1001 |
+
"step": 6200
|
| 1002 |
+
},
|
| 1003 |
+
{
|
| 1004 |
+
"epoch": 18.26086956521739,
|
| 1005 |
+
"grad_norm": 0.6736889481544495,
|
| 1006 |
+
"learning_rate": 0.00020656281407035174,
|
| 1007 |
+
"loss": 0.2448,
|
| 1008 |
+
"step": 6300
|
| 1009 |
+
},
|
| 1010 |
+
{
|
| 1011 |
+
"epoch": 18.26086956521739,
|
| 1012 |
+
"eval_loss": 0.21374870836734772,
|
| 1013 |
+
"eval_runtime": 148.8537,
|
| 1014 |
+
"eval_samples_per_second": 17.494,
|
| 1015 |
+
"eval_steps_per_second": 4.373,
|
| 1016 |
+
"eval_wer": 0.20344512073835383,
|
| 1017 |
+
"step": 6300
|
| 1018 |
+
},
|
| 1019 |
+
{
|
| 1020 |
+
"epoch": 18.55072463768116,
|
| 1021 |
+
"grad_norm": 0.6890196204185486,
|
| 1022 |
+
"learning_rate": 0.00020505527638190954,
|
| 1023 |
+
"loss": 0.2418,
|
| 1024 |
+
"step": 6400
|
| 1025 |
+
},
|
| 1026 |
+
{
|
| 1027 |
+
"epoch": 18.55072463768116,
|
| 1028 |
+
"eval_loss": 0.20951078832149506,
|
| 1029 |
+
"eval_runtime": 148.479,
|
| 1030 |
+
"eval_samples_per_second": 17.538,
|
| 1031 |
+
"eval_steps_per_second": 4.384,
|
| 1032 |
+
"eval_wer": 0.20284839081831563,
|
| 1033 |
+
"step": 6400
|
| 1034 |
+
},
|
| 1035 |
+
{
|
| 1036 |
+
"epoch": 18.840579710144926,
|
| 1037 |
+
"grad_norm": 0.5898504853248596,
|
| 1038 |
+
"learning_rate": 0.0002035477386934673,
|
| 1039 |
+
"loss": 0.246,
|
| 1040 |
+
"step": 6500
|
| 1041 |
+
},
|
| 1042 |
+
{
|
| 1043 |
+
"epoch": 18.840579710144926,
|
| 1044 |
+
"eval_loss": 0.20731888711452484,
|
| 1045 |
+
"eval_runtime": 149.6587,
|
| 1046 |
+
"eval_samples_per_second": 17.4,
|
| 1047 |
+
"eval_steps_per_second": 4.35,
|
| 1048 |
+
"eval_wer": 0.19853204439670605,
|
| 1049 |
+
"step": 6500
|
| 1050 |
+
},
|
| 1051 |
+
{
|
| 1052 |
+
"epoch": 19.130434782608695,
|
| 1053 |
+
"grad_norm": 0.721211314201355,
|
| 1054 |
+
"learning_rate": 0.0002020402010050251,
|
| 1055 |
+
"loss": 0.2389,
|
| 1056 |
+
"step": 6600
|
| 1057 |
+
},
|
| 1058 |
+
{
|
| 1059 |
+
"epoch": 19.130434782608695,
|
| 1060 |
+
"eval_loss": 0.21186397969722748,
|
| 1061 |
+
"eval_runtime": 149.3428,
|
| 1062 |
+
"eval_samples_per_second": 17.436,
|
| 1063 |
+
"eval_steps_per_second": 4.359,
|
| 1064 |
+
"eval_wer": 0.19763694951664876,
|
| 1065 |
+
"step": 6600
|
| 1066 |
+
},
|
| 1067 |
+
{
|
| 1068 |
+
"epoch": 19.420289855072465,
|
| 1069 |
+
"grad_norm": 0.9309697151184082,
|
| 1070 |
+
"learning_rate": 0.00020053266331658288,
|
| 1071 |
+
"loss": 0.2332,
|
| 1072 |
+
"step": 6700
|
| 1073 |
+
},
|
| 1074 |
+
{
|
| 1075 |
+
"epoch": 19.420289855072465,
|
| 1076 |
+
"eval_loss": 0.2146858423948288,
|
| 1077 |
+
"eval_runtime": 149.4385,
|
| 1078 |
+
"eval_samples_per_second": 17.425,
|
| 1079 |
+
"eval_steps_per_second": 4.356,
|
| 1080 |
+
"eval_wer": 0.1992680112980865,
|
| 1081 |
+
"step": 6700
|
| 1082 |
+
},
|
| 1083 |
+
{
|
| 1084 |
+
"epoch": 19.71014492753623,
|
| 1085 |
+
"grad_norm": 0.6736502051353455,
|
| 1086 |
+
"learning_rate": 0.0001990251256281407,
|
| 1087 |
+
"loss": 0.2364,
|
| 1088 |
+
"step": 6800
|
| 1089 |
+
},
|
| 1090 |
+
{
|
| 1091 |
+
"epoch": 19.71014492753623,
|
| 1092 |
+
"eval_loss": 0.2084682285785675,
|
| 1093 |
+
"eval_runtime": 149.2563,
|
| 1094 |
+
"eval_samples_per_second": 17.447,
|
| 1095 |
+
"eval_steps_per_second": 4.362,
|
| 1096 |
+
"eval_wer": 0.19831324342602538,
|
| 1097 |
+
"step": 6800
|
| 1098 |
+
},
|
| 1099 |
+
{
|
| 1100 |
+
"epoch": 20.0,
|
| 1101 |
+
"grad_norm": 1.5724033117294312,
|
| 1102 |
+
"learning_rate": 0.00019751758793969847,
|
| 1103 |
+
"loss": 0.244,
|
| 1104 |
+
"step": 6900
|
| 1105 |
+
},
|
| 1106 |
+
{
|
| 1107 |
+
"epoch": 20.0,
|
| 1108 |
+
"eval_loss": 0.21120484173297882,
|
| 1109 |
+
"eval_runtime": 149.1916,
|
| 1110 |
+
"eval_samples_per_second": 17.454,
|
| 1111 |
+
"eval_steps_per_second": 4.364,
|
| 1112 |
+
"eval_wer": 0.1955086128018459,
|
| 1113 |
+
"step": 6900
|
| 1114 |
+
},
|
| 1115 |
+
{
|
| 1116 |
+
"epoch": 20.28985507246377,
|
| 1117 |
+
"grad_norm": 1.5974398851394653,
|
| 1118 |
+
"learning_rate": 0.00019601005025125627,
|
| 1119 |
+
"loss": 0.231,
|
| 1120 |
+
"step": 7000
|
| 1121 |
+
},
|
| 1122 |
+
{
|
| 1123 |
+
"epoch": 20.28985507246377,
|
| 1124 |
+
"eval_loss": 0.2139531373977661,
|
| 1125 |
+
"eval_runtime": 149.1228,
|
| 1126 |
+
"eval_samples_per_second": 17.462,
|
| 1127 |
+
"eval_steps_per_second": 4.366,
|
| 1128 |
+
"eval_wer": 0.19568763177785733,
|
| 1129 |
+
"step": 7000
|
| 1130 |
+
},
|
| 1131 |
+
{
|
| 1132 |
+
"epoch": 20.579710144927535,
|
| 1133 |
+
"grad_norm": 1.7866404056549072,
|
| 1134 |
+
"learning_rate": 0.00019450251256281404,
|
| 1135 |
+
"loss": 0.2321,
|
| 1136 |
+
"step": 7100
|
| 1137 |
+
},
|
| 1138 |
+
{
|
| 1139 |
+
"epoch": 20.579710144927535,
|
| 1140 |
+
"eval_loss": 0.21654276549816132,
|
| 1141 |
+
"eval_runtime": 148.6885,
|
| 1142 |
+
"eval_samples_per_second": 17.513,
|
| 1143 |
+
"eval_steps_per_second": 4.378,
|
| 1144 |
+
"eval_wer": 0.19644348967657238,
|
| 1145 |
+
"step": 7100
|
| 1146 |
+
},
|
| 1147 |
+
{
|
| 1148 |
+
"epoch": 20.869565217391305,
|
| 1149 |
+
"grad_norm": 1.2823141813278198,
|
| 1150 |
+
"learning_rate": 0.00019299497487437184,
|
| 1151 |
+
"loss": 0.2288,
|
| 1152 |
+
"step": 7200
|
| 1153 |
+
},
|
| 1154 |
+
{
|
| 1155 |
+
"epoch": 20.869565217391305,
|
| 1156 |
+
"eval_loss": 0.2096690535545349,
|
| 1157 |
+
"eval_runtime": 149.16,
|
| 1158 |
+
"eval_samples_per_second": 17.458,
|
| 1159 |
+
"eval_steps_per_second": 4.364,
|
| 1160 |
+
"eval_wer": 0.1921271432549628,
|
| 1161 |
+
"step": 7200
|
| 1162 |
+
},
|
| 1163 |
+
{
|
| 1164 |
+
"epoch": 21.159420289855074,
|
| 1165 |
+
"grad_norm": 0.3959142863750458,
|
| 1166 |
+
"learning_rate": 0.0001914874371859296,
|
| 1167 |
+
"loss": 0.2373,
|
| 1168 |
+
"step": 7300
|
| 1169 |
+
},
|
| 1170 |
+
{
|
| 1171 |
+
"epoch": 21.159420289855074,
|
| 1172 |
+
"eval_loss": 0.20639722049236298,
|
| 1173 |
+
"eval_runtime": 148.8351,
|
| 1174 |
+
"eval_samples_per_second": 17.496,
|
| 1175 |
+
"eval_steps_per_second": 4.374,
|
| 1176 |
+
"eval_wer": 0.1895810955961332,
|
| 1177 |
+
"step": 7300
|
| 1178 |
+
},
|
| 1179 |
+
{
|
| 1180 |
+
"epoch": 21.44927536231884,
|
| 1181 |
+
"grad_norm": 0.4794645309448242,
|
| 1182 |
+
"learning_rate": 0.00018997989949748743,
|
| 1183 |
+
"loss": 0.222,
|
| 1184 |
+
"step": 7400
|
| 1185 |
+
},
|
| 1186 |
+
{
|
| 1187 |
+
"epoch": 21.44927536231884,
|
| 1188 |
+
"eval_loss": 0.20799419283866882,
|
| 1189 |
+
"eval_runtime": 149.7147,
|
| 1190 |
+
"eval_samples_per_second": 17.393,
|
| 1191 |
+
"eval_steps_per_second": 4.348,
|
| 1192 |
+
"eval_wer": 0.18979989656681387,
|
| 1193 |
+
"step": 7400
|
| 1194 |
+
},
|
| 1195 |
+
{
|
| 1196 |
+
"epoch": 21.73913043478261,
|
| 1197 |
+
"grad_norm": 0.42001789808273315,
|
| 1198 |
+
"learning_rate": 0.0001884723618090452,
|
| 1199 |
+
"loss": 0.2254,
|
| 1200 |
+
"step": 7500
|
| 1201 |
+
},
|
| 1202 |
+
{
|
| 1203 |
+
"epoch": 21.73913043478261,
|
| 1204 |
+
"eval_loss": 0.21223315596580505,
|
| 1205 |
+
"eval_runtime": 148.6287,
|
| 1206 |
+
"eval_samples_per_second": 17.52,
|
| 1207 |
+
"eval_steps_per_second": 4.38,
|
| 1208 |
+
"eval_wer": 0.19033695349484822,
|
| 1209 |
+
"step": 7500
|
| 1210 |
+
},
|
| 1211 |
+
{
|
| 1212 |
+
"epoch": 22.028985507246375,
|
| 1213 |
+
"grad_norm": 0.5453273057937622,
|
| 1214 |
+
"learning_rate": 0.000186964824120603,
|
| 1215 |
+
"loss": 0.2271,
|
| 1216 |
+
"step": 7600
|
| 1217 |
+
},
|
| 1218 |
+
{
|
| 1219 |
+
"epoch": 22.028985507246375,
|
| 1220 |
+
"eval_loss": 0.20491968095302582,
|
| 1221 |
+
"eval_runtime": 153.1957,
|
| 1222 |
+
"eval_samples_per_second": 16.998,
|
| 1223 |
+
"eval_steps_per_second": 4.249,
|
| 1224 |
+
"eval_wer": 0.1892628396387795,
|
| 1225 |
+
"step": 7600
|
| 1226 |
+
},
|
| 1227 |
+
{
|
| 1228 |
+
"epoch": 22.318840579710145,
|
| 1229 |
+
"grad_norm": 0.42996543645858765,
|
| 1230 |
+
"learning_rate": 0.00018545728643216077,
|
| 1231 |
+
"loss": 0.2232,
|
| 1232 |
+
"step": 7700
|
| 1233 |
+
},
|
| 1234 |
+
{
|
| 1235 |
+
"epoch": 22.318840579710145,
|
| 1236 |
+
"eval_loss": 0.20111997425556183,
|
| 1237 |
+
"eval_runtime": 149.8154,
|
| 1238 |
+
"eval_samples_per_second": 17.381,
|
| 1239 |
+
"eval_steps_per_second": 4.345,
|
| 1240 |
+
"eval_wer": 0.18699526594263435,
|
| 1241 |
+
"step": 7700
|
| 1242 |
+
},
|
| 1243 |
+
{
|
| 1244 |
+
"epoch": 22.608695652173914,
|
| 1245 |
+
"grad_norm": 0.559751033782959,
|
| 1246 |
+
"learning_rate": 0.00018394974874371857,
|
| 1247 |
+
"loss": 0.217,
|
| 1248 |
+
"step": 7800
|
| 1249 |
+
},
|
| 1250 |
+
{
|
| 1251 |
+
"epoch": 22.608695652173914,
|
| 1252 |
+
"eval_loss": 0.20821848511695862,
|
| 1253 |
+
"eval_runtime": 149.8048,
|
| 1254 |
+
"eval_samples_per_second": 17.383,
|
| 1255 |
+
"eval_steps_per_second": 4.346,
|
| 1256 |
+
"eval_wer": 0.1867963559692883,
|
| 1257 |
+
"step": 7800
|
| 1258 |
+
},
|
| 1259 |
+
{
|
| 1260 |
+
"epoch": 22.89855072463768,
|
| 1261 |
+
"grad_norm": 0.5080951452255249,
|
| 1262 |
+
"learning_rate": 0.00018245728643216078,
|
| 1263 |
+
"loss": 0.2279,
|
| 1264 |
+
"step": 7900
|
| 1265 |
+
},
|
| 1266 |
+
{
|
| 1267 |
+
"epoch": 22.89855072463768,
|
| 1268 |
+
"eval_loss": 0.20747588574886322,
|
| 1269 |
+
"eval_runtime": 148.8562,
|
| 1270 |
+
"eval_samples_per_second": 17.493,
|
| 1271 |
+
"eval_steps_per_second": 4.373,
|
| 1272 |
+
"eval_wer": 0.188327962764053,
|
| 1273 |
+
"step": 7900
|
| 1274 |
+
},
|
| 1275 |
+
{
|
| 1276 |
+
"epoch": 23.18840579710145,
|
| 1277 |
+
"grad_norm": 0.5432471036911011,
|
| 1278 |
+
"learning_rate": 0.0001809497487437186,
|
| 1279 |
+
"loss": 0.2227,
|
| 1280 |
+
"step": 8000
|
| 1281 |
+
},
|
| 1282 |
+
{
|
| 1283 |
+
"epoch": 23.18840579710145,
|
| 1284 |
+
"eval_loss": 0.20593281090259552,
|
| 1285 |
+
"eval_runtime": 148.78,
|
| 1286 |
+
"eval_samples_per_second": 17.502,
|
| 1287 |
+
"eval_steps_per_second": 4.376,
|
| 1288 |
+
"eval_wer": 0.18407128933444722,
|
| 1289 |
+
"step": 8000
|
| 1290 |
+
},
|
| 1291 |
+
{
|
| 1292 |
+
"epoch": 23.47826086956522,
|
| 1293 |
+
"grad_norm": 0.45922884345054626,
|
| 1294 |
+
"learning_rate": 0.00017944221105527637,
|
| 1295 |
+
"loss": 0.2123,
|
| 1296 |
+
"step": 8100
|
| 1297 |
+
},
|
| 1298 |
+
{
|
| 1299 |
+
"epoch": 23.47826086956522,
|
| 1300 |
+
"eval_loss": 0.2026965320110321,
|
| 1301 |
+
"eval_runtime": 148.8956,
|
| 1302 |
+
"eval_samples_per_second": 17.489,
|
| 1303 |
+
"eval_steps_per_second": 4.372,
|
| 1304 |
+
"eval_wer": 0.18202251660898278,
|
| 1305 |
+
"step": 8100
|
| 1306 |
+
},
|
| 1307 |
+
{
|
| 1308 |
+
"epoch": 23.768115942028984,
|
| 1309 |
+
"grad_norm": 0.5219236612319946,
|
| 1310 |
+
"learning_rate": 0.00017793467336683417,
|
| 1311 |
+
"loss": 0.2173,
|
| 1312 |
+
"step": 8200
|
| 1313 |
+
},
|
| 1314 |
+
{
|
| 1315 |
+
"epoch": 23.768115942028984,
|
| 1316 |
+
"eval_loss": 0.20592840015888214,
|
| 1317 |
+
"eval_runtime": 149.7695,
|
| 1318 |
+
"eval_samples_per_second": 17.387,
|
| 1319 |
+
"eval_steps_per_second": 4.347,
|
| 1320 |
+
"eval_wer": 0.18204240760631737,
|
| 1321 |
+
"step": 8200
|
| 1322 |
+
},
|
| 1323 |
+
{
|
| 1324 |
+
"epoch": 24.057971014492754,
|
| 1325 |
+
"grad_norm": 0.49628207087516785,
|
| 1326 |
+
"learning_rate": 0.00017642713567839194,
|
| 1327 |
+
"loss": 0.2134,
|
| 1328 |
+
"step": 8300
|
| 1329 |
+
},
|
| 1330 |
+
{
|
| 1331 |
+
"epoch": 24.057971014492754,
|
| 1332 |
+
"eval_loss": 0.2052663117647171,
|
| 1333 |
+
"eval_runtime": 149.7801,
|
| 1334 |
+
"eval_samples_per_second": 17.385,
|
| 1335 |
+
"eval_steps_per_second": 4.346,
|
| 1336 |
+
"eval_wer": 0.18166447865695987,
|
| 1337 |
+
"step": 8300
|
| 1338 |
+
},
|
| 1339 |
+
{
|
| 1340 |
+
"epoch": 24.347826086956523,
|
| 1341 |
+
"grad_norm": 0.521787166595459,
|
| 1342 |
+
"learning_rate": 0.0001749195979899497,
|
| 1343 |
+
"loss": 0.2092,
|
| 1344 |
+
"step": 8400
|
| 1345 |
+
},
|
| 1346 |
+
{
|
| 1347 |
+
"epoch": 24.347826086956523,
|
| 1348 |
+
"eval_loss": 0.2022761106491089,
|
| 1349 |
+
"eval_runtime": 149.3054,
|
| 1350 |
+
"eval_samples_per_second": 17.441,
|
| 1351 |
+
"eval_steps_per_second": 4.36,
|
| 1352 |
+
"eval_wer": 0.17699009428332738,
|
| 1353 |
+
"step": 8400
|
| 1354 |
+
},
|
| 1355 |
+
{
|
| 1356 |
+
"epoch": 24.63768115942029,
|
| 1357 |
+
"grad_norm": 0.4815770387649536,
|
| 1358 |
+
"learning_rate": 0.00017341206030150753,
|
| 1359 |
+
"loss": 0.2097,
|
| 1360 |
+
"step": 8500
|
| 1361 |
+
},
|
| 1362 |
+
{
|
| 1363 |
+
"epoch": 24.63768115942029,
|
| 1364 |
+
"eval_loss": 0.20263001322746277,
|
| 1365 |
+
"eval_runtime": 150.3389,
|
| 1366 |
+
"eval_samples_per_second": 17.321,
|
| 1367 |
+
"eval_steps_per_second": 4.33,
|
| 1368 |
+
"eval_wer": 0.18001352587818753,
|
| 1369 |
+
"step": 8500
|
| 1370 |
+
},
|
| 1371 |
+
{
|
| 1372 |
+
"epoch": 24.92753623188406,
|
| 1373 |
+
"grad_norm": 0.4982714056968689,
|
| 1374 |
+
"learning_rate": 0.00017190452261306533,
|
| 1375 |
+
"loss": 0.2199,
|
| 1376 |
+
"step": 8600
|
| 1377 |
+
},
|
| 1378 |
+
{
|
| 1379 |
+
"epoch": 24.92753623188406,
|
| 1380 |
+
"eval_loss": 0.20027528703212738,
|
| 1381 |
+
"eval_runtime": 149.586,
|
| 1382 |
+
"eval_samples_per_second": 17.408,
|
| 1383 |
+
"eval_steps_per_second": 4.352,
|
| 1384 |
+
"eval_wer": 0.1743048096431555,
|
| 1385 |
+
"step": 8600
|
| 1386 |
+
},
|
| 1387 |
+
{
|
| 1388 |
+
"epoch": 25.217391304347824,
|
| 1389 |
+
"grad_norm": 0.5470702648162842,
|
| 1390 |
+
"learning_rate": 0.0001703969849246231,
|
| 1391 |
+
"loss": 0.2059,
|
| 1392 |
+
"step": 8700
|
| 1393 |
+
},
|
| 1394 |
+
{
|
| 1395 |
+
"epoch": 25.217391304347824,
|
| 1396 |
+
"eval_loss": 0.20195870101451874,
|
| 1397 |
+
"eval_runtime": 150.6367,
|
| 1398 |
+
"eval_samples_per_second": 17.287,
|
| 1399 |
+
"eval_steps_per_second": 4.322,
|
| 1400 |
+
"eval_wer": 0.17754704220869635,
|
| 1401 |
+
"step": 8700
|
| 1402 |
+
},
|
| 1403 |
+
{
|
| 1404 |
+
"epoch": 25.507246376811594,
|
| 1405 |
+
"grad_norm": 0.4693375825881958,
|
| 1406 |
+
"learning_rate": 0.0001688894472361809,
|
| 1407 |
+
"loss": 0.2058,
|
| 1408 |
+
"step": 8800
|
| 1409 |
+
},
|
| 1410 |
+
{
|
| 1411 |
+
"epoch": 25.507246376811594,
|
| 1412 |
+
"eval_loss": 0.19962410628795624,
|
| 1413 |
+
"eval_runtime": 149.5537,
|
| 1414 |
+
"eval_samples_per_second": 17.412,
|
| 1415 |
+
"eval_steps_per_second": 4.353,
|
| 1416 |
+
"eval_wer": 0.17627401837928153,
|
| 1417 |
+
"step": 8800
|
| 1418 |
+
},
|
| 1419 |
+
{
|
| 1420 |
+
"epoch": 25.797101449275363,
|
| 1421 |
+
"grad_norm": 0.6513779759407043,
|
| 1422 |
+
"learning_rate": 0.00016738190954773867,
|
| 1423 |
+
"loss": 0.2048,
|
| 1424 |
+
"step": 8900
|
| 1425 |
+
},
|
| 1426 |
+
{
|
| 1427 |
+
"epoch": 25.797101449275363,
|
| 1428 |
+
"eval_loss": 0.19703231751918793,
|
| 1429 |
+
"eval_runtime": 150.269,
|
| 1430 |
+
"eval_samples_per_second": 17.329,
|
| 1431 |
+
"eval_steps_per_second": 4.332,
|
| 1432 |
+
"eval_wer": 0.17792497115805386,
|
| 1433 |
+
"step": 8900
|
| 1434 |
+
},
|
| 1435 |
+
{
|
| 1436 |
+
"epoch": 26.08695652173913,
|
| 1437 |
+
"grad_norm": 0.6113376617431641,
|
| 1438 |
+
"learning_rate": 0.0001658743718592965,
|
| 1439 |
+
"loss": 0.207,
|
| 1440 |
+
"step": 9000
|
| 1441 |
+
},
|
| 1442 |
+
{
|
| 1443 |
+
"epoch": 26.08695652173913,
|
| 1444 |
+
"eval_loss": 0.19729487597942352,
|
| 1445 |
+
"eval_runtime": 149.9591,
|
| 1446 |
+
"eval_samples_per_second": 17.365,
|
| 1447 |
+
"eval_steps_per_second": 4.341,
|
| 1448 |
+
"eval_wer": 0.1733301507737598,
|
| 1449 |
+
"step": 9000
|
| 1450 |
+
},
|
| 1451 |
+
{
|
| 1452 |
+
"epoch": 26.3768115942029,
|
| 1453 |
+
"grad_norm": 0.4407137930393219,
|
| 1454 |
+
"learning_rate": 0.00016436683417085426,
|
| 1455 |
+
"loss": 0.2024,
|
| 1456 |
+
"step": 9100
|
| 1457 |
+
},
|
| 1458 |
+
{
|
| 1459 |
+
"epoch": 26.3768115942029,
|
| 1460 |
+
"eval_loss": 0.19801633059978485,
|
| 1461 |
+
"eval_runtime": 150.9313,
|
| 1462 |
+
"eval_samples_per_second": 17.253,
|
| 1463 |
+
"eval_steps_per_second": 4.313,
|
| 1464 |
+
"eval_wer": 0.17506066754187055,
|
| 1465 |
+
"step": 9100
|
| 1466 |
+
},
|
| 1467 |
+
{
|
| 1468 |
+
"epoch": 26.666666666666668,
|
| 1469 |
+
"grad_norm": 0.5502184629440308,
|
| 1470 |
+
"learning_rate": 0.00016285929648241206,
|
| 1471 |
+
"loss": 0.2049,
|
| 1472 |
+
"step": 9200
|
| 1473 |
+
},
|
| 1474 |
+
{
|
| 1475 |
+
"epoch": 26.666666666666668,
|
| 1476 |
+
"eval_loss": 0.19622650742530823,
|
| 1477 |
+
"eval_runtime": 150.3185,
|
| 1478 |
+
"eval_samples_per_second": 17.323,
|
| 1479 |
+
"eval_steps_per_second": 4.331,
|
| 1480 |
+
"eval_wer": 0.17187810796833353,
|
| 1481 |
+
"step": 9200
|
| 1482 |
+
},
|
| 1483 |
+
{
|
| 1484 |
+
"epoch": 26.956521739130434,
|
| 1485 |
+
"grad_norm": 0.5984034538269043,
|
| 1486 |
+
"learning_rate": 0.00016135175879396983,
|
| 1487 |
+
"loss": 0.1991,
|
| 1488 |
+
"step": 9300
|
| 1489 |
+
},
|
| 1490 |
+
{
|
| 1491 |
+
"epoch": 26.956521739130434,
|
| 1492 |
+
"eval_loss": 0.19447383284568787,
|
| 1493 |
+
"eval_runtime": 149.7796,
|
| 1494 |
+
"eval_samples_per_second": 17.386,
|
| 1495 |
+
"eval_steps_per_second": 4.346,
|
| 1496 |
+
"eval_wer": 0.17193778096033735,
|
| 1497 |
+
"step": 9300
|
| 1498 |
+
},
|
| 1499 |
+
{
|
| 1500 |
+
"epoch": 27.246376811594203,
|
| 1501 |
+
"grad_norm": 0.556126594543457,
|
| 1502 |
+
"learning_rate": 0.00015984422110552763,
|
| 1503 |
+
"loss": 0.209,
|
| 1504 |
+
"step": 9400
|
| 1505 |
+
},
|
| 1506 |
+
{
|
| 1507 |
+
"epoch": 27.246376811594203,
|
| 1508 |
+
"eval_loss": 0.19995641708374023,
|
| 1509 |
+
"eval_runtime": 149.7895,
|
| 1510 |
+
"eval_samples_per_second": 17.384,
|
| 1511 |
+
"eval_steps_per_second": 4.346,
|
| 1512 |
+
"eval_wer": 0.1742849186458209,
|
| 1513 |
+
"step": 9400
|
| 1514 |
+
},
|
| 1515 |
+
{
|
| 1516 |
+
"epoch": 27.536231884057973,
|
| 1517 |
+
"grad_norm": 0.5679728984832764,
|
| 1518 |
+
"learning_rate": 0.0001583366834170854,
|
| 1519 |
+
"loss": 0.1993,
|
| 1520 |
+
"step": 9500
|
| 1521 |
+
},
|
| 1522 |
+
{
|
| 1523 |
+
"epoch": 27.536231884057973,
|
| 1524 |
+
"eval_loss": 0.19617217779159546,
|
| 1525 |
+
"eval_runtime": 150.0128,
|
| 1526 |
+
"eval_samples_per_second": 17.359,
|
| 1527 |
+
"eval_steps_per_second": 4.34,
|
| 1528 |
+
"eval_wer": 0.17446393762183235,
|
| 1529 |
+
"step": 9500
|
| 1530 |
+
},
|
| 1531 |
+
{
|
| 1532 |
+
"epoch": 27.82608695652174,
|
| 1533 |
+
"grad_norm": 0.5760718584060669,
|
| 1534 |
+
"learning_rate": 0.00015682914572864322,
|
| 1535 |
+
"loss": 0.1961,
|
| 1536 |
+
"step": 9600
|
| 1537 |
+
},
|
| 1538 |
+
{
|
| 1539 |
+
"epoch": 27.82608695652174,
|
| 1540 |
+
"eval_loss": 0.2008921355009079,
|
| 1541 |
+
"eval_runtime": 149.6147,
|
| 1542 |
+
"eval_samples_per_second": 17.405,
|
| 1543 |
+
"eval_steps_per_second": 4.351,
|
| 1544 |
+
"eval_wer": 0.17265385686438317,
|
| 1545 |
+
"step": 9600
|
| 1546 |
+
},
|
| 1547 |
+
{
|
| 1548 |
+
"epoch": 28.115942028985508,
|
| 1549 |
+
"grad_norm": 0.4592433571815491,
|
| 1550 |
+
"learning_rate": 0.000155321608040201,
|
| 1551 |
+
"loss": 0.2005,
|
| 1552 |
+
"step": 9700
|
| 1553 |
+
},
|
| 1554 |
+
{
|
| 1555 |
+
"epoch": 28.115942028985508,
|
| 1556 |
+
"eval_loss": 0.19731920957565308,
|
| 1557 |
+
"eval_runtime": 150.014,
|
| 1558 |
+
"eval_samples_per_second": 17.358,
|
| 1559 |
+
"eval_steps_per_second": 4.34,
|
| 1560 |
+
"eval_wer": 0.17235549190436408,
|
| 1561 |
+
"step": 9700
|
| 1562 |
+
},
|
| 1563 |
+
{
|
| 1564 |
+
"epoch": 28.405797101449274,
|
| 1565 |
+
"grad_norm": 0.613828718662262,
|
| 1566 |
+
"learning_rate": 0.0001538140703517588,
|
| 1567 |
+
"loss": 0.1959,
|
| 1568 |
+
"step": 9800
|
| 1569 |
+
},
|
| 1570 |
+
{
|
| 1571 |
+
"epoch": 28.405797101449274,
|
| 1572 |
+
"eval_loss": 0.19310329854488373,
|
| 1573 |
+
"eval_runtime": 150.0935,
|
| 1574 |
+
"eval_samples_per_second": 17.349,
|
| 1575 |
+
"eval_steps_per_second": 4.337,
|
| 1576 |
+
"eval_wer": 0.16636830170664757,
|
| 1577 |
+
"step": 9800
|
| 1578 |
+
},
|
| 1579 |
+
{
|
| 1580 |
+
"epoch": 28.695652173913043,
|
| 1581 |
+
"grad_norm": 0.42406728863716125,
|
| 1582 |
+
"learning_rate": 0.00015230653266331656,
|
| 1583 |
+
"loss": 0.1976,
|
| 1584 |
+
"step": 9900
|
| 1585 |
+
},
|
| 1586 |
+
{
|
| 1587 |
+
"epoch": 28.695652173913043,
|
| 1588 |
+
"eval_loss": 0.20253592729568481,
|
| 1589 |
+
"eval_runtime": 150.3535,
|
| 1590 |
+
"eval_samples_per_second": 17.319,
|
| 1591 |
+
"eval_steps_per_second": 4.33,
|
| 1592 |
+
"eval_wer": 0.17028682818156501,
|
| 1593 |
+
"step": 9900
|
| 1594 |
+
},
|
| 1595 |
+
{
|
| 1596 |
+
"epoch": 28.985507246376812,
|
| 1597 |
+
"grad_norm": 0.6821871399879456,
|
| 1598 |
+
"learning_rate": 0.00015081407035175877,
|
| 1599 |
+
"loss": 0.1918,
|
| 1600 |
+
"step": 10000
|
| 1601 |
+
},
|
| 1602 |
+
{
|
| 1603 |
+
"epoch": 28.985507246376812,
|
| 1604 |
+
"eval_loss": 0.1978112906217575,
|
| 1605 |
+
"eval_runtime": 150.1656,
|
| 1606 |
+
"eval_samples_per_second": 17.341,
|
| 1607 |
+
"eval_steps_per_second": 4.335,
|
| 1608 |
+
"eval_wer": 0.16933206030950393,
|
| 1609 |
+
"step": 10000
|
| 1610 |
+
},
|
| 1611 |
+
{
|
| 1612 |
+
"epoch": 29.27536231884058,
|
| 1613 |
+
"grad_norm": 0.733881413936615,
|
| 1614 |
+
"learning_rate": 0.00014930653266331657,
|
| 1615 |
+
"loss": 0.1932,
|
| 1616 |
+
"step": 10100
|
| 1617 |
+
},
|
| 1618 |
+
{
|
| 1619 |
+
"epoch": 29.27536231884058,
|
| 1620 |
+
"eval_loss": 0.19511191546916962,
|
| 1621 |
+
"eval_runtime": 150.0272,
|
| 1622 |
+
"eval_samples_per_second": 17.357,
|
| 1623 |
+
"eval_steps_per_second": 4.339,
|
| 1624 |
+
"eval_wer": 0.16714405060269721,
|
| 1625 |
+
"step": 10100
|
| 1626 |
+
},
|
| 1627 |
+
{
|
| 1628 |
+
"epoch": 29.565217391304348,
|
| 1629 |
+
"grad_norm": 0.6013245582580566,
|
| 1630 |
+
"learning_rate": 0.00014779899497487437,
|
| 1631 |
+
"loss": 0.194,
|
| 1632 |
+
"step": 10200
|
| 1633 |
+
},
|
| 1634 |
+
{
|
| 1635 |
+
"epoch": 29.565217391304348,
|
| 1636 |
+
"eval_loss": 0.19978828728199005,
|
| 1637 |
+
"eval_runtime": 150.159,
|
| 1638 |
+
"eval_samples_per_second": 17.342,
|
| 1639 |
+
"eval_steps_per_second": 4.335,
|
| 1640 |
+
"eval_wer": 0.16672633965867048,
|
| 1641 |
+
"step": 10200
|
| 1642 |
+
},
|
| 1643 |
+
{
|
| 1644 |
+
"epoch": 29.855072463768117,
|
| 1645 |
+
"grad_norm": 0.8047321438789368,
|
| 1646 |
+
"learning_rate": 0.00014629145728643214,
|
| 1647 |
+
"loss": 0.1862,
|
| 1648 |
+
"step": 10300
|
| 1649 |
+
},
|
| 1650 |
+
{
|
| 1651 |
+
"epoch": 29.855072463768117,
|
| 1652 |
+
"eval_loss": 0.19975517690181732,
|
| 1653 |
+
"eval_runtime": 150.1129,
|
| 1654 |
+
"eval_samples_per_second": 17.347,
|
| 1655 |
+
"eval_steps_per_second": 4.337,
|
| 1656 |
+
"eval_wer": 0.1674424155627163,
|
| 1657 |
+
"step": 10300
|
| 1658 |
+
},
|
| 1659 |
+
{
|
| 1660 |
+
"epoch": 30.144927536231883,
|
| 1661 |
+
"grad_norm": 1.9314672946929932,
|
| 1662 |
+
"learning_rate": 0.00014478391959798993,
|
| 1663 |
+
"loss": 0.1916,
|
| 1664 |
+
"step": 10400
|
| 1665 |
+
},
|
| 1666 |
+
{
|
| 1667 |
+
"epoch": 30.144927536231883,
|
| 1668 |
+
"eval_loss": 0.20524050295352936,
|
| 1669 |
+
"eval_runtime": 150.327,
|
| 1670 |
+
"eval_samples_per_second": 17.322,
|
| 1671 |
+
"eval_steps_per_second": 4.331,
|
| 1672 |
+
"eval_wer": 0.16610971874129768,
|
| 1673 |
+
"step": 10400
|
| 1674 |
+
},
|
| 1675 |
+
{
|
| 1676 |
+
"epoch": 30.434782608695652,
|
| 1677 |
+
"grad_norm": 1.7343331575393677,
|
| 1678 |
+
"learning_rate": 0.00014327638190954773,
|
| 1679 |
+
"loss": 0.1945,
|
| 1680 |
+
"step": 10500
|
| 1681 |
+
},
|
| 1682 |
+
{
|
| 1683 |
+
"epoch": 30.434782608695652,
|
| 1684 |
+
"eval_loss": 0.19761110842227936,
|
| 1685 |
+
"eval_runtime": 150.5791,
|
| 1686 |
+
"eval_samples_per_second": 17.293,
|
| 1687 |
+
"eval_steps_per_second": 4.323,
|
| 1688 |
+
"eval_wer": 0.16513505987190197,
|
| 1689 |
+
"step": 10500
|
| 1690 |
+
},
|
| 1691 |
+
{
|
| 1692 |
+
"epoch": 30.72463768115942,
|
| 1693 |
+
"grad_norm": 1.779826283454895,
|
| 1694 |
+
"learning_rate": 0.00014176884422110553,
|
| 1695 |
+
"loss": 0.1864,
|
| 1696 |
+
"step": 10600
|
| 1697 |
+
},
|
| 1698 |
+
{
|
| 1699 |
+
"epoch": 30.72463768115942,
|
| 1700 |
+
"eval_loss": 0.20148740708827972,
|
| 1701 |
+
"eval_runtime": 150.4391,
|
| 1702 |
+
"eval_samples_per_second": 17.309,
|
| 1703 |
+
"eval_steps_per_second": 4.327,
|
| 1704 |
+
"eval_wer": 0.1651549508692366,
|
| 1705 |
+
"step": 10600
|
| 1706 |
+
},
|
| 1707 |
+
{
|
| 1708 |
+
"epoch": 31.014492753623188,
|
| 1709 |
+
"grad_norm": 0.46111172437667847,
|
| 1710 |
+
"learning_rate": 0.0001402613065326633,
|
| 1711 |
+
"loss": 0.1887,
|
| 1712 |
+
"step": 10700
|
| 1713 |
+
},
|
| 1714 |
+
{
|
| 1715 |
+
"epoch": 31.014492753623188,
|
| 1716 |
+
"eval_loss": 0.19817914068698883,
|
| 1717 |
+
"eval_runtime": 149.7345,
|
| 1718 |
+
"eval_samples_per_second": 17.391,
|
| 1719 |
+
"eval_steps_per_second": 4.348,
|
| 1720 |
+
"eval_wer": 0.16451843895452917,
|
| 1721 |
+
"step": 10700
|
| 1722 |
+
},
|
| 1723 |
+
{
|
| 1724 |
+
"epoch": 31.304347826086957,
|
| 1725 |
+
"grad_norm": 0.4922316074371338,
|
| 1726 |
+
"learning_rate": 0.0001387537688442211,
|
| 1727 |
+
"loss": 0.1871,
|
| 1728 |
+
"step": 10800
|
| 1729 |
+
},
|
| 1730 |
+
{
|
| 1731 |
+
"epoch": 31.304347826086957,
|
| 1732 |
+
"eval_loss": 0.20190125703811646,
|
| 1733 |
+
"eval_runtime": 150.4015,
|
| 1734 |
+
"eval_samples_per_second": 17.314,
|
| 1735 |
+
"eval_steps_per_second": 4.328,
|
| 1736 |
+
"eval_wer": 0.16599037275729006,
|
| 1737 |
+
"step": 10800
|
| 1738 |
+
},
|
| 1739 |
+
{
|
| 1740 |
+
"epoch": 31.594202898550726,
|
| 1741 |
+
"grad_norm": 0.5234766602516174,
|
| 1742 |
+
"learning_rate": 0.0001372462311557789,
|
| 1743 |
+
"loss": 0.1913,
|
| 1744 |
+
"step": 10900
|
| 1745 |
+
},
|
| 1746 |
+
{
|
| 1747 |
+
"epoch": 31.594202898550726,
|
| 1748 |
+
"eval_loss": 0.2058933973312378,
|
| 1749 |
+
"eval_runtime": 150.6356,
|
| 1750 |
+
"eval_samples_per_second": 17.287,
|
| 1751 |
+
"eval_steps_per_second": 4.322,
|
| 1752 |
+
"eval_wer": 0.16784023550940844,
|
| 1753 |
+
"step": 10900
|
| 1754 |
+
},
|
| 1755 |
+
{
|
| 1756 |
+
"epoch": 31.884057971014492,
|
| 1757 |
+
"grad_norm": 0.5543705224990845,
|
| 1758 |
+
"learning_rate": 0.00013573869346733666,
|
| 1759 |
+
"loss": 0.1937,
|
| 1760 |
+
"step": 11000
|
| 1761 |
+
},
|
| 1762 |
+
{
|
| 1763 |
+
"epoch": 31.884057971014492,
|
| 1764 |
+
"eval_loss": 0.19291311502456665,
|
| 1765 |
+
"eval_runtime": 149.9941,
|
| 1766 |
+
"eval_samples_per_second": 17.361,
|
| 1767 |
+
"eval_steps_per_second": 4.34,
|
| 1768 |
+
"eval_wer": 0.1636830170664757,
|
| 1769 |
+
"step": 11000
|
| 1770 |
+
},
|
| 1771 |
+
{
|
| 1772 |
+
"epoch": 32.17391304347826,
|
| 1773 |
+
"grad_norm": 0.5391444563865662,
|
| 1774 |
+
"learning_rate": 0.00013423115577889446,
|
| 1775 |
+
"loss": 0.1813,
|
| 1776 |
+
"step": 11100
|
| 1777 |
+
},
|
| 1778 |
+
{
|
| 1779 |
+
"epoch": 32.17391304347826,
|
| 1780 |
+
"eval_loss": 0.200283020734787,
|
| 1781 |
+
"eval_runtime": 150.1119,
|
| 1782 |
+
"eval_samples_per_second": 17.347,
|
| 1783 |
+
"eval_steps_per_second": 4.337,
|
| 1784 |
+
"eval_wer": 0.1646576759358714,
|
| 1785 |
+
"step": 11100
|
| 1786 |
+
},
|
| 1787 |
+
{
|
| 1788 |
+
"epoch": 32.46376811594203,
|
| 1789 |
+
"grad_norm": 0.6253378391265869,
|
| 1790 |
+
"learning_rate": 0.00013272361809045226,
|
| 1791 |
+
"loss": 0.1838,
|
| 1792 |
+
"step": 11200
|
| 1793 |
+
},
|
| 1794 |
+
{
|
| 1795 |
+
"epoch": 32.46376811594203,
|
| 1796 |
+
"eval_loss": 0.20407435297966003,
|
| 1797 |
+
"eval_runtime": 149.6169,
|
| 1798 |
+
"eval_samples_per_second": 17.404,
|
| 1799 |
+
"eval_steps_per_second": 4.351,
|
| 1800 |
+
"eval_wer": 0.16593069976528624,
|
| 1801 |
+
"step": 11200
|
| 1802 |
+
},
|
| 1803 |
+
{
|
| 1804 |
+
"epoch": 32.7536231884058,
|
| 1805 |
+
"grad_norm": 0.8853746652603149,
|
| 1806 |
+
"learning_rate": 0.00013121608040201003,
|
| 1807 |
+
"loss": 0.1815,
|
| 1808 |
+
"step": 11300
|
| 1809 |
+
},
|
| 1810 |
+
{
|
| 1811 |
+
"epoch": 32.7536231884058,
|
| 1812 |
+
"eval_loss": 0.19644393026828766,
|
| 1813 |
+
"eval_runtime": 151.2753,
|
| 1814 |
+
"eval_samples_per_second": 17.214,
|
| 1815 |
+
"eval_steps_per_second": 4.303,
|
| 1816 |
+
"eval_wer": 0.16155468035167284,
|
| 1817 |
+
"step": 11300
|
| 1818 |
+
},
|
| 1819 |
+
{
|
| 1820 |
+
"epoch": 33.04347826086956,
|
| 1821 |
+
"grad_norm": 0.6834578514099121,
|
| 1822 |
+
"learning_rate": 0.00012970854271356782,
|
| 1823 |
+
"loss": 0.1843,
|
| 1824 |
+
"step": 11400
|
| 1825 |
+
},
|
| 1826 |
+
{
|
| 1827 |
+
"epoch": 33.04347826086956,
|
| 1828 |
+
"eval_loss": 0.20076803863048553,
|
| 1829 |
+
"eval_runtime": 150.4942,
|
| 1830 |
+
"eval_samples_per_second": 17.303,
|
| 1831 |
+
"eval_steps_per_second": 4.326,
|
| 1832 |
+
"eval_wer": 0.16308628714643753,
|
| 1833 |
+
"step": 11400
|
| 1834 |
+
},
|
| 1835 |
+
{
|
| 1836 |
+
"epoch": 33.333333333333336,
|
| 1837 |
+
"grad_norm": 0.5597058534622192,
|
| 1838 |
+
"learning_rate": 0.00012820100502512562,
|
| 1839 |
+
"loss": 0.1831,
|
| 1840 |
+
"step": 11500
|
| 1841 |
+
},
|
| 1842 |
+
{
|
| 1843 |
+
"epoch": 33.333333333333336,
|
| 1844 |
+
"eval_loss": 0.20445819199085236,
|
| 1845 |
+
"eval_runtime": 150.129,
|
| 1846 |
+
"eval_samples_per_second": 17.345,
|
| 1847 |
+
"eval_steps_per_second": 4.336,
|
| 1848 |
+
"eval_wer": 0.16370290806381033,
|
| 1849 |
+
"step": 11500
|
| 1850 |
+
},
|
| 1851 |
+
{
|
| 1852 |
+
"epoch": 33.6231884057971,
|
| 1853 |
+
"grad_norm": 0.6372693181037903,
|
| 1854 |
+
"learning_rate": 0.00012669346733668342,
|
| 1855 |
+
"loss": 0.1823,
|
| 1856 |
+
"step": 11600
|
| 1857 |
+
},
|
| 1858 |
+
{
|
| 1859 |
+
"epoch": 33.6231884057971,
|
| 1860 |
+
"eval_loss": 0.19614148139953613,
|
| 1861 |
+
"eval_runtime": 150.5634,
|
| 1862 |
+
"eval_samples_per_second": 17.295,
|
| 1863 |
+
"eval_steps_per_second": 4.324,
|
| 1864 |
+
"eval_wer": 0.1632851971197836,
|
| 1865 |
+
"step": 11600
|
| 1866 |
+
},
|
| 1867 |
+
{
|
| 1868 |
+
"epoch": 33.91304347826087,
|
| 1869 |
+
"grad_norm": 0.6953226327896118,
|
| 1870 |
+
"learning_rate": 0.0001251859296482412,
|
| 1871 |
+
"loss": 0.1825,
|
| 1872 |
+
"step": 11700
|
| 1873 |
+
},
|
| 1874 |
+
{
|
| 1875 |
+
"epoch": 33.91304347826087,
|
| 1876 |
+
"eval_loss": 0.200005903840065,
|
| 1877 |
+
"eval_runtime": 150.8381,
|
| 1878 |
+
"eval_samples_per_second": 17.264,
|
| 1879 |
+
"eval_steps_per_second": 4.316,
|
| 1880 |
+
"eval_wer": 0.16511516887456737,
|
| 1881 |
+
"step": 11700
|
| 1882 |
+
},
|
| 1883 |
+
{
|
| 1884 |
+
"epoch": 34.20289855072464,
|
| 1885 |
+
"grad_norm": 0.581469714641571,
|
| 1886 |
+
"learning_rate": 0.00012367839195979899,
|
| 1887 |
+
"loss": 0.1809,
|
| 1888 |
+
"step": 11800
|
| 1889 |
+
},
|
| 1890 |
+
{
|
| 1891 |
+
"epoch": 34.20289855072464,
|
| 1892 |
+
"eval_loss": 0.19282744824886322,
|
| 1893 |
+
"eval_runtime": 150.9011,
|
| 1894 |
+
"eval_samples_per_second": 17.256,
|
| 1895 |
+
"eval_steps_per_second": 4.314,
|
| 1896 |
+
"eval_wer": 0.16153478935433824,
|
| 1897 |
+
"step": 11800
|
| 1898 |
+
},
|
| 1899 |
+
{
|
| 1900 |
+
"epoch": 34.492753623188406,
|
| 1901 |
+
"grad_norm": 0.5484752058982849,
|
| 1902 |
+
"learning_rate": 0.00012217085427135678,
|
| 1903 |
+
"loss": 0.1841,
|
| 1904 |
+
"step": 11900
|
| 1905 |
+
},
|
| 1906 |
+
{
|
| 1907 |
+
"epoch": 34.492753623188406,
|
| 1908 |
+
"eval_loss": 0.19993117451667786,
|
| 1909 |
+
"eval_runtime": 150.2778,
|
| 1910 |
+
"eval_samples_per_second": 17.328,
|
| 1911 |
+
"eval_steps_per_second": 4.332,
|
| 1912 |
+
"eval_wer": 0.16378247205314875,
|
| 1913 |
+
"step": 11900
|
| 1914 |
+
},
|
| 1915 |
+
{
|
| 1916 |
+
"epoch": 34.78260869565217,
|
| 1917 |
+
"grad_norm": 0.7815059423446655,
|
| 1918 |
+
"learning_rate": 0.00012066331658291455,
|
| 1919 |
+
"loss": 0.1757,
|
| 1920 |
+
"step": 12000
|
| 1921 |
+
},
|
| 1922 |
+
{
|
| 1923 |
+
"epoch": 34.78260869565217,
|
| 1924 |
+
"eval_loss": 0.2015853375196457,
|
| 1925 |
+
"eval_runtime": 150.5729,
|
| 1926 |
+
"eval_samples_per_second": 17.294,
|
| 1927 |
+
"eval_steps_per_second": 4.323,
|
| 1928 |
+
"eval_wer": 0.16306639614910293,
|
| 1929 |
+
"step": 12000
|
| 1930 |
+
},
|
| 1931 |
+
{
|
| 1932 |
+
"epoch": 35.072463768115945,
|
| 1933 |
+
"grad_norm": 0.5722095966339111,
|
| 1934 |
+
"learning_rate": 0.00011915577889447236,
|
| 1935 |
+
"loss": 0.1873,
|
| 1936 |
+
"step": 12100
|
| 1937 |
+
},
|
| 1938 |
+
{
|
| 1939 |
+
"epoch": 35.072463768115945,
|
| 1940 |
+
"eval_loss": 0.1939040720462799,
|
| 1941 |
+
"eval_runtime": 151.4034,
|
| 1942 |
+
"eval_samples_per_second": 17.199,
|
| 1943 |
+
"eval_steps_per_second": 4.3,
|
| 1944 |
+
"eval_wer": 0.16028165652225804,
|
| 1945 |
+
"step": 12100
|
| 1946 |
+
},
|
| 1947 |
+
{
|
| 1948 |
+
"epoch": 35.36231884057971,
|
| 1949 |
+
"grad_norm": 0.4075194001197815,
|
| 1950 |
+
"learning_rate": 0.00011764824120603015,
|
| 1951 |
+
"loss": 0.1735,
|
| 1952 |
+
"step": 12200
|
| 1953 |
+
},
|
| 1954 |
+
{
|
| 1955 |
+
"epoch": 35.36231884057971,
|
| 1956 |
+
"eval_loss": 0.1967676430940628,
|
| 1957 |
+
"eval_runtime": 150.192,
|
| 1958 |
+
"eval_samples_per_second": 17.338,
|
| 1959 |
+
"eval_steps_per_second": 4.334,
|
| 1960 |
+
"eval_wer": 0.1589091777061702,
|
| 1961 |
+
"step": 12200
|
| 1962 |
+
},
|
| 1963 |
+
{
|
| 1964 |
+
"epoch": 35.65217391304348,
|
| 1965 |
+
"grad_norm": 0.6874520182609558,
|
| 1966 |
+
"learning_rate": 0.00011614070351758792,
|
| 1967 |
+
"loss": 0.1751,
|
| 1968 |
+
"step": 12300
|
| 1969 |
+
},
|
| 1970 |
+
{
|
| 1971 |
+
"epoch": 35.65217391304348,
|
| 1972 |
+
"eval_loss": 0.19623582065105438,
|
| 1973 |
+
"eval_runtime": 150.5346,
|
| 1974 |
+
"eval_samples_per_second": 17.298,
|
| 1975 |
+
"eval_steps_per_second": 4.325,
|
| 1976 |
+
"eval_wer": 0.1589290687035048,
|
| 1977 |
+
"step": 12300
|
| 1978 |
+
},
|
| 1979 |
+
{
|
| 1980 |
+
"epoch": 35.94202898550725,
|
| 1981 |
+
"grad_norm": 0.5243326425552368,
|
| 1982 |
+
"learning_rate": 0.00011463316582914573,
|
| 1983 |
+
"loss": 0.1763,
|
| 1984 |
+
"step": 12400
|
| 1985 |
+
},
|
| 1986 |
+
{
|
| 1987 |
+
"epoch": 35.94202898550725,
|
| 1988 |
+
"eval_loss": 0.1922197937965393,
|
| 1989 |
+
"eval_runtime": 150.9274,
|
| 1990 |
+
"eval_samples_per_second": 17.253,
|
| 1991 |
+
"eval_steps_per_second": 4.313,
|
| 1992 |
+
"eval_wer": 0.1573775709114055,
|
| 1993 |
+
"step": 12400
|
| 1994 |
+
},
|
| 1995 |
+
{
|
| 1996 |
+
"epoch": 36.231884057971016,
|
| 1997 |
+
"grad_norm": 0.4569384455680847,
|
| 1998 |
+
"learning_rate": 0.00011312562814070351,
|
| 1999 |
+
"loss": 0.1819,
|
| 2000 |
+
"step": 12500
|
| 2001 |
+
},
|
| 2002 |
+
{
|
| 2003 |
+
"epoch": 36.231884057971016,
|
| 2004 |
+
"eval_loss": 0.19286799430847168,
|
| 2005 |
+
"eval_runtime": 150.6138,
|
| 2006 |
+
"eval_samples_per_second": 17.289,
|
| 2007 |
+
"eval_steps_per_second": 4.322,
|
| 2008 |
+
"eval_wer": 0.158312447786132,
|
| 2009 |
+
"step": 12500
|
| 2010 |
+
},
|
| 2011 |
+
{
|
| 2012 |
+
"epoch": 36.52173913043478,
|
| 2013 |
+
"grad_norm": 0.4915807843208313,
|
| 2014 |
+
"learning_rate": 0.00011161809045226128,
|
| 2015 |
+
"loss": 0.1732,
|
| 2016 |
+
"step": 12600
|
| 2017 |
+
},
|
| 2018 |
+
{
|
| 2019 |
+
"epoch": 36.52173913043478,
|
| 2020 |
+
"eval_loss": 0.19591785967350006,
|
| 2021 |
+
"eval_runtime": 150.6206,
|
| 2022 |
+
"eval_samples_per_second": 17.288,
|
| 2023 |
+
"eval_steps_per_second": 4.322,
|
| 2024 |
+
"eval_wer": 0.15690018697537494,
|
| 2025 |
+
"step": 12600
|
| 2026 |
+
},
|
| 2027 |
+
{
|
| 2028 |
+
"epoch": 36.81159420289855,
|
| 2029 |
+
"grad_norm": 0.5279297232627869,
|
| 2030 |
+
"learning_rate": 0.0001101105527638191,
|
| 2031 |
+
"loss": 0.1716,
|
| 2032 |
+
"step": 12700
|
| 2033 |
+
},
|
| 2034 |
+
{
|
| 2035 |
+
"epoch": 36.81159420289855,
|
| 2036 |
+
"eval_loss": 0.19358539581298828,
|
| 2037 |
+
"eval_runtime": 149.8266,
|
| 2038 |
+
"eval_samples_per_second": 17.38,
|
| 2039 |
+
"eval_steps_per_second": 4.345,
|
| 2040 |
+
"eval_wer": 0.15692007797270954,
|
| 2041 |
+
"step": 12700
|
| 2042 |
+
},
|
| 2043 |
+
{
|
| 2044 |
+
"epoch": 37.10144927536232,
|
| 2045 |
+
"grad_norm": 0.514346718788147,
|
| 2046 |
+
"learning_rate": 0.00010860301507537686,
|
| 2047 |
+
"loss": 0.1718,
|
| 2048 |
+
"step": 12800
|
| 2049 |
+
},
|
| 2050 |
+
{
|
| 2051 |
+
"epoch": 37.10144927536232,
|
| 2052 |
+
"eval_loss": 0.19505684077739716,
|
| 2053 |
+
"eval_runtime": 150.4836,
|
| 2054 |
+
"eval_samples_per_second": 17.304,
|
| 2055 |
+
"eval_steps_per_second": 4.326,
|
| 2056 |
+
"eval_wer": 0.15747702589807852,
|
| 2057 |
+
"step": 12800
|
| 2058 |
+
},
|
| 2059 |
+
{
|
| 2060 |
+
"epoch": 37.391304347826086,
|
| 2061 |
+
"grad_norm": 0.534958004951477,
|
| 2062 |
+
"learning_rate": 0.00010709547738693467,
|
| 2063 |
+
"loss": 0.1688,
|
| 2064 |
+
"step": 12900
|
| 2065 |
+
},
|
| 2066 |
+
{
|
| 2067 |
+
"epoch": 37.391304347826086,
|
| 2068 |
+
"eval_loss": 0.19502244889736176,
|
| 2069 |
+
"eval_runtime": 150.9887,
|
| 2070 |
+
"eval_samples_per_second": 17.246,
|
| 2071 |
+
"eval_steps_per_second": 4.312,
|
| 2072 |
+
"eval_wer": 0.15809364681545132,
|
| 2073 |
+
"step": 12900
|
| 2074 |
+
},
|
| 2075 |
+
{
|
| 2076 |
+
"epoch": 37.68115942028985,
|
| 2077 |
+
"grad_norm": 0.5444557070732117,
|
| 2078 |
+
"learning_rate": 0.00010558793969849246,
|
| 2079 |
+
"loss": 0.171,
|
| 2080 |
+
"step": 13000
|
| 2081 |
+
},
|
| 2082 |
+
{
|
| 2083 |
+
"epoch": 37.68115942028985,
|
| 2084 |
+
"eval_loss": 0.19441960752010345,
|
| 2085 |
+
"eval_runtime": 151.1091,
|
| 2086 |
+
"eval_samples_per_second": 17.233,
|
| 2087 |
+
"eval_steps_per_second": 4.308,
|
| 2088 |
+
"eval_wer": 0.1567410589966981,
|
| 2089 |
+
"step": 13000
|
| 2090 |
+
},
|
| 2091 |
+
{
|
| 2092 |
+
"epoch": 37.971014492753625,
|
| 2093 |
+
"grad_norm": 0.5509995222091675,
|
| 2094 |
+
"learning_rate": 0.00010408040201005023,
|
| 2095 |
+
"loss": 0.1732,
|
| 2096 |
+
"step": 13100
|
| 2097 |
+
},
|
| 2098 |
+
{
|
| 2099 |
+
"epoch": 37.971014492753625,
|
| 2100 |
+
"eval_loss": 0.19966420531272888,
|
| 2101 |
+
"eval_runtime": 150.8532,
|
| 2102 |
+
"eval_samples_per_second": 17.262,
|
| 2103 |
+
"eval_steps_per_second": 4.315,
|
| 2104 |
+
"eval_wer": 0.1586307037434857,
|
| 2105 |
+
"step": 13100
|
| 2106 |
+
},
|
| 2107 |
+
{
|
| 2108 |
+
"epoch": 38.26086956521739,
|
| 2109 |
+
"grad_norm": 1.5159767866134644,
|
| 2110 |
+
"learning_rate": 0.00010257286432160804,
|
| 2111 |
+
"loss": 0.1646,
|
| 2112 |
+
"step": 13200
|
| 2113 |
+
},
|
| 2114 |
+
{
|
| 2115 |
+
"epoch": 38.26086956521739,
|
| 2116 |
+
"eval_loss": 0.19946832954883575,
|
| 2117 |
+
"eval_runtime": 150.9123,
|
| 2118 |
+
"eval_samples_per_second": 17.255,
|
| 2119 |
+
"eval_steps_per_second": 4.314,
|
| 2120 |
+
"eval_wer": 0.16034132951426183,
|
| 2121 |
+
"step": 13200
|
| 2122 |
+
},
|
| 2123 |
+
{
|
| 2124 |
+
"epoch": 38.55072463768116,
|
| 2125 |
+
"grad_norm": 0.5519012808799744,
|
| 2126 |
+
"learning_rate": 0.00010106532663316582,
|
| 2127 |
+
"loss": 0.1775,
|
| 2128 |
+
"step": 13300
|
| 2129 |
+
},
|
| 2130 |
+
{
|
| 2131 |
+
"epoch": 38.55072463768116,
|
| 2132 |
+
"eval_loss": 0.19182883203029633,
|
| 2133 |
+
"eval_runtime": 150.2357,
|
| 2134 |
+
"eval_samples_per_second": 17.333,
|
| 2135 |
+
"eval_steps_per_second": 4.333,
|
| 2136 |
+
"eval_wer": 0.15670127700202888,
|
| 2137 |
+
"step": 13300
|
| 2138 |
+
},
|
| 2139 |
+
{
|
| 2140 |
+
"epoch": 38.84057971014493,
|
| 2141 |
+
"grad_norm": 0.6432836651802063,
|
| 2142 |
+
"learning_rate": 9.955778894472362e-05,
|
| 2143 |
+
"loss": 0.1702,
|
| 2144 |
+
"step": 13400
|
| 2145 |
+
},
|
| 2146 |
+
{
|
| 2147 |
+
"epoch": 38.84057971014493,
|
| 2148 |
+
"eval_loss": 0.1931132823228836,
|
| 2149 |
+
"eval_runtime": 150.1239,
|
| 2150 |
+
"eval_samples_per_second": 17.346,
|
| 2151 |
+
"eval_steps_per_second": 4.336,
|
| 2152 |
+
"eval_wer": 0.15606476508732148,
|
| 2153 |
+
"step": 13400
|
| 2154 |
+
},
|
| 2155 |
+
{
|
| 2156 |
+
"epoch": 39.130434782608695,
|
| 2157 |
+
"grad_norm": 0.8637715578079224,
|
| 2158 |
+
"learning_rate": 9.80502512562814e-05,
|
| 2159 |
+
"loss": 0.1683,
|
| 2160 |
+
"step": 13500
|
| 2161 |
+
},
|
| 2162 |
+
{
|
| 2163 |
+
"epoch": 39.130434782608695,
|
| 2164 |
+
"eval_loss": 0.20001940429210663,
|
| 2165 |
+
"eval_runtime": 150.2986,
|
| 2166 |
+
"eval_samples_per_second": 17.326,
|
| 2167 |
+
"eval_steps_per_second": 4.331,
|
| 2168 |
+
"eval_wer": 0.1580140828261129,
|
| 2169 |
+
"step": 13500
|
| 2170 |
+
},
|
| 2171 |
+
{
|
| 2172 |
+
"epoch": 39.42028985507246,
|
| 2173 |
+
"grad_norm": 0.8066322207450867,
|
| 2174 |
+
"learning_rate": 9.654271356783919e-05,
|
| 2175 |
+
"loss": 0.164,
|
| 2176 |
+
"step": 13600
|
| 2177 |
+
},
|
| 2178 |
+
{
|
| 2179 |
+
"epoch": 39.42028985507246,
|
| 2180 |
+
"eval_loss": 0.1947360634803772,
|
| 2181 |
+
"eval_runtime": 150.9523,
|
| 2182 |
+
"eval_samples_per_second": 17.25,
|
| 2183 |
+
"eval_steps_per_second": 4.313,
|
| 2184 |
+
"eval_wer": 0.15469228627123363,
|
| 2185 |
+
"step": 13600
|
| 2186 |
+
},
|
| 2187 |
+
{
|
| 2188 |
+
"epoch": 39.710144927536234,
|
| 2189 |
+
"grad_norm": 0.522996723651886,
|
| 2190 |
+
"learning_rate": 9.503517587939698e-05,
|
| 2191 |
+
"loss": 0.1683,
|
| 2192 |
+
"step": 13700
|
| 2193 |
+
},
|
| 2194 |
+
{
|
| 2195 |
+
"epoch": 39.710144927536234,
|
| 2196 |
+
"eval_loss": 0.19472628831863403,
|
| 2197 |
+
"eval_runtime": 150.4547,
|
| 2198 |
+
"eval_samples_per_second": 17.308,
|
| 2199 |
+
"eval_steps_per_second": 4.327,
|
| 2200 |
+
"eval_wer": 0.1561443290766599,
|
| 2201 |
+
"step": 13700
|
| 2202 |
+
},
|
| 2203 |
+
{
|
| 2204 |
+
"epoch": 40.0,
|
| 2205 |
+
"grad_norm": 1.0164830684661865,
|
| 2206 |
+
"learning_rate": 9.352763819095477e-05,
|
| 2207 |
+
"loss": 0.1694,
|
| 2208 |
+
"step": 13800
|
| 2209 |
+
},
|
| 2210 |
+
{
|
| 2211 |
+
"epoch": 40.0,
|
| 2212 |
+
"eval_loss": 0.19553135335445404,
|
| 2213 |
+
"eval_runtime": 154.0901,
|
| 2214 |
+
"eval_samples_per_second": 16.899,
|
| 2215 |
+
"eval_steps_per_second": 4.225,
|
| 2216 |
+
"eval_wer": 0.1548514142499105,
|
| 2217 |
+
"step": 13800
|
| 2218 |
+
},
|
| 2219 |
+
{
|
| 2220 |
+
"epoch": 40.289855072463766,
|
| 2221 |
+
"grad_norm": 1.6812098026275635,
|
| 2222 |
+
"learning_rate": 9.202010050251257e-05,
|
| 2223 |
+
"loss": 0.1683,
|
| 2224 |
+
"step": 13900
|
| 2225 |
+
},
|
| 2226 |
+
{
|
| 2227 |
+
"epoch": 40.289855072463766,
|
| 2228 |
+
"eval_loss": 0.1967693269252777,
|
| 2229 |
+
"eval_runtime": 149.7768,
|
| 2230 |
+
"eval_samples_per_second": 17.386,
|
| 2231 |
+
"eval_steps_per_second": 4.346,
|
| 2232 |
+
"eval_wer": 0.1577356088634284,
|
| 2233 |
+
"step": 13900
|
| 2234 |
+
},
|
| 2235 |
+
{
|
| 2236 |
+
"epoch": 40.57971014492754,
|
| 2237 |
+
"grad_norm": 0.9731245040893555,
|
| 2238 |
+
"learning_rate": 9.052763819095476e-05,
|
| 2239 |
+
"loss": 0.1689,
|
| 2240 |
+
"step": 14000
|
| 2241 |
+
},
|
| 2242 |
+
{
|
| 2243 |
+
"epoch": 40.57971014492754,
|
| 2244 |
+
"eval_loss": 0.19680753350257874,
|
| 2245 |
+
"eval_runtime": 151.1604,
|
| 2246 |
+
"eval_samples_per_second": 17.227,
|
| 2247 |
+
"eval_steps_per_second": 4.307,
|
| 2248 |
+
"eval_wer": 0.15475195926323745,
|
| 2249 |
+
"step": 14000
|
| 2250 |
+
},
|
| 2251 |
+
{
|
| 2252 |
+
"epoch": 40.869565217391305,
|
| 2253 |
+
"grad_norm": 1.847554326057434,
|
| 2254 |
+
"learning_rate": 8.903517587939697e-05,
|
| 2255 |
+
"loss": 0.1678,
|
| 2256 |
+
"step": 14100
|
| 2257 |
+
},
|
| 2258 |
+
{
|
| 2259 |
+
"epoch": 40.869565217391305,
|
| 2260 |
+
"eval_loss": 0.201131209731102,
|
| 2261 |
+
"eval_runtime": 150.5156,
|
| 2262 |
+
"eval_samples_per_second": 17.301,
|
| 2263 |
+
"eval_steps_per_second": 4.325,
|
| 2264 |
+
"eval_wer": 0.15664160401002505,
|
| 2265 |
+
"step": 14100
|
| 2266 |
+
},
|
| 2267 |
+
{
|
| 2268 |
+
"epoch": 41.15942028985507,
|
| 2269 |
+
"grad_norm": 0.4787213206291199,
|
| 2270 |
+
"learning_rate": 8.752763819095477e-05,
|
| 2271 |
+
"loss": 0.1626,
|
| 2272 |
+
"step": 14200
|
| 2273 |
+
},
|
| 2274 |
+
{
|
| 2275 |
+
"epoch": 41.15942028985507,
|
| 2276 |
+
"eval_loss": 0.19733218848705292,
|
| 2277 |
+
"eval_runtime": 150.8887,
|
| 2278 |
+
"eval_samples_per_second": 17.258,
|
| 2279 |
+
"eval_steps_per_second": 4.314,
|
| 2280 |
+
"eval_wer": 0.15355849942316108,
|
| 2281 |
+
"step": 14200
|
| 2282 |
+
},
|
| 2283 |
+
{
|
| 2284 |
+
"epoch": 41.44927536231884,
|
| 2285 |
+
"grad_norm": 0.6790758371353149,
|
| 2286 |
+
"learning_rate": 8.602010050251256e-05,
|
| 2287 |
+
"loss": 0.1677,
|
| 2288 |
+
"step": 14300
|
| 2289 |
+
},
|
| 2290 |
+
{
|
| 2291 |
+
"epoch": 41.44927536231884,
|
| 2292 |
+
"eval_loss": 0.19702668488025665,
|
| 2293 |
+
"eval_runtime": 150.5236,
|
| 2294 |
+
"eval_samples_per_second": 17.3,
|
| 2295 |
+
"eval_steps_per_second": 4.325,
|
| 2296 |
+
"eval_wer": 0.1535982814178303,
|
| 2297 |
+
"step": 14300
|
| 2298 |
+
},
|
| 2299 |
+
{
|
| 2300 |
+
"epoch": 41.73913043478261,
|
| 2301 |
+
"grad_norm": 0.4391827881336212,
|
| 2302 |
+
"learning_rate": 8.451256281407034e-05,
|
| 2303 |
+
"loss": 0.1643,
|
| 2304 |
+
"step": 14400
|
| 2305 |
+
},
|
| 2306 |
+
{
|
| 2307 |
+
"epoch": 41.73913043478261,
|
| 2308 |
+
"eval_loss": 0.19612571597099304,
|
| 2309 |
+
"eval_runtime": 150.1133,
|
| 2310 |
+
"eval_samples_per_second": 17.347,
|
| 2311 |
+
"eval_steps_per_second": 4.337,
|
| 2312 |
+
"eval_wer": 0.15419501133786848,
|
| 2313 |
+
"step": 14400
|
| 2314 |
+
},
|
| 2315 |
+
{
|
| 2316 |
+
"epoch": 42.028985507246375,
|
| 2317 |
+
"grad_norm": 0.5288633108139038,
|
| 2318 |
+
"learning_rate": 8.300502512562814e-05,
|
| 2319 |
+
"loss": 0.1632,
|
| 2320 |
+
"step": 14500
|
| 2321 |
+
},
|
| 2322 |
+
{
|
| 2323 |
+
"epoch": 42.028985507246375,
|
| 2324 |
+
"eval_loss": 0.19461919367313385,
|
| 2325 |
+
"eval_runtime": 150.447,
|
| 2326 |
+
"eval_samples_per_second": 17.308,
|
| 2327 |
+
"eval_steps_per_second": 4.327,
|
| 2328 |
+
"eval_wer": 0.15477185026057205,
|
| 2329 |
+
"step": 14500
|
| 2330 |
+
},
|
| 2331 |
+
{
|
| 2332 |
+
"epoch": 42.31884057971015,
|
| 2333 |
+
"grad_norm": 0.5337244272232056,
|
| 2334 |
+
"learning_rate": 8.149748743718592e-05,
|
| 2335 |
+
"loss": 0.1648,
|
| 2336 |
+
"step": 14600
|
| 2337 |
+
},
|
| 2338 |
+
{
|
| 2339 |
+
"epoch": 42.31884057971015,
|
| 2340 |
+
"eval_loss": 0.20134814083576202,
|
| 2341 |
+
"eval_runtime": 149.9106,
|
| 2342 |
+
"eval_samples_per_second": 17.37,
|
| 2343 |
+
"eval_steps_per_second": 4.343,
|
| 2344 |
+
"eval_wer": 0.15572661813263317,
|
| 2345 |
+
"step": 14600
|
| 2346 |
+
},
|
| 2347 |
+
{
|
| 2348 |
+
"epoch": 42.608695652173914,
|
| 2349 |
+
"grad_norm": 0.4042835533618927,
|
| 2350 |
+
"learning_rate": 7.99899497487437e-05,
|
| 2351 |
+
"loss": 0.1681,
|
| 2352 |
+
"step": 14700
|
| 2353 |
+
},
|
| 2354 |
+
{
|
| 2355 |
+
"epoch": 42.608695652173914,
|
| 2356 |
+
"eval_loss": 0.18993638455867767,
|
| 2357 |
+
"eval_runtime": 150.6382,
|
| 2358 |
+
"eval_samples_per_second": 17.286,
|
| 2359 |
+
"eval_steps_per_second": 4.322,
|
| 2360 |
+
"eval_wer": 0.15341926244181883,
|
| 2361 |
+
"step": 14700
|
| 2362 |
+
},
|
| 2363 |
+
{
|
| 2364 |
+
"epoch": 42.89855072463768,
|
| 2365 |
+
"grad_norm": 0.46454140543937683,
|
| 2366 |
+
"learning_rate": 7.84824120603015e-05,
|
| 2367 |
+
"loss": 0.1632,
|
| 2368 |
+
"step": 14800
|
| 2369 |
+
},
|
| 2370 |
+
{
|
| 2371 |
+
"epoch": 42.89855072463768,
|
| 2372 |
+
"eval_loss": 0.19958487153053284,
|
| 2373 |
+
"eval_runtime": 150.1539,
|
| 2374 |
+
"eval_samples_per_second": 17.342,
|
| 2375 |
+
"eval_steps_per_second": 4.336,
|
| 2376 |
+
"eval_wer": 0.15421490233520307,
|
| 2377 |
+
"step": 14800
|
| 2378 |
+
},
|
| 2379 |
+
{
|
| 2380 |
+
"epoch": 43.18840579710145,
|
| 2381 |
+
"grad_norm": 0.5660408735275269,
|
| 2382 |
+
"learning_rate": 7.697487437185928e-05,
|
| 2383 |
+
"loss": 0.1635,
|
| 2384 |
+
"step": 14900
|
| 2385 |
+
},
|
| 2386 |
+
{
|
| 2387 |
+
"epoch": 43.18840579710145,
|
| 2388 |
+
"eval_loss": 0.19778329133987427,
|
| 2389 |
+
"eval_runtime": 150.7927,
|
| 2390 |
+
"eval_samples_per_second": 17.269,
|
| 2391 |
+
"eval_steps_per_second": 4.317,
|
| 2392 |
+
"eval_wer": 0.1536380634124995,
|
| 2393 |
+
"step": 14900
|
| 2394 |
+
},
|
| 2395 |
+
{
|
| 2396 |
+
"epoch": 43.47826086956522,
|
| 2397 |
+
"grad_norm": 0.5456855893135071,
|
| 2398 |
+
"learning_rate": 7.546733668341708e-05,
|
| 2399 |
+
"loss": 0.1592,
|
| 2400 |
+
"step": 15000
|
| 2401 |
+
},
|
| 2402 |
+
{
|
| 2403 |
+
"epoch": 43.47826086956522,
|
| 2404 |
+
"eval_loss": 0.20038050413131714,
|
| 2405 |
+
"eval_runtime": 150.9242,
|
| 2406 |
+
"eval_samples_per_second": 17.254,
|
| 2407 |
+
"eval_steps_per_second": 4.313,
|
| 2408 |
+
"eval_wer": 0.1530015514977921,
|
| 2409 |
+
"step": 15000
|
| 2410 |
+
},
|
| 2411 |
+
{
|
| 2412 |
+
"epoch": 43.768115942028984,
|
| 2413 |
+
"grad_norm": 0.4355134963989258,
|
| 2414 |
+
"learning_rate": 7.395979899497487e-05,
|
| 2415 |
+
"loss": 0.1624,
|
| 2416 |
+
"step": 15100
|
| 2417 |
+
},
|
| 2418 |
+
{
|
| 2419 |
+
"epoch": 43.768115942028984,
|
| 2420 |
+
"eval_loss": 0.19985315203666687,
|
| 2421 |
+
"eval_runtime": 150.4394,
|
| 2422 |
+
"eval_samples_per_second": 17.309,
|
| 2423 |
+
"eval_steps_per_second": 4.327,
|
| 2424 |
+
"eval_wer": 0.15546803516728327,
|
| 2425 |
+
"step": 15100
|
| 2426 |
+
},
|
| 2427 |
+
{
|
| 2428 |
+
"epoch": 44.05797101449275,
|
| 2429 |
+
"grad_norm": 0.5090399384498596,
|
| 2430 |
+
"learning_rate": 7.245226130653266e-05,
|
| 2431 |
+
"loss": 0.1637,
|
| 2432 |
+
"step": 15200
|
| 2433 |
+
},
|
| 2434 |
+
{
|
| 2435 |
+
"epoch": 44.05797101449275,
|
| 2436 |
+
"eval_loss": 0.19995567202568054,
|
| 2437 |
+
"eval_runtime": 150.709,
|
| 2438 |
+
"eval_samples_per_second": 17.278,
|
| 2439 |
+
"eval_steps_per_second": 4.32,
|
| 2440 |
+
"eval_wer": 0.15588574611131,
|
| 2441 |
+
"step": 15200
|
| 2442 |
+
},
|
| 2443 |
+
{
|
| 2444 |
+
"epoch": 44.34782608695652,
|
| 2445 |
+
"grad_norm": 0.6206871271133423,
|
| 2446 |
+
"learning_rate": 7.094472361809045e-05,
|
| 2447 |
+
"loss": 0.1583,
|
| 2448 |
+
"step": 15300
|
| 2449 |
+
},
|
| 2450 |
+
{
|
| 2451 |
+
"epoch": 44.34782608695652,
|
| 2452 |
+
"eval_loss": 0.20139965415000916,
|
| 2453 |
+
"eval_runtime": 150.7257,
|
| 2454 |
+
"eval_samples_per_second": 17.276,
|
| 2455 |
+
"eval_steps_per_second": 4.319,
|
| 2456 |
+
"eval_wer": 0.15367784540716872,
|
| 2457 |
+
"step": 15300
|
| 2458 |
+
},
|
| 2459 |
+
{
|
| 2460 |
+
"epoch": 44.63768115942029,
|
| 2461 |
+
"grad_norm": 0.5422778129577637,
|
| 2462 |
+
"learning_rate": 6.943718592964823e-05,
|
| 2463 |
+
"loss": 0.1595,
|
| 2464 |
+
"step": 15400
|
| 2465 |
+
},
|
| 2466 |
+
{
|
| 2467 |
+
"epoch": 44.63768115942029,
|
| 2468 |
+
"eval_loss": 0.19820022583007812,
|
| 2469 |
+
"eval_runtime": 150.8522,
|
| 2470 |
+
"eval_samples_per_second": 17.262,
|
| 2471 |
+
"eval_steps_per_second": 4.315,
|
| 2472 |
+
"eval_wer": 0.15441381230854914,
|
| 2473 |
+
"step": 15400
|
| 2474 |
+
},
|
| 2475 |
+
{
|
| 2476 |
+
"epoch": 44.927536231884055,
|
| 2477 |
+
"grad_norm": 0.5393760800361633,
|
| 2478 |
+
"learning_rate": 6.792964824120603e-05,
|
| 2479 |
+
"loss": 0.164,
|
| 2480 |
+
"step": 15500
|
| 2481 |
+
},
|
| 2482 |
+
{
|
| 2483 |
+
"epoch": 44.927536231884055,
|
| 2484 |
+
"eval_loss": 0.1983390897512436,
|
| 2485 |
+
"eval_runtime": 150.8937,
|
| 2486 |
+
"eval_samples_per_second": 17.257,
|
| 2487 |
+
"eval_steps_per_second": 4.314,
|
| 2488 |
+
"eval_wer": 0.152703186537773,
|
| 2489 |
+
"step": 15500
|
| 2490 |
+
},
|
| 2491 |
+
{
|
| 2492 |
+
"epoch": 45.21739130434783,
|
| 2493 |
+
"grad_norm": 0.5035853385925293,
|
| 2494 |
+
"learning_rate": 6.642211055276381e-05,
|
| 2495 |
+
"loss": 0.158,
|
| 2496 |
+
"step": 15600
|
| 2497 |
+
},
|
| 2498 |
+
{
|
| 2499 |
+
"epoch": 45.21739130434783,
|
| 2500 |
+
"eval_loss": 0.19686628878116608,
|
| 2501 |
+
"eval_runtime": 151.0924,
|
| 2502 |
+
"eval_samples_per_second": 17.234,
|
| 2503 |
+
"eval_steps_per_second": 4.309,
|
| 2504 |
+
"eval_wer": 0.15409555635119546,
|
| 2505 |
+
"step": 15600
|
| 2506 |
+
},
|
| 2507 |
+
{
|
| 2508 |
+
"epoch": 45.507246376811594,
|
| 2509 |
+
"grad_norm": 0.5787107348442078,
|
| 2510 |
+
"learning_rate": 6.491457286432161e-05,
|
| 2511 |
+
"loss": 0.1585,
|
| 2512 |
+
"step": 15700
|
| 2513 |
+
},
|
| 2514 |
+
{
|
| 2515 |
+
"epoch": 45.507246376811594,
|
| 2516 |
+
"eval_loss": 0.2003917545080185,
|
| 2517 |
+
"eval_runtime": 151.9231,
|
| 2518 |
+
"eval_samples_per_second": 17.14,
|
| 2519 |
+
"eval_steps_per_second": 4.285,
|
| 2520 |
+
"eval_wer": 0.15216612960973863,
|
| 2521 |
+
"step": 15700
|
| 2522 |
+
},
|
| 2523 |
+
{
|
| 2524 |
+
"epoch": 45.79710144927536,
|
| 2525 |
+
"grad_norm": 0.6162687540054321,
|
| 2526 |
+
"learning_rate": 6.340703517587939e-05,
|
| 2527 |
+
"loss": 0.1551,
|
| 2528 |
+
"step": 15800
|
| 2529 |
+
},
|
| 2530 |
+
{
|
| 2531 |
+
"epoch": 45.79710144927536,
|
| 2532 |
+
"eval_loss": 0.19759398698806763,
|
| 2533 |
+
"eval_runtime": 150.7462,
|
| 2534 |
+
"eval_samples_per_second": 17.274,
|
| 2535 |
+
"eval_steps_per_second": 4.319,
|
| 2536 |
+
"eval_wer": 0.15117157974300832,
|
| 2537 |
+
"step": 15800
|
| 2538 |
+
},
|
| 2539 |
+
{
|
| 2540 |
+
"epoch": 46.08695652173913,
|
| 2541 |
+
"grad_norm": 0.5009045004844666,
|
| 2542 |
+
"learning_rate": 6.189949748743718e-05,
|
| 2543 |
+
"loss": 0.1583,
|
| 2544 |
+
"step": 15900
|
| 2545 |
+
},
|
| 2546 |
+
{
|
| 2547 |
+
"epoch": 46.08695652173913,
|
| 2548 |
+
"eval_loss": 0.19603191316127777,
|
| 2549 |
+
"eval_runtime": 150.3144,
|
| 2550 |
+
"eval_samples_per_second": 17.324,
|
| 2551 |
+
"eval_steps_per_second": 4.331,
|
| 2552 |
+
"eval_wer": 0.1526435135457692,
|
| 2553 |
+
"step": 15900
|
| 2554 |
+
},
|
| 2555 |
+
{
|
| 2556 |
+
"epoch": 46.3768115942029,
|
| 2557 |
+
"grad_norm": 0.6324445009231567,
|
| 2558 |
+
"learning_rate": 6.0391959798994966e-05,
|
| 2559 |
+
"loss": 0.1587,
|
| 2560 |
+
"step": 16000
|
| 2561 |
+
},
|
| 2562 |
+
{
|
| 2563 |
+
"epoch": 46.3768115942029,
|
| 2564 |
+
"eval_loss": 0.19517464935779572,
|
| 2565 |
+
"eval_runtime": 150.663,
|
| 2566 |
+
"eval_samples_per_second": 17.284,
|
| 2567 |
+
"eval_steps_per_second": 4.321,
|
| 2568 |
+
"eval_wer": 0.15127103472968134,
|
| 2569 |
+
"step": 16000
|
| 2570 |
+
},
|
| 2571 |
+
{
|
| 2572 |
+
"epoch": 46.666666666666664,
|
| 2573 |
+
"grad_norm": 0.5137718319892883,
|
| 2574 |
+
"learning_rate": 5.8899497487437184e-05,
|
| 2575 |
+
"loss": 0.1596,
|
| 2576 |
+
"step": 16100
|
| 2577 |
+
},
|
| 2578 |
+
{
|
| 2579 |
+
"epoch": 46.666666666666664,
|
| 2580 |
+
"eval_loss": 0.19590044021606445,
|
| 2581 |
+
"eval_runtime": 151.0573,
|
| 2582 |
+
"eval_samples_per_second": 17.238,
|
| 2583 |
+
"eval_steps_per_second": 4.31,
|
| 2584 |
+
"eval_wer": 0.15214623861240403,
|
| 2585 |
+
"step": 16100
|
| 2586 |
+
},
|
| 2587 |
+
{
|
| 2588 |
+
"epoch": 46.95652173913044,
|
| 2589 |
+
"grad_norm": 0.46284008026123047,
|
| 2590 |
+
"learning_rate": 5.739195979899497e-05,
|
| 2591 |
+
"loss": 0.1543,
|
| 2592 |
+
"step": 16200
|
| 2593 |
+
},
|
| 2594 |
+
{
|
| 2595 |
+
"epoch": 46.95652173913044,
|
| 2596 |
+
"eval_loss": 0.19321778416633606,
|
| 2597 |
+
"eval_runtime": 149.3218,
|
| 2598 |
+
"eval_samples_per_second": 17.439,
|
| 2599 |
+
"eval_steps_per_second": 4.36,
|
| 2600 |
+
"eval_wer": 0.14880455106019017,
|
| 2601 |
+
"step": 16200
|
| 2602 |
+
},
|
| 2603 |
+
{
|
| 2604 |
+
"epoch": 47.2463768115942,
|
| 2605 |
+
"grad_norm": 0.7941703796386719,
|
| 2606 |
+
"learning_rate": 5.588442211055276e-05,
|
| 2607 |
+
"loss": 0.1559,
|
| 2608 |
+
"step": 16300
|
| 2609 |
+
},
|
| 2610 |
+
{
|
| 2611 |
+
"epoch": 47.2463768115942,
|
| 2612 |
+
"eval_loss": 0.19706296920776367,
|
| 2613 |
+
"eval_runtime": 150.5032,
|
| 2614 |
+
"eval_samples_per_second": 17.302,
|
| 2615 |
+
"eval_steps_per_second": 4.325,
|
| 2616 |
+
"eval_wer": 0.1504753948362971,
|
| 2617 |
+
"step": 16300
|
| 2618 |
+
},
|
| 2619 |
+
{
|
| 2620 |
+
"epoch": 47.53623188405797,
|
| 2621 |
+
"grad_norm": 0.5882352590560913,
|
| 2622 |
+
"learning_rate": 5.437688442211055e-05,
|
| 2623 |
+
"loss": 0.1573,
|
| 2624 |
+
"step": 16400
|
| 2625 |
+
},
|
| 2626 |
+
{
|
| 2627 |
+
"epoch": 47.53623188405797,
|
| 2628 |
+
"eval_loss": 0.19471529126167297,
|
| 2629 |
+
"eval_runtime": 150.1514,
|
| 2630 |
+
"eval_samples_per_second": 17.342,
|
| 2631 |
+
"eval_steps_per_second": 4.336,
|
| 2632 |
+
"eval_wer": 0.15057484982297012,
|
| 2633 |
+
"step": 16400
|
| 2634 |
+
},
|
| 2635 |
+
{
|
| 2636 |
+
"epoch": 47.82608695652174,
|
| 2637 |
+
"grad_norm": 0.4913437068462372,
|
| 2638 |
+
"learning_rate": 5.286934673366834e-05,
|
| 2639 |
+
"loss": 0.1578,
|
| 2640 |
+
"step": 16500
|
| 2641 |
+
},
|
| 2642 |
+
{
|
| 2643 |
+
"epoch": 47.82608695652174,
|
| 2644 |
+
"eval_loss": 0.1927487701177597,
|
| 2645 |
+
"eval_runtime": 150.1261,
|
| 2646 |
+
"eval_samples_per_second": 17.345,
|
| 2647 |
+
"eval_steps_per_second": 4.336,
|
| 2648 |
+
"eval_wer": 0.15039583084695868,
|
| 2649 |
+
"step": 16500
|
| 2650 |
+
},
|
| 2651 |
+
{
|
| 2652 |
+
"epoch": 48.11594202898551,
|
| 2653 |
+
"grad_norm": 0.6036613583564758,
|
| 2654 |
+
"learning_rate": 5.136180904522613e-05,
|
| 2655 |
+
"loss": 0.1541,
|
| 2656 |
+
"step": 16600
|
| 2657 |
+
},
|
| 2658 |
+
{
|
| 2659 |
+
"epoch": 48.11594202898551,
|
| 2660 |
+
"eval_loss": 0.19660724699497223,
|
| 2661 |
+
"eval_runtime": 151.0113,
|
| 2662 |
+
"eval_samples_per_second": 17.244,
|
| 2663 |
+
"eval_steps_per_second": 4.311,
|
| 2664 |
+
"eval_wer": 0.1498985559135935,
|
| 2665 |
+
"step": 16600
|
| 2666 |
+
},
|
| 2667 |
+
{
|
| 2668 |
+
"epoch": 48.405797101449274,
|
| 2669 |
+
"grad_norm": 0.45591381192207336,
|
| 2670 |
+
"learning_rate": 4.985427135678391e-05,
|
| 2671 |
+
"loss": 0.1522,
|
| 2672 |
+
"step": 16700
|
| 2673 |
+
},
|
| 2674 |
+
{
|
| 2675 |
+
"epoch": 48.405797101449274,
|
| 2676 |
+
"eval_loss": 0.19638657569885254,
|
| 2677 |
+
"eval_runtime": 150.2963,
|
| 2678 |
+
"eval_samples_per_second": 17.326,
|
| 2679 |
+
"eval_steps_per_second": 4.331,
|
| 2680 |
+
"eval_wer": 0.14975931893225125,
|
| 2681 |
+
"step": 16700
|
| 2682 |
+
},
|
| 2683 |
+
{
|
| 2684 |
+
"epoch": 48.69565217391305,
|
| 2685 |
+
"grad_norm": 0.5124571323394775,
|
| 2686 |
+
"learning_rate": 4.8346733668341704e-05,
|
| 2687 |
+
"loss": 0.1565,
|
| 2688 |
+
"step": 16800
|
| 2689 |
+
},
|
| 2690 |
+
{
|
| 2691 |
+
"epoch": 48.69565217391305,
|
| 2692 |
+
"eval_loss": 0.19512411952018738,
|
| 2693 |
+
"eval_runtime": 150.989,
|
| 2694 |
+
"eval_samples_per_second": 17.246,
|
| 2695 |
+
"eval_steps_per_second": 4.312,
|
| 2696 |
+
"eval_wer": 0.15009746588693956,
|
| 2697 |
+
"step": 16800
|
| 2698 |
+
},
|
| 2699 |
+
{
|
| 2700 |
+
"epoch": 48.98550724637681,
|
| 2701 |
+
"grad_norm": 0.80686354637146,
|
| 2702 |
+
"learning_rate": 4.6839195979899494e-05,
|
| 2703 |
+
"loss": 0.1522,
|
| 2704 |
+
"step": 16900
|
| 2705 |
+
},
|
| 2706 |
+
{
|
| 2707 |
+
"epoch": 48.98550724637681,
|
| 2708 |
+
"eval_loss": 0.195089191198349,
|
| 2709 |
+
"eval_runtime": 151.6726,
|
| 2710 |
+
"eval_samples_per_second": 17.169,
|
| 2711 |
+
"eval_steps_per_second": 4.292,
|
| 2712 |
+
"eval_wer": 0.14874487806818634,
|
| 2713 |
+
"step": 16900
|
| 2714 |
+
},
|
| 2715 |
+
{
|
| 2716 |
+
"epoch": 49.27536231884058,
|
| 2717 |
+
"grad_norm": 0.5458950400352478,
|
| 2718 |
+
"learning_rate": 4.5331658291457285e-05,
|
| 2719 |
+
"loss": 0.1544,
|
| 2720 |
+
"step": 17000
|
| 2721 |
+
},
|
| 2722 |
+
{
|
| 2723 |
+
"epoch": 49.27536231884058,
|
| 2724 |
+
"eval_loss": 0.19665881991386414,
|
| 2725 |
+
"eval_runtime": 150.3912,
|
| 2726 |
+
"eval_samples_per_second": 17.315,
|
| 2727 |
+
"eval_steps_per_second": 4.329,
|
| 2728 |
+
"eval_wer": 0.1502367028682818,
|
| 2729 |
+
"step": 17000
|
| 2730 |
+
},
|
| 2731 |
+
{
|
| 2732 |
+
"epoch": 49.56521739130435,
|
| 2733 |
+
"grad_norm": 0.7740549445152283,
|
| 2734 |
+
"learning_rate": 4.3824120603015075e-05,
|
| 2735 |
+
"loss": 0.1501,
|
| 2736 |
+
"step": 17100
|
| 2737 |
+
},
|
| 2738 |
+
{
|
| 2739 |
+
"epoch": 49.56521739130435,
|
| 2740 |
+
"eval_loss": 0.19731196761131287,
|
| 2741 |
+
"eval_runtime": 150.5913,
|
| 2742 |
+
"eval_samples_per_second": 17.292,
|
| 2743 |
+
"eval_steps_per_second": 4.323,
|
| 2744 |
+
"eval_wer": 0.14878466006285554,
|
| 2745 |
+
"step": 17100
|
| 2746 |
+
},
|
| 2747 |
+
{
|
| 2748 |
+
"epoch": 49.85507246376812,
|
| 2749 |
+
"grad_norm": 0.72871333360672,
|
| 2750 |
+
"learning_rate": 4.231658291457286e-05,
|
| 2751 |
+
"loss": 0.1556,
|
| 2752 |
+
"step": 17200
|
| 2753 |
+
},
|
| 2754 |
+
{
|
| 2755 |
+
"epoch": 49.85507246376812,
|
| 2756 |
+
"eval_loss": 0.19586071372032166,
|
| 2757 |
+
"eval_runtime": 149.9817,
|
| 2758 |
+
"eval_samples_per_second": 17.362,
|
| 2759 |
+
"eval_steps_per_second": 4.341,
|
| 2760 |
+
"eval_wer": 0.14912280701754385,
|
| 2761 |
+
"step": 17200
|
| 2762 |
+
},
|
| 2763 |
+
{
|
| 2764 |
+
"epoch": 50.14492753623188,
|
| 2765 |
+
"grad_norm": 1.8948681354522705,
|
| 2766 |
+
"learning_rate": 4.080904522613065e-05,
|
| 2767 |
+
"loss": 0.1548,
|
| 2768 |
+
"step": 17300
|
| 2769 |
+
},
|
| 2770 |
+
{
|
| 2771 |
+
"epoch": 50.14492753623188,
|
| 2772 |
+
"eval_loss": 0.19828546047210693,
|
| 2773 |
+
"eval_runtime": 150.304,
|
| 2774 |
+
"eval_samples_per_second": 17.325,
|
| 2775 |
+
"eval_steps_per_second": 4.331,
|
| 2776 |
+
"eval_wer": 0.1501969208736126,
|
| 2777 |
+
"step": 17300
|
| 2778 |
+
},
|
| 2779 |
+
{
|
| 2780 |
+
"epoch": 50.43478260869565,
|
| 2781 |
+
"grad_norm": 1.8404796123504639,
|
| 2782 |
+
"learning_rate": 3.930150753768844e-05,
|
| 2783 |
+
"loss": 0.1532,
|
| 2784 |
+
"step": 17400
|
| 2785 |
+
},
|
| 2786 |
+
{
|
| 2787 |
+
"epoch": 50.43478260869565,
|
| 2788 |
+
"eval_loss": 0.2000737339258194,
|
| 2789 |
+
"eval_runtime": 150.2978,
|
| 2790 |
+
"eval_samples_per_second": 17.326,
|
| 2791 |
+
"eval_steps_per_second": 4.331,
|
| 2792 |
+
"eval_wer": 0.1498985559135935,
|
| 2793 |
+
"step": 17400
|
| 2794 |
+
},
|
| 2795 |
+
{
|
| 2796 |
+
"epoch": 50.72463768115942,
|
| 2797 |
+
"grad_norm": 1.8255242109298706,
|
| 2798 |
+
"learning_rate": 3.779396984924623e-05,
|
| 2799 |
+
"loss": 0.153,
|
| 2800 |
+
"step": 17500
|
| 2801 |
+
},
|
| 2802 |
+
{
|
| 2803 |
+
"epoch": 50.72463768115942,
|
| 2804 |
+
"eval_loss": 0.19718687236309052,
|
| 2805 |
+
"eval_runtime": 150.0456,
|
| 2806 |
+
"eval_samples_per_second": 17.355,
|
| 2807 |
+
"eval_steps_per_second": 4.339,
|
| 2808 |
+
"eval_wer": 0.1498587739189243,
|
| 2809 |
+
"step": 17500
|
| 2810 |
+
},
|
| 2811 |
+
{
|
| 2812 |
+
"epoch": 51.01449275362319,
|
| 2813 |
+
"grad_norm": 0.49974119663238525,
|
| 2814 |
+
"learning_rate": 3.6286432160804014e-05,
|
| 2815 |
+
"loss": 0.1511,
|
| 2816 |
+
"step": 17600
|
| 2817 |
+
},
|
| 2818 |
+
{
|
| 2819 |
+
"epoch": 51.01449275362319,
|
| 2820 |
+
"eval_loss": 0.20012781023979187,
|
| 2821 |
+
"eval_runtime": 150.9723,
|
| 2822 |
+
"eval_samples_per_second": 17.248,
|
| 2823 |
+
"eval_steps_per_second": 4.312,
|
| 2824 |
+
"eval_wer": 0.14938138998289374,
|
| 2825 |
+
"step": 17600
|
| 2826 |
+
},
|
| 2827 |
+
{
|
| 2828 |
+
"epoch": 51.30434782608695,
|
| 2829 |
+
"grad_norm": 0.4928853213787079,
|
| 2830 |
+
"learning_rate": 3.4778894472361804e-05,
|
| 2831 |
+
"loss": 0.1545,
|
| 2832 |
+
"step": 17700
|
| 2833 |
+
},
|
| 2834 |
+
{
|
| 2835 |
+
"epoch": 51.30434782608695,
|
| 2836 |
+
"eval_loss": 0.19761820137500763,
|
| 2837 |
+
"eval_runtime": 151.4411,
|
| 2838 |
+
"eval_samples_per_second": 17.195,
|
| 2839 |
+
"eval_steps_per_second": 4.299,
|
| 2840 |
+
"eval_wer": 0.15061463181763934,
|
| 2841 |
+
"step": 17700
|
| 2842 |
+
},
|
| 2843 |
+
{
|
| 2844 |
+
"epoch": 51.594202898550726,
|
| 2845 |
+
"grad_norm": 0.4174346625804901,
|
| 2846 |
+
"learning_rate": 3.3271356783919595e-05,
|
| 2847 |
+
"loss": 0.151,
|
| 2848 |
+
"step": 17800
|
| 2849 |
+
},
|
| 2850 |
+
{
|
| 2851 |
+
"epoch": 51.594202898550726,
|
| 2852 |
+
"eval_loss": 0.19907838106155396,
|
| 2853 |
+
"eval_runtime": 150.2239,
|
| 2854 |
+
"eval_samples_per_second": 17.334,
|
| 2855 |
+
"eval_steps_per_second": 4.334,
|
| 2856 |
+
"eval_wer": 0.14973942793491665,
|
| 2857 |
+
"step": 17800
|
| 2858 |
+
},
|
| 2859 |
+
{
|
| 2860 |
+
"epoch": 51.88405797101449,
|
| 2861 |
+
"grad_norm": 0.5709109902381897,
|
| 2862 |
+
"learning_rate": 3.1763819095477385e-05,
|
| 2863 |
+
"loss": 0.1513,
|
| 2864 |
+
"step": 17900
|
| 2865 |
+
},
|
| 2866 |
+
{
|
| 2867 |
+
"epoch": 51.88405797101449,
|
| 2868 |
+
"eval_loss": 0.19772109389305115,
|
| 2869 |
+
"eval_runtime": 151.1838,
|
| 2870 |
+
"eval_samples_per_second": 17.224,
|
| 2871 |
+
"eval_steps_per_second": 4.306,
|
| 2872 |
+
"eval_wer": 0.1521263476150694,
|
| 2873 |
+
"step": 17900
|
| 2874 |
+
},
|
| 2875 |
+
{
|
| 2876 |
+
"epoch": 52.17391304347826,
|
| 2877 |
+
"grad_norm": 0.4559178650379181,
|
| 2878 |
+
"learning_rate": 3.0256281407035173e-05,
|
| 2879 |
+
"loss": 0.1555,
|
| 2880 |
+
"step": 18000
|
| 2881 |
+
},
|
| 2882 |
+
{
|
| 2883 |
+
"epoch": 52.17391304347826,
|
| 2884 |
+
"eval_loss": 0.19882014393806458,
|
| 2885 |
+
"eval_runtime": 150.796,
|
| 2886 |
+
"eval_samples_per_second": 17.268,
|
| 2887 |
+
"eval_steps_per_second": 4.317,
|
| 2888 |
+
"eval_wer": 0.14874487806818634,
|
| 2889 |
+
"step": 18000
|
| 2890 |
+
},
|
| 2891 |
+
{
|
| 2892 |
+
"epoch": 52.46376811594203,
|
| 2893 |
+
"grad_norm": 0.6781191229820251,
|
| 2894 |
+
"learning_rate": 2.8763819095477384e-05,
|
| 2895 |
+
"loss": 0.1483,
|
| 2896 |
+
"step": 18100
|
| 2897 |
+
},
|
| 2898 |
+
{
|
| 2899 |
+
"epoch": 52.46376811594203,
|
| 2900 |
+
"eval_loss": 0.19811075925827026,
|
| 2901 |
+
"eval_runtime": 150.9776,
|
| 2902 |
+
"eval_samples_per_second": 17.248,
|
| 2903 |
+
"eval_steps_per_second": 4.312,
|
| 2904 |
+
"eval_wer": 0.14874487806818634,
|
| 2905 |
+
"step": 18100
|
| 2906 |
+
},
|
| 2907 |
+
{
|
| 2908 |
+
"epoch": 52.7536231884058,
|
| 2909 |
+
"grad_norm": 0.47036242485046387,
|
| 2910 |
+
"learning_rate": 2.7256281407035174e-05,
|
| 2911 |
+
"loss": 0.1498,
|
| 2912 |
+
"step": 18200
|
| 2913 |
+
},
|
| 2914 |
+
{
|
| 2915 |
+
"epoch": 52.7536231884058,
|
| 2916 |
+
"eval_loss": 0.19891373813152313,
|
| 2917 |
+
"eval_runtime": 150.1998,
|
| 2918 |
+
"eval_samples_per_second": 17.337,
|
| 2919 |
+
"eval_steps_per_second": 4.334,
|
| 2920 |
+
"eval_wer": 0.1498985559135935,
|
| 2921 |
+
"step": 18200
|
| 2922 |
+
},
|
| 2923 |
+
{
|
| 2924 |
+
"epoch": 53.04347826086956,
|
| 2925 |
+
"grad_norm": 0.40730613470077515,
|
| 2926 |
+
"learning_rate": 2.574874371859296e-05,
|
| 2927 |
+
"loss": 0.1484,
|
| 2928 |
+
"step": 18300
|
| 2929 |
+
},
|
| 2930 |
+
{
|
| 2931 |
+
"epoch": 53.04347826086956,
|
| 2932 |
+
"eval_loss": 0.19684499502182007,
|
| 2933 |
+
"eval_runtime": 150.5275,
|
| 2934 |
+
"eval_samples_per_second": 17.299,
|
| 2935 |
+
"eval_steps_per_second": 4.325,
|
| 2936 |
+
"eval_wer": 0.14850618610017105,
|
| 2937 |
+
"step": 18300
|
| 2938 |
+
},
|
| 2939 |
+
{
|
| 2940 |
+
"epoch": 53.333333333333336,
|
| 2941 |
+
"grad_norm": 0.5591597557067871,
|
| 2942 |
+
"learning_rate": 2.424120603015075e-05,
|
| 2943 |
+
"loss": 0.1579,
|
| 2944 |
+
"step": 18400
|
| 2945 |
+
},
|
| 2946 |
+
{
|
| 2947 |
+
"epoch": 53.333333333333336,
|
| 2948 |
+
"eval_loss": 0.19650055468082428,
|
| 2949 |
+
"eval_runtime": 150.0643,
|
| 2950 |
+
"eval_samples_per_second": 17.353,
|
| 2951 |
+
"eval_steps_per_second": 4.338,
|
| 2952 |
+
"eval_wer": 0.14876476906552094,
|
| 2953 |
+
"step": 18400
|
| 2954 |
+
},
|
| 2955 |
+
{
|
| 2956 |
+
"epoch": 53.6231884057971,
|
| 2957 |
+
"grad_norm": 0.5747944712638855,
|
| 2958 |
+
"learning_rate": 2.2733668341708542e-05,
|
| 2959 |
+
"loss": 0.1488,
|
| 2960 |
+
"step": 18500
|
| 2961 |
+
},
|
| 2962 |
+
{
|
| 2963 |
+
"epoch": 53.6231884057971,
|
| 2964 |
+
"eval_loss": 0.19772984087467194,
|
| 2965 |
+
"eval_runtime": 150.2351,
|
| 2966 |
+
"eval_samples_per_second": 17.333,
|
| 2967 |
+
"eval_steps_per_second": 4.333,
|
| 2968 |
+
"eval_wer": 0.14844651310816726,
|
| 2969 |
+
"step": 18500
|
| 2970 |
+
},
|
| 2971 |
+
{
|
| 2972 |
+
"epoch": 53.91304347826087,
|
| 2973 |
+
"grad_norm": 0.7890971899032593,
|
| 2974 |
+
"learning_rate": 2.122613065326633e-05,
|
| 2975 |
+
"loss": 0.1465,
|
| 2976 |
+
"step": 18600
|
| 2977 |
+
},
|
| 2978 |
+
{
|
| 2979 |
+
"epoch": 53.91304347826087,
|
| 2980 |
+
"eval_loss": 0.19878774881362915,
|
| 2981 |
+
"eval_runtime": 149.5777,
|
| 2982 |
+
"eval_samples_per_second": 17.409,
|
| 2983 |
+
"eval_steps_per_second": 4.352,
|
| 2984 |
+
"eval_wer": 0.1486255320841787,
|
| 2985 |
+
"step": 18600
|
| 2986 |
+
},
|
| 2987 |
+
{
|
| 2988 |
+
"epoch": 54.20289855072464,
|
| 2989 |
+
"grad_norm": 0.4769749641418457,
|
| 2990 |
+
"learning_rate": 1.971859296482412e-05,
|
| 2991 |
+
"loss": 0.1562,
|
| 2992 |
+
"step": 18700
|
| 2993 |
+
},
|
| 2994 |
+
{
|
| 2995 |
+
"epoch": 54.20289855072464,
|
| 2996 |
+
"eval_loss": 0.19685682654380798,
|
| 2997 |
+
"eval_runtime": 150.394,
|
| 2998 |
+
"eval_samples_per_second": 17.315,
|
| 2999 |
+
"eval_steps_per_second": 4.329,
|
| 3000 |
+
"eval_wer": 0.14842662211083263,
|
| 3001 |
+
"step": 18700
|
| 3002 |
+
},
|
| 3003 |
+
{
|
| 3004 |
+
"epoch": 54.492753623188406,
|
| 3005 |
+
"grad_norm": 0.5691216588020325,
|
| 3006 |
+
"learning_rate": 1.8211055276381907e-05,
|
| 3007 |
+
"loss": 0.1549,
|
| 3008 |
+
"step": 18800
|
| 3009 |
+
},
|
| 3010 |
+
{
|
| 3011 |
+
"epoch": 54.492753623188406,
|
| 3012 |
+
"eval_loss": 0.19770777225494385,
|
| 3013 |
+
"eval_runtime": 153.5999,
|
| 3014 |
+
"eval_samples_per_second": 16.953,
|
| 3015 |
+
"eval_steps_per_second": 4.238,
|
| 3016 |
+
"eval_wer": 0.14810836615347894,
|
| 3017 |
+
"step": 18800
|
| 3018 |
+
},
|
| 3019 |
+
{
|
| 3020 |
+
"epoch": 54.78260869565217,
|
| 3021 |
+
"grad_norm": 0.4381687343120575,
|
| 3022 |
+
"learning_rate": 1.6703517587939697e-05,
|
| 3023 |
+
"loss": 0.1475,
|
| 3024 |
+
"step": 18900
|
| 3025 |
+
},
|
| 3026 |
+
{
|
| 3027 |
+
"epoch": 54.78260869565217,
|
| 3028 |
+
"eval_loss": 0.19628171622753143,
|
| 3029 |
+
"eval_runtime": 154.2596,
|
| 3030 |
+
"eval_samples_per_second": 16.881,
|
| 3031 |
+
"eval_steps_per_second": 4.22,
|
| 3032 |
+
"eval_wer": 0.1479691291721367,
|
| 3033 |
+
"step": 18900
|
| 3034 |
+
},
|
| 3035 |
+
{
|
| 3036 |
+
"epoch": 55.072463768115945,
|
| 3037 |
+
"grad_norm": 0.5168628692626953,
|
| 3038 |
+
"learning_rate": 1.5195979899497486e-05,
|
| 3039 |
+
"loss": 0.151,
|
| 3040 |
+
"step": 19000
|
| 3041 |
+
},
|
| 3042 |
+
{
|
| 3043 |
+
"epoch": 55.072463768115945,
|
| 3044 |
+
"eval_loss": 0.196857288479805,
|
| 3045 |
+
"eval_runtime": 149.3476,
|
| 3046 |
+
"eval_samples_per_second": 17.436,
|
| 3047 |
+
"eval_steps_per_second": 4.359,
|
| 3048 |
+
"eval_wer": 0.14806858415880972,
|
| 3049 |
+
"step": 19000
|
| 3050 |
+
},
|
| 3051 |
+
{
|
| 3052 |
+
"epoch": 55.36231884057971,
|
| 3053 |
+
"grad_norm": 0.6839106678962708,
|
| 3054 |
+
"learning_rate": 1.3688442211055275e-05,
|
| 3055 |
+
"loss": 0.1501,
|
| 3056 |
+
"step": 19100
|
| 3057 |
+
},
|
| 3058 |
+
{
|
| 3059 |
+
"epoch": 55.36231884057971,
|
| 3060 |
+
"eval_loss": 0.19586588442325592,
|
| 3061 |
+
"eval_runtime": 150.2879,
|
| 3062 |
+
"eval_samples_per_second": 17.327,
|
| 3063 |
+
"eval_steps_per_second": 4.332,
|
| 3064 |
+
"eval_wer": 0.14810836615347894,
|
| 3065 |
+
"step": 19100
|
| 3066 |
+
},
|
| 3067 |
+
{
|
| 3068 |
+
"epoch": 55.65217391304348,
|
| 3069 |
+
"grad_norm": 0.5769901275634766,
|
| 3070 |
+
"learning_rate": 1.2180904522613064e-05,
|
| 3071 |
+
"loss": 0.1528,
|
| 3072 |
+
"step": 19200
|
| 3073 |
+
},
|
| 3074 |
+
{
|
| 3075 |
+
"epoch": 55.65217391304348,
|
| 3076 |
+
"eval_loss": 0.19648610055446625,
|
| 3077 |
+
"eval_runtime": 149.6106,
|
| 3078 |
+
"eval_samples_per_second": 17.405,
|
| 3079 |
+
"eval_steps_per_second": 4.351,
|
| 3080 |
+
"eval_wer": 0.14854596809484027,
|
| 3081 |
+
"step": 19200
|
| 3082 |
+
},
|
| 3083 |
+
{
|
| 3084 |
+
"epoch": 55.94202898550725,
|
| 3085 |
+
"grad_norm": 0.5911151170730591,
|
| 3086 |
+
"learning_rate": 1.0673366834170852e-05,
|
| 3087 |
+
"loss": 0.1435,
|
| 3088 |
+
"step": 19300
|
| 3089 |
+
},
|
| 3090 |
+
{
|
| 3091 |
+
"epoch": 55.94202898550725,
|
| 3092 |
+
"eval_loss": 0.19634607434272766,
|
| 3093 |
+
"eval_runtime": 150.4393,
|
| 3094 |
+
"eval_samples_per_second": 17.309,
|
| 3095 |
+
"eval_steps_per_second": 4.327,
|
| 3096 |
+
"eval_wer": 0.14747185423877154,
|
| 3097 |
+
"step": 19300
|
| 3098 |
+
},
|
| 3099 |
+
{
|
| 3100 |
+
"epoch": 56.231884057971016,
|
| 3101 |
+
"grad_norm": 0.5163128972053528,
|
| 3102 |
+
"learning_rate": 9.165829145728643e-06,
|
| 3103 |
+
"loss": 0.1564,
|
| 3104 |
+
"step": 19400
|
| 3105 |
+
},
|
| 3106 |
+
{
|
| 3107 |
+
"epoch": 56.231884057971016,
|
| 3108 |
+
"eval_loss": 0.19604472815990448,
|
| 3109 |
+
"eval_runtime": 150.9658,
|
| 3110 |
+
"eval_samples_per_second": 17.249,
|
| 3111 |
+
"eval_steps_per_second": 4.312,
|
| 3112 |
+
"eval_wer": 0.1479691291721367,
|
| 3113 |
+
"step": 19400
|
| 3114 |
+
},
|
| 3115 |
+
{
|
| 3116 |
+
"epoch": 56.52173913043478,
|
| 3117 |
+
"grad_norm": 0.6075275540351868,
|
| 3118 |
+
"learning_rate": 7.658291457286432e-06,
|
| 3119 |
+
"loss": 0.1485,
|
| 3120 |
+
"step": 19500
|
| 3121 |
+
},
|
| 3122 |
+
{
|
| 3123 |
+
"epoch": 56.52173913043478,
|
| 3124 |
+
"eval_loss": 0.19642896950244904,
|
| 3125 |
+
"eval_runtime": 151.1813,
|
| 3126 |
+
"eval_samples_per_second": 17.224,
|
| 3127 |
+
"eval_steps_per_second": 4.306,
|
| 3128 |
+
"eval_wer": 0.1476906552094522,
|
| 3129 |
+
"step": 19500
|
| 3130 |
+
},
|
| 3131 |
+
{
|
| 3132 |
+
"epoch": 56.81159420289855,
|
| 3133 |
+
"grad_norm": 0.5421344637870789,
|
| 3134 |
+
"learning_rate": 6.1507537688442204e-06,
|
| 3135 |
+
"loss": 0.1529,
|
| 3136 |
+
"step": 19600
|
| 3137 |
+
},
|
| 3138 |
+
{
|
| 3139 |
+
"epoch": 56.81159420289855,
|
| 3140 |
+
"eval_loss": 0.19620098173618317,
|
| 3141 |
+
"eval_runtime": 150.2568,
|
| 3142 |
+
"eval_samples_per_second": 17.33,
|
| 3143 |
+
"eval_steps_per_second": 4.333,
|
| 3144 |
+
"eval_wer": 0.14745196324143692,
|
| 3145 |
+
"step": 19600
|
| 3146 |
+
},
|
| 3147 |
+
{
|
| 3148 |
+
"epoch": 57.10144927536232,
|
| 3149 |
+
"grad_norm": 0.4514584541320801,
|
| 3150 |
+
"learning_rate": 4.64321608040201e-06,
|
| 3151 |
+
"loss": 0.1485,
|
| 3152 |
+
"step": 19700
|
| 3153 |
+
},
|
| 3154 |
+
{
|
| 3155 |
+
"epoch": 57.10144927536232,
|
| 3156 |
+
"eval_loss": 0.1970347762107849,
|
| 3157 |
+
"eval_runtime": 150.57,
|
| 3158 |
+
"eval_samples_per_second": 17.294,
|
| 3159 |
+
"eval_steps_per_second": 4.324,
|
| 3160 |
+
"eval_wer": 0.14777021919879063,
|
| 3161 |
+
"step": 19700
|
| 3162 |
+
},
|
| 3163 |
+
{
|
| 3164 |
+
"epoch": 57.391304347826086,
|
| 3165 |
+
"grad_norm": 0.5390310287475586,
|
| 3166 |
+
"learning_rate": 3.135678391959799e-06,
|
| 3167 |
+
"loss": 0.1478,
|
| 3168 |
+
"step": 19800
|
| 3169 |
+
},
|
| 3170 |
+
{
|
| 3171 |
+
"epoch": 57.391304347826086,
|
| 3172 |
+
"eval_loss": 0.1968718320131302,
|
| 3173 |
+
"eval_runtime": 149.8169,
|
| 3174 |
+
"eval_samples_per_second": 17.381,
|
| 3175 |
+
"eval_steps_per_second": 4.345,
|
| 3176 |
+
"eval_wer": 0.14775032820145603,
|
| 3177 |
+
"step": 19800
|
| 3178 |
+
},
|
| 3179 |
+
{
|
| 3180 |
+
"epoch": 57.68115942028985,
|
| 3181 |
+
"grad_norm": 0.618834912776947,
|
| 3182 |
+
"learning_rate": 1.6281407035175876e-06,
|
| 3183 |
+
"loss": 0.1501,
|
| 3184 |
+
"step": 19900
|
| 3185 |
+
},
|
| 3186 |
+
{
|
| 3187 |
+
"epoch": 57.68115942028985,
|
| 3188 |
+
"eval_loss": 0.19653910398483276,
|
| 3189 |
+
"eval_runtime": 150.49,
|
| 3190 |
+
"eval_samples_per_second": 17.303,
|
| 3191 |
+
"eval_steps_per_second": 4.326,
|
| 3192 |
+
"eval_wer": 0.14767076421211758,
|
| 3193 |
+
"step": 19900
|
| 3194 |
+
},
|
| 3195 |
+
{
|
| 3196 |
+
"epoch": 57.971014492753625,
|
| 3197 |
+
"grad_norm": 0.5765581727027893,
|
| 3198 |
+
"learning_rate": 1.2060301507537687e-07,
|
| 3199 |
+
"loss": 0.1522,
|
| 3200 |
+
"step": 20000
|
| 3201 |
+
},
|
| 3202 |
+
{
|
| 3203 |
+
"epoch": 57.971014492753625,
|
| 3204 |
+
"eval_loss": 0.19661328196525574,
|
| 3205 |
+
"eval_runtime": 150.3563,
|
| 3206 |
+
"eval_samples_per_second": 17.319,
|
| 3207 |
+
"eval_steps_per_second": 4.33,
|
| 3208 |
+
"eval_wer": 0.14755141822810997,
|
| 3209 |
+
"step": 20000
|
| 3210 |
+
},
|
| 3211 |
+
{
|
| 3212 |
+
"epoch": 57.971014492753625,
|
| 3213 |
+
"step": 20000,
|
| 3214 |
+
"total_flos": 6.201114678692461e+19,
|
| 3215 |
+
"train_loss": 0.3545982142448425,
|
| 3216 |
+
"train_runtime": 69759.9962,
|
| 3217 |
+
"train_samples_per_second": 2.294,
|
| 3218 |
+
"train_steps_per_second": 0.287
|
| 3219 |
+
}
|
| 3220 |
+
],
|
| 3221 |
+
"logging_steps": 100,
|
| 3222 |
+
"max_steps": 20000,
|
| 3223 |
+
"num_input_tokens_seen": 0,
|
| 3224 |
+
"num_train_epochs": 58,
|
| 3225 |
+
"save_steps": 400,
|
| 3226 |
+
"stateful_callbacks": {
|
| 3227 |
+
"TrainerControl": {
|
| 3228 |
+
"args": {
|
| 3229 |
+
"should_epoch_stop": false,
|
| 3230 |
+
"should_evaluate": false,
|
| 3231 |
+
"should_log": false,
|
| 3232 |
+
"should_save": true,
|
| 3233 |
+
"should_training_stop": true
|
| 3234 |
+
},
|
| 3235 |
+
"attributes": {}
|
| 3236 |
+
}
|
| 3237 |
+
},
|
| 3238 |
+
"total_flos": 6.201114678692461e+19,
|
| 3239 |
+
"train_batch_size": 4,
|
| 3240 |
+
"trial_name": null,
|
| 3241 |
+
"trial_params": null
|
| 3242 |
+
}
|