Commit
•
ddfd34c
1
Parent(s):
39b2f20
End of training
Browse files- README.md +3 -1
- all_results.json +15 -0
- eval_results.json +9 -0
- runs/Jun26_08-42-00_b5ff05b1c2c9/events.out.tfevents.1719407922.b5ff05b1c2c9.31.1 +3 -0
- train_results.json +9 -0
- trainer_state.json +1082 -0
README.md
CHANGED
|
@@ -2,6 +2,8 @@
|
|
| 2 |
license: apache-2.0
|
| 3 |
base_model: facebook/wav2vec2-large-xlsr-53
|
| 4 |
tags:
|
|
|
|
|
|
|
| 5 |
- generated_from_trainer
|
| 6 |
metrics:
|
| 7 |
- wer
|
|
@@ -15,7 +17,7 @@ should probably proofread and complete it, then remove this comment. -->
|
|
| 15 |
|
| 16 |
# wav2vec2-xlsr-53-ft-btb-ccv-enc-cy
|
| 17 |
|
| 18 |
-
This model is a fine-tuned version of [facebook/wav2vec2-large-xlsr-53](https://huggingface.co/facebook/wav2vec2-large-xlsr-53) on
|
| 19 |
It achieves the following results on the evaluation set:
|
| 20 |
- Loss: 0.4095
|
| 21 |
- Wer: 0.3271
|
|
|
|
| 2 |
license: apache-2.0
|
| 3 |
base_model: facebook/wav2vec2-large-xlsr-53
|
| 4 |
tags:
|
| 5 |
+
- automatic-speech-recognition
|
| 6 |
+
- DewiBrynJones/banc-trawsgrifiadau-bangor-clean-with-ccv
|
| 7 |
- generated_from_trainer
|
| 8 |
metrics:
|
| 9 |
- wer
|
|
|
|
| 17 |
|
| 18 |
# wav2vec2-xlsr-53-ft-btb-ccv-enc-cy
|
| 19 |
|
| 20 |
+
This model is a fine-tuned version of [facebook/wav2vec2-large-xlsr-53](https://huggingface.co/facebook/wav2vec2-large-xlsr-53) on the DEWIBRYNJONES/BANC-TRAWSGRIFIADAU-BANGOR-CLEAN-WITH-CCV - DEFAULT dataset.
|
| 21 |
It achieves the following results on the evaluation set:
|
| 22 |
- Loss: 0.4095
|
| 23 |
- Wer: 0.3271
|
all_results.json
ADDED
|
@@ -0,0 +1,15 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"epoch": 1.9357336430507162,
|
| 3 |
+
"eval_loss": 0.409473717212677,
|
| 4 |
+
"eval_runtime": 147.313,
|
| 5 |
+
"eval_samples": 5656,
|
| 6 |
+
"eval_samples_per_second": 38.394,
|
| 7 |
+
"eval_steps_per_second": 4.799,
|
| 8 |
+
"eval_wer": 0.3270690568278474,
|
| 9 |
+
"total_flos": 1.1255918428180738e+19,
|
| 10 |
+
"train_loss": 0.7365739318847656,
|
| 11 |
+
"train_runtime": 19473.5524,
|
| 12 |
+
"train_samples": 41326,
|
| 13 |
+
"train_samples_per_second": 4.108,
|
| 14 |
+
"train_steps_per_second": 0.514
|
| 15 |
+
}
|
eval_results.json
ADDED
|
@@ -0,0 +1,9 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"epoch": 1.9357336430507162,
|
| 3 |
+
"eval_loss": 0.409473717212677,
|
| 4 |
+
"eval_runtime": 147.313,
|
| 5 |
+
"eval_samples": 5656,
|
| 6 |
+
"eval_samples_per_second": 38.394,
|
| 7 |
+
"eval_steps_per_second": 4.799,
|
| 8 |
+
"eval_wer": 0.3270690568278474
|
| 9 |
+
}
|
runs/Jun26_08-42-00_b5ff05b1c2c9/events.out.tfevents.1719407922.b5ff05b1c2c9.31.1
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:7dc754c2fbdec3308e9a76e572e132d646e204d217a81300d72cb7aba427f5b1
|
| 3 |
+
size 406
|
train_results.json
ADDED
|
@@ -0,0 +1,9 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"epoch": 1.9357336430507162,
|
| 3 |
+
"total_flos": 1.1255918428180738e+19,
|
| 4 |
+
"train_loss": 0.7365739318847656,
|
| 5 |
+
"train_runtime": 19473.5524,
|
| 6 |
+
"train_samples": 41326,
|
| 7 |
+
"train_samples_per_second": 4.108,
|
| 8 |
+
"train_steps_per_second": 0.514
|
| 9 |
+
}
|
trainer_state.json
ADDED
|
@@ -0,0 +1,1082 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"best_metric": null,
|
| 3 |
+
"best_model_checkpoint": null,
|
| 4 |
+
"epoch": 1.9357336430507162,
|
| 5 |
+
"eval_steps": 100,
|
| 6 |
+
"global_step": 10000,
|
| 7 |
+
"is_hyper_param_search": false,
|
| 8 |
+
"is_local_process_zero": true,
|
| 9 |
+
"is_world_process_zero": true,
|
| 10 |
+
"log_history": [
|
| 11 |
+
{
|
| 12 |
+
"epoch": 0.019357336430507164,
|
| 13 |
+
"eval_loss": 3.547485828399658,
|
| 14 |
+
"eval_runtime": 144.8623,
|
| 15 |
+
"eval_samples_per_second": 39.044,
|
| 16 |
+
"eval_steps_per_second": 4.88,
|
| 17 |
+
"eval_wer": 1.0,
|
| 18 |
+
"step": 100
|
| 19 |
+
},
|
| 20 |
+
{
|
| 21 |
+
"epoch": 0.03871467286101433,
|
| 22 |
+
"eval_loss": 3.0259251594543457,
|
| 23 |
+
"eval_runtime": 142.9174,
|
| 24 |
+
"eval_samples_per_second": 39.575,
|
| 25 |
+
"eval_steps_per_second": 4.947,
|
| 26 |
+
"eval_wer": 1.0,
|
| 27 |
+
"step": 200
|
| 28 |
+
},
|
| 29 |
+
{
|
| 30 |
+
"epoch": 0.05807200929152149,
|
| 31 |
+
"eval_loss": 3.0886833667755127,
|
| 32 |
+
"eval_runtime": 141.7177,
|
| 33 |
+
"eval_samples_per_second": 39.91,
|
| 34 |
+
"eval_steps_per_second": 4.989,
|
| 35 |
+
"eval_wer": 1.0,
|
| 36 |
+
"step": 300
|
| 37 |
+
},
|
| 38 |
+
{
|
| 39 |
+
"epoch": 0.07742934572202866,
|
| 40 |
+
"eval_loss": 2.3821566104888916,
|
| 41 |
+
"eval_runtime": 143.4279,
|
| 42 |
+
"eval_samples_per_second": 39.434,
|
| 43 |
+
"eval_steps_per_second": 4.929,
|
| 44 |
+
"eval_wer": 0.9971915071175234,
|
| 45 |
+
"step": 400
|
| 46 |
+
},
|
| 47 |
+
{
|
| 48 |
+
"epoch": 0.09678668215253582,
|
| 49 |
+
"grad_norm": 3.033390760421753,
|
| 50 |
+
"learning_rate": 0.0002982,
|
| 51 |
+
"loss": 4.0938,
|
| 52 |
+
"step": 500
|
| 53 |
+
},
|
| 54 |
+
{
|
| 55 |
+
"epoch": 0.09678668215253582,
|
| 56 |
+
"eval_loss": 1.4546788930892944,
|
| 57 |
+
"eval_runtime": 142.7727,
|
| 58 |
+
"eval_samples_per_second": 39.615,
|
| 59 |
+
"eval_steps_per_second": 4.952,
|
| 60 |
+
"eval_wer": 0.9020076711977019,
|
| 61 |
+
"step": 500
|
| 62 |
+
},
|
| 63 |
+
{
|
| 64 |
+
"epoch": 0.11614401858304298,
|
| 65 |
+
"eval_loss": 1.2602813243865967,
|
| 66 |
+
"eval_runtime": 143.0991,
|
| 67 |
+
"eval_samples_per_second": 39.525,
|
| 68 |
+
"eval_steps_per_second": 4.941,
|
| 69 |
+
"eval_wer": 0.8509733433904126,
|
| 70 |
+
"step": 600
|
| 71 |
+
},
|
| 72 |
+
{
|
| 73 |
+
"epoch": 0.13550135501355012,
|
| 74 |
+
"eval_loss": 1.0939536094665527,
|
| 75 |
+
"eval_runtime": 145.4158,
|
| 76 |
+
"eval_samples_per_second": 38.895,
|
| 77 |
+
"eval_steps_per_second": 4.862,
|
| 78 |
+
"eval_wer": 0.7654667715170677,
|
| 79 |
+
"step": 700
|
| 80 |
+
},
|
| 81 |
+
{
|
| 82 |
+
"epoch": 0.1548586914440573,
|
| 83 |
+
"eval_loss": 1.0704576969146729,
|
| 84 |
+
"eval_runtime": 148.1277,
|
| 85 |
+
"eval_samples_per_second": 38.183,
|
| 86 |
+
"eval_steps_per_second": 4.773,
|
| 87 |
+
"eval_wer": 0.7601547078364975,
|
| 88 |
+
"step": 800
|
| 89 |
+
},
|
| 90 |
+
{
|
| 91 |
+
"epoch": 0.17421602787456447,
|
| 92 |
+
"eval_loss": 0.9356458187103271,
|
| 93 |
+
"eval_runtime": 143.6296,
|
| 94 |
+
"eval_samples_per_second": 39.379,
|
| 95 |
+
"eval_steps_per_second": 4.922,
|
| 96 |
+
"eval_wer": 0.6972926128612925,
|
| 97 |
+
"step": 900
|
| 98 |
+
},
|
| 99 |
+
{
|
| 100 |
+
"epoch": 0.19357336430507163,
|
| 101 |
+
"grad_norm": 3.2989861965179443,
|
| 102 |
+
"learning_rate": 0.0002843684210526315,
|
| 103 |
+
"loss": 1.0597,
|
| 104 |
+
"step": 1000
|
| 105 |
+
},
|
| 106 |
+
{
|
| 107 |
+
"epoch": 0.19357336430507163,
|
| 108 |
+
"eval_loss": 0.9103516936302185,
|
| 109 |
+
"eval_runtime": 146.7237,
|
| 110 |
+
"eval_samples_per_second": 38.549,
|
| 111 |
+
"eval_steps_per_second": 4.819,
|
| 112 |
+
"eval_wer": 0.6765579111232367,
|
| 113 |
+
"step": 1000
|
| 114 |
+
},
|
| 115 |
+
{
|
| 116 |
+
"epoch": 0.2129307007355788,
|
| 117 |
+
"eval_loss": 0.8879104256629944,
|
| 118 |
+
"eval_runtime": 153.6385,
|
| 119 |
+
"eval_samples_per_second": 36.814,
|
| 120 |
+
"eval_steps_per_second": 4.602,
|
| 121 |
+
"eval_wer": 0.6569947521304424,
|
| 122 |
+
"step": 1100
|
| 123 |
+
},
|
| 124 |
+
{
|
| 125 |
+
"epoch": 0.23228803716608595,
|
| 126 |
+
"eval_loss": 0.8594633936882019,
|
| 127 |
+
"eval_runtime": 147.9212,
|
| 128 |
+
"eval_samples_per_second": 38.237,
|
| 129 |
+
"eval_steps_per_second": 4.78,
|
| 130 |
+
"eval_wer": 0.6611834186580219,
|
| 131 |
+
"step": 1200
|
| 132 |
+
},
|
| 133 |
+
{
|
| 134 |
+
"epoch": 0.2516453735965931,
|
| 135 |
+
"eval_loss": 0.8351845145225525,
|
| 136 |
+
"eval_runtime": 148.1861,
|
| 137 |
+
"eval_samples_per_second": 38.168,
|
| 138 |
+
"eval_steps_per_second": 4.771,
|
| 139 |
+
"eval_wer": 0.6075331803373402,
|
| 140 |
+
"step": 1300
|
| 141 |
+
},
|
| 142 |
+
{
|
| 143 |
+
"epoch": 0.27100271002710025,
|
| 144 |
+
"eval_loss": 0.791232168674469,
|
| 145 |
+
"eval_runtime": 148.9749,
|
| 146 |
+
"eval_samples_per_second": 37.966,
|
| 147 |
+
"eval_steps_per_second": 4.746,
|
| 148 |
+
"eval_wer": 0.6033124167482466,
|
| 149 |
+
"step": 1400
|
| 150 |
+
},
|
| 151 |
+
{
|
| 152 |
+
"epoch": 0.29036004645760743,
|
| 153 |
+
"grad_norm": 6.474522113800049,
|
| 154 |
+
"learning_rate": 0.000268578947368421,
|
| 155 |
+
"loss": 0.8484,
|
| 156 |
+
"step": 1500
|
| 157 |
+
},
|
| 158 |
+
{
|
| 159 |
+
"epoch": 0.29036004645760743,
|
| 160 |
+
"eval_loss": 0.7862286567687988,
|
| 161 |
+
"eval_runtime": 146.7521,
|
| 162 |
+
"eval_samples_per_second": 38.541,
|
| 163 |
+
"eval_steps_per_second": 4.818,
|
| 164 |
+
"eval_wer": 0.6067468023302467,
|
| 165 |
+
"step": 1500
|
| 166 |
+
},
|
| 167 |
+
{
|
| 168 |
+
"epoch": 0.3097173828881146,
|
| 169 |
+
"eval_loss": 0.7790109515190125,
|
| 170 |
+
"eval_runtime": 147.7062,
|
| 171 |
+
"eval_samples_per_second": 38.292,
|
| 172 |
+
"eval_steps_per_second": 4.787,
|
| 173 |
+
"eval_wer": 0.6009051371346953,
|
| 174 |
+
"step": 1600
|
| 175 |
+
},
|
| 176 |
+
{
|
| 177 |
+
"epoch": 0.32907471931862176,
|
| 178 |
+
"eval_loss": 0.7678210735321045,
|
| 179 |
+
"eval_runtime": 148.3951,
|
| 180 |
+
"eval_samples_per_second": 38.114,
|
| 181 |
+
"eval_steps_per_second": 4.764,
|
| 182 |
+
"eval_wer": 0.5629182648328546,
|
| 183 |
+
"step": 1700
|
| 184 |
+
},
|
| 185 |
+
{
|
| 186 |
+
"epoch": 0.34843205574912894,
|
| 187 |
+
"eval_loss": 0.7514644861221313,
|
| 188 |
+
"eval_runtime": 149.2674,
|
| 189 |
+
"eval_samples_per_second": 37.892,
|
| 190 |
+
"eval_steps_per_second": 4.736,
|
| 191 |
+
"eval_wer": 0.5798655133122562,
|
| 192 |
+
"step": 1800
|
| 193 |
+
},
|
| 194 |
+
{
|
| 195 |
+
"epoch": 0.3677893921796361,
|
| 196 |
+
"eval_loss": 0.7423551678657532,
|
| 197 |
+
"eval_runtime": 149.3427,
|
| 198 |
+
"eval_samples_per_second": 37.873,
|
| 199 |
+
"eval_steps_per_second": 4.734,
|
| 200 |
+
"eval_wer": 0.5859158094076488,
|
| 201 |
+
"step": 1900
|
| 202 |
+
},
|
| 203 |
+
{
|
| 204 |
+
"epoch": 0.38714672861014326,
|
| 205 |
+
"grad_norm": 2.573913335800171,
|
| 206 |
+
"learning_rate": 0.0002527894736842105,
|
| 207 |
+
"loss": 0.764,
|
| 208 |
+
"step": 2000
|
| 209 |
+
},
|
| 210 |
+
{
|
| 211 |
+
"epoch": 0.38714672861014326,
|
| 212 |
+
"eval_loss": 0.7129915356636047,
|
| 213 |
+
"eval_runtime": 148.3711,
|
| 214 |
+
"eval_samples_per_second": 38.121,
|
| 215 |
+
"eval_steps_per_second": 4.765,
|
| 216 |
+
"eval_wer": 0.5520855065718734,
|
| 217 |
+
"step": 2000
|
| 218 |
+
},
|
| 219 |
+
{
|
| 220 |
+
"epoch": 0.4065040650406504,
|
| 221 |
+
"eval_loss": 0.7114368677139282,
|
| 222 |
+
"eval_runtime": 148.2007,
|
| 223 |
+
"eval_samples_per_second": 38.164,
|
| 224 |
+
"eval_steps_per_second": 4.771,
|
| 225 |
+
"eval_wer": 0.5407712923881819,
|
| 226 |
+
"step": 2100
|
| 227 |
+
},
|
| 228 |
+
{
|
| 229 |
+
"epoch": 0.4258614014711576,
|
| 230 |
+
"eval_loss": 0.7228682637214661,
|
| 231 |
+
"eval_runtime": 149.1432,
|
| 232 |
+
"eval_samples_per_second": 37.923,
|
| 233 |
+
"eval_steps_per_second": 4.74,
|
| 234 |
+
"eval_wer": 0.5577024923368266,
|
| 235 |
+
"step": 2200
|
| 236 |
+
},
|
| 237 |
+
{
|
| 238 |
+
"epoch": 0.4452187379016647,
|
| 239 |
+
"eval_loss": 0.6773180961608887,
|
| 240 |
+
"eval_runtime": 154.221,
|
| 241 |
+
"eval_samples_per_second": 36.675,
|
| 242 |
+
"eval_steps_per_second": 4.584,
|
| 243 |
+
"eval_wer": 0.5160084094301167,
|
| 244 |
+
"step": 2300
|
| 245 |
+
},
|
| 246 |
+
{
|
| 247 |
+
"epoch": 0.4645760743321719,
|
| 248 |
+
"eval_loss": 0.6784498691558838,
|
| 249 |
+
"eval_runtime": 149.2744,
|
| 250 |
+
"eval_samples_per_second": 37.89,
|
| 251 |
+
"eval_steps_per_second": 4.736,
|
| 252 |
+
"eval_wer": 0.5177897963441447,
|
| 253 |
+
"step": 2400
|
| 254 |
+
},
|
| 255 |
+
{
|
| 256 |
+
"epoch": 0.48393341076267904,
|
| 257 |
+
"grad_norm": 3.0869553089141846,
|
| 258 |
+
"learning_rate": 0.000237,
|
| 259 |
+
"loss": 0.6868,
|
| 260 |
+
"step": 2500
|
| 261 |
+
},
|
| 262 |
+
{
|
| 263 |
+
"epoch": 0.48393341076267904,
|
| 264 |
+
"eval_loss": 0.672030508518219,
|
| 265 |
+
"eval_runtime": 149.4453,
|
| 266 |
+
"eval_samples_per_second": 37.847,
|
| 267 |
+
"eval_steps_per_second": 4.731,
|
| 268 |
+
"eval_wer": 0.5261831779300605,
|
| 269 |
+
"step": 2500
|
| 270 |
+
},
|
| 271 |
+
{
|
| 272 |
+
"epoch": 0.5032907471931862,
|
| 273 |
+
"eval_loss": 0.6804332137107849,
|
| 274 |
+
"eval_runtime": 151.0327,
|
| 275 |
+
"eval_samples_per_second": 37.449,
|
| 276 |
+
"eval_steps_per_second": 4.681,
|
| 277 |
+
"eval_wer": 0.5336617932628268,
|
| 278 |
+
"step": 2600
|
| 279 |
+
},
|
| 280 |
+
{
|
| 281 |
+
"epoch": 0.5226480836236934,
|
| 282 |
+
"eval_loss": 0.6598911285400391,
|
| 283 |
+
"eval_runtime": 149.0299,
|
| 284 |
+
"eval_samples_per_second": 37.952,
|
| 285 |
+
"eval_steps_per_second": 4.744,
|
| 286 |
+
"eval_wer": 0.5023832068174159,
|
| 287 |
+
"step": 2700
|
| 288 |
+
},
|
| 289 |
+
{
|
| 290 |
+
"epoch": 0.5420054200542005,
|
| 291 |
+
"eval_loss": 0.6287100911140442,
|
| 292 |
+
"eval_runtime": 149.9845,
|
| 293 |
+
"eval_samples_per_second": 37.711,
|
| 294 |
+
"eval_steps_per_second": 4.714,
|
| 295 |
+
"eval_wer": 0.4902023719728459,
|
| 296 |
+
"step": 2800
|
| 297 |
+
},
|
| 298 |
+
{
|
| 299 |
+
"epoch": 0.5613627564847077,
|
| 300 |
+
"eval_loss": 0.6304338574409485,
|
| 301 |
+
"eval_runtime": 150.016,
|
| 302 |
+
"eval_samples_per_second": 37.703,
|
| 303 |
+
"eval_steps_per_second": 4.713,
|
| 304 |
+
"eval_wer": 0.49471200911556545,
|
| 305 |
+
"step": 2900
|
| 306 |
+
},
|
| 307 |
+
{
|
| 308 |
+
"epoch": 0.5807200929152149,
|
| 309 |
+
"grad_norm": 5.678714275360107,
|
| 310 |
+
"learning_rate": 0.00022121052631578946,
|
| 311 |
+
"loss": 0.6761,
|
| 312 |
+
"step": 3000
|
| 313 |
+
},
|
| 314 |
+
{
|
| 315 |
+
"epoch": 0.5807200929152149,
|
| 316 |
+
"eval_loss": 0.6258472204208374,
|
| 317 |
+
"eval_runtime": 149.8088,
|
| 318 |
+
"eval_samples_per_second": 37.755,
|
| 319 |
+
"eval_steps_per_second": 4.719,
|
| 320 |
+
"eval_wer": 0.48513103625363097,
|
| 321 |
+
"step": 3000
|
| 322 |
+
},
|
| 323 |
+
{
|
| 324 |
+
"epoch": 0.6000774293457221,
|
| 325 |
+
"eval_loss": 0.6310975551605225,
|
| 326 |
+
"eval_runtime": 148.9286,
|
| 327 |
+
"eval_samples_per_second": 37.978,
|
| 328 |
+
"eval_steps_per_second": 4.747,
|
| 329 |
+
"eval_wer": 0.4989809182969299,
|
| 330 |
+
"step": 3100
|
| 331 |
+
},
|
| 332 |
+
{
|
| 333 |
+
"epoch": 0.6194347657762292,
|
| 334 |
+
"eval_loss": 0.6171565651893616,
|
| 335 |
+
"eval_runtime": 148.6924,
|
| 336 |
+
"eval_samples_per_second": 38.038,
|
| 337 |
+
"eval_steps_per_second": 4.755,
|
| 338 |
+
"eval_wer": 0.4901060807883038,
|
| 339 |
+
"step": 3200
|
| 340 |
+
},
|
| 341 |
+
{
|
| 342 |
+
"epoch": 0.6387921022067363,
|
| 343 |
+
"eval_loss": 0.6187321543693542,
|
| 344 |
+
"eval_runtime": 149.7679,
|
| 345 |
+
"eval_samples_per_second": 37.765,
|
| 346 |
+
"eval_steps_per_second": 4.721,
|
| 347 |
+
"eval_wer": 0.46661103176004237,
|
| 348 |
+
"step": 3300
|
| 349 |
+
},
|
| 350 |
+
{
|
| 351 |
+
"epoch": 0.6581494386372435,
|
| 352 |
+
"eval_loss": 0.6044796109199524,
|
| 353 |
+
"eval_runtime": 149.5983,
|
| 354 |
+
"eval_samples_per_second": 37.808,
|
| 355 |
+
"eval_steps_per_second": 4.726,
|
| 356 |
+
"eval_wer": 0.4725489881401358,
|
| 357 |
+
"step": 3400
|
| 358 |
+
},
|
| 359 |
+
{
|
| 360 |
+
"epoch": 0.6775067750677507,
|
| 361 |
+
"grad_norm": 4.122500419616699,
|
| 362 |
+
"learning_rate": 0.00020542105263157893,
|
| 363 |
+
"loss": 0.6462,
|
| 364 |
+
"step": 3500
|
| 365 |
+
},
|
| 366 |
+
{
|
| 367 |
+
"epoch": 0.6775067750677507,
|
| 368 |
+
"eval_loss": 0.5950499773025513,
|
| 369 |
+
"eval_runtime": 148.2511,
|
| 370 |
+
"eval_samples_per_second": 38.151,
|
| 371 |
+
"eval_steps_per_second": 4.769,
|
| 372 |
+
"eval_wer": 0.4716823674792573,
|
| 373 |
+
"step": 3500
|
| 374 |
+
},
|
| 375 |
+
{
|
| 376 |
+
"epoch": 0.6968641114982579,
|
| 377 |
+
"eval_loss": 0.5902624726295471,
|
| 378 |
+
"eval_runtime": 149.3094,
|
| 379 |
+
"eval_samples_per_second": 37.881,
|
| 380 |
+
"eval_steps_per_second": 4.735,
|
| 381 |
+
"eval_wer": 0.4602237165187527,
|
| 382 |
+
"step": 3600
|
| 383 |
+
},
|
| 384 |
+
{
|
| 385 |
+
"epoch": 0.716221447928765,
|
| 386 |
+
"eval_loss": 0.5864866375923157,
|
| 387 |
+
"eval_runtime": 149.6434,
|
| 388 |
+
"eval_samples_per_second": 37.797,
|
| 389 |
+
"eval_steps_per_second": 4.725,
|
| 390 |
+
"eval_wer": 0.47267737638619184,
|
| 391 |
+
"step": 3700
|
| 392 |
+
},
|
| 393 |
+
{
|
| 394 |
+
"epoch": 0.7355787843592722,
|
| 395 |
+
"eval_loss": 0.5820363759994507,
|
| 396 |
+
"eval_runtime": 148.886,
|
| 397 |
+
"eval_samples_per_second": 37.989,
|
| 398 |
+
"eval_steps_per_second": 4.749,
|
| 399 |
+
"eval_wer": 0.459036125242734,
|
| 400 |
+
"step": 3800
|
| 401 |
+
},
|
| 402 |
+
{
|
| 403 |
+
"epoch": 0.7549361207897793,
|
| 404 |
+
"eval_loss": 0.6025602221488953,
|
| 405 |
+
"eval_runtime": 148.9627,
|
| 406 |
+
"eval_samples_per_second": 37.969,
|
| 407 |
+
"eval_steps_per_second": 4.746,
|
| 408 |
+
"eval_wer": 0.48296448460143476,
|
| 409 |
+
"step": 3900
|
| 410 |
+
},
|
| 411 |
+
{
|
| 412 |
+
"epoch": 0.7742934572202865,
|
| 413 |
+
"grad_norm": 5.146019458770752,
|
| 414 |
+
"learning_rate": 0.0001896315789473684,
|
| 415 |
+
"loss": 0.6193,
|
| 416 |
+
"step": 4000
|
| 417 |
+
},
|
| 418 |
+
{
|
| 419 |
+
"epoch": 0.7742934572202865,
|
| 420 |
+
"eval_loss": 0.5807139277458191,
|
| 421 |
+
"eval_runtime": 147.9966,
|
| 422 |
+
"eval_samples_per_second": 38.217,
|
| 423 |
+
"eval_steps_per_second": 4.777,
|
| 424 |
+
"eval_wer": 0.44963168621912664,
|
| 425 |
+
"step": 4000
|
| 426 |
+
},
|
| 427 |
+
{
|
| 428 |
+
"epoch": 0.7936507936507936,
|
| 429 |
+
"eval_loss": 0.5620962977409363,
|
| 430 |
+
"eval_runtime": 148.8391,
|
| 431 |
+
"eval_samples_per_second": 38.001,
|
| 432 |
+
"eval_steps_per_second": 4.75,
|
| 433 |
+
"eval_wer": 0.44857248318916404,
|
| 434 |
+
"step": 4100
|
| 435 |
+
},
|
| 436 |
+
{
|
| 437 |
+
"epoch": 0.8130081300813008,
|
| 438 |
+
"eval_loss": 0.5730157494544983,
|
| 439 |
+
"eval_runtime": 148.9808,
|
| 440 |
+
"eval_samples_per_second": 37.965,
|
| 441 |
+
"eval_steps_per_second": 4.746,
|
| 442 |
+
"eval_wer": 0.4593410473271172,
|
| 443 |
+
"step": 4200
|
| 444 |
+
},
|
| 445 |
+
{
|
| 446 |
+
"epoch": 0.832365466511808,
|
| 447 |
+
"eval_loss": 0.5592055916786194,
|
| 448 |
+
"eval_runtime": 147.8897,
|
| 449 |
+
"eval_samples_per_second": 38.245,
|
| 450 |
+
"eval_steps_per_second": 4.781,
|
| 451 |
+
"eval_wer": 0.43741875431304267,
|
| 452 |
+
"step": 4300
|
| 453 |
+
},
|
| 454 |
+
{
|
| 455 |
+
"epoch": 0.8517228029423152,
|
| 456 |
+
"eval_loss": 0.5621338486671448,
|
| 457 |
+
"eval_runtime": 148.7799,
|
| 458 |
+
"eval_samples_per_second": 38.016,
|
| 459 |
+
"eval_steps_per_second": 4.752,
|
| 460 |
+
"eval_wer": 0.42387379435412686,
|
| 461 |
+
"step": 4400
|
| 462 |
+
},
|
| 463 |
+
{
|
| 464 |
+
"epoch": 0.8710801393728222,
|
| 465 |
+
"grad_norm": 2.8218295574188232,
|
| 466 |
+
"learning_rate": 0.0001738421052631579,
|
| 467 |
+
"loss": 0.59,
|
| 468 |
+
"step": 4500
|
| 469 |
+
},
|
| 470 |
+
{
|
| 471 |
+
"epoch": 0.8710801393728222,
|
| 472 |
+
"eval_loss": 0.545798659324646,
|
| 473 |
+
"eval_runtime": 150.2397,
|
| 474 |
+
"eval_samples_per_second": 37.647,
|
| 475 |
+
"eval_steps_per_second": 4.706,
|
| 476 |
+
"eval_wer": 0.4304055463722296,
|
| 477 |
+
"step": 4500
|
| 478 |
+
},
|
| 479 |
+
{
|
| 480 |
+
"epoch": 0.8904374758033294,
|
| 481 |
+
"eval_loss": 0.5406409502029419,
|
| 482 |
+
"eval_runtime": 148.931,
|
| 483 |
+
"eval_samples_per_second": 37.977,
|
| 484 |
+
"eval_steps_per_second": 4.747,
|
| 485 |
+
"eval_wer": 0.4270674519747717,
|
| 486 |
+
"step": 4600
|
| 487 |
+
},
|
| 488 |
+
{
|
| 489 |
+
"epoch": 0.9097948122338366,
|
| 490 |
+
"eval_loss": 0.5268651247024536,
|
| 491 |
+
"eval_runtime": 148.7725,
|
| 492 |
+
"eval_samples_per_second": 38.018,
|
| 493 |
+
"eval_steps_per_second": 4.752,
|
| 494 |
+
"eval_wer": 0.41315337580844474,
|
| 495 |
+
"step": 4700
|
| 496 |
+
},
|
| 497 |
+
{
|
| 498 |
+
"epoch": 0.9291521486643438,
|
| 499 |
+
"eval_loss": 0.5362106561660767,
|
| 500 |
+
"eval_runtime": 147.8165,
|
| 501 |
+
"eval_samples_per_second": 38.264,
|
| 502 |
+
"eval_steps_per_second": 4.783,
|
| 503 |
+
"eval_wer": 0.4214665147405755,
|
| 504 |
+
"step": 4800
|
| 505 |
+
},
|
| 506 |
+
{
|
| 507 |
+
"epoch": 0.948509485094851,
|
| 508 |
+
"eval_loss": 0.5226009488105774,
|
| 509 |
+
"eval_runtime": 149.4387,
|
| 510 |
+
"eval_samples_per_second": 37.848,
|
| 511 |
+
"eval_steps_per_second": 4.731,
|
| 512 |
+
"eval_wer": 0.41626679077530454,
|
| 513 |
+
"step": 4900
|
| 514 |
+
},
|
| 515 |
+
{
|
| 516 |
+
"epoch": 0.9678668215253581,
|
| 517 |
+
"grad_norm": 7.241621017456055,
|
| 518 |
+
"learning_rate": 0.00015808421052631577,
|
| 519 |
+
"loss": 0.5636,
|
| 520 |
+
"step": 5000
|
| 521 |
+
},
|
| 522 |
+
{
|
| 523 |
+
"epoch": 0.9678668215253581,
|
| 524 |
+
"eval_loss": 0.5297274589538574,
|
| 525 |
+
"eval_runtime": 149.3726,
|
| 526 |
+
"eval_samples_per_second": 37.865,
|
| 527 |
+
"eval_steps_per_second": 4.733,
|
| 528 |
+
"eval_wer": 0.4148384715379307,
|
| 529 |
+
"step": 5000
|
| 530 |
+
},
|
| 531 |
+
{
|
| 532 |
+
"epoch": 0.9872241579558653,
|
| 533 |
+
"eval_loss": 0.5225785970687866,
|
| 534 |
+
"eval_runtime": 149.3039,
|
| 535 |
+
"eval_samples_per_second": 37.882,
|
| 536 |
+
"eval_steps_per_second": 4.735,
|
| 537 |
+
"eval_wer": 0.413634831731155,
|
| 538 |
+
"step": 5100
|
| 539 |
+
},
|
| 540 |
+
{
|
| 541 |
+
"epoch": 1.0065814943863725,
|
| 542 |
+
"eval_loss": 0.5239331722259521,
|
| 543 |
+
"eval_runtime": 149.3008,
|
| 544 |
+
"eval_samples_per_second": 37.883,
|
| 545 |
+
"eval_steps_per_second": 4.735,
|
| 546 |
+
"eval_wer": 0.4054179839835663,
|
| 547 |
+
"step": 5200
|
| 548 |
+
},
|
| 549 |
+
{
|
| 550 |
+
"epoch": 1.0259388308168795,
|
| 551 |
+
"eval_loss": 0.5382751226425171,
|
| 552 |
+
"eval_runtime": 148.0837,
|
| 553 |
+
"eval_samples_per_second": 38.195,
|
| 554 |
+
"eval_steps_per_second": 4.774,
|
| 555 |
+
"eval_wer": 0.4057871001909775,
|
| 556 |
+
"step": 5300
|
| 557 |
+
},
|
| 558 |
+
{
|
| 559 |
+
"epoch": 1.0452961672473868,
|
| 560 |
+
"eval_loss": 0.5125272274017334,
|
| 561 |
+
"eval_runtime": 149.1205,
|
| 562 |
+
"eval_samples_per_second": 37.929,
|
| 563 |
+
"eval_steps_per_second": 4.741,
|
| 564 |
+
"eval_wer": 0.4067179149748841,
|
| 565 |
+
"step": 5400
|
| 566 |
+
},
|
| 567 |
+
{
|
| 568 |
+
"epoch": 1.064653503677894,
|
| 569 |
+
"grad_norm": 1.124423623085022,
|
| 570 |
+
"learning_rate": 0.00014232631578947366,
|
| 571 |
+
"loss": 0.4924,
|
| 572 |
+
"step": 5500
|
| 573 |
+
},
|
| 574 |
+
{
|
| 575 |
+
"epoch": 1.064653503677894,
|
| 576 |
+
"eval_loss": 0.5029215812683105,
|
| 577 |
+
"eval_runtime": 147.7988,
|
| 578 |
+
"eval_samples_per_second": 38.268,
|
| 579 |
+
"eval_steps_per_second": 4.784,
|
| 580 |
+
"eval_wer": 0.39533950666816453,
|
| 581 |
+
"step": 5500
|
| 582 |
+
},
|
| 583 |
+
{
|
| 584 |
+
"epoch": 1.084010840108401,
|
| 585 |
+
"eval_loss": 0.505442202091217,
|
| 586 |
+
"eval_runtime": 149.3815,
|
| 587 |
+
"eval_samples_per_second": 37.863,
|
| 588 |
+
"eval_steps_per_second": 4.733,
|
| 589 |
+
"eval_wer": 0.3932050520774823,
|
| 590 |
+
"step": 5600
|
| 591 |
+
},
|
| 592 |
+
{
|
| 593 |
+
"epoch": 1.1033681765389083,
|
| 594 |
+
"eval_loss": 0.4968744218349457,
|
| 595 |
+
"eval_runtime": 149.7956,
|
| 596 |
+
"eval_samples_per_second": 37.758,
|
| 597 |
+
"eval_steps_per_second": 4.72,
|
| 598 |
+
"eval_wer": 0.3894015502880711,
|
| 599 |
+
"step": 5700
|
| 600 |
+
},
|
| 601 |
+
{
|
| 602 |
+
"epoch": 1.1227255129694154,
|
| 603 |
+
"eval_loss": 0.49354633688926697,
|
| 604 |
+
"eval_runtime": 148.4196,
|
| 605 |
+
"eval_samples_per_second": 38.108,
|
| 606 |
+
"eval_steps_per_second": 4.764,
|
| 607 |
+
"eval_wer": 0.38508449551443563,
|
| 608 |
+
"step": 5800
|
| 609 |
+
},
|
| 610 |
+
{
|
| 611 |
+
"epoch": 1.1420828493999227,
|
| 612 |
+
"eval_loss": 0.49766939878463745,
|
| 613 |
+
"eval_runtime": 149.5302,
|
| 614 |
+
"eval_samples_per_second": 37.825,
|
| 615 |
+
"eval_steps_per_second": 4.728,
|
| 616 |
+
"eval_wer": 0.3816501099324357,
|
| 617 |
+
"step": 5900
|
| 618 |
+
},
|
| 619 |
+
{
|
| 620 |
+
"epoch": 1.1614401858304297,
|
| 621 |
+
"grad_norm": 1.9677255153656006,
|
| 622 |
+
"learning_rate": 0.00012653684210526316,
|
| 623 |
+
"loss": 0.4602,
|
| 624 |
+
"step": 6000
|
| 625 |
+
},
|
| 626 |
+
{
|
| 627 |
+
"epoch": 1.1614401858304297,
|
| 628 |
+
"eval_loss": 0.4862758219242096,
|
| 629 |
+
"eval_runtime": 150.9135,
|
| 630 |
+
"eval_samples_per_second": 37.478,
|
| 631 |
+
"eval_steps_per_second": 4.685,
|
| 632 |
+
"eval_wer": 0.387395483943445,
|
| 633 |
+
"step": 6000
|
| 634 |
+
},
|
| 635 |
+
{
|
| 636 |
+
"epoch": 1.1807975222609368,
|
| 637 |
+
"eval_loss": 0.4906172454357147,
|
| 638 |
+
"eval_runtime": 148.5353,
|
| 639 |
+
"eval_samples_per_second": 38.078,
|
| 640 |
+
"eval_steps_per_second": 4.76,
|
| 641 |
+
"eval_wer": 0.3776700743046974,
|
| 642 |
+
"step": 6100
|
| 643 |
+
},
|
| 644 |
+
{
|
| 645 |
+
"epoch": 1.2001548586914441,
|
| 646 |
+
"eval_loss": 0.4891129434108734,
|
| 647 |
+
"eval_runtime": 149.6289,
|
| 648 |
+
"eval_samples_per_second": 37.8,
|
| 649 |
+
"eval_steps_per_second": 4.725,
|
| 650 |
+
"eval_wer": 0.3763861918441367,
|
| 651 |
+
"step": 6200
|
| 652 |
+
},
|
| 653 |
+
{
|
| 654 |
+
"epoch": 1.2195121951219512,
|
| 655 |
+
"eval_loss": 0.488125741481781,
|
| 656 |
+
"eval_runtime": 148.288,
|
| 657 |
+
"eval_samples_per_second": 38.142,
|
| 658 |
+
"eval_steps_per_second": 4.768,
|
| 659 |
+
"eval_wer": 0.3800934024490058,
|
| 660 |
+
"step": 6300
|
| 661 |
+
},
|
| 662 |
+
{
|
| 663 |
+
"epoch": 1.2388695315524583,
|
| 664 |
+
"eval_loss": 0.48135778307914734,
|
| 665 |
+
"eval_runtime": 147.6048,
|
| 666 |
+
"eval_samples_per_second": 38.319,
|
| 667 |
+
"eval_steps_per_second": 4.79,
|
| 668 |
+
"eval_wer": 0.37266293270851053,
|
| 669 |
+
"step": 6400
|
| 670 |
+
},
|
| 671 |
+
{
|
| 672 |
+
"epoch": 1.2582268679829656,
|
| 673 |
+
"grad_norm": 1.2907174825668335,
|
| 674 |
+
"learning_rate": 0.00011074736842105263,
|
| 675 |
+
"loss": 0.4407,
|
| 676 |
+
"step": 6500
|
| 677 |
+
},
|
| 678 |
+
{
|
| 679 |
+
"epoch": 1.2582268679829656,
|
| 680 |
+
"eval_loss": 0.47142112255096436,
|
| 681 |
+
"eval_runtime": 147.9064,
|
| 682 |
+
"eval_samples_per_second": 38.24,
|
| 683 |
+
"eval_steps_per_second": 4.78,
|
| 684 |
+
"eval_wer": 0.37723676397425815,
|
| 685 |
+
"step": 6500
|
| 686 |
+
},
|
| 687 |
+
{
|
| 688 |
+
"epoch": 1.2775842044134726,
|
| 689 |
+
"eval_loss": 0.47389352321624756,
|
| 690 |
+
"eval_runtime": 146.8545,
|
| 691 |
+
"eval_samples_per_second": 38.514,
|
| 692 |
+
"eval_steps_per_second": 4.814,
|
| 693 |
+
"eval_wer": 0.3705605751793423,
|
| 694 |
+
"step": 6600
|
| 695 |
+
},
|
| 696 |
+
{
|
| 697 |
+
"epoch": 1.29694154084398,
|
| 698 |
+
"eval_loss": 0.4691925644874573,
|
| 699 |
+
"eval_runtime": 146.6568,
|
| 700 |
+
"eval_samples_per_second": 38.566,
|
| 701 |
+
"eval_steps_per_second": 4.821,
|
| 702 |
+
"eval_wer": 0.3713790502479498,
|
| 703 |
+
"step": 6700
|
| 704 |
+
},
|
| 705 |
+
{
|
| 706 |
+
"epoch": 1.316298877274487,
|
| 707 |
+
"eval_loss": 0.4672953486442566,
|
| 708 |
+
"eval_runtime": 146.8165,
|
| 709 |
+
"eval_samples_per_second": 38.524,
|
| 710 |
+
"eval_steps_per_second": 4.816,
|
| 711 |
+
"eval_wer": 0.3728073694853236,
|
| 712 |
+
"step": 6800
|
| 713 |
+
},
|
| 714 |
+
{
|
| 715 |
+
"epoch": 1.3356562137049943,
|
| 716 |
+
"eval_loss": 0.46098417043685913,
|
| 717 |
+
"eval_runtime": 147.3051,
|
| 718 |
+
"eval_samples_per_second": 38.397,
|
| 719 |
+
"eval_steps_per_second": 4.8,
|
| 720 |
+
"eval_wer": 0.36780022788913674,
|
| 721 |
+
"step": 6900
|
| 722 |
+
},
|
| 723 |
+
{
|
| 724 |
+
"epoch": 1.3550135501355014,
|
| 725 |
+
"grad_norm": 0.8472552299499512,
|
| 726 |
+
"learning_rate": 9.49578947368421e-05,
|
| 727 |
+
"loss": 0.4284,
|
| 728 |
+
"step": 7000
|
| 729 |
+
},
|
| 730 |
+
{
|
| 731 |
+
"epoch": 1.3550135501355014,
|
| 732 |
+
"eval_loss": 0.47299668192863464,
|
| 733 |
+
"eval_runtime": 151.323,
|
| 734 |
+
"eval_samples_per_second": 37.377,
|
| 735 |
+
"eval_steps_per_second": 4.672,
|
| 736 |
+
"eval_wer": 0.36531270562180035,
|
| 737 |
+
"step": 7000
|
| 738 |
+
},
|
| 739 |
+
{
|
| 740 |
+
"epoch": 1.3743708865660085,
|
| 741 |
+
"eval_loss": 0.46056076884269714,
|
| 742 |
+
"eval_runtime": 146.0139,
|
| 743 |
+
"eval_samples_per_second": 38.736,
|
| 744 |
+
"eval_steps_per_second": 4.842,
|
| 745 |
+
"eval_wer": 0.36399672609972555,
|
| 746 |
+
"step": 7100
|
| 747 |
+
},
|
| 748 |
+
{
|
| 749 |
+
"epoch": 1.3937282229965158,
|
| 750 |
+
"eval_loss": 0.4571812152862549,
|
| 751 |
+
"eval_runtime": 146.7792,
|
| 752 |
+
"eval_samples_per_second": 38.534,
|
| 753 |
+
"eval_steps_per_second": 4.817,
|
| 754 |
+
"eval_wer": 0.3620067082858564,
|
| 755 |
+
"step": 7200
|
| 756 |
+
},
|
| 757 |
+
{
|
| 758 |
+
"epoch": 1.4130855594270229,
|
| 759 |
+
"eval_loss": 0.45746785402297974,
|
| 760 |
+
"eval_runtime": 146.8097,
|
| 761 |
+
"eval_samples_per_second": 38.526,
|
| 762 |
+
"eval_steps_per_second": 4.816,
|
| 763 |
+
"eval_wer": 0.362969620131277,
|
| 764 |
+
"step": 7300
|
| 765 |
+
},
|
| 766 |
+
{
|
| 767 |
+
"epoch": 1.43244289585753,
|
| 768 |
+
"eval_loss": 0.45778077840805054,
|
| 769 |
+
"eval_runtime": 146.9433,
|
| 770 |
+
"eval_samples_per_second": 38.491,
|
| 771 |
+
"eval_steps_per_second": 4.811,
|
| 772 |
+
"eval_wer": 0.3590216815650527,
|
| 773 |
+
"step": 7400
|
| 774 |
+
},
|
| 775 |
+
{
|
| 776 |
+
"epoch": 1.4518002322880372,
|
| 777 |
+
"grad_norm": 0.9635696411132812,
|
| 778 |
+
"learning_rate": 7.916842105263156e-05,
|
| 779 |
+
"loss": 0.4299,
|
| 780 |
+
"step": 7500
|
| 781 |
+
},
|
| 782 |
+
{
|
| 783 |
+
"epoch": 1.4518002322880372,
|
| 784 |
+
"eval_loss": 0.4477390646934509,
|
| 785 |
+
"eval_runtime": 146.7454,
|
| 786 |
+
"eval_samples_per_second": 38.543,
|
| 787 |
+
"eval_steps_per_second": 4.818,
|
| 788 |
+
"eval_wer": 0.3569193240358845,
|
| 789 |
+
"step": 7500
|
| 790 |
+
},
|
| 791 |
+
{
|
| 792 |
+
"epoch": 1.4711575687185443,
|
| 793 |
+
"eval_loss": 0.4441732168197632,
|
| 794 |
+
"eval_runtime": 147.4263,
|
| 795 |
+
"eval_samples_per_second": 38.365,
|
| 796 |
+
"eval_steps_per_second": 4.796,
|
| 797 |
+
"eval_wer": 0.3551700341833705,
|
| 798 |
+
"step": 7600
|
| 799 |
+
},
|
| 800 |
+
{
|
| 801 |
+
"epoch": 1.4905149051490514,
|
| 802 |
+
"eval_loss": 0.4420062303543091,
|
| 803 |
+
"eval_runtime": 146.725,
|
| 804 |
+
"eval_samples_per_second": 38.548,
|
| 805 |
+
"eval_steps_per_second": 4.819,
|
| 806 |
+
"eval_wer": 0.3546083356068752,
|
| 807 |
+
"step": 7700
|
| 808 |
+
},
|
| 809 |
+
{
|
| 810 |
+
"epoch": 1.5098722415795587,
|
| 811 |
+
"eval_loss": 0.4436999559402466,
|
| 812 |
+
"eval_runtime": 145.7818,
|
| 813 |
+
"eval_samples_per_second": 38.798,
|
| 814 |
+
"eval_steps_per_second": 4.85,
|
| 815 |
+
"eval_wer": 0.3482531174270995,
|
| 816 |
+
"step": 7800
|
| 817 |
+
},
|
| 818 |
+
{
|
| 819 |
+
"epoch": 1.5292295780100658,
|
| 820 |
+
"eval_loss": 0.43728071451187134,
|
| 821 |
+
"eval_runtime": 146.721,
|
| 822 |
+
"eval_samples_per_second": 38.549,
|
| 823 |
+
"eval_steps_per_second": 4.819,
|
| 824 |
+
"eval_wer": 0.3485740880422397,
|
| 825 |
+
"step": 7900
|
| 826 |
+
},
|
| 827 |
+
{
|
| 828 |
+
"epoch": 1.5485869144405728,
|
| 829 |
+
"grad_norm": 1.1358890533447266,
|
| 830 |
+
"learning_rate": 6.341052631578946e-05,
|
| 831 |
+
"loss": 0.408,
|
| 832 |
+
"step": 8000
|
| 833 |
+
},
|
| 834 |
+
{
|
| 835 |
+
"epoch": 1.5485869144405728,
|
| 836 |
+
"eval_loss": 0.4335756301879883,
|
| 837 |
+
"eval_runtime": 146.7599,
|
| 838 |
+
"eval_samples_per_second": 38.539,
|
| 839 |
+
"eval_steps_per_second": 4.817,
|
| 840 |
+
"eval_wer": 0.3464075363900435,
|
| 841 |
+
"step": 8000
|
| 842 |
+
},
|
| 843 |
+
{
|
| 844 |
+
"epoch": 1.5679442508710801,
|
| 845 |
+
"eval_loss": 0.4347936511039734,
|
| 846 |
+
"eval_runtime": 146.7423,
|
| 847 |
+
"eval_samples_per_second": 38.544,
|
| 848 |
+
"eval_steps_per_second": 4.818,
|
| 849 |
+
"eval_wer": 0.34475453772207154,
|
| 850 |
+
"step": 8100
|
| 851 |
+
},
|
| 852 |
+
{
|
| 853 |
+
"epoch": 1.5873015873015874,
|
| 854 |
+
"eval_loss": 0.42762240767478943,
|
| 855 |
+
"eval_runtime": 151.2432,
|
| 856 |
+
"eval_samples_per_second": 37.397,
|
| 857 |
+
"eval_steps_per_second": 4.675,
|
| 858 |
+
"eval_wer": 0.34180160806278187,
|
| 859 |
+
"step": 8200
|
| 860 |
+
},
|
| 861 |
+
{
|
| 862 |
+
"epoch": 1.6066589237320945,
|
| 863 |
+
"eval_loss": 0.42939648032188416,
|
| 864 |
+
"eval_runtime": 146.0228,
|
| 865 |
+
"eval_samples_per_second": 38.734,
|
| 866 |
+
"eval_steps_per_second": 4.842,
|
| 867 |
+
"eval_wer": 0.3399078814334548,
|
| 868 |
+
"step": 8300
|
| 869 |
+
},
|
| 870 |
+
{
|
| 871 |
+
"epoch": 1.6260162601626016,
|
| 872 |
+
"eval_loss": 0.42716294527053833,
|
| 873 |
+
"eval_runtime": 145.901,
|
| 874 |
+
"eval_samples_per_second": 38.766,
|
| 875 |
+
"eval_steps_per_second": 4.846,
|
| 876 |
+
"eval_wer": 0.3387523872189501,
|
| 877 |
+
"step": 8400
|
| 878 |
+
},
|
| 879 |
+
{
|
| 880 |
+
"epoch": 1.645373596593109,
|
| 881 |
+
"grad_norm": 1.037522792816162,
|
| 882 |
+
"learning_rate": 4.762105263157894e-05,
|
| 883 |
+
"loss": 0.3964,
|
| 884 |
+
"step": 8500
|
| 885 |
+
},
|
| 886 |
+
{
|
| 887 |
+
"epoch": 1.645373596593109,
|
| 888 |
+
"eval_loss": 0.4310940206050873,
|
| 889 |
+
"eval_runtime": 145.7507,
|
| 890 |
+
"eval_samples_per_second": 38.806,
|
| 891 |
+
"eval_steps_per_second": 4.851,
|
| 892 |
+
"eval_wer": 0.3408707932788753,
|
| 893 |
+
"step": 8500
|
| 894 |
+
},
|
| 895 |
+
{
|
| 896 |
+
"epoch": 1.664730933023616,
|
| 897 |
+
"eval_loss": 0.4260464608669281,
|
| 898 |
+
"eval_runtime": 146.3966,
|
| 899 |
+
"eval_samples_per_second": 38.635,
|
| 900 |
+
"eval_steps_per_second": 4.829,
|
| 901 |
+
"eval_wer": 0.3381264945194267,
|
| 902 |
+
"step": 8600
|
| 903 |
+
},
|
| 904 |
+
{
|
| 905 |
+
"epoch": 1.684088269454123,
|
| 906 |
+
"eval_loss": 0.4260489046573639,
|
| 907 |
+
"eval_runtime": 146.6331,
|
| 908 |
+
"eval_samples_per_second": 38.572,
|
| 909 |
+
"eval_steps_per_second": 4.822,
|
| 910 |
+
"eval_wer": 0.3370672914894641,
|
| 911 |
+
"step": 8700
|
| 912 |
+
},
|
| 913 |
+
{
|
| 914 |
+
"epoch": 1.7034456058846303,
|
| 915 |
+
"eval_loss": 0.4259546101093292,
|
| 916 |
+
"eval_runtime": 146.1762,
|
| 917 |
+
"eval_samples_per_second": 38.693,
|
| 918 |
+
"eval_steps_per_second": 4.837,
|
| 919 |
+
"eval_wer": 0.33636115613615575,
|
| 920 |
+
"step": 8800
|
| 921 |
+
},
|
| 922 |
+
{
|
| 923 |
+
"epoch": 1.7228029423151374,
|
| 924 |
+
"eval_loss": 0.42149877548217773,
|
| 925 |
+
"eval_runtime": 147.5316,
|
| 926 |
+
"eval_samples_per_second": 38.338,
|
| 927 |
+
"eval_steps_per_second": 4.792,
|
| 928 |
+
"eval_wer": 0.335109370737109,
|
| 929 |
+
"step": 8900
|
| 930 |
+
},
|
| 931 |
+
{
|
| 932 |
+
"epoch": 1.7421602787456445,
|
| 933 |
+
"grad_norm": 1.4968059062957764,
|
| 934 |
+
"learning_rate": 3.186315789473684e-05,
|
| 935 |
+
"loss": 0.3866,
|
| 936 |
+
"step": 9000
|
| 937 |
+
},
|
| 938 |
+
{
|
| 939 |
+
"epoch": 1.7421602787456445,
|
| 940 |
+
"eval_loss": 0.4234353303909302,
|
| 941 |
+
"eval_runtime": 146.8779,
|
| 942 |
+
"eval_samples_per_second": 38.508,
|
| 943 |
+
"eval_steps_per_second": 4.814,
|
| 944 |
+
"eval_wer": 0.3330391102694548,
|
| 945 |
+
"step": 9000
|
| 946 |
+
},
|
| 947 |
+
{
|
| 948 |
+
"epoch": 1.7615176151761518,
|
| 949 |
+
"eval_loss": 0.4210032522678375,
|
| 950 |
+
"eval_runtime": 146.0169,
|
| 951 |
+
"eval_samples_per_second": 38.735,
|
| 952 |
+
"eval_steps_per_second": 4.842,
|
| 953 |
+
"eval_wer": 0.3318515189934362,
|
| 954 |
+
"step": 9100
|
| 955 |
+
},
|
| 956 |
+
{
|
| 957 |
+
"epoch": 1.7808749516066589,
|
| 958 |
+
"eval_loss": 0.41560646891593933,
|
| 959 |
+
"eval_runtime": 145.9957,
|
| 960 |
+
"eval_samples_per_second": 38.741,
|
| 961 |
+
"eval_steps_per_second": 4.843,
|
| 962 |
+
"eval_wer": 0.3300540835486511,
|
| 963 |
+
"step": 9200
|
| 964 |
+
},
|
| 965 |
+
{
|
| 966 |
+
"epoch": 1.800232288037166,
|
| 967 |
+
"eval_loss": 0.41584905982017517,
|
| 968 |
+
"eval_runtime": 147.0182,
|
| 969 |
+
"eval_samples_per_second": 38.471,
|
| 970 |
+
"eval_steps_per_second": 4.809,
|
| 971 |
+
"eval_wer": 0.33032690857152025,
|
| 972 |
+
"step": 9300
|
| 973 |
+
},
|
| 974 |
+
{
|
| 975 |
+
"epoch": 1.8195896244676733,
|
| 976 |
+
"eval_loss": 0.41545388102531433,
|
| 977 |
+
"eval_runtime": 147.1819,
|
| 978 |
+
"eval_samples_per_second": 38.429,
|
| 979 |
+
"eval_steps_per_second": 4.804,
|
| 980 |
+
"eval_wer": 0.32944423937988476,
|
| 981 |
+
"step": 9400
|
| 982 |
+
},
|
| 983 |
+
{
|
| 984 |
+
"epoch": 1.8389469608981805,
|
| 985 |
+
"grad_norm": 0.9967782497406006,
|
| 986 |
+
"learning_rate": 1.6073684210526313e-05,
|
| 987 |
+
"loss": 0.37,
|
| 988 |
+
"step": 9500
|
| 989 |
+
},
|
| 990 |
+
{
|
| 991 |
+
"epoch": 1.8389469608981805,
|
| 992 |
+
"eval_loss": 0.41372692584991455,
|
| 993 |
+
"eval_runtime": 146.2893,
|
| 994 |
+
"eval_samples_per_second": 38.663,
|
| 995 |
+
"eval_steps_per_second": 4.833,
|
| 996 |
+
"eval_wer": 0.32921955994928664,
|
| 997 |
+
"step": 9500
|
| 998 |
+
},
|
| 999 |
+
{
|
| 1000 |
+
"epoch": 1.8583042973286876,
|
| 1001 |
+
"eval_loss": 0.4120025932788849,
|
| 1002 |
+
"eval_runtime": 146.1391,
|
| 1003 |
+
"eval_samples_per_second": 38.703,
|
| 1004 |
+
"eval_steps_per_second": 4.838,
|
| 1005 |
+
"eval_wer": 0.3284492304729502,
|
| 1006 |
+
"step": 9600
|
| 1007 |
+
},
|
| 1008 |
+
{
|
| 1009 |
+
"epoch": 1.8776616337591947,
|
| 1010 |
+
"eval_loss": 0.4108966886997223,
|
| 1011 |
+
"eval_runtime": 146.9334,
|
| 1012 |
+
"eval_samples_per_second": 38.494,
|
| 1013 |
+
"eval_steps_per_second": 4.812,
|
| 1014 |
+
"eval_wer": 0.3300701320794081,
|
| 1015 |
+
"step": 9700
|
| 1016 |
+
},
|
| 1017 |
+
{
|
| 1018 |
+
"epoch": 1.897018970189702,
|
| 1019 |
+
"eval_loss": 0.4100329577922821,
|
| 1020 |
+
"eval_runtime": 146.8452,
|
| 1021 |
+
"eval_samples_per_second": 38.517,
|
| 1022 |
+
"eval_steps_per_second": 4.815,
|
| 1023 |
+
"eval_wer": 0.32785543483494084,
|
| 1024 |
+
"step": 9800
|
| 1025 |
+
},
|
| 1026 |
+
{
|
| 1027 |
+
"epoch": 1.916376306620209,
|
| 1028 |
+
"eval_loss": 0.4094770848751068,
|
| 1029 |
+
"eval_runtime": 146.5252,
|
| 1030 |
+
"eval_samples_per_second": 38.601,
|
| 1031 |
+
"eval_steps_per_second": 4.825,
|
| 1032 |
+
"eval_wer": 0.3266999406204362,
|
| 1033 |
+
"step": 9900
|
| 1034 |
+
},
|
| 1035 |
+
{
|
| 1036 |
+
"epoch": 1.9357336430507162,
|
| 1037 |
+
"grad_norm": 0.7779282927513123,
|
| 1038 |
+
"learning_rate": 2.842105263157894e-07,
|
| 1039 |
+
"loss": 0.371,
|
| 1040 |
+
"step": 10000
|
| 1041 |
+
},
|
| 1042 |
+
{
|
| 1043 |
+
"epoch": 1.9357336430507162,
|
| 1044 |
+
"eval_loss": 0.409473717212677,
|
| 1045 |
+
"eval_runtime": 146.0247,
|
| 1046 |
+
"eval_samples_per_second": 38.733,
|
| 1047 |
+
"eval_steps_per_second": 4.842,
|
| 1048 |
+
"eval_wer": 0.3270690568278474,
|
| 1049 |
+
"step": 10000
|
| 1050 |
+
},
|
| 1051 |
+
{
|
| 1052 |
+
"epoch": 1.9357336430507162,
|
| 1053 |
+
"step": 10000,
|
| 1054 |
+
"total_flos": 1.1255918428180738e+19,
|
| 1055 |
+
"train_loss": 0.7365739318847656,
|
| 1056 |
+
"train_runtime": 19473.5524,
|
| 1057 |
+
"train_samples_per_second": 4.108,
|
| 1058 |
+
"train_steps_per_second": 0.514
|
| 1059 |
+
}
|
| 1060 |
+
],
|
| 1061 |
+
"logging_steps": 500,
|
| 1062 |
+
"max_steps": 10000,
|
| 1063 |
+
"num_input_tokens_seen": 0,
|
| 1064 |
+
"num_train_epochs": 2,
|
| 1065 |
+
"save_steps": 400,
|
| 1066 |
+
"stateful_callbacks": {
|
| 1067 |
+
"TrainerControl": {
|
| 1068 |
+
"args": {
|
| 1069 |
+
"should_epoch_stop": false,
|
| 1070 |
+
"should_evaluate": false,
|
| 1071 |
+
"should_log": false,
|
| 1072 |
+
"should_save": true,
|
| 1073 |
+
"should_training_stop": true
|
| 1074 |
+
},
|
| 1075 |
+
"attributes": {}
|
| 1076 |
+
}
|
| 1077 |
+
},
|
| 1078 |
+
"total_flos": 1.1255918428180738e+19,
|
| 1079 |
+
"train_batch_size": 8,
|
| 1080 |
+
"trial_name": null,
|
| 1081 |
+
"trial_params": null
|
| 1082 |
+
}
|