Commit
•
d9d123d
1
Parent(s):
5051cf0
End of training
Browse files- README.md +3 -1
- all_results.json +11 -11
- eval_results.json +6 -6
- runs/May10_10-45-37_09e070d6a7b1/events.out.tfevents.1715362857.09e070d6a7b1.138.1 +3 -0
- train_results.json +6 -6
- trainer_state.json +1090 -60
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-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: 0.9669
|
| 21 |
- Wer: 0.6125
|
|
|
|
| 2 |
license: apache-2.0
|
| 3 |
base_model: facebook/wav2vec2-large-xlsr-53
|
| 4 |
tags:
|
| 5 |
+
- automatic-speech-recognition
|
| 6 |
+
- DewiBrynJones/banc-trawsgrifiadau-bangor-normalized
|
| 7 |
- generated_from_trainer
|
| 8 |
metrics:
|
| 9 |
- wer
|
|
|
|
| 17 |
|
| 18 |
# wav2vec2-xlsr-53-ft-btb-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-NORMALIZED - DEFAULT dataset.
|
| 21 |
It achieves the following results on the evaluation set:
|
| 22 |
- Loss: 0.9669
|
| 23 |
- Wer: 0.6125
|
all_results.json
CHANGED
|
@@ -1,15 +1,15 @@
|
|
| 1 |
{
|
| 2 |
-
"epoch":
|
| 3 |
-
"eval_loss": 0.
|
| 4 |
-
"eval_runtime":
|
| 5 |
"eval_samples": 5656,
|
| 6 |
-
"eval_samples_per_second":
|
| 7 |
-
"eval_steps_per_second":
|
| 8 |
-
"eval_wer": 0.
|
| 9 |
-
"total_flos":
|
| 10 |
-
"train_loss":
|
| 11 |
-
"train_runtime":
|
| 12 |
"train_samples": 22621,
|
| 13 |
-
"train_samples_per_second":
|
| 14 |
-
"train_steps_per_second": 0.
|
| 15 |
}
|
|
|
|
| 1 |
{
|
| 2 |
+
"epoch": 15.0,
|
| 3 |
+
"eval_loss": 0.9668737053871155,
|
| 4 |
+
"eval_runtime": 156.2848,
|
| 5 |
"eval_samples": 5656,
|
| 6 |
+
"eval_samples_per_second": 36.19,
|
| 7 |
+
"eval_steps_per_second": 4.524,
|
| 8 |
+
"eval_wer": 0.6124682055958152,
|
| 9 |
+
"total_flos": 4.036432448909298e+19,
|
| 10 |
+
"train_loss": 1.1022147617942597,
|
| 11 |
+
"train_runtime": 28048.236,
|
| 12 |
"train_samples": 22621,
|
| 13 |
+
"train_samples_per_second": 12.098,
|
| 14 |
+
"train_steps_per_second": 0.378
|
| 15 |
}
|
eval_results.json
CHANGED
|
@@ -1,9 +1,9 @@
|
|
| 1 |
{
|
| 2 |
-
"epoch":
|
| 3 |
-
"eval_loss": 0.
|
| 4 |
-
"eval_runtime":
|
| 5 |
"eval_samples": 5656,
|
| 6 |
-
"eval_samples_per_second":
|
| 7 |
-
"eval_steps_per_second":
|
| 8 |
-
"eval_wer": 0.
|
| 9 |
}
|
|
|
|
| 1 |
{
|
| 2 |
+
"epoch": 15.0,
|
| 3 |
+
"eval_loss": 0.9668737053871155,
|
| 4 |
+
"eval_runtime": 156.2848,
|
| 5 |
"eval_samples": 5656,
|
| 6 |
+
"eval_samples_per_second": 36.19,
|
| 7 |
+
"eval_steps_per_second": 4.524,
|
| 8 |
+
"eval_wer": 0.6124682055958152
|
| 9 |
}
|
runs/May10_10-45-37_09e070d6a7b1/events.out.tfevents.1715362857.09e070d6a7b1.138.1
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:a09963d0cd0801af8ea207a91e7c71fb957c6e6f33b740be94e09fe49065b4cc
|
| 3 |
+
size 406
|
train_results.json
CHANGED
|
@@ -1,9 +1,9 @@
|
|
| 1 |
{
|
| 2 |
-
"epoch":
|
| 3 |
-
"total_flos":
|
| 4 |
-
"train_loss":
|
| 5 |
-
"train_runtime":
|
| 6 |
"train_samples": 22621,
|
| 7 |
-
"train_samples_per_second":
|
| 8 |
-
"train_steps_per_second": 0.
|
| 9 |
}
|
|
|
|
| 1 |
{
|
| 2 |
+
"epoch": 15.0,
|
| 3 |
+
"total_flos": 4.036432448909298e+19,
|
| 4 |
+
"train_loss": 1.1022147617942597,
|
| 5 |
+
"train_runtime": 28048.236,
|
| 6 |
"train_samples": 22621,
|
| 7 |
+
"train_samples_per_second": 12.098,
|
| 8 |
+
"train_steps_per_second": 0.378
|
| 9 |
}
|
trainer_state.json
CHANGED
|
@@ -1,101 +1,1131 @@
|
|
| 1 |
{
|
| 2 |
-
"best_metric":
|
| 3 |
-
"best_model_checkpoint":
|
| 4 |
-
"epoch":
|
| 5 |
-
"eval_steps":
|
| 6 |
-
"global_step":
|
| 7 |
"is_hyper_param_search": false,
|
| 8 |
"is_local_process_zero": true,
|
| 9 |
"is_world_process_zero": true,
|
| 10 |
"log_history": [
|
| 11 |
{
|
| 12 |
-
"epoch":
|
| 13 |
-
"
|
| 14 |
-
"
|
| 15 |
-
"
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 16 |
"step": 400
|
| 17 |
},
|
| 18 |
{
|
| 19 |
-
"epoch":
|
| 20 |
-
"
|
| 21 |
-
"
|
| 22 |
-
"
|
| 23 |
-
"eval_steps_per_second": 1.014,
|
| 24 |
-
"eval_wer": 0.802891945242413,
|
| 25 |
"step": 500
|
| 26 |
},
|
| 27 |
{
|
| 28 |
-
"epoch":
|
| 29 |
-
"
|
| 30 |
-
"
|
| 31 |
-
"
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 32 |
"step": 800
|
| 33 |
},
|
| 34 |
{
|
| 35 |
-
"epoch":
|
| 36 |
-
"eval_loss": 0.
|
| 37 |
-
"eval_runtime":
|
| 38 |
-
"eval_samples_per_second":
|
| 39 |
-
"eval_steps_per_second":
|
| 40 |
-
"eval_wer": 0.
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 41 |
"step": 1000
|
| 42 |
},
|
| 43 |
{
|
| 44 |
-
"epoch":
|
| 45 |
-
"
|
| 46 |
-
"
|
| 47 |
-
"
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 48 |
"step": 1200
|
| 49 |
},
|
| 50 |
{
|
| 51 |
-
"epoch":
|
| 52 |
-
"eval_loss":
|
| 53 |
-
"eval_runtime":
|
| 54 |
-
"eval_samples_per_second":
|
| 55 |
-
"eval_steps_per_second":
|
| 56 |
-
"eval_wer": 0.
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 57 |
"step": 1500
|
| 58 |
},
|
| 59 |
{
|
| 60 |
-
"epoch":
|
| 61 |
-
"
|
| 62 |
-
"
|
| 63 |
-
"
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 64 |
"step": 1600
|
| 65 |
},
|
| 66 |
{
|
| 67 |
-
"epoch":
|
| 68 |
-
"
|
| 69 |
-
"
|
| 70 |
-
"
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 71 |
"step": 2000
|
| 72 |
},
|
| 73 |
{
|
| 74 |
-
"epoch":
|
| 75 |
-
"eval_loss":
|
| 76 |
-
"eval_runtime":
|
| 77 |
-
"eval_samples_per_second":
|
| 78 |
-
"eval_steps_per_second":
|
| 79 |
-
"eval_wer": 0.
|
| 80 |
"step": 2000
|
| 81 |
},
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 82 |
{
|
| 83 |
"epoch": 5.657708628005658,
|
| 84 |
-
"
|
| 85 |
-
"
|
| 86 |
-
"
|
| 87 |
-
"
|
| 88 |
-
"
|
| 89 |
-
"
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 90 |
}
|
| 91 |
],
|
| 92 |
-
"logging_steps":
|
| 93 |
-
"max_steps":
|
| 94 |
"num_input_tokens_seen": 0,
|
| 95 |
"num_train_epochs": 15,
|
| 96 |
-
"save_steps":
|
| 97 |
-
"total_flos":
|
| 98 |
-
"train_batch_size":
|
| 99 |
"trial_name": null,
|
| 100 |
"trial_params": null
|
| 101 |
}
|
|
|
|
| 1 |
{
|
| 2 |
+
"best_metric": null,
|
| 3 |
+
"best_model_checkpoint": null,
|
| 4 |
+
"epoch": 15.0,
|
| 5 |
+
"eval_steps": 100,
|
| 6 |
+
"global_step": 10605,
|
| 7 |
"is_hyper_param_search": false,
|
| 8 |
"is_local_process_zero": true,
|
| 9 |
"is_world_process_zero": true,
|
| 10 |
"log_history": [
|
| 11 |
{
|
| 12 |
+
"epoch": 0.14144271570014144,
|
| 13 |
+
"eval_loss": 3.7426819801330566,
|
| 14 |
+
"eval_runtime": 154.1912,
|
| 15 |
+
"eval_samples_per_second": 36.682,
|
| 16 |
+
"eval_steps_per_second": 4.585,
|
| 17 |
+
"eval_wer": 1.0,
|
| 18 |
+
"step": 100
|
| 19 |
+
},
|
| 20 |
+
{
|
| 21 |
+
"epoch": 0.2828854314002829,
|
| 22 |
+
"eval_loss": 2.9179046154022217,
|
| 23 |
+
"eval_runtime": 149.8513,
|
| 24 |
+
"eval_samples_per_second": 37.744,
|
| 25 |
+
"eval_steps_per_second": 4.718,
|
| 26 |
+
"eval_wer": 1.0,
|
| 27 |
+
"step": 200
|
| 28 |
+
},
|
| 29 |
+
{
|
| 30 |
+
"epoch": 0.4243281471004243,
|
| 31 |
+
"eval_loss": 2.8035507202148438,
|
| 32 |
+
"eval_runtime": 151.9434,
|
| 33 |
+
"eval_samples_per_second": 37.224,
|
| 34 |
+
"eval_steps_per_second": 4.653,
|
| 35 |
+
"eval_wer": 1.0,
|
| 36 |
+
"step": 300
|
| 37 |
+
},
|
| 38 |
+
{
|
| 39 |
+
"epoch": 0.5657708628005658,
|
| 40 |
+
"eval_loss": 1.2195814847946167,
|
| 41 |
+
"eval_runtime": 152.8075,
|
| 42 |
+
"eval_samples_per_second": 37.014,
|
| 43 |
+
"eval_steps_per_second": 4.627,
|
| 44 |
+
"eval_wer": 0.8933627681528051,
|
| 45 |
"step": 400
|
| 46 |
},
|
| 47 |
{
|
| 48 |
+
"epoch": 0.7072135785007072,
|
| 49 |
+
"grad_norm": 2.533352851867676,
|
| 50 |
+
"learning_rate": 0.00029699999999999996,
|
| 51 |
+
"loss": 3.574,
|
|
|
|
|
|
|
| 52 |
"step": 500
|
| 53 |
},
|
| 54 |
{
|
| 55 |
+
"epoch": 0.7072135785007072,
|
| 56 |
+
"eval_loss": 0.9860211610794067,
|
| 57 |
+
"eval_runtime": 151.7482,
|
| 58 |
+
"eval_samples_per_second": 37.272,
|
| 59 |
+
"eval_steps_per_second": 4.659,
|
| 60 |
+
"eval_wer": 0.7276159395946313,
|
| 61 |
+
"step": 500
|
| 62 |
+
},
|
| 63 |
+
{
|
| 64 |
+
"epoch": 0.8486562942008486,
|
| 65 |
+
"eval_loss": 0.8391533493995667,
|
| 66 |
+
"eval_runtime": 153.0126,
|
| 67 |
+
"eval_samples_per_second": 36.964,
|
| 68 |
+
"eval_steps_per_second": 4.621,
|
| 69 |
+
"eval_wer": 0.6504295244037049,
|
| 70 |
+
"step": 600
|
| 71 |
+
},
|
| 72 |
+
{
|
| 73 |
+
"epoch": 0.9900990099009901,
|
| 74 |
+
"eval_loss": 0.7803803086280823,
|
| 75 |
+
"eval_runtime": 152.7414,
|
| 76 |
+
"eval_samples_per_second": 37.03,
|
| 77 |
+
"eval_steps_per_second": 4.629,
|
| 78 |
+
"eval_wer": 0.6029178864519844,
|
| 79 |
+
"step": 700
|
| 80 |
+
},
|
| 81 |
+
{
|
| 82 |
+
"epoch": 1.1315417256011315,
|
| 83 |
+
"eval_loss": 0.6122242212295532,
|
| 84 |
+
"eval_runtime": 153.2031,
|
| 85 |
+
"eval_samples_per_second": 36.918,
|
| 86 |
+
"eval_steps_per_second": 4.615,
|
| 87 |
+
"eval_wer": 0.4909056006142919,
|
| 88 |
"step": 800
|
| 89 |
},
|
| 90 |
{
|
| 91 |
+
"epoch": 1.272984441301273,
|
| 92 |
+
"eval_loss": 0.5901117920875549,
|
| 93 |
+
"eval_runtime": 152.5961,
|
| 94 |
+
"eval_samples_per_second": 37.065,
|
| 95 |
+
"eval_steps_per_second": 4.633,
|
| 96 |
+
"eval_wer": 0.48828206235702515,
|
| 97 |
+
"step": 900
|
| 98 |
+
},
|
| 99 |
+
{
|
| 100 |
+
"epoch": 1.4144271570014144,
|
| 101 |
+
"grad_norm": 0.9860969185829163,
|
| 102 |
+
"learning_rate": 0.0002853043047996041,
|
| 103 |
+
"loss": 0.811,
|
| 104 |
"step": 1000
|
| 105 |
},
|
| 106 |
{
|
| 107 |
+
"epoch": 1.4144271570014144,
|
| 108 |
+
"eval_loss": 0.5500049591064453,
|
| 109 |
+
"eval_runtime": 153.3918,
|
| 110 |
+
"eval_samples_per_second": 36.873,
|
| 111 |
+
"eval_steps_per_second": 4.609,
|
| 112 |
+
"eval_wer": 0.45078466189950567,
|
| 113 |
+
"step": 1000
|
| 114 |
+
},
|
| 115 |
+
{
|
| 116 |
+
"epoch": 1.5558698727015559,
|
| 117 |
+
"eval_loss": 0.5231846570968628,
|
| 118 |
+
"eval_runtime": 151.7862,
|
| 119 |
+
"eval_samples_per_second": 37.263,
|
| 120 |
+
"eval_steps_per_second": 4.658,
|
| 121 |
+
"eval_wer": 0.41418310377373585,
|
| 122 |
+
"step": 1100
|
| 123 |
+
},
|
| 124 |
+
{
|
| 125 |
+
"epoch": 1.6973125884016973,
|
| 126 |
+
"eval_loss": 0.518621027469635,
|
| 127 |
+
"eval_runtime": 152.9089,
|
| 128 |
+
"eval_samples_per_second": 36.989,
|
| 129 |
+
"eval_steps_per_second": 4.624,
|
| 130 |
+
"eval_wer": 0.4064724608468909,
|
| 131 |
"step": 1200
|
| 132 |
},
|
| 133 |
{
|
| 134 |
+
"epoch": 1.8387553041018387,
|
| 135 |
+
"eval_loss": 0.49534252285957336,
|
| 136 |
+
"eval_runtime": 153.7928,
|
| 137 |
+
"eval_samples_per_second": 36.777,
|
| 138 |
+
"eval_steps_per_second": 4.597,
|
| 139 |
+
"eval_wer": 0.3929388427636736,
|
| 140 |
+
"step": 1300
|
| 141 |
+
},
|
| 142 |
+
{
|
| 143 |
+
"epoch": 1.9801980198019802,
|
| 144 |
+
"eval_loss": 0.48800522089004517,
|
| 145 |
+
"eval_runtime": 152.9541,
|
| 146 |
+
"eval_samples_per_second": 36.978,
|
| 147 |
+
"eval_steps_per_second": 4.622,
|
| 148 |
+
"eval_wer": 0.39279486810321385,
|
| 149 |
+
"step": 1400
|
| 150 |
+
},
|
| 151 |
+
{
|
| 152 |
+
"epoch": 2.1216407355021216,
|
| 153 |
+
"grad_norm": 0.8855465650558472,
|
| 154 |
+
"learning_rate": 0.00027054923305294405,
|
| 155 |
+
"loss": 0.6459,
|
| 156 |
"step": 1500
|
| 157 |
},
|
| 158 |
{
|
| 159 |
+
"epoch": 2.1216407355021216,
|
| 160 |
+
"eval_loss": 0.46446114778518677,
|
| 161 |
+
"eval_runtime": 152.609,
|
| 162 |
+
"eval_samples_per_second": 37.062,
|
| 163 |
+
"eval_steps_per_second": 4.633,
|
| 164 |
+
"eval_wer": 0.3691990209723089,
|
| 165 |
+
"step": 1500
|
| 166 |
+
},
|
| 167 |
+
{
|
| 168 |
+
"epoch": 2.263083451202263,
|
| 169 |
+
"eval_loss": 0.4666256904602051,
|
| 170 |
+
"eval_runtime": 153.127,
|
| 171 |
+
"eval_samples_per_second": 36.937,
|
| 172 |
+
"eval_steps_per_second": 4.617,
|
| 173 |
+
"eval_wer": 0.3585928876517733,
|
| 174 |
"step": 1600
|
| 175 |
},
|
| 176 |
{
|
| 177 |
+
"epoch": 2.4045261669024045,
|
| 178 |
+
"eval_loss": 0.45017901062965393,
|
| 179 |
+
"eval_runtime": 154.5149,
|
| 180 |
+
"eval_samples_per_second": 36.605,
|
| 181 |
+
"eval_steps_per_second": 4.576,
|
| 182 |
+
"eval_wer": 0.35934475532306315,
|
| 183 |
+
"step": 1700
|
| 184 |
+
},
|
| 185 |
+
{
|
| 186 |
+
"epoch": 2.545968882602546,
|
| 187 |
+
"eval_loss": 0.45280084013938904,
|
| 188 |
+
"eval_runtime": 152.9066,
|
| 189 |
+
"eval_samples_per_second": 36.99,
|
| 190 |
+
"eval_steps_per_second": 4.624,
|
| 191 |
+
"eval_wer": 0.3637759754283246,
|
| 192 |
+
"step": 1800
|
| 193 |
+
},
|
| 194 |
+
{
|
| 195 |
+
"epoch": 2.6874115983026874,
|
| 196 |
+
"eval_loss": 0.46647289395332336,
|
| 197 |
+
"eval_runtime": 153.103,
|
| 198 |
+
"eval_samples_per_second": 36.942,
|
| 199 |
+
"eval_steps_per_second": 4.618,
|
| 200 |
+
"eval_wer": 0.392618899073763,
|
| 201 |
+
"step": 1900
|
| 202 |
+
},
|
| 203 |
+
{
|
| 204 |
+
"epoch": 2.828854314002829,
|
| 205 |
+
"grad_norm": 0.6227492690086365,
|
| 206 |
+
"learning_rate": 0.00025579416130628403,
|
| 207 |
+
"loss": 0.5306,
|
| 208 |
"step": 2000
|
| 209 |
},
|
| 210 |
{
|
| 211 |
+
"epoch": 2.828854314002829,
|
| 212 |
+
"eval_loss": 0.4328605532646179,
|
| 213 |
+
"eval_runtime": 153.5492,
|
| 214 |
+
"eval_samples_per_second": 36.835,
|
| 215 |
+
"eval_steps_per_second": 4.604,
|
| 216 |
+
"eval_wer": 0.3505143094815312,
|
| 217 |
"step": 2000
|
| 218 |
},
|
| 219 |
+
{
|
| 220 |
+
"epoch": 2.9702970297029703,
|
| 221 |
+
"eval_loss": 0.4245360791683197,
|
| 222 |
+
"eval_runtime": 153.9467,
|
| 223 |
+
"eval_samples_per_second": 36.74,
|
| 224 |
+
"eval_steps_per_second": 4.592,
|
| 225 |
+
"eval_wer": 0.3373806210107021,
|
| 226 |
+
"step": 2100
|
| 227 |
+
},
|
| 228 |
+
{
|
| 229 |
+
"epoch": 3.1117397454031117,
|
| 230 |
+
"eval_loss": 0.4376748204231262,
|
| 231 |
+
"eval_runtime": 152.9335,
|
| 232 |
+
"eval_samples_per_second": 36.983,
|
| 233 |
+
"eval_steps_per_second": 4.623,
|
| 234 |
+
"eval_wer": 0.3340372094511366,
|
| 235 |
+
"step": 2200
|
| 236 |
+
},
|
| 237 |
+
{
|
| 238 |
+
"epoch": 3.253182461103253,
|
| 239 |
+
"eval_loss": 0.4271674156188965,
|
| 240 |
+
"eval_runtime": 153.1738,
|
| 241 |
+
"eval_samples_per_second": 36.925,
|
| 242 |
+
"eval_steps_per_second": 4.616,
|
| 243 |
+
"eval_wer": 0.33373326294572153,
|
| 244 |
+
"step": 2300
|
| 245 |
+
},
|
| 246 |
+
{
|
| 247 |
+
"epoch": 3.3946251768033946,
|
| 248 |
+
"eval_loss": 0.43350949883461,
|
| 249 |
+
"eval_runtime": 153.2683,
|
| 250 |
+
"eval_samples_per_second": 36.903,
|
| 251 |
+
"eval_steps_per_second": 4.613,
|
| 252 |
+
"eval_wer": 0.332597462846539,
|
| 253 |
+
"step": 2400
|
| 254 |
+
},
|
| 255 |
+
{
|
| 256 |
+
"epoch": 3.536067892503536,
|
| 257 |
+
"grad_norm": 0.6211841106414795,
|
| 258 |
+
"learning_rate": 0.0002410094012864918,
|
| 259 |
+
"loss": 0.4628,
|
| 260 |
+
"step": 2500
|
| 261 |
+
},
|
| 262 |
+
{
|
| 263 |
+
"epoch": 3.536067892503536,
|
| 264 |
+
"eval_loss": 0.42679545283317566,
|
| 265 |
+
"eval_runtime": 154.4777,
|
| 266 |
+
"eval_samples_per_second": 36.614,
|
| 267 |
+
"eval_steps_per_second": 4.577,
|
| 268 |
+
"eval_wer": 0.3274783638079698,
|
| 269 |
+
"step": 2500
|
| 270 |
+
},
|
| 271 |
+
{
|
| 272 |
+
"epoch": 3.6775106082036775,
|
| 273 |
+
"eval_loss": 0.4502430558204651,
|
| 274 |
+
"eval_runtime": 154.6864,
|
| 275 |
+
"eval_samples_per_second": 36.564,
|
| 276 |
+
"eval_steps_per_second": 4.571,
|
| 277 |
+
"eval_wer": 0.34091599878421397,
|
| 278 |
+
"step": 2600
|
| 279 |
+
},
|
| 280 |
+
{
|
| 281 |
+
"epoch": 3.818953323903819,
|
| 282 |
+
"eval_loss": 0.6344878673553467,
|
| 283 |
+
"eval_runtime": 153.1131,
|
| 284 |
+
"eval_samples_per_second": 36.94,
|
| 285 |
+
"eval_steps_per_second": 4.618,
|
| 286 |
+
"eval_wer": 0.43904272847978754,
|
| 287 |
+
"step": 2700
|
| 288 |
+
},
|
| 289 |
+
{
|
| 290 |
+
"epoch": 3.9603960396039604,
|
| 291 |
+
"eval_loss": 1.0202795267105103,
|
| 292 |
+
"eval_runtime": 153.5324,
|
| 293 |
+
"eval_samples_per_second": 36.839,
|
| 294 |
+
"eval_steps_per_second": 4.605,
|
| 295 |
+
"eval_wer": 0.6403193038025308,
|
| 296 |
+
"step": 2800
|
| 297 |
+
},
|
| 298 |
+
{
|
| 299 |
+
"epoch": 4.101838755304102,
|
| 300 |
+
"eval_loss": 1.2207801342010498,
|
| 301 |
+
"eval_runtime": 154.3697,
|
| 302 |
+
"eval_samples_per_second": 36.639,
|
| 303 |
+
"eval_steps_per_second": 4.58,
|
| 304 |
+
"eval_wer": 0.7921805762185855,
|
| 305 |
+
"step": 2900
|
| 306 |
+
},
|
| 307 |
+
{
|
| 308 |
+
"epoch": 4.243281471004243,
|
| 309 |
+
"grad_norm": 2.4671809673309326,
|
| 310 |
+
"learning_rate": 0.00022619495299356753,
|
| 311 |
+
"loss": 0.8685,
|
| 312 |
+
"step": 3000
|
| 313 |
+
},
|
| 314 |
+
{
|
| 315 |
+
"epoch": 4.243281471004243,
|
| 316 |
+
"eval_loss": 1.101838231086731,
|
| 317 |
+
"eval_runtime": 155.2474,
|
| 318 |
+
"eval_samples_per_second": 36.432,
|
| 319 |
+
"eval_steps_per_second": 4.554,
|
| 320 |
+
"eval_wer": 0.7387019884500328,
|
| 321 |
+
"step": 3000
|
| 322 |
+
},
|
| 323 |
+
{
|
| 324 |
+
"epoch": 4.384724186704385,
|
| 325 |
+
"eval_loss": 1.24972665309906,
|
| 326 |
+
"eval_runtime": 154.4564,
|
| 327 |
+
"eval_samples_per_second": 36.619,
|
| 328 |
+
"eval_steps_per_second": 4.577,
|
| 329 |
+
"eval_wer": 0.8061941098366687,
|
| 330 |
+
"step": 3100
|
| 331 |
+
},
|
| 332 |
+
{
|
| 333 |
+
"epoch": 4.526166902404526,
|
| 334 |
+
"eval_loss": 1.6164859533309937,
|
| 335 |
+
"eval_runtime": 153.9159,
|
| 336 |
+
"eval_samples_per_second": 36.747,
|
| 337 |
+
"eval_steps_per_second": 4.593,
|
| 338 |
+
"eval_wer": 0.9616227543952265,
|
| 339 |
+
"step": 3200
|
| 340 |
+
},
|
| 341 |
+
{
|
| 342 |
+
"epoch": 4.667609618104668,
|
| 343 |
+
"eval_loss": 1.4655081033706665,
|
| 344 |
+
"eval_runtime": 153.645,
|
| 345 |
+
"eval_samples_per_second": 36.812,
|
| 346 |
+
"eval_steps_per_second": 4.602,
|
| 347 |
+
"eval_wer": 0.9216617875253955,
|
| 348 |
+
"step": 3300
|
| 349 |
+
},
|
| 350 |
+
{
|
| 351 |
+
"epoch": 4.809052333804809,
|
| 352 |
+
"eval_loss": 1.0288450717926025,
|
| 353 |
+
"eval_runtime": 152.6496,
|
| 354 |
+
"eval_samples_per_second": 37.052,
|
| 355 |
+
"eval_steps_per_second": 4.632,
|
| 356 |
+
"eval_wer": 0.7464606229303643,
|
| 357 |
+
"step": 3400
|
| 358 |
+
},
|
| 359 |
+
{
|
| 360 |
+
"epoch": 4.9504950495049505,
|
| 361 |
+
"grad_norm": 2.009023666381836,
|
| 362 |
+
"learning_rate": 0.00021143988124690746,
|
| 363 |
+
"loss": 1.3918,
|
| 364 |
+
"step": 3500
|
| 365 |
+
},
|
| 366 |
+
{
|
| 367 |
+
"epoch": 4.9504950495049505,
|
| 368 |
+
"eval_loss": 0.9067263603210449,
|
| 369 |
+
"eval_runtime": 156.1428,
|
| 370 |
+
"eval_samples_per_second": 36.223,
|
| 371 |
+
"eval_steps_per_second": 4.528,
|
| 372 |
+
"eval_wer": 0.5948393082817424,
|
| 373 |
+
"step": 3500
|
| 374 |
+
},
|
| 375 |
+
{
|
| 376 |
+
"epoch": 5.091937765205092,
|
| 377 |
+
"eval_loss": 0.9486163258552551,
|
| 378 |
+
"eval_runtime": 154.3538,
|
| 379 |
+
"eval_samples_per_second": 36.643,
|
| 380 |
+
"eval_steps_per_second": 4.58,
|
| 381 |
+
"eval_wer": 0.6352801906864392,
|
| 382 |
+
"step": 3600
|
| 383 |
+
},
|
| 384 |
+
{
|
| 385 |
+
"epoch": 5.233380480905233,
|
| 386 |
+
"eval_loss": 0.8674383163452148,
|
| 387 |
+
"eval_runtime": 155.8543,
|
| 388 |
+
"eval_samples_per_second": 36.29,
|
| 389 |
+
"eval_steps_per_second": 4.536,
|
| 390 |
+
"eval_wer": 0.5428324614867783,
|
| 391 |
+
"step": 3700
|
| 392 |
+
},
|
| 393 |
+
{
|
| 394 |
+
"epoch": 5.374823196605375,
|
| 395 |
+
"eval_loss": 0.9402504563331604,
|
| 396 |
+
"eval_runtime": 154.1915,
|
| 397 |
+
"eval_samples_per_second": 36.682,
|
| 398 |
+
"eval_steps_per_second": 4.585,
|
| 399 |
+
"eval_wer": 0.5792900449520885,
|
| 400 |
+
"step": 3800
|
| 401 |
+
},
|
| 402 |
+
{
|
| 403 |
+
"epoch": 5.516265912305516,
|
| 404 |
+
"eval_loss": 0.948098361492157,
|
| 405 |
+
"eval_runtime": 159.4305,
|
| 406 |
+
"eval_samples_per_second": 35.476,
|
| 407 |
+
"eval_steps_per_second": 4.435,
|
| 408 |
+
"eval_wer": 0.5763625601894067,
|
| 409 |
+
"step": 3900
|
| 410 |
+
},
|
| 411 |
+
{
|
| 412 |
+
"epoch": 5.657708628005658,
|
| 413 |
+
"grad_norm": 1.2370288372039795,
|
| 414 |
+
"learning_rate": 0.0001966848095002474,
|
| 415 |
+
"loss": 1.0402,
|
| 416 |
+
"step": 4000
|
| 417 |
+
},
|
| 418 |
{
|
| 419 |
"epoch": 5.657708628005658,
|
| 420 |
+
"eval_loss": 1.0175639390945435,
|
| 421 |
+
"eval_runtime": 154.2179,
|
| 422 |
+
"eval_samples_per_second": 36.675,
|
| 423 |
+
"eval_steps_per_second": 4.584,
|
| 424 |
+
"eval_wer": 0.8256946777367183,
|
| 425 |
+
"step": 4000
|
| 426 |
+
},
|
| 427 |
+
{
|
| 428 |
+
"epoch": 5.799151343705799,
|
| 429 |
+
"eval_loss": 0.9857003092765808,
|
| 430 |
+
"eval_runtime": 154.4099,
|
| 431 |
+
"eval_samples_per_second": 36.63,
|
| 432 |
+
"eval_steps_per_second": 4.579,
|
| 433 |
+
"eval_wer": 0.6342883652477164,
|
| 434 |
+
"step": 4100
|
| 435 |
+
},
|
| 436 |
+
{
|
| 437 |
+
"epoch": 5.9405940594059405,
|
| 438 |
+
"eval_loss": 1.3289028406143188,
|
| 439 |
+
"eval_runtime": 155.7394,
|
| 440 |
+
"eval_samples_per_second": 36.317,
|
| 441 |
+
"eval_steps_per_second": 4.54,
|
| 442 |
+
"eval_wer": 0.9014093519540561,
|
| 443 |
+
"step": 4200
|
| 444 |
+
},
|
| 445 |
+
{
|
| 446 |
+
"epoch": 6.082036775106082,
|
| 447 |
+
"eval_loss": 2.0890820026397705,
|
| 448 |
+
"eval_runtime": 154.2749,
|
| 449 |
+
"eval_samples_per_second": 36.662,
|
| 450 |
+
"eval_steps_per_second": 4.583,
|
| 451 |
+
"eval_wer": 0.7125305946153477,
|
| 452 |
+
"step": 4300
|
| 453 |
+
},
|
| 454 |
+
{
|
| 455 |
+
"epoch": 6.223479490806223,
|
| 456 |
+
"eval_loss": 1.256324291229248,
|
| 457 |
+
"eval_runtime": 154.5375,
|
| 458 |
+
"eval_samples_per_second": 36.6,
|
| 459 |
+
"eval_steps_per_second": 4.575,
|
| 460 |
+
"eval_wer": 0.7696085488953944,
|
| 461 |
+
"step": 4400
|
| 462 |
+
},
|
| 463 |
+
{
|
| 464 |
+
"epoch": 6.364922206506365,
|
| 465 |
+
"grad_norm": 0.9906980395317078,
|
| 466 |
+
"learning_rate": 0.0001819000494804552,
|
| 467 |
+
"loss": 1.2886,
|
| 468 |
+
"step": 4500
|
| 469 |
+
},
|
| 470 |
+
{
|
| 471 |
+
"epoch": 6.364922206506365,
|
| 472 |
+
"eval_loss": 1.1441457271575928,
|
| 473 |
+
"eval_runtime": 155.0449,
|
| 474 |
+
"eval_samples_per_second": 36.48,
|
| 475 |
+
"eval_steps_per_second": 4.56,
|
| 476 |
+
"eval_wer": 0.692726080209883,
|
| 477 |
+
"step": 4500
|
| 478 |
+
},
|
| 479 |
+
{
|
| 480 |
+
"epoch": 6.506364922206506,
|
| 481 |
+
"eval_loss": 1.0626095533370972,
|
| 482 |
+
"eval_runtime": 155.5445,
|
| 483 |
+
"eval_samples_per_second": 36.363,
|
| 484 |
+
"eval_steps_per_second": 4.545,
|
| 485 |
+
"eval_wer": 0.6573083137367823,
|
| 486 |
+
"step": 4600
|
| 487 |
+
},
|
| 488 |
+
{
|
| 489 |
+
"epoch": 6.647807637906648,
|
| 490 |
+
"eval_loss": 0.9997339248657227,
|
| 491 |
+
"eval_runtime": 155.8362,
|
| 492 |
+
"eval_samples_per_second": 36.295,
|
| 493 |
+
"eval_steps_per_second": 4.537,
|
| 494 |
+
"eval_wer": 0.6422869574954808,
|
| 495 |
+
"step": 4700
|
| 496 |
+
},
|
| 497 |
+
{
|
| 498 |
+
"epoch": 6.789250353606789,
|
| 499 |
+
"eval_loss": 0.9813728928565979,
|
| 500 |
+
"eval_runtime": 155.2999,
|
| 501 |
+
"eval_samples_per_second": 36.42,
|
| 502 |
+
"eval_steps_per_second": 4.552,
|
| 503 |
+
"eval_wer": 0.6380317064196701,
|
| 504 |
+
"step": 4800
|
| 505 |
+
},
|
| 506 |
+
{
|
| 507 |
+
"epoch": 6.930693069306931,
|
| 508 |
+
"eval_loss": 1.0955251455307007,
|
| 509 |
+
"eval_runtime": 153.746,
|
| 510 |
+
"eval_samples_per_second": 36.788,
|
| 511 |
+
"eval_steps_per_second": 4.598,
|
| 512 |
+
"eval_wer": 0.7651133400521508,
|
| 513 |
+
"step": 4900
|
| 514 |
+
},
|
| 515 |
+
{
|
| 516 |
+
"epoch": 7.072135785007072,
|
| 517 |
+
"grad_norm": 1.6002442836761475,
|
| 518 |
+
"learning_rate": 0.00016708560118753091,
|
| 519 |
+
"loss": 1.0984,
|
| 520 |
+
"step": 5000
|
| 521 |
+
},
|
| 522 |
+
{
|
| 523 |
+
"epoch": 7.072135785007072,
|
| 524 |
+
"eval_loss": 0.9212619066238403,
|
| 525 |
+
"eval_runtime": 155.7654,
|
| 526 |
+
"eval_samples_per_second": 36.311,
|
| 527 |
+
"eval_steps_per_second": 4.539,
|
| 528 |
+
"eval_wer": 0.5882964598230711,
|
| 529 |
+
"step": 5000
|
| 530 |
+
},
|
| 531 |
+
{
|
| 532 |
+
"epoch": 7.2135785007072135,
|
| 533 |
+
"eval_loss": 0.8884870409965515,
|
| 534 |
+
"eval_runtime": 153.6804,
|
| 535 |
+
"eval_samples_per_second": 36.804,
|
| 536 |
+
"eval_steps_per_second": 4.6,
|
| 537 |
+
"eval_wer": 0.5932715842011806,
|
| 538 |
+
"step": 5100
|
| 539 |
+
},
|
| 540 |
+
{
|
| 541 |
+
"epoch": 7.355021216407355,
|
| 542 |
+
"eval_loss": 0.900116503238678,
|
| 543 |
+
"eval_runtime": 155.054,
|
| 544 |
+
"eval_samples_per_second": 36.478,
|
| 545 |
+
"eval_steps_per_second": 4.56,
|
| 546 |
+
"eval_wer": 0.5898641839036329,
|
| 547 |
+
"step": 5200
|
| 548 |
+
},
|
| 549 |
+
{
|
| 550 |
+
"epoch": 7.496463932107496,
|
| 551 |
+
"eval_loss": 0.8783684372901917,
|
| 552 |
+
"eval_runtime": 156.2294,
|
| 553 |
+
"eval_samples_per_second": 36.203,
|
| 554 |
+
"eval_steps_per_second": 4.525,
|
| 555 |
+
"eval_wer": 0.5858968821487418,
|
| 556 |
+
"step": 5300
|
| 557 |
+
},
|
| 558 |
+
{
|
| 559 |
+
"epoch": 7.637906647807638,
|
| 560 |
+
"eval_loss": 0.9072028398513794,
|
| 561 |
+
"eval_runtime": 154.0671,
|
| 562 |
+
"eval_samples_per_second": 36.711,
|
| 563 |
+
"eval_steps_per_second": 4.589,
|
| 564 |
+
"eval_wer": 0.5897522036121643,
|
| 565 |
+
"step": 5400
|
| 566 |
+
},
|
| 567 |
+
{
|
| 568 |
+
"epoch": 7.779349363507779,
|
| 569 |
+
"grad_norm": 1.3490877151489258,
|
| 570 |
+
"learning_rate": 0.00015230084116773872,
|
| 571 |
+
"loss": 0.9659,
|
| 572 |
+
"step": 5500
|
| 573 |
+
},
|
| 574 |
+
{
|
| 575 |
+
"epoch": 7.779349363507779,
|
| 576 |
+
"eval_loss": 0.8811922669410706,
|
| 577 |
+
"eval_runtime": 153.8934,
|
| 578 |
+
"eval_samples_per_second": 36.753,
|
| 579 |
+
"eval_steps_per_second": 4.594,
|
| 580 |
+
"eval_wer": 0.5841051974852426,
|
| 581 |
+
"step": 5500
|
| 582 |
+
},
|
| 583 |
+
{
|
| 584 |
+
"epoch": 7.920792079207921,
|
| 585 |
+
"eval_loss": 0.891165018081665,
|
| 586 |
+
"eval_runtime": 153.7325,
|
| 587 |
+
"eval_samples_per_second": 36.791,
|
| 588 |
+
"eval_steps_per_second": 4.599,
|
| 589 |
+
"eval_wer": 0.5855129497208491,
|
| 590 |
+
"step": 5600
|
| 591 |
+
},
|
| 592 |
+
{
|
| 593 |
+
"epoch": 8.062234794908063,
|
| 594 |
+
"eval_loss": 0.8815582990646362,
|
| 595 |
+
"eval_runtime": 154.6799,
|
| 596 |
+
"eval_samples_per_second": 36.566,
|
| 597 |
+
"eval_steps_per_second": 4.571,
|
| 598 |
+
"eval_wer": 0.5807137943721905,
|
| 599 |
+
"step": 5700
|
| 600 |
+
},
|
| 601 |
+
{
|
| 602 |
+
"epoch": 8.203677510608204,
|
| 603 |
+
"eval_loss": 0.891440749168396,
|
| 604 |
+
"eval_runtime": 155.4972,
|
| 605 |
+
"eval_samples_per_second": 36.374,
|
| 606 |
+
"eval_steps_per_second": 4.547,
|
| 607 |
+
"eval_wer": 0.5803138647598023,
|
| 608 |
+
"step": 5800
|
| 609 |
+
},
|
| 610 |
+
{
|
| 611 |
+
"epoch": 8.345120226308346,
|
| 612 |
+
"eval_loss": 0.8956438899040222,
|
| 613 |
+
"eval_runtime": 154.6865,
|
| 614 |
+
"eval_samples_per_second": 36.564,
|
| 615 |
+
"eval_steps_per_second": 4.571,
|
| 616 |
+
"eval_wer": 0.5810337380621011,
|
| 617 |
+
"step": 5900
|
| 618 |
+
},
|
| 619 |
+
{
|
| 620 |
+
"epoch": 8.486562942008486,
|
| 621 |
+
"grad_norm": 1.257432460784912,
|
| 622 |
+
"learning_rate": 0.00013754576942107867,
|
| 623 |
+
"loss": 0.9679,
|
| 624 |
+
"step": 6000
|
| 625 |
+
},
|
| 626 |
+
{
|
| 627 |
+
"epoch": 8.486562942008486,
|
| 628 |
+
"eval_loss": 0.9162164330482483,
|
| 629 |
+
"eval_runtime": 155.1949,
|
| 630 |
+
"eval_samples_per_second": 36.444,
|
| 631 |
+
"eval_steps_per_second": 4.556,
|
| 632 |
+
"eval_wer": 0.5780262673769416,
|
| 633 |
+
"step": 6000
|
| 634 |
+
},
|
| 635 |
+
{
|
| 636 |
+
"epoch": 8.628005657708629,
|
| 637 |
+
"eval_loss": 0.9409377574920654,
|
| 638 |
+
"eval_runtime": 154.6732,
|
| 639 |
+
"eval_samples_per_second": 36.567,
|
| 640 |
+
"eval_steps_per_second": 4.571,
|
| 641 |
+
"eval_wer": 0.5810177408776055,
|
| 642 |
+
"step": 6100
|
| 643 |
+
},
|
| 644 |
+
{
|
| 645 |
+
"epoch": 8.76944837340877,
|
| 646 |
+
"eval_loss": 0.9370973706245422,
|
| 647 |
+
"eval_runtime": 155.5326,
|
| 648 |
+
"eval_samples_per_second": 36.365,
|
| 649 |
+
"eval_steps_per_second": 4.546,
|
| 650 |
+
"eval_wer": 0.5780742589304283,
|
| 651 |
+
"step": 6200
|
| 652 |
+
},
|
| 653 |
+
{
|
| 654 |
+
"epoch": 8.910891089108912,
|
| 655 |
+
"eval_loss": 0.941677987575531,
|
| 656 |
+
"eval_runtime": 155.4822,
|
| 657 |
+
"eval_samples_per_second": 36.377,
|
| 658 |
+
"eval_steps_per_second": 4.547,
|
| 659 |
+
"eval_wer": 0.5790020956311689,
|
| 660 |
+
"step": 6300
|
| 661 |
+
},
|
| 662 |
+
{
|
| 663 |
+
"epoch": 9.052333804809052,
|
| 664 |
+
"eval_loss": 0.9663541913032532,
|
| 665 |
+
"eval_runtime": 155.844,
|
| 666 |
+
"eval_samples_per_second": 36.293,
|
| 667 |
+
"eval_steps_per_second": 4.537,
|
| 668 |
+
"eval_wer": 0.5783782054358433,
|
| 669 |
+
"step": 6400
|
| 670 |
+
},
|
| 671 |
+
{
|
| 672 |
+
"epoch": 9.193776520509195,
|
| 673 |
+
"grad_norm": 1.5867938995361328,
|
| 674 |
+
"learning_rate": 0.00012273132112815437,
|
| 675 |
+
"loss": 1.0241,
|
| 676 |
+
"step": 6500
|
| 677 |
+
},
|
| 678 |
+
{
|
| 679 |
+
"epoch": 9.193776520509195,
|
| 680 |
+
"eval_loss": 0.9720383286476135,
|
| 681 |
+
"eval_runtime": 155.604,
|
| 682 |
+
"eval_samples_per_second": 36.349,
|
| 683 |
+
"eval_steps_per_second": 4.544,
|
| 684 |
+
"eval_wer": 0.5775143574730848,
|
| 685 |
+
"step": 6500
|
| 686 |
+
},
|
| 687 |
+
{
|
| 688 |
+
"epoch": 9.335219236209335,
|
| 689 |
+
"eval_loss": 0.9840742349624634,
|
| 690 |
+
"eval_runtime": 155.8175,
|
| 691 |
+
"eval_samples_per_second": 36.299,
|
| 692 |
+
"eval_steps_per_second": 4.537,
|
| 693 |
+
"eval_wer": 0.5783942026203388,
|
| 694 |
+
"step": 6600
|
| 695 |
+
},
|
| 696 |
+
{
|
| 697 |
+
"epoch": 9.476661951909477,
|
| 698 |
+
"eval_loss": 0.9574136137962341,
|
| 699 |
+
"eval_runtime": 155.2609,
|
| 700 |
+
"eval_samples_per_second": 36.429,
|
| 701 |
+
"eval_steps_per_second": 4.554,
|
| 702 |
+
"eval_wer": 0.5886803922509638,
|
| 703 |
+
"step": 6700
|
| 704 |
+
},
|
| 705 |
+
{
|
| 706 |
+
"epoch": 9.618104667609618,
|
| 707 |
+
"eval_loss": 1.0725222826004028,
|
| 708 |
+
"eval_runtime": 154.2708,
|
| 709 |
+
"eval_samples_per_second": 36.663,
|
| 710 |
+
"eval_steps_per_second": 4.583,
|
| 711 |
+
"eval_wer": 0.606837196653389,
|
| 712 |
+
"step": 6800
|
| 713 |
+
},
|
| 714 |
+
{
|
| 715 |
+
"epoch": 9.75954738330976,
|
| 716 |
+
"eval_loss": 1.0362112522125244,
|
| 717 |
+
"eval_runtime": 155.3381,
|
| 718 |
+
"eval_samples_per_second": 36.411,
|
| 719 |
+
"eval_steps_per_second": 4.551,
|
| 720 |
+
"eval_wer": 0.5999584073203116,
|
| 721 |
+
"step": 6900
|
| 722 |
+
},
|
| 723 |
+
{
|
| 724 |
+
"epoch": 9.900990099009901,
|
| 725 |
+
"grad_norm": 0.6058325171470642,
|
| 726 |
+
"learning_rate": 0.00010794656110836219,
|
| 727 |
+
"loss": 1.0797,
|
| 728 |
+
"step": 7000
|
| 729 |
+
},
|
| 730 |
+
{
|
| 731 |
+
"epoch": 9.900990099009901,
|
| 732 |
+
"eval_loss": 1.0116764307022095,
|
| 733 |
+
"eval_runtime": 155.5442,
|
| 734 |
+
"eval_samples_per_second": 36.363,
|
| 735 |
+
"eval_steps_per_second": 4.545,
|
| 736 |
+
"eval_wer": 0.5914319079841948,
|
| 737 |
+
"step": 7000
|
| 738 |
+
},
|
| 739 |
+
{
|
| 740 |
+
"epoch": 10.042432814710043,
|
| 741 |
+
"eval_loss": 0.9563263058662415,
|
| 742 |
+
"eval_runtime": 155.5047,
|
| 743 |
+
"eval_samples_per_second": 36.372,
|
| 744 |
+
"eval_steps_per_second": 4.546,
|
| 745 |
+
"eval_wer": 0.6058293740301707,
|
| 746 |
+
"step": 7100
|
| 747 |
+
},
|
| 748 |
+
{
|
| 749 |
+
"epoch": 10.183875530410184,
|
| 750 |
+
"eval_loss": 0.9663692116737366,
|
| 751 |
+
"eval_runtime": 155.3578,
|
| 752 |
+
"eval_samples_per_second": 36.406,
|
| 753 |
+
"eval_steps_per_second": 4.551,
|
| 754 |
+
"eval_wer": 0.5978307817824063,
|
| 755 |
+
"step": 7200
|
| 756 |
+
},
|
| 757 |
+
{
|
| 758 |
+
"epoch": 10.325318246110326,
|
| 759 |
+
"eval_loss": 1.0209406614303589,
|
| 760 |
+
"eval_runtime": 155.4685,
|
| 761 |
+
"eval_samples_per_second": 36.38,
|
| 762 |
+
"eval_steps_per_second": 4.548,
|
| 763 |
+
"eval_wer": 0.6022140103341812,
|
| 764 |
+
"step": 7300
|
| 765 |
+
},
|
| 766 |
+
{
|
| 767 |
+
"epoch": 10.466760961810467,
|
| 768 |
+
"eval_loss": 0.9848981499671936,
|
| 769 |
+
"eval_runtime": 156.2008,
|
| 770 |
+
"eval_samples_per_second": 36.21,
|
| 771 |
+
"eval_steps_per_second": 4.526,
|
| 772 |
+
"eval_wer": 0.5974788437235047,
|
| 773 |
+
"step": 7400
|
| 774 |
+
},
|
| 775 |
+
{
|
| 776 |
+
"epoch": 10.608203677510609,
|
| 777 |
+
"grad_norm": 0.0,
|
| 778 |
+
"learning_rate": 9.319148936170212e-05,
|
| 779 |
+
"loss": 1.0701,
|
| 780 |
+
"step": 7500
|
| 781 |
+
},
|
| 782 |
+
{
|
| 783 |
+
"epoch": 10.608203677510609,
|
| 784 |
+
"eval_loss": 0.9718888401985168,
|
| 785 |
+
"eval_runtime": 156.5494,
|
| 786 |
+
"eval_samples_per_second": 36.129,
|
| 787 |
+
"eval_steps_per_second": 4.516,
|
| 788 |
+
"eval_wer": 0.6057013965542064,
|
| 789 |
+
"step": 7500
|
| 790 |
+
},
|
| 791 |
+
{
|
| 792 |
+
"epoch": 10.74964639321075,
|
| 793 |
+
"eval_loss": 0.966968834400177,
|
| 794 |
+
"eval_runtime": 155.045,
|
| 795 |
+
"eval_samples_per_second": 36.48,
|
| 796 |
+
"eval_steps_per_second": 4.56,
|
| 797 |
+
"eval_wer": 0.6122602421973733,
|
| 798 |
+
"step": 7600
|
| 799 |
+
},
|
| 800 |
+
{
|
| 801 |
+
"epoch": 10.891089108910892,
|
| 802 |
+
"eval_loss": 0.9668737053871155,
|
| 803 |
+
"eval_runtime": 156.2212,
|
| 804 |
+
"eval_samples_per_second": 36.205,
|
| 805 |
+
"eval_steps_per_second": 4.526,
|
| 806 |
+
"eval_wer": 0.6124682055958152,
|
| 807 |
+
"step": 7700
|
| 808 |
+
},
|
| 809 |
+
{
|
| 810 |
+
"epoch": 11.032531824611032,
|
| 811 |
+
"eval_loss": 0.9668737053871155,
|
| 812 |
+
"eval_runtime": 156.7045,
|
| 813 |
+
"eval_samples_per_second": 36.093,
|
| 814 |
+
"eval_steps_per_second": 4.512,
|
| 815 |
+
"eval_wer": 0.6124682055958152,
|
| 816 |
+
"step": 7800
|
| 817 |
+
},
|
| 818 |
+
{
|
| 819 |
+
"epoch": 11.173974540311175,
|
| 820 |
+
"eval_loss": 0.9668737053871155,
|
| 821 |
+
"eval_runtime": 156.3156,
|
| 822 |
+
"eval_samples_per_second": 36.183,
|
| 823 |
+
"eval_steps_per_second": 4.523,
|
| 824 |
+
"eval_wer": 0.6124682055958152,
|
| 825 |
+
"step": 7900
|
| 826 |
+
},
|
| 827 |
+
{
|
| 828 |
+
"epoch": 11.315417256011315,
|
| 829 |
+
"grad_norm": 0.0,
|
| 830 |
+
"learning_rate": 7.843641761504205e-05,
|
| 831 |
+
"loss": 1.0518,
|
| 832 |
+
"step": 8000
|
| 833 |
+
},
|
| 834 |
+
{
|
| 835 |
+
"epoch": 11.315417256011315,
|
| 836 |
+
"eval_loss": 0.9668737053871155,
|
| 837 |
+
"eval_runtime": 156.4393,
|
| 838 |
+
"eval_samples_per_second": 36.155,
|
| 839 |
+
"eval_steps_per_second": 4.519,
|
| 840 |
+
"eval_wer": 0.6124682055958152,
|
| 841 |
+
"step": 8000
|
| 842 |
+
},
|
| 843 |
+
{
|
| 844 |
+
"epoch": 11.456859971711458,
|
| 845 |
+
"eval_loss": 0.9668737053871155,
|
| 846 |
+
"eval_runtime": 155.12,
|
| 847 |
+
"eval_samples_per_second": 36.462,
|
| 848 |
+
"eval_steps_per_second": 4.558,
|
| 849 |
+
"eval_wer": 0.6124682055958152,
|
| 850 |
+
"step": 8100
|
| 851 |
+
},
|
| 852 |
+
{
|
| 853 |
+
"epoch": 11.598302687411598,
|
| 854 |
+
"eval_loss": 0.9668737053871155,
|
| 855 |
+
"eval_runtime": 155.4065,
|
| 856 |
+
"eval_samples_per_second": 36.395,
|
| 857 |
+
"eval_steps_per_second": 4.549,
|
| 858 |
+
"eval_wer": 0.6124682055958152,
|
| 859 |
+
"step": 8200
|
| 860 |
+
},
|
| 861 |
+
{
|
| 862 |
+
"epoch": 11.73974540311174,
|
| 863 |
+
"eval_loss": 0.9668737053871155,
|
| 864 |
+
"eval_runtime": 154.2056,
|
| 865 |
+
"eval_samples_per_second": 36.678,
|
| 866 |
+
"eval_steps_per_second": 4.585,
|
| 867 |
+
"eval_wer": 0.6124682055958152,
|
| 868 |
+
"step": 8300
|
| 869 |
+
},
|
| 870 |
+
{
|
| 871 |
+
"epoch": 11.881188118811881,
|
| 872 |
+
"eval_loss": 0.9668737053871155,
|
| 873 |
+
"eval_runtime": 155.9204,
|
| 874 |
+
"eval_samples_per_second": 36.275,
|
| 875 |
+
"eval_steps_per_second": 4.534,
|
| 876 |
+
"eval_wer": 0.6124682055958152,
|
| 877 |
+
"step": 8400
|
| 878 |
+
},
|
| 879 |
+
{
|
| 880 |
+
"epoch": 12.022630834512023,
|
| 881 |
+
"grad_norm": 0.0,
|
| 882 |
+
"learning_rate": 6.362196932211776e-05,
|
| 883 |
+
"loss": 1.0594,
|
| 884 |
+
"step": 8500
|
| 885 |
+
},
|
| 886 |
+
{
|
| 887 |
+
"epoch": 12.022630834512023,
|
| 888 |
+
"eval_loss": 0.9668737053871155,
|
| 889 |
+
"eval_runtime": 154.4966,
|
| 890 |
+
"eval_samples_per_second": 36.609,
|
| 891 |
+
"eval_steps_per_second": 4.576,
|
| 892 |
+
"eval_wer": 0.6124682055958152,
|
| 893 |
+
"step": 8500
|
| 894 |
+
},
|
| 895 |
+
{
|
| 896 |
+
"epoch": 12.164073550212164,
|
| 897 |
+
"eval_loss": 0.9668737053871155,
|
| 898 |
+
"eval_runtime": 154.0012,
|
| 899 |
+
"eval_samples_per_second": 36.727,
|
| 900 |
+
"eval_steps_per_second": 4.591,
|
| 901 |
+
"eval_wer": 0.6124682055958152,
|
| 902 |
+
"step": 8600
|
| 903 |
+
},
|
| 904 |
+
{
|
| 905 |
+
"epoch": 12.305516265912306,
|
| 906 |
+
"eval_loss": 0.9668737053871155,
|
| 907 |
+
"eval_runtime": 155.297,
|
| 908 |
+
"eval_samples_per_second": 36.421,
|
| 909 |
+
"eval_steps_per_second": 4.553,
|
| 910 |
+
"eval_wer": 0.6124682055958152,
|
| 911 |
+
"step": 8700
|
| 912 |
+
},
|
| 913 |
+
{
|
| 914 |
+
"epoch": 12.446958981612447,
|
| 915 |
+
"eval_loss": 0.9668737053871155,
|
| 916 |
+
"eval_runtime": 156.7077,
|
| 917 |
+
"eval_samples_per_second": 36.093,
|
| 918 |
+
"eval_steps_per_second": 4.512,
|
| 919 |
+
"eval_wer": 0.6124682055958152,
|
| 920 |
+
"step": 8800
|
| 921 |
+
},
|
| 922 |
+
{
|
| 923 |
+
"epoch": 12.58840169731259,
|
| 924 |
+
"eval_loss": 0.9668737053871155,
|
| 925 |
+
"eval_runtime": 155.8279,
|
| 926 |
+
"eval_samples_per_second": 36.296,
|
| 927 |
+
"eval_steps_per_second": 4.537,
|
| 928 |
+
"eval_wer": 0.6124682055958152,
|
| 929 |
+
"step": 8900
|
| 930 |
+
},
|
| 931 |
+
{
|
| 932 |
+
"epoch": 12.72984441301273,
|
| 933 |
+
"grad_norm": 0.0,
|
| 934 |
+
"learning_rate": 4.8866897575457694e-05,
|
| 935 |
+
"loss": 1.0584,
|
| 936 |
+
"step": 9000
|
| 937 |
+
},
|
| 938 |
+
{
|
| 939 |
+
"epoch": 12.72984441301273,
|
| 940 |
+
"eval_loss": 0.9668737053871155,
|
| 941 |
+
"eval_runtime": 155.3599,
|
| 942 |
+
"eval_samples_per_second": 36.406,
|
| 943 |
+
"eval_steps_per_second": 4.551,
|
| 944 |
+
"eval_wer": 0.6124682055958152,
|
| 945 |
+
"step": 9000
|
| 946 |
+
},
|
| 947 |
+
{
|
| 948 |
+
"epoch": 12.871287128712872,
|
| 949 |
+
"eval_loss": 0.9668737053871155,
|
| 950 |
+
"eval_runtime": 155.3076,
|
| 951 |
+
"eval_samples_per_second": 36.418,
|
| 952 |
+
"eval_steps_per_second": 4.552,
|
| 953 |
+
"eval_wer": 0.6124682055958152,
|
| 954 |
+
"step": 9100
|
| 955 |
+
},
|
| 956 |
+
{
|
| 957 |
+
"epoch": 13.012729844413013,
|
| 958 |
+
"eval_loss": 0.9668737053871155,
|
| 959 |
+
"eval_runtime": 155.448,
|
| 960 |
+
"eval_samples_per_second": 36.385,
|
| 961 |
+
"eval_steps_per_second": 4.548,
|
| 962 |
+
"eval_wer": 0.6124682055958152,
|
| 963 |
+
"step": 9200
|
| 964 |
+
},
|
| 965 |
+
{
|
| 966 |
+
"epoch": 13.154172560113155,
|
| 967 |
+
"eval_loss": 0.9668737053871155,
|
| 968 |
+
"eval_runtime": 155.6022,
|
| 969 |
+
"eval_samples_per_second": 36.349,
|
| 970 |
+
"eval_steps_per_second": 4.544,
|
| 971 |
+
"eval_wer": 0.6124682055958152,
|
| 972 |
+
"step": 9300
|
| 973 |
+
},
|
| 974 |
+
{
|
| 975 |
+
"epoch": 13.295615275813295,
|
| 976 |
+
"eval_loss": 0.9668737053871155,
|
| 977 |
+
"eval_runtime": 154.7567,
|
| 978 |
+
"eval_samples_per_second": 36.548,
|
| 979 |
+
"eval_steps_per_second": 4.568,
|
| 980 |
+
"eval_wer": 0.6124682055958152,
|
| 981 |
+
"step": 9400
|
| 982 |
+
},
|
| 983 |
+
{
|
| 984 |
+
"epoch": 13.437057991513438,
|
| 985 |
+
"grad_norm": 0.0,
|
| 986 |
+
"learning_rate": 3.408213755566551e-05,
|
| 987 |
+
"loss": 1.0556,
|
| 988 |
+
"step": 9500
|
| 989 |
+
},
|
| 990 |
+
{
|
| 991 |
+
"epoch": 13.437057991513438,
|
| 992 |
+
"eval_loss": 0.9668737053871155,
|
| 993 |
+
"eval_runtime": 154.3187,
|
| 994 |
+
"eval_samples_per_second": 36.651,
|
| 995 |
+
"eval_steps_per_second": 4.581,
|
| 996 |
+
"eval_wer": 0.6124682055958152,
|
| 997 |
+
"step": 9500
|
| 998 |
+
},
|
| 999 |
+
{
|
| 1000 |
+
"epoch": 13.578500707213578,
|
| 1001 |
+
"eval_loss": 0.9668737053871155,
|
| 1002 |
+
"eval_runtime": 155.5249,
|
| 1003 |
+
"eval_samples_per_second": 36.367,
|
| 1004 |
+
"eval_steps_per_second": 4.546,
|
| 1005 |
+
"eval_wer": 0.6124682055958152,
|
| 1006 |
+
"step": 9600
|
| 1007 |
+
},
|
| 1008 |
+
{
|
| 1009 |
+
"epoch": 13.71994342291372,
|
| 1010 |
+
"eval_loss": 0.9668737053871155,
|
| 1011 |
+
"eval_runtime": 155.212,
|
| 1012 |
+
"eval_samples_per_second": 36.44,
|
| 1013 |
+
"eval_steps_per_second": 4.555,
|
| 1014 |
+
"eval_wer": 0.6124682055958152,
|
| 1015 |
+
"step": 9700
|
| 1016 |
+
},
|
| 1017 |
+
{
|
| 1018 |
+
"epoch": 13.861386138613861,
|
| 1019 |
+
"eval_loss": 0.9668737053871155,
|
| 1020 |
+
"eval_runtime": 154.6321,
|
| 1021 |
+
"eval_samples_per_second": 36.577,
|
| 1022 |
+
"eval_steps_per_second": 4.572,
|
| 1023 |
+
"eval_wer": 0.6124682055958152,
|
| 1024 |
+
"step": 9800
|
| 1025 |
+
},
|
| 1026 |
+
{
|
| 1027 |
+
"epoch": 14.002828854314004,
|
| 1028 |
+
"eval_loss": 0.9668737053871155,
|
| 1029 |
+
"eval_runtime": 155.2988,
|
| 1030 |
+
"eval_samples_per_second": 36.42,
|
| 1031 |
+
"eval_steps_per_second": 4.553,
|
| 1032 |
+
"eval_wer": 0.6124682055958152,
|
| 1033 |
+
"step": 9900
|
| 1034 |
+
},
|
| 1035 |
+
{
|
| 1036 |
+
"epoch": 14.144271570014144,
|
| 1037 |
+
"grad_norm": 0.0,
|
| 1038 |
+
"learning_rate": 1.9297377535873327e-05,
|
| 1039 |
+
"loss": 1.0511,
|
| 1040 |
+
"step": 10000
|
| 1041 |
+
},
|
| 1042 |
+
{
|
| 1043 |
+
"epoch": 14.144271570014144,
|
| 1044 |
+
"eval_loss": 0.9668737053871155,
|
| 1045 |
+
"eval_runtime": 155.9222,
|
| 1046 |
+
"eval_samples_per_second": 36.274,
|
| 1047 |
+
"eval_steps_per_second": 4.534,
|
| 1048 |
+
"eval_wer": 0.6124682055958152,
|
| 1049 |
+
"step": 10000
|
| 1050 |
+
},
|
| 1051 |
+
{
|
| 1052 |
+
"epoch": 14.285714285714286,
|
| 1053 |
+
"eval_loss": 0.9668737053871155,
|
| 1054 |
+
"eval_runtime": 156.6698,
|
| 1055 |
+
"eval_samples_per_second": 36.101,
|
| 1056 |
+
"eval_steps_per_second": 4.513,
|
| 1057 |
+
"eval_wer": 0.6124682055958152,
|
| 1058 |
+
"step": 10100
|
| 1059 |
+
},
|
| 1060 |
+
{
|
| 1061 |
+
"epoch": 14.427157001414427,
|
| 1062 |
+
"eval_loss": 0.9668737053871155,
|
| 1063 |
+
"eval_runtime": 155.8474,
|
| 1064 |
+
"eval_samples_per_second": 36.292,
|
| 1065 |
+
"eval_steps_per_second": 4.536,
|
| 1066 |
+
"eval_wer": 0.6124682055958152,
|
| 1067 |
+
"step": 10200
|
| 1068 |
+
},
|
| 1069 |
+
{
|
| 1070 |
+
"epoch": 14.56859971711457,
|
| 1071 |
+
"eval_loss": 0.9668737053871155,
|
| 1072 |
+
"eval_runtime": 154.9914,
|
| 1073 |
+
"eval_samples_per_second": 36.492,
|
| 1074 |
+
"eval_steps_per_second": 4.562,
|
| 1075 |
+
"eval_wer": 0.6124682055958152,
|
| 1076 |
+
"step": 10300
|
| 1077 |
+
},
|
| 1078 |
+
{
|
| 1079 |
+
"epoch": 14.71004243281471,
|
| 1080 |
+
"eval_loss": 0.9668737053871155,
|
| 1081 |
+
"eval_runtime": 155.7638,
|
| 1082 |
+
"eval_samples_per_second": 36.311,
|
| 1083 |
+
"eval_steps_per_second": 4.539,
|
| 1084 |
+
"eval_wer": 0.6124682055958152,
|
| 1085 |
+
"step": 10400
|
| 1086 |
+
},
|
| 1087 |
+
{
|
| 1088 |
+
"epoch": 14.851485148514852,
|
| 1089 |
+
"grad_norm": 0.0,
|
| 1090 |
+
"learning_rate": 4.512617516081148e-06,
|
| 1091 |
+
"loss": 1.0585,
|
| 1092 |
+
"step": 10500
|
| 1093 |
+
},
|
| 1094 |
+
{
|
| 1095 |
+
"epoch": 14.851485148514852,
|
| 1096 |
+
"eval_loss": 0.9668737053871155,
|
| 1097 |
+
"eval_runtime": 154.954,
|
| 1098 |
+
"eval_samples_per_second": 36.501,
|
| 1099 |
+
"eval_steps_per_second": 4.563,
|
| 1100 |
+
"eval_wer": 0.6124682055958152,
|
| 1101 |
+
"step": 10500
|
| 1102 |
+
},
|
| 1103 |
+
{
|
| 1104 |
+
"epoch": 14.992927864214993,
|
| 1105 |
+
"eval_loss": 0.9668737053871155,
|
| 1106 |
+
"eval_runtime": 154.7299,
|
| 1107 |
+
"eval_samples_per_second": 36.554,
|
| 1108 |
+
"eval_steps_per_second": 4.569,
|
| 1109 |
+
"eval_wer": 0.6124682055958152,
|
| 1110 |
+
"step": 10600
|
| 1111 |
+
},
|
| 1112 |
+
{
|
| 1113 |
+
"epoch": 15.0,
|
| 1114 |
+
"step": 10605,
|
| 1115 |
+
"total_flos": 4.036432448909298e+19,
|
| 1116 |
+
"train_loss": 1.1022147617942597,
|
| 1117 |
+
"train_runtime": 28048.236,
|
| 1118 |
+
"train_samples_per_second": 12.098,
|
| 1119 |
+
"train_steps_per_second": 0.378
|
| 1120 |
}
|
| 1121 |
],
|
| 1122 |
+
"logging_steps": 500,
|
| 1123 |
+
"max_steps": 10605,
|
| 1124 |
"num_input_tokens_seen": 0,
|
| 1125 |
"num_train_epochs": 15,
|
| 1126 |
+
"save_steps": 400,
|
| 1127 |
+
"total_flos": 4.036432448909298e+19,
|
| 1128 |
+
"train_batch_size": 16,
|
| 1129 |
"trial_name": null,
|
| 1130 |
"trial_params": null
|
| 1131 |
}
|