End of training
Browse files- README.md +16 -4
- all_results.json +15 -0
- eval_results.json +9 -0
- runs/Dec31_03-20-54_srvrocgpu011.uct.ac.za/events.out.tfevents.1735623028.srvrocgpu011.uct.ac.za +3 -0
- train_results.json +9 -0
- trainer_state.json +1026 -0
README.md
CHANGED
|
@@ -4,11 +4,23 @@ license: apache-2.0
|
|
| 4 |
base_model: openai/whisper-medium
|
| 5 |
tags:
|
| 6 |
- generated_from_trainer
|
|
|
|
|
|
|
| 7 |
metrics:
|
| 8 |
- wer
|
| 9 |
model-index:
|
| 10 |
- name: whisper-medium-bigcgen-combined-20hrs-model
|
| 11 |
-
results:
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 12 |
---
|
| 13 |
|
| 14 |
<!-- This model card has been generated automatically according to the information the Trainer had access to. You
|
|
@@ -16,10 +28,10 @@ should probably proofread and complete it, then remove this comment. -->
|
|
| 16 |
|
| 17 |
# whisper-medium-bigcgen-combined-20hrs-model
|
| 18 |
|
| 19 |
-
This model is a fine-tuned version of [openai/whisper-medium](https://huggingface.co/openai/whisper-medium) on
|
| 20 |
It achieves the following results on the evaluation set:
|
| 21 |
-
- Loss: 0.
|
| 22 |
-
- Wer: 0.
|
| 23 |
|
| 24 |
## Model description
|
| 25 |
|
|
|
|
| 4 |
base_model: openai/whisper-medium
|
| 5 |
tags:
|
| 6 |
- generated_from_trainer
|
| 7 |
+
datasets:
|
| 8 |
+
- bigcgen
|
| 9 |
metrics:
|
| 10 |
- wer
|
| 11 |
model-index:
|
| 12 |
- name: whisper-medium-bigcgen-combined-20hrs-model
|
| 13 |
+
results:
|
| 14 |
+
- task:
|
| 15 |
+
name: Automatic Speech Recognition
|
| 16 |
+
type: automatic-speech-recognition
|
| 17 |
+
dataset:
|
| 18 |
+
name: bigcgen
|
| 19 |
+
type: bigcgen
|
| 20 |
+
metrics:
|
| 21 |
+
- name: Wer
|
| 22 |
+
type: wer
|
| 23 |
+
value: 0.4210649229332088
|
| 24 |
---
|
| 25 |
|
| 26 |
<!-- This model card has been generated automatically according to the information the Trainer had access to. You
|
|
|
|
| 28 |
|
| 29 |
# whisper-medium-bigcgen-combined-20hrs-model
|
| 30 |
|
| 31 |
+
This model is a fine-tuned version of [openai/whisper-medium](https://huggingface.co/openai/whisper-medium) on the bigcgen dataset.
|
| 32 |
It achieves the following results on the evaluation set:
|
| 33 |
+
- Loss: 0.5368
|
| 34 |
+
- Wer: 0.4211
|
| 35 |
|
| 36 |
## Model description
|
| 37 |
|
all_results.json
ADDED
|
@@ -0,0 +1,15 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"epoch": 2.282943525385054,
|
| 3 |
+
"eval_loss": 0.5367517471313477,
|
| 4 |
+
"eval_runtime": 533.3404,
|
| 5 |
+
"eval_samples": 916,
|
| 6 |
+
"eval_samples_per_second": 1.717,
|
| 7 |
+
"eval_steps_per_second": 0.859,
|
| 8 |
+
"eval_wer": 0.4210649229332088,
|
| 9 |
+
"total_flos": 2.450268261974016e+19,
|
| 10 |
+
"train_loss": 2.4239137687683106,
|
| 11 |
+
"train_runtime": 14144.0718,
|
| 12 |
+
"train_samples": 10518,
|
| 13 |
+
"train_samples_per_second": 2.828,
|
| 14 |
+
"train_steps_per_second": 0.354
|
| 15 |
+
}
|
eval_results.json
ADDED
|
@@ -0,0 +1,9 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"epoch": 2.282943525385054,
|
| 3 |
+
"eval_loss": 0.5367517471313477,
|
| 4 |
+
"eval_runtime": 533.3404,
|
| 5 |
+
"eval_samples": 916,
|
| 6 |
+
"eval_samples_per_second": 1.717,
|
| 7 |
+
"eval_steps_per_second": 0.859,
|
| 8 |
+
"eval_wer": 0.4210649229332088
|
| 9 |
+
}
|
runs/Dec31_03-20-54_srvrocgpu011.uct.ac.za/events.out.tfevents.1735623028.srvrocgpu011.uct.ac.za
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:61a0ac3d5bf846281aceead9126504c5db13e5403aa5a2c19d52e0f22588e8b7
|
| 3 |
+
size 40
|
train_results.json
ADDED
|
@@ -0,0 +1,9 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"epoch": 2.282943525385054,
|
| 3 |
+
"total_flos": 2.450268261974016e+19,
|
| 4 |
+
"train_loss": 2.4239137687683106,
|
| 5 |
+
"train_runtime": 14144.0718,
|
| 6 |
+
"train_samples": 10518,
|
| 7 |
+
"train_samples_per_second": 2.828,
|
| 8 |
+
"train_steps_per_second": 0.354
|
| 9 |
+
}
|
trainer_state.json
ADDED
|
@@ -0,0 +1,1026 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"best_metric": 0.5367517471313477,
|
| 3 |
+
"best_model_checkpoint": "/scratch/skscla001/speech/results/whisper-medium-bigcgen-combined-20hrs-model/checkpoint-2400",
|
| 4 |
+
"epoch": 2.282943525385054,
|
| 5 |
+
"eval_steps": 200,
|
| 6 |
+
"global_step": 3000,
|
| 7 |
+
"is_hyper_param_search": false,
|
| 8 |
+
"is_local_process_zero": true,
|
| 9 |
+
"is_world_process_zero": true,
|
| 10 |
+
"log_history": [
|
| 11 |
+
{
|
| 12 |
+
"epoch": 0.019015021867275148,
|
| 13 |
+
"grad_norm": 131.49729919433594,
|
| 14 |
+
"learning_rate": 6.649999999999999e-07,
|
| 15 |
+
"loss": 13.9909,
|
| 16 |
+
"step": 25
|
| 17 |
+
},
|
| 18 |
+
{
|
| 19 |
+
"epoch": 0.038030043734550295,
|
| 20 |
+
"grad_norm": 104.89710235595703,
|
| 21 |
+
"learning_rate": 1.5399999999999999e-06,
|
| 22 |
+
"loss": 10.094,
|
| 23 |
+
"step": 50
|
| 24 |
+
},
|
| 25 |
+
{
|
| 26 |
+
"epoch": 0.05704506560182544,
|
| 27 |
+
"grad_norm": 103.4134292602539,
|
| 28 |
+
"learning_rate": 2.415e-06,
|
| 29 |
+
"loss": 6.8224,
|
| 30 |
+
"step": 75
|
| 31 |
+
},
|
| 32 |
+
{
|
| 33 |
+
"epoch": 0.07606008746910059,
|
| 34 |
+
"grad_norm": 58.73570251464844,
|
| 35 |
+
"learning_rate": 3.29e-06,
|
| 36 |
+
"loss": 5.5754,
|
| 37 |
+
"step": 100
|
| 38 |
+
},
|
| 39 |
+
{
|
| 40 |
+
"epoch": 0.09507510933637574,
|
| 41 |
+
"grad_norm": 84.27243041992188,
|
| 42 |
+
"learning_rate": 4.164999999999999e-06,
|
| 43 |
+
"loss": 4.2985,
|
| 44 |
+
"step": 125
|
| 45 |
+
},
|
| 46 |
+
{
|
| 47 |
+
"epoch": 0.11409013120365089,
|
| 48 |
+
"grad_norm": 49.697967529296875,
|
| 49 |
+
"learning_rate": 5.039999999999999e-06,
|
| 50 |
+
"loss": 4.0458,
|
| 51 |
+
"step": 150
|
| 52 |
+
},
|
| 53 |
+
{
|
| 54 |
+
"epoch": 0.13310515307092602,
|
| 55 |
+
"grad_norm": 77.4672622680664,
|
| 56 |
+
"learning_rate": 5.915e-06,
|
| 57 |
+
"loss": 4.0797,
|
| 58 |
+
"step": 175
|
| 59 |
+
},
|
| 60 |
+
{
|
| 61 |
+
"epoch": 0.15212017493820118,
|
| 62 |
+
"grad_norm": 56.156681060791016,
|
| 63 |
+
"learning_rate": 6.789999999999999e-06,
|
| 64 |
+
"loss": 3.7679,
|
| 65 |
+
"step": 200
|
| 66 |
+
},
|
| 67 |
+
{
|
| 68 |
+
"epoch": 0.15212017493820118,
|
| 69 |
+
"eval_loss": 0.8942375183105469,
|
| 70 |
+
"eval_runtime": 545.4273,
|
| 71 |
+
"eval_samples_per_second": 1.679,
|
| 72 |
+
"eval_steps_per_second": 0.84,
|
| 73 |
+
"eval_wer": 0.64549276039234,
|
| 74 |
+
"step": 200
|
| 75 |
+
},
|
| 76 |
+
{
|
| 77 |
+
"epoch": 0.17113519680547631,
|
| 78 |
+
"grad_norm": 71.11776733398438,
|
| 79 |
+
"learning_rate": 7.664999999999999e-06,
|
| 80 |
+
"loss": 4.0009,
|
| 81 |
+
"step": 225
|
| 82 |
+
},
|
| 83 |
+
{
|
| 84 |
+
"epoch": 0.19015021867275148,
|
| 85 |
+
"grad_norm": 64.66726684570312,
|
| 86 |
+
"learning_rate": 8.54e-06,
|
| 87 |
+
"loss": 3.4266,
|
| 88 |
+
"step": 250
|
| 89 |
+
},
|
| 90 |
+
{
|
| 91 |
+
"epoch": 0.2091652405400266,
|
| 92 |
+
"grad_norm": 58.9884033203125,
|
| 93 |
+
"learning_rate": 9.415e-06,
|
| 94 |
+
"loss": 3.8021,
|
| 95 |
+
"step": 275
|
| 96 |
+
},
|
| 97 |
+
{
|
| 98 |
+
"epoch": 0.22818026240730177,
|
| 99 |
+
"grad_norm": 61.482627868652344,
|
| 100 |
+
"learning_rate": 1.0289999999999998e-05,
|
| 101 |
+
"loss": 3.167,
|
| 102 |
+
"step": 300
|
| 103 |
+
},
|
| 104 |
+
{
|
| 105 |
+
"epoch": 0.2471952842745769,
|
| 106 |
+
"grad_norm": 57.6241569519043,
|
| 107 |
+
"learning_rate": 1.1164999999999999e-05,
|
| 108 |
+
"loss": 3.0943,
|
| 109 |
+
"step": 325
|
| 110 |
+
},
|
| 111 |
+
{
|
| 112 |
+
"epoch": 0.26621030614185204,
|
| 113 |
+
"grad_norm": 55.91292953491211,
|
| 114 |
+
"learning_rate": 1.2039999999999998e-05,
|
| 115 |
+
"loss": 3.1799,
|
| 116 |
+
"step": 350
|
| 117 |
+
},
|
| 118 |
+
{
|
| 119 |
+
"epoch": 0.2852253280091272,
|
| 120 |
+
"grad_norm": 64.21235656738281,
|
| 121 |
+
"learning_rate": 1.2915e-05,
|
| 122 |
+
"loss": 3.17,
|
| 123 |
+
"step": 375
|
| 124 |
+
},
|
| 125 |
+
{
|
| 126 |
+
"epoch": 0.30424034987640236,
|
| 127 |
+
"grad_norm": 38.54878616333008,
|
| 128 |
+
"learning_rate": 1.379e-05,
|
| 129 |
+
"loss": 3.4733,
|
| 130 |
+
"step": 400
|
| 131 |
+
},
|
| 132 |
+
{
|
| 133 |
+
"epoch": 0.30424034987640236,
|
| 134 |
+
"eval_loss": 0.7609195113182068,
|
| 135 |
+
"eval_runtime": 542.694,
|
| 136 |
+
"eval_samples_per_second": 1.688,
|
| 137 |
+
"eval_steps_per_second": 0.844,
|
| 138 |
+
"eval_wer": 0.5650397010742644,
|
| 139 |
+
"step": 400
|
| 140 |
+
},
|
| 141 |
+
{
|
| 142 |
+
"epoch": 0.3232553717436775,
|
| 143 |
+
"grad_norm": 62.31101608276367,
|
| 144 |
+
"learning_rate": 1.4664999999999998e-05,
|
| 145 |
+
"loss": 3.2382,
|
| 146 |
+
"step": 425
|
| 147 |
+
},
|
| 148 |
+
{
|
| 149 |
+
"epoch": 0.34227039361095263,
|
| 150 |
+
"grad_norm": 44.90293502807617,
|
| 151 |
+
"learning_rate": 1.554e-05,
|
| 152 |
+
"loss": 3.476,
|
| 153 |
+
"step": 450
|
| 154 |
+
},
|
| 155 |
+
{
|
| 156 |
+
"epoch": 0.3612854154782278,
|
| 157 |
+
"grad_norm": 61.89760971069336,
|
| 158 |
+
"learning_rate": 1.6414999999999998e-05,
|
| 159 |
+
"loss": 2.7826,
|
| 160 |
+
"step": 475
|
| 161 |
+
},
|
| 162 |
+
{
|
| 163 |
+
"epoch": 0.38030043734550295,
|
| 164 |
+
"grad_norm": 64.2639389038086,
|
| 165 |
+
"learning_rate": 1.729e-05,
|
| 166 |
+
"loss": 3.1842,
|
| 167 |
+
"step": 500
|
| 168 |
+
},
|
| 169 |
+
{
|
| 170 |
+
"epoch": 0.3993154592127781,
|
| 171 |
+
"grad_norm": 58.504661560058594,
|
| 172 |
+
"learning_rate": 1.7426111111111108e-05,
|
| 173 |
+
"loss": 2.7755,
|
| 174 |
+
"step": 525
|
| 175 |
+
},
|
| 176 |
+
{
|
| 177 |
+
"epoch": 0.4183304810800532,
|
| 178 |
+
"grad_norm": 52.18159484863281,
|
| 179 |
+
"learning_rate": 1.7328888888888887e-05,
|
| 180 |
+
"loss": 2.8868,
|
| 181 |
+
"step": 550
|
| 182 |
+
},
|
| 183 |
+
{
|
| 184 |
+
"epoch": 0.4373455029473284,
|
| 185 |
+
"grad_norm": 63.206600189208984,
|
| 186 |
+
"learning_rate": 1.7231666666666666e-05,
|
| 187 |
+
"loss": 3.1849,
|
| 188 |
+
"step": 575
|
| 189 |
+
},
|
| 190 |
+
{
|
| 191 |
+
"epoch": 0.45636052481460354,
|
| 192 |
+
"grad_norm": 43.44758605957031,
|
| 193 |
+
"learning_rate": 1.7134444444444442e-05,
|
| 194 |
+
"loss": 2.6698,
|
| 195 |
+
"step": 600
|
| 196 |
+
},
|
| 197 |
+
{
|
| 198 |
+
"epoch": 0.45636052481460354,
|
| 199 |
+
"eval_loss": 0.6957600712776184,
|
| 200 |
+
"eval_runtime": 538.2917,
|
| 201 |
+
"eval_samples_per_second": 1.702,
|
| 202 |
+
"eval_steps_per_second": 0.851,
|
| 203 |
+
"eval_wer": 0.5374824848201775,
|
| 204 |
+
"step": 600
|
| 205 |
+
},
|
| 206 |
+
{
|
| 207 |
+
"epoch": 0.4753755466818787,
|
| 208 |
+
"grad_norm": 59.37651062011719,
|
| 209 |
+
"learning_rate": 1.703722222222222e-05,
|
| 210 |
+
"loss": 2.9604,
|
| 211 |
+
"step": 625
|
| 212 |
+
},
|
| 213 |
+
{
|
| 214 |
+
"epoch": 0.4943905685491538,
|
| 215 |
+
"grad_norm": 49.53156661987305,
|
| 216 |
+
"learning_rate": 1.6939999999999997e-05,
|
| 217 |
+
"loss": 2.6304,
|
| 218 |
+
"step": 650
|
| 219 |
+
},
|
| 220 |
+
{
|
| 221 |
+
"epoch": 0.513405590416429,
|
| 222 |
+
"grad_norm": 47.221038818359375,
|
| 223 |
+
"learning_rate": 1.6842777777777776e-05,
|
| 224 |
+
"loss": 2.9998,
|
| 225 |
+
"step": 675
|
| 226 |
+
},
|
| 227 |
+
{
|
| 228 |
+
"epoch": 0.5324206122837041,
|
| 229 |
+
"grad_norm": 55.7159309387207,
|
| 230 |
+
"learning_rate": 1.6745555555555555e-05,
|
| 231 |
+
"loss": 2.7585,
|
| 232 |
+
"step": 700
|
| 233 |
+
},
|
| 234 |
+
{
|
| 235 |
+
"epoch": 0.5514356341509793,
|
| 236 |
+
"grad_norm": 48.83000946044922,
|
| 237 |
+
"learning_rate": 1.664833333333333e-05,
|
| 238 |
+
"loss": 2.7687,
|
| 239 |
+
"step": 725
|
| 240 |
+
},
|
| 241 |
+
{
|
| 242 |
+
"epoch": 0.5704506560182544,
|
| 243 |
+
"grad_norm": 41.48577880859375,
|
| 244 |
+
"learning_rate": 1.655111111111111e-05,
|
| 245 |
+
"loss": 2.4255,
|
| 246 |
+
"step": 750
|
| 247 |
+
},
|
| 248 |
+
{
|
| 249 |
+
"epoch": 0.5894656778855296,
|
| 250 |
+
"grad_norm": 58.95222091674805,
|
| 251 |
+
"learning_rate": 1.6453888888888886e-05,
|
| 252 |
+
"loss": 2.9675,
|
| 253 |
+
"step": 775
|
| 254 |
+
},
|
| 255 |
+
{
|
| 256 |
+
"epoch": 0.6084806997528047,
|
| 257 |
+
"grad_norm": 65.72980499267578,
|
| 258 |
+
"learning_rate": 1.6356666666666665e-05,
|
| 259 |
+
"loss": 2.5647,
|
| 260 |
+
"step": 800
|
| 261 |
+
},
|
| 262 |
+
{
|
| 263 |
+
"epoch": 0.6084806997528047,
|
| 264 |
+
"eval_loss": 0.6684515476226807,
|
| 265 |
+
"eval_runtime": 543.3556,
|
| 266 |
+
"eval_samples_per_second": 1.686,
|
| 267 |
+
"eval_steps_per_second": 0.843,
|
| 268 |
+
"eval_wer": 0.5496263428304531,
|
| 269 |
+
"step": 800
|
| 270 |
+
},
|
| 271 |
+
{
|
| 272 |
+
"epoch": 0.6274957216200798,
|
| 273 |
+
"grad_norm": 55.358070373535156,
|
| 274 |
+
"learning_rate": 1.6259444444444445e-05,
|
| 275 |
+
"loss": 2.713,
|
| 276 |
+
"step": 825
|
| 277 |
+
},
|
| 278 |
+
{
|
| 279 |
+
"epoch": 0.646510743487355,
|
| 280 |
+
"grad_norm": 61.67774963378906,
|
| 281 |
+
"learning_rate": 1.616222222222222e-05,
|
| 282 |
+
"loss": 3.1045,
|
| 283 |
+
"step": 850
|
| 284 |
+
},
|
| 285 |
+
{
|
| 286 |
+
"epoch": 0.6655257653546301,
|
| 287 |
+
"grad_norm": 58.04518127441406,
|
| 288 |
+
"learning_rate": 1.6065e-05,
|
| 289 |
+
"loss": 2.8299,
|
| 290 |
+
"step": 875
|
| 291 |
+
},
|
| 292 |
+
{
|
| 293 |
+
"epoch": 0.6845407872219053,
|
| 294 |
+
"grad_norm": 57.893192291259766,
|
| 295 |
+
"learning_rate": 1.5967777777777775e-05,
|
| 296 |
+
"loss": 2.7055,
|
| 297 |
+
"step": 900
|
| 298 |
+
},
|
| 299 |
+
{
|
| 300 |
+
"epoch": 0.7035558090891805,
|
| 301 |
+
"grad_norm": 53.013309478759766,
|
| 302 |
+
"learning_rate": 1.5870555555555554e-05,
|
| 303 |
+
"loss": 2.6981,
|
| 304 |
+
"step": 925
|
| 305 |
+
},
|
| 306 |
+
{
|
| 307 |
+
"epoch": 0.7225708309564556,
|
| 308 |
+
"grad_norm": 45.20479202270508,
|
| 309 |
+
"learning_rate": 1.577333333333333e-05,
|
| 310 |
+
"loss": 3.0628,
|
| 311 |
+
"step": 950
|
| 312 |
+
},
|
| 313 |
+
{
|
| 314 |
+
"epoch": 0.7415858528237308,
|
| 315 |
+
"grad_norm": 46.2863655090332,
|
| 316 |
+
"learning_rate": 1.567611111111111e-05,
|
| 317 |
+
"loss": 2.6797,
|
| 318 |
+
"step": 975
|
| 319 |
+
},
|
| 320 |
+
{
|
| 321 |
+
"epoch": 0.7606008746910059,
|
| 322 |
+
"grad_norm": 42.149932861328125,
|
| 323 |
+
"learning_rate": 1.557888888888889e-05,
|
| 324 |
+
"loss": 2.1636,
|
| 325 |
+
"step": 1000
|
| 326 |
+
},
|
| 327 |
+
{
|
| 328 |
+
"epoch": 0.7606008746910059,
|
| 329 |
+
"eval_loss": 0.6228325963020325,
|
| 330 |
+
"eval_runtime": 556.097,
|
| 331 |
+
"eval_samples_per_second": 1.647,
|
| 332 |
+
"eval_steps_per_second": 0.824,
|
| 333 |
+
"eval_wer": 0.5102755721625408,
|
| 334 |
+
"step": 1000
|
| 335 |
+
},
|
| 336 |
+
{
|
| 337 |
+
"epoch": 0.779615896558281,
|
| 338 |
+
"grad_norm": 49.531471252441406,
|
| 339 |
+
"learning_rate": 1.5481666666666668e-05,
|
| 340 |
+
"loss": 2.8739,
|
| 341 |
+
"step": 1025
|
| 342 |
+
},
|
| 343 |
+
{
|
| 344 |
+
"epoch": 0.7986309184255562,
|
| 345 |
+
"grad_norm": 39.97299575805664,
|
| 346 |
+
"learning_rate": 1.5384444444444444e-05,
|
| 347 |
+
"loss": 2.595,
|
| 348 |
+
"step": 1050
|
| 349 |
+
},
|
| 350 |
+
{
|
| 351 |
+
"epoch": 0.8176459402928313,
|
| 352 |
+
"grad_norm": 42.683372497558594,
|
| 353 |
+
"learning_rate": 1.528722222222222e-05,
|
| 354 |
+
"loss": 2.4377,
|
| 355 |
+
"step": 1075
|
| 356 |
+
},
|
| 357 |
+
{
|
| 358 |
+
"epoch": 0.8366609621601064,
|
| 359 |
+
"grad_norm": 27.87632942199707,
|
| 360 |
+
"learning_rate": 1.5189999999999999e-05,
|
| 361 |
+
"loss": 2.2694,
|
| 362 |
+
"step": 1100
|
| 363 |
+
},
|
| 364 |
+
{
|
| 365 |
+
"epoch": 0.8556759840273817,
|
| 366 |
+
"grad_norm": 40.27122116088867,
|
| 367 |
+
"learning_rate": 1.5092777777777776e-05,
|
| 368 |
+
"loss": 2.6472,
|
| 369 |
+
"step": 1125
|
| 370 |
+
},
|
| 371 |
+
{
|
| 372 |
+
"epoch": 0.8746910058946568,
|
| 373 |
+
"grad_norm": 53.14582824707031,
|
| 374 |
+
"learning_rate": 1.4995555555555555e-05,
|
| 375 |
+
"loss": 2.5136,
|
| 376 |
+
"step": 1150
|
| 377 |
+
},
|
| 378 |
+
{
|
| 379 |
+
"epoch": 0.893706027761932,
|
| 380 |
+
"grad_norm": 66.0662612915039,
|
| 381 |
+
"learning_rate": 1.4898333333333333e-05,
|
| 382 |
+
"loss": 2.5476,
|
| 383 |
+
"step": 1175
|
| 384 |
+
},
|
| 385 |
+
{
|
| 386 |
+
"epoch": 0.9127210496292071,
|
| 387 |
+
"grad_norm": 48.282222747802734,
|
| 388 |
+
"learning_rate": 1.4801111111111109e-05,
|
| 389 |
+
"loss": 2.7265,
|
| 390 |
+
"step": 1200
|
| 391 |
+
},
|
| 392 |
+
{
|
| 393 |
+
"epoch": 0.9127210496292071,
|
| 394 |
+
"eval_loss": 0.5869264602661133,
|
| 395 |
+
"eval_runtime": 539.0867,
|
| 396 |
+
"eval_samples_per_second": 1.699,
|
| 397 |
+
"eval_steps_per_second": 0.85,
|
| 398 |
+
"eval_wer": 0.4705744978981784,
|
| 399 |
+
"step": 1200
|
| 400 |
+
},
|
| 401 |
+
{
|
| 402 |
+
"epoch": 0.9317360714964822,
|
| 403 |
+
"grad_norm": 36.0697135925293,
|
| 404 |
+
"learning_rate": 1.4703888888888888e-05,
|
| 405 |
+
"loss": 2.1289,
|
| 406 |
+
"step": 1225
|
| 407 |
+
},
|
| 408 |
+
{
|
| 409 |
+
"epoch": 0.9507510933637574,
|
| 410 |
+
"grad_norm": 36.52973556518555,
|
| 411 |
+
"learning_rate": 1.4606666666666665e-05,
|
| 412 |
+
"loss": 2.2548,
|
| 413 |
+
"step": 1250
|
| 414 |
+
},
|
| 415 |
+
{
|
| 416 |
+
"epoch": 0.9697661152310325,
|
| 417 |
+
"grad_norm": 36.967796325683594,
|
| 418 |
+
"learning_rate": 1.4509444444444444e-05,
|
| 419 |
+
"loss": 2.263,
|
| 420 |
+
"step": 1275
|
| 421 |
+
},
|
| 422 |
+
{
|
| 423 |
+
"epoch": 0.9887811370983076,
|
| 424 |
+
"grad_norm": 44.840248107910156,
|
| 425 |
+
"learning_rate": 1.4412222222222222e-05,
|
| 426 |
+
"loss": 2.6958,
|
| 427 |
+
"step": 1300
|
| 428 |
+
},
|
| 429 |
+
{
|
| 430 |
+
"epoch": 1.008366609621601,
|
| 431 |
+
"grad_norm": 45.30537796020508,
|
| 432 |
+
"learning_rate": 1.4314999999999998e-05,
|
| 433 |
+
"loss": 2.3112,
|
| 434 |
+
"step": 1325
|
| 435 |
+
},
|
| 436 |
+
{
|
| 437 |
+
"epoch": 1.0273816314888762,
|
| 438 |
+
"grad_norm": 41.10464859008789,
|
| 439 |
+
"learning_rate": 1.4217777777777777e-05,
|
| 440 |
+
"loss": 2.1152,
|
| 441 |
+
"step": 1350
|
| 442 |
+
},
|
| 443 |
+
{
|
| 444 |
+
"epoch": 1.0463966533561513,
|
| 445 |
+
"grad_norm": 38.10401153564453,
|
| 446 |
+
"learning_rate": 1.4120555555555554e-05,
|
| 447 |
+
"loss": 1.7989,
|
| 448 |
+
"step": 1375
|
| 449 |
+
},
|
| 450 |
+
{
|
| 451 |
+
"epoch": 1.0654116752234266,
|
| 452 |
+
"grad_norm": 30.451261520385742,
|
| 453 |
+
"learning_rate": 1.4023333333333332e-05,
|
| 454 |
+
"loss": 1.5404,
|
| 455 |
+
"step": 1400
|
| 456 |
+
},
|
| 457 |
+
{
|
| 458 |
+
"epoch": 1.0654116752234266,
|
| 459 |
+
"eval_loss": 0.598966121673584,
|
| 460 |
+
"eval_runtime": 530.9607,
|
| 461 |
+
"eval_samples_per_second": 1.725,
|
| 462 |
+
"eval_steps_per_second": 0.863,
|
| 463 |
+
"eval_wer": 0.45422699673049977,
|
| 464 |
+
"step": 1400
|
| 465 |
+
},
|
| 466 |
+
{
|
| 467 |
+
"epoch": 1.0844266970907017,
|
| 468 |
+
"grad_norm": 33.9229736328125,
|
| 469 |
+
"learning_rate": 1.3926111111111111e-05,
|
| 470 |
+
"loss": 1.6963,
|
| 471 |
+
"step": 1425
|
| 472 |
+
},
|
| 473 |
+
{
|
| 474 |
+
"epoch": 1.1034417189579768,
|
| 475 |
+
"grad_norm": 37.29084396362305,
|
| 476 |
+
"learning_rate": 1.3828888888888888e-05,
|
| 477 |
+
"loss": 1.9026,
|
| 478 |
+
"step": 1450
|
| 479 |
+
},
|
| 480 |
+
{
|
| 481 |
+
"epoch": 1.122456740825252,
|
| 482 |
+
"grad_norm": 37.309139251708984,
|
| 483 |
+
"learning_rate": 1.3731666666666664e-05,
|
| 484 |
+
"loss": 1.7798,
|
| 485 |
+
"step": 1475
|
| 486 |
+
},
|
| 487 |
+
{
|
| 488 |
+
"epoch": 1.141471762692527,
|
| 489 |
+
"grad_norm": 27.37919807434082,
|
| 490 |
+
"learning_rate": 1.3634444444444443e-05,
|
| 491 |
+
"loss": 1.8952,
|
| 492 |
+
"step": 1500
|
| 493 |
+
},
|
| 494 |
+
{
|
| 495 |
+
"epoch": 1.1604867845598021,
|
| 496 |
+
"grad_norm": 47.561519622802734,
|
| 497 |
+
"learning_rate": 1.3537222222222221e-05,
|
| 498 |
+
"loss": 1.7325,
|
| 499 |
+
"step": 1525
|
| 500 |
+
},
|
| 501 |
+
{
|
| 502 |
+
"epoch": 1.1795018064270775,
|
| 503 |
+
"grad_norm": 46.943992614746094,
|
| 504 |
+
"learning_rate": 1.3439999999999998e-05,
|
| 505 |
+
"loss": 1.71,
|
| 506 |
+
"step": 1550
|
| 507 |
+
},
|
| 508 |
+
{
|
| 509 |
+
"epoch": 1.1985168282943526,
|
| 510 |
+
"grad_norm": 48.82526397705078,
|
| 511 |
+
"learning_rate": 1.3342777777777778e-05,
|
| 512 |
+
"loss": 1.8316,
|
| 513 |
+
"step": 1575
|
| 514 |
+
},
|
| 515 |
+
{
|
| 516 |
+
"epoch": 1.2175318501616277,
|
| 517 |
+
"grad_norm": 45.330326080322266,
|
| 518 |
+
"learning_rate": 1.3245555555555555e-05,
|
| 519 |
+
"loss": 1.9844,
|
| 520 |
+
"step": 1600
|
| 521 |
+
},
|
| 522 |
+
{
|
| 523 |
+
"epoch": 1.2175318501616277,
|
| 524 |
+
"eval_loss": 0.5892675518989563,
|
| 525 |
+
"eval_runtime": 532.0087,
|
| 526 |
+
"eval_samples_per_second": 1.722,
|
| 527 |
+
"eval_steps_per_second": 0.861,
|
| 528 |
+
"eval_wer": 0.4642690331620738,
|
| 529 |
+
"step": 1600
|
| 530 |
+
},
|
| 531 |
+
{
|
| 532 |
+
"epoch": 1.2365468720289028,
|
| 533 |
+
"grad_norm": 34.17546463012695,
|
| 534 |
+
"learning_rate": 1.314833333333333e-05,
|
| 535 |
+
"loss": 1.7767,
|
| 536 |
+
"step": 1625
|
| 537 |
+
},
|
| 538 |
+
{
|
| 539 |
+
"epoch": 1.2555618938961781,
|
| 540 |
+
"grad_norm": 34.69303512573242,
|
| 541 |
+
"learning_rate": 1.305111111111111e-05,
|
| 542 |
+
"loss": 1.8796,
|
| 543 |
+
"step": 1650
|
| 544 |
+
},
|
| 545 |
+
{
|
| 546 |
+
"epoch": 1.2745769157634532,
|
| 547 |
+
"grad_norm": 35.748783111572266,
|
| 548 |
+
"learning_rate": 1.2953888888888887e-05,
|
| 549 |
+
"loss": 1.7917,
|
| 550 |
+
"step": 1675
|
| 551 |
+
},
|
| 552 |
+
{
|
| 553 |
+
"epoch": 1.2935919376307283,
|
| 554 |
+
"grad_norm": 51.4835319519043,
|
| 555 |
+
"learning_rate": 1.2856666666666667e-05,
|
| 556 |
+
"loss": 2.0249,
|
| 557 |
+
"step": 1700
|
| 558 |
+
},
|
| 559 |
+
{
|
| 560 |
+
"epoch": 1.3126069594980034,
|
| 561 |
+
"grad_norm": 21.467344284057617,
|
| 562 |
+
"learning_rate": 1.2759444444444444e-05,
|
| 563 |
+
"loss": 1.8292,
|
| 564 |
+
"step": 1725
|
| 565 |
+
},
|
| 566 |
+
{
|
| 567 |
+
"epoch": 1.3316219813652785,
|
| 568 |
+
"grad_norm": 35.42878723144531,
|
| 569 |
+
"learning_rate": 1.266222222222222e-05,
|
| 570 |
+
"loss": 1.852,
|
| 571 |
+
"step": 1750
|
| 572 |
+
},
|
| 573 |
+
{
|
| 574 |
+
"epoch": 1.3506370032325536,
|
| 575 |
+
"grad_norm": 47.91001510620117,
|
| 576 |
+
"learning_rate": 1.2564999999999999e-05,
|
| 577 |
+
"loss": 1.7353,
|
| 578 |
+
"step": 1775
|
| 579 |
+
},
|
| 580 |
+
{
|
| 581 |
+
"epoch": 1.3696520250998288,
|
| 582 |
+
"grad_norm": 30.44899559020996,
|
| 583 |
+
"learning_rate": 1.2467777777777777e-05,
|
| 584 |
+
"loss": 1.6926,
|
| 585 |
+
"step": 1800
|
| 586 |
+
},
|
| 587 |
+
{
|
| 588 |
+
"epoch": 1.3696520250998288,
|
| 589 |
+
"eval_loss": 0.5730497241020203,
|
| 590 |
+
"eval_runtime": 512.6986,
|
| 591 |
+
"eval_samples_per_second": 1.787,
|
| 592 |
+
"eval_steps_per_second": 0.893,
|
| 593 |
+
"eval_wer": 0.4412657636618403,
|
| 594 |
+
"step": 1800
|
| 595 |
+
},
|
| 596 |
+
{
|
| 597 |
+
"epoch": 1.388667046967104,
|
| 598 |
+
"grad_norm": 34.98470687866211,
|
| 599 |
+
"learning_rate": 1.2370555555555554e-05,
|
| 600 |
+
"loss": 1.956,
|
| 601 |
+
"step": 1825
|
| 602 |
+
},
|
| 603 |
+
{
|
| 604 |
+
"epoch": 1.4076820688343792,
|
| 605 |
+
"grad_norm": 28.451519012451172,
|
| 606 |
+
"learning_rate": 1.2273333333333333e-05,
|
| 607 |
+
"loss": 1.6875,
|
| 608 |
+
"step": 1850
|
| 609 |
+
},
|
| 610 |
+
{
|
| 611 |
+
"epoch": 1.4266970907016543,
|
| 612 |
+
"grad_norm": 54.90264129638672,
|
| 613 |
+
"learning_rate": 1.217611111111111e-05,
|
| 614 |
+
"loss": 1.68,
|
| 615 |
+
"step": 1875
|
| 616 |
+
},
|
| 617 |
+
{
|
| 618 |
+
"epoch": 1.4457121125689294,
|
| 619 |
+
"grad_norm": 51.0279655456543,
|
| 620 |
+
"learning_rate": 1.2078888888888887e-05,
|
| 621 |
+
"loss": 1.7196,
|
| 622 |
+
"step": 1900
|
| 623 |
+
},
|
| 624 |
+
{
|
| 625 |
+
"epoch": 1.4647271344362047,
|
| 626 |
+
"grad_norm": 37.92000961303711,
|
| 627 |
+
"learning_rate": 1.1981666666666666e-05,
|
| 628 |
+
"loss": 1.9185,
|
| 629 |
+
"step": 1925
|
| 630 |
+
},
|
| 631 |
+
{
|
| 632 |
+
"epoch": 1.4837421563034798,
|
| 633 |
+
"grad_norm": 37.440956115722656,
|
| 634 |
+
"learning_rate": 1.1884444444444443e-05,
|
| 635 |
+
"loss": 1.7995,
|
| 636 |
+
"step": 1950
|
| 637 |
+
},
|
| 638 |
+
{
|
| 639 |
+
"epoch": 1.502757178170755,
|
| 640 |
+
"grad_norm": 29.802404403686523,
|
| 641 |
+
"learning_rate": 1.178722222222222e-05,
|
| 642 |
+
"loss": 1.9844,
|
| 643 |
+
"step": 1975
|
| 644 |
+
},
|
| 645 |
+
{
|
| 646 |
+
"epoch": 1.52177220003803,
|
| 647 |
+
"grad_norm": 44.70022964477539,
|
| 648 |
+
"learning_rate": 1.169e-05,
|
| 649 |
+
"loss": 1.8654,
|
| 650 |
+
"step": 2000
|
| 651 |
+
},
|
| 652 |
+
{
|
| 653 |
+
"epoch": 1.52177220003803,
|
| 654 |
+
"eval_loss": 0.5549585223197937,
|
| 655 |
+
"eval_runtime": 524.5769,
|
| 656 |
+
"eval_samples_per_second": 1.746,
|
| 657 |
+
"eval_steps_per_second": 0.873,
|
| 658 |
+
"eval_wer": 0.4599486221391873,
|
| 659 |
+
"step": 2000
|
| 660 |
+
},
|
| 661 |
+
{
|
| 662 |
+
"epoch": 1.5407872219053051,
|
| 663 |
+
"grad_norm": 37.87806701660156,
|
| 664 |
+
"learning_rate": 1.1592777777777776e-05,
|
| 665 |
+
"loss": 1.6249,
|
| 666 |
+
"step": 2025
|
| 667 |
+
},
|
| 668 |
+
{
|
| 669 |
+
"epoch": 1.5598022437725803,
|
| 670 |
+
"grad_norm": 48.078773498535156,
|
| 671 |
+
"learning_rate": 1.1495555555555553e-05,
|
| 672 |
+
"loss": 1.8834,
|
| 673 |
+
"step": 2050
|
| 674 |
+
},
|
| 675 |
+
{
|
| 676 |
+
"epoch": 1.5788172656398554,
|
| 677 |
+
"grad_norm": 33.26875305175781,
|
| 678 |
+
"learning_rate": 1.1398333333333332e-05,
|
| 679 |
+
"loss": 1.8992,
|
| 680 |
+
"step": 2075
|
| 681 |
+
},
|
| 682 |
+
{
|
| 683 |
+
"epoch": 1.5978322875071307,
|
| 684 |
+
"grad_norm": 34.35662841796875,
|
| 685 |
+
"learning_rate": 1.130111111111111e-05,
|
| 686 |
+
"loss": 1.8728,
|
| 687 |
+
"step": 2100
|
| 688 |
+
},
|
| 689 |
+
{
|
| 690 |
+
"epoch": 1.6168473093744058,
|
| 691 |
+
"grad_norm": 29.405532836914062,
|
| 692 |
+
"learning_rate": 1.1203888888888889e-05,
|
| 693 |
+
"loss": 1.7595,
|
| 694 |
+
"step": 2125
|
| 695 |
+
},
|
| 696 |
+
{
|
| 697 |
+
"epoch": 1.635862331241681,
|
| 698 |
+
"grad_norm": 33.51266098022461,
|
| 699 |
+
"learning_rate": 1.1106666666666666e-05,
|
| 700 |
+
"loss": 1.8762,
|
| 701 |
+
"step": 2150
|
| 702 |
+
},
|
| 703 |
+
{
|
| 704 |
+
"epoch": 1.6548773531089562,
|
| 705 |
+
"grad_norm": 38.80997848510742,
|
| 706 |
+
"learning_rate": 1.1009444444444442e-05,
|
| 707 |
+
"loss": 2.0687,
|
| 708 |
+
"step": 2175
|
| 709 |
+
},
|
| 710 |
+
{
|
| 711 |
+
"epoch": 1.6738923749762313,
|
| 712 |
+
"grad_norm": 43.26702880859375,
|
| 713 |
+
"learning_rate": 1.0912222222222221e-05,
|
| 714 |
+
"loss": 1.8045,
|
| 715 |
+
"step": 2200
|
| 716 |
+
},
|
| 717 |
+
{
|
| 718 |
+
"epoch": 1.6738923749762313,
|
| 719 |
+
"eval_loss": 0.5445375442504883,
|
| 720 |
+
"eval_runtime": 530.1331,
|
| 721 |
+
"eval_samples_per_second": 1.728,
|
| 722 |
+
"eval_steps_per_second": 0.864,
|
| 723 |
+
"eval_wer": 0.41779542269967307,
|
| 724 |
+
"step": 2200
|
| 725 |
+
},
|
| 726 |
+
{
|
| 727 |
+
"epoch": 1.6929073968435064,
|
| 728 |
+
"grad_norm": 36.24801254272461,
|
| 729 |
+
"learning_rate": 1.0814999999999999e-05,
|
| 730 |
+
"loss": 2.0172,
|
| 731 |
+
"step": 2225
|
| 732 |
+
},
|
| 733 |
+
{
|
| 734 |
+
"epoch": 1.7119224187107815,
|
| 735 |
+
"grad_norm": 32.35186767578125,
|
| 736 |
+
"learning_rate": 1.0717777777777776e-05,
|
| 737 |
+
"loss": 1.6788,
|
| 738 |
+
"step": 2250
|
| 739 |
+
},
|
| 740 |
+
{
|
| 741 |
+
"epoch": 1.7309374405780567,
|
| 742 |
+
"grad_norm": 39.30363845825195,
|
| 743 |
+
"learning_rate": 1.0620555555555556e-05,
|
| 744 |
+
"loss": 1.5118,
|
| 745 |
+
"step": 2275
|
| 746 |
+
},
|
| 747 |
+
{
|
| 748 |
+
"epoch": 1.7499524624453318,
|
| 749 |
+
"grad_norm": 37.63472366333008,
|
| 750 |
+
"learning_rate": 1.0523333333333333e-05,
|
| 751 |
+
"loss": 1.8138,
|
| 752 |
+
"step": 2300
|
| 753 |
+
},
|
| 754 |
+
{
|
| 755 |
+
"epoch": 1.7689674843126069,
|
| 756 |
+
"grad_norm": 35.43053436279297,
|
| 757 |
+
"learning_rate": 1.0426111111111109e-05,
|
| 758 |
+
"loss": 1.7971,
|
| 759 |
+
"step": 2325
|
| 760 |
+
},
|
| 761 |
+
{
|
| 762 |
+
"epoch": 1.787982506179882,
|
| 763 |
+
"grad_norm": 41.22832107543945,
|
| 764 |
+
"learning_rate": 1.0328888888888888e-05,
|
| 765 |
+
"loss": 1.7849,
|
| 766 |
+
"step": 2350
|
| 767 |
+
},
|
| 768 |
+
{
|
| 769 |
+
"epoch": 1.8069975280471573,
|
| 770 |
+
"grad_norm": 34.19541931152344,
|
| 771 |
+
"learning_rate": 1.0231666666666666e-05,
|
| 772 |
+
"loss": 1.6543,
|
| 773 |
+
"step": 2375
|
| 774 |
+
},
|
| 775 |
+
{
|
| 776 |
+
"epoch": 1.8260125499144324,
|
| 777 |
+
"grad_norm": 44.1766242980957,
|
| 778 |
+
"learning_rate": 1.0134444444444445e-05,
|
| 779 |
+
"loss": 1.8258,
|
| 780 |
+
"step": 2400
|
| 781 |
+
},
|
| 782 |
+
{
|
| 783 |
+
"epoch": 1.8260125499144324,
|
| 784 |
+
"eval_loss": 0.5367517471313477,
|
| 785 |
+
"eval_runtime": 534.8861,
|
| 786 |
+
"eval_samples_per_second": 1.713,
|
| 787 |
+
"eval_steps_per_second": 0.856,
|
| 788 |
+
"eval_wer": 0.4210649229332088,
|
| 789 |
+
"step": 2400
|
| 790 |
+
},
|
| 791 |
+
{
|
| 792 |
+
"epoch": 1.8450275717817075,
|
| 793 |
+
"grad_norm": 53.05873107910156,
|
| 794 |
+
"learning_rate": 1.0037222222222222e-05,
|
| 795 |
+
"loss": 1.651,
|
| 796 |
+
"step": 2425
|
| 797 |
+
},
|
| 798 |
+
{
|
| 799 |
+
"epoch": 1.8640425936489828,
|
| 800 |
+
"grad_norm": 49.550086975097656,
|
| 801 |
+
"learning_rate": 9.939999999999998e-06,
|
| 802 |
+
"loss": 1.8205,
|
| 803 |
+
"step": 2450
|
| 804 |
+
},
|
| 805 |
+
{
|
| 806 |
+
"epoch": 1.883057615516258,
|
| 807 |
+
"grad_norm": 39.620155334472656,
|
| 808 |
+
"learning_rate": 9.842777777777777e-06,
|
| 809 |
+
"loss": 1.8859,
|
| 810 |
+
"step": 2475
|
| 811 |
+
},
|
| 812 |
+
{
|
| 813 |
+
"epoch": 1.902072637383533,
|
| 814 |
+
"grad_norm": 24.356233596801758,
|
| 815 |
+
"learning_rate": 9.745555555555555e-06,
|
| 816 |
+
"loss": 1.8407,
|
| 817 |
+
"step": 2500
|
| 818 |
+
},
|
| 819 |
+
{
|
| 820 |
+
"epoch": 1.9210876592508082,
|
| 821 |
+
"grad_norm": 36.06525421142578,
|
| 822 |
+
"learning_rate": 9.648333333333332e-06,
|
| 823 |
+
"loss": 1.5886,
|
| 824 |
+
"step": 2525
|
| 825 |
+
},
|
| 826 |
+
{
|
| 827 |
+
"epoch": 1.9401026811180833,
|
| 828 |
+
"grad_norm": 32.340152740478516,
|
| 829 |
+
"learning_rate": 9.551111111111111e-06,
|
| 830 |
+
"loss": 1.6434,
|
| 831 |
+
"step": 2550
|
| 832 |
+
},
|
| 833 |
+
{
|
| 834 |
+
"epoch": 1.9591177029853584,
|
| 835 |
+
"grad_norm": 22.598873138427734,
|
| 836 |
+
"learning_rate": 9.453888888888889e-06,
|
| 837 |
+
"loss": 1.9292,
|
| 838 |
+
"step": 2575
|
| 839 |
+
},
|
| 840 |
+
{
|
| 841 |
+
"epoch": 1.9781327248526335,
|
| 842 |
+
"grad_norm": 28.097835540771484,
|
| 843 |
+
"learning_rate": 9.356666666666665e-06,
|
| 844 |
+
"loss": 1.543,
|
| 845 |
+
"step": 2600
|
| 846 |
+
},
|
| 847 |
+
{
|
| 848 |
+
"epoch": 1.9781327248526335,
|
| 849 |
+
"eval_loss": 0.5370936393737793,
|
| 850 |
+
"eval_runtime": 535.3346,
|
| 851 |
+
"eval_samples_per_second": 1.711,
|
| 852 |
+
"eval_steps_per_second": 0.856,
|
| 853 |
+
"eval_wer": 0.42445119103222795,
|
| 854 |
+
"step": 2600
|
| 855 |
+
},
|
| 856 |
+
{
|
| 857 |
+
"epoch": 1.9971477467199086,
|
| 858 |
+
"grad_norm": 40.16139221191406,
|
| 859 |
+
"learning_rate": 9.259444444444444e-06,
|
| 860 |
+
"loss": 2.2658,
|
| 861 |
+
"step": 2625
|
| 862 |
+
},
|
| 863 |
+
{
|
| 864 |
+
"epoch": 2.016733219243202,
|
| 865 |
+
"grad_norm": 34.17202377319336,
|
| 866 |
+
"learning_rate": 9.162222222222221e-06,
|
| 867 |
+
"loss": 1.237,
|
| 868 |
+
"step": 2650
|
| 869 |
+
},
|
| 870 |
+
{
|
| 871 |
+
"epoch": 2.0357482411104773,
|
| 872 |
+
"grad_norm": 45.333221435546875,
|
| 873 |
+
"learning_rate": 9.064999999999999e-06,
|
| 874 |
+
"loss": 0.977,
|
| 875 |
+
"step": 2675
|
| 876 |
+
},
|
| 877 |
+
{
|
| 878 |
+
"epoch": 2.0547632629777524,
|
| 879 |
+
"grad_norm": 39.80405807495117,
|
| 880 |
+
"learning_rate": 8.967777777777778e-06,
|
| 881 |
+
"loss": 1.0369,
|
| 882 |
+
"step": 2700
|
| 883 |
+
},
|
| 884 |
+
{
|
| 885 |
+
"epoch": 2.0737782848450275,
|
| 886 |
+
"grad_norm": 37.61433029174805,
|
| 887 |
+
"learning_rate": 8.870555555555555e-06,
|
| 888 |
+
"loss": 0.9978,
|
| 889 |
+
"step": 2725
|
| 890 |
+
},
|
| 891 |
+
{
|
| 892 |
+
"epoch": 2.0927933067123026,
|
| 893 |
+
"grad_norm": 44.44002914428711,
|
| 894 |
+
"learning_rate": 8.773333333333331e-06,
|
| 895 |
+
"loss": 1.0163,
|
| 896 |
+
"step": 2750
|
| 897 |
+
},
|
| 898 |
+
{
|
| 899 |
+
"epoch": 2.1118083285795777,
|
| 900 |
+
"grad_norm": 35.349544525146484,
|
| 901 |
+
"learning_rate": 8.67611111111111e-06,
|
| 902 |
+
"loss": 0.9823,
|
| 903 |
+
"step": 2775
|
| 904 |
+
},
|
| 905 |
+
{
|
| 906 |
+
"epoch": 2.1308233504468532,
|
| 907 |
+
"grad_norm": 25.33350372314453,
|
| 908 |
+
"learning_rate": 8.578888888888888e-06,
|
| 909 |
+
"loss": 0.9667,
|
| 910 |
+
"step": 2800
|
| 911 |
+
},
|
| 912 |
+
{
|
| 913 |
+
"epoch": 2.1308233504468532,
|
| 914 |
+
"eval_loss": 0.5587082505226135,
|
| 915 |
+
"eval_runtime": 547.2043,
|
| 916 |
+
"eval_samples_per_second": 1.674,
|
| 917 |
+
"eval_steps_per_second": 0.837,
|
| 918 |
+
"eval_wer": 0.4544605324614666,
|
| 919 |
+
"step": 2800
|
| 920 |
+
},
|
| 921 |
+
{
|
| 922 |
+
"epoch": 2.1498383723141283,
|
| 923 |
+
"grad_norm": 31.81471061706543,
|
| 924 |
+
"learning_rate": 8.481666666666667e-06,
|
| 925 |
+
"loss": 1.3269,
|
| 926 |
+
"step": 2825
|
| 927 |
+
},
|
| 928 |
+
{
|
| 929 |
+
"epoch": 2.1688533941814034,
|
| 930 |
+
"grad_norm": 18.468244552612305,
|
| 931 |
+
"learning_rate": 8.384444444444443e-06,
|
| 932 |
+
"loss": 1.0429,
|
| 933 |
+
"step": 2850
|
| 934 |
+
},
|
| 935 |
+
{
|
| 936 |
+
"epoch": 2.1878684160486785,
|
| 937 |
+
"grad_norm": 20.408119201660156,
|
| 938 |
+
"learning_rate": 8.287222222222222e-06,
|
| 939 |
+
"loss": 1.2085,
|
| 940 |
+
"step": 2875
|
| 941 |
+
},
|
| 942 |
+
{
|
| 943 |
+
"epoch": 2.2068834379159536,
|
| 944 |
+
"grad_norm": 45.54349136352539,
|
| 945 |
+
"learning_rate": 8.19e-06,
|
| 946 |
+
"loss": 0.9542,
|
| 947 |
+
"step": 2900
|
| 948 |
+
},
|
| 949 |
+
{
|
| 950 |
+
"epoch": 2.2258984597832288,
|
| 951 |
+
"grad_norm": 38.3132209777832,
|
| 952 |
+
"learning_rate": 8.092777777777777e-06,
|
| 953 |
+
"loss": 1.0642,
|
| 954 |
+
"step": 2925
|
| 955 |
+
},
|
| 956 |
+
{
|
| 957 |
+
"epoch": 2.244913481650504,
|
| 958 |
+
"grad_norm": 29.104040145874023,
|
| 959 |
+
"learning_rate": 7.995555555555554e-06,
|
| 960 |
+
"loss": 1.0053,
|
| 961 |
+
"step": 2950
|
| 962 |
+
},
|
| 963 |
+
{
|
| 964 |
+
"epoch": 2.263928503517779,
|
| 965 |
+
"grad_norm": 33.763031005859375,
|
| 966 |
+
"learning_rate": 7.898333333333332e-06,
|
| 967 |
+
"loss": 1.3275,
|
| 968 |
+
"step": 2975
|
| 969 |
+
},
|
| 970 |
+
{
|
| 971 |
+
"epoch": 2.282943525385054,
|
| 972 |
+
"grad_norm": 32.66128921508789,
|
| 973 |
+
"learning_rate": 7.801111111111111e-06,
|
| 974 |
+
"loss": 1.0216,
|
| 975 |
+
"step": 3000
|
| 976 |
+
},
|
| 977 |
+
{
|
| 978 |
+
"epoch": 2.282943525385054,
|
| 979 |
+
"eval_loss": 0.5617040395736694,
|
| 980 |
+
"eval_runtime": 528.9401,
|
| 981 |
+
"eval_samples_per_second": 1.732,
|
| 982 |
+
"eval_steps_per_second": 0.866,
|
| 983 |
+
"eval_wer": 0.4096216721158337,
|
| 984 |
+
"step": 3000
|
| 985 |
+
},
|
| 986 |
+
{
|
| 987 |
+
"epoch": 2.282943525385054,
|
| 988 |
+
"step": 3000,
|
| 989 |
+
"total_flos": 2.450268261974016e+19,
|
| 990 |
+
"train_loss": 2.4239137687683106,
|
| 991 |
+
"train_runtime": 14144.0718,
|
| 992 |
+
"train_samples_per_second": 2.828,
|
| 993 |
+
"train_steps_per_second": 0.354
|
| 994 |
+
}
|
| 995 |
+
],
|
| 996 |
+
"logging_steps": 25,
|
| 997 |
+
"max_steps": 5000,
|
| 998 |
+
"num_input_tokens_seen": 0,
|
| 999 |
+
"num_train_epochs": 4,
|
| 1000 |
+
"save_steps": 200,
|
| 1001 |
+
"stateful_callbacks": {
|
| 1002 |
+
"EarlyStoppingCallback": {
|
| 1003 |
+
"args": {
|
| 1004 |
+
"early_stopping_patience": 3,
|
| 1005 |
+
"early_stopping_threshold": 0.0
|
| 1006 |
+
},
|
| 1007 |
+
"attributes": {
|
| 1008 |
+
"early_stopping_patience_counter": 3
|
| 1009 |
+
}
|
| 1010 |
+
},
|
| 1011 |
+
"TrainerControl": {
|
| 1012 |
+
"args": {
|
| 1013 |
+
"should_epoch_stop": false,
|
| 1014 |
+
"should_evaluate": false,
|
| 1015 |
+
"should_log": false,
|
| 1016 |
+
"should_save": true,
|
| 1017 |
+
"should_training_stop": true
|
| 1018 |
+
},
|
| 1019 |
+
"attributes": {}
|
| 1020 |
+
}
|
| 1021 |
+
},
|
| 1022 |
+
"total_flos": 2.450268261974016e+19,
|
| 1023 |
+
"train_batch_size": 2,
|
| 1024 |
+
"trial_name": null,
|
| 1025 |
+
"trial_params": null
|
| 1026 |
+
}
|