End of training
Browse files- README.md +10 -9
- all_results.json +12 -0
- eval_results.json +8 -0
- runs/Apr06_17-46-59_ip-172-31-23-125/events.out.tfevents.1712448513.ip-172-31-23-125.8590.1 +3 -0
- train_results.json +7 -0
- trainer_state.json +1320 -0
README.md
CHANGED
|
@@ -2,38 +2,39 @@
|
|
| 2 |
license: apache-2.0
|
| 3 |
base_model: arun100/whisper-base-th-1
|
| 4 |
tags:
|
|
|
|
| 5 |
- generated_from_trainer
|
| 6 |
datasets:
|
| 7 |
-
- fleurs
|
| 8 |
metrics:
|
| 9 |
- wer
|
| 10 |
model-index:
|
| 11 |
-
- name:
|
| 12 |
results:
|
| 13 |
- task:
|
| 14 |
name: Automatic Speech Recognition
|
| 15 |
type: automatic-speech-recognition
|
| 16 |
dataset:
|
| 17 |
-
name: fleurs
|
| 18 |
-
type: fleurs
|
| 19 |
config: th_th
|
| 20 |
split: test
|
| 21 |
args: th_th
|
| 22 |
metrics:
|
| 23 |
- name: Wer
|
| 24 |
type: wer
|
| 25 |
-
value:
|
| 26 |
---
|
| 27 |
|
| 28 |
<!-- This model card has been generated automatically according to the information the Trainer had access to. You
|
| 29 |
should probably proofread and complete it, then remove this comment. -->
|
| 30 |
|
| 31 |
-
#
|
| 32 |
|
| 33 |
-
This model is a fine-tuned version of [arun100/whisper-base-th-1](https://huggingface.co/arun100/whisper-base-th-1) on the fleurs dataset.
|
| 34 |
It achieves the following results on the evaluation set:
|
| 35 |
-
- Loss: 0.
|
| 36 |
-
- Wer:
|
| 37 |
|
| 38 |
## Model description
|
| 39 |
|
|
|
|
| 2 |
license: apache-2.0
|
| 3 |
base_model: arun100/whisper-base-th-1
|
| 4 |
tags:
|
| 5 |
+
- whisper-event
|
| 6 |
- generated_from_trainer
|
| 7 |
datasets:
|
| 8 |
+
- google/fleurs
|
| 9 |
metrics:
|
| 10 |
- wer
|
| 11 |
model-index:
|
| 12 |
+
- name: Whisper Base Thai (2)
|
| 13 |
results:
|
| 14 |
- task:
|
| 15 |
name: Automatic Speech Recognition
|
| 16 |
type: automatic-speech-recognition
|
| 17 |
dataset:
|
| 18 |
+
name: google/fleurs th_th
|
| 19 |
+
type: google/fleurs
|
| 20 |
config: th_th
|
| 21 |
split: test
|
| 22 |
args: th_th
|
| 23 |
metrics:
|
| 24 |
- name: Wer
|
| 25 |
type: wer
|
| 26 |
+
value: 53.662828506943114
|
| 27 |
---
|
| 28 |
|
| 29 |
<!-- This model card has been generated automatically according to the information the Trainer had access to. You
|
| 30 |
should probably proofread and complete it, then remove this comment. -->
|
| 31 |
|
| 32 |
+
# Whisper Base Thai (2)
|
| 33 |
|
| 34 |
+
This model is a fine-tuned version of [arun100/whisper-base-th-1](https://huggingface.co/arun100/whisper-base-th-1) on the google/fleurs th_th dataset.
|
| 35 |
It achieves the following results on the evaluation set:
|
| 36 |
+
- Loss: 0.5628
|
| 37 |
+
- Wer: 53.6628
|
| 38 |
|
| 39 |
## Model description
|
| 40 |
|
all_results.json
ADDED
|
@@ -0,0 +1,12 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"epoch": 357.0,
|
| 3 |
+
"eval_loss": 0.5628067255020142,
|
| 4 |
+
"eval_runtime": 220.2375,
|
| 5 |
+
"eval_samples_per_second": 4.636,
|
| 6 |
+
"eval_steps_per_second": 0.145,
|
| 7 |
+
"eval_wer": 53.662828506943114,
|
| 8 |
+
"train_loss": 0.22935251636505127,
|
| 9 |
+
"train_runtime": 22650.8672,
|
| 10 |
+
"train_samples_per_second": 14.127,
|
| 11 |
+
"train_steps_per_second": 0.221
|
| 12 |
+
}
|
eval_results.json
ADDED
|
@@ -0,0 +1,8 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"epoch": 357.0,
|
| 3 |
+
"eval_loss": 0.5628067255020142,
|
| 4 |
+
"eval_runtime": 220.2375,
|
| 5 |
+
"eval_samples_per_second": 4.636,
|
| 6 |
+
"eval_steps_per_second": 0.145,
|
| 7 |
+
"eval_wer": 53.662828506943114
|
| 8 |
+
}
|
runs/Apr06_17-46-59_ip-172-31-23-125/events.out.tfevents.1712448513.ip-172-31-23-125.8590.1
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:dc535d22d96ae2e4ec1a4ff0979763a820ba92493a44232e9d220f5432146e12
|
| 3 |
+
size 406
|
train_results.json
ADDED
|
@@ -0,0 +1,7 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"epoch": 357.0,
|
| 3 |
+
"train_loss": 0.22935251636505127,
|
| 4 |
+
"train_runtime": 22650.8672,
|
| 5 |
+
"train_samples_per_second": 14.127,
|
| 6 |
+
"train_steps_per_second": 0.221
|
| 7 |
+
}
|
trainer_state.json
ADDED
|
@@ -0,0 +1,1320 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"best_metric": 53.662828506943114,
|
| 3 |
+
"best_model_checkpoint": "./checkpoint-3000",
|
| 4 |
+
"epoch": 357.0004,
|
| 5 |
+
"eval_steps": 500,
|
| 6 |
+
"global_step": 5000,
|
| 7 |
+
"is_hyper_param_search": false,
|
| 8 |
+
"is_local_process_zero": true,
|
| 9 |
+
"is_world_process_zero": true,
|
| 10 |
+
"log_history": [
|
| 11 |
+
{
|
| 12 |
+
"epoch": 1.0,
|
| 13 |
+
"learning_rate": 2.5e-08,
|
| 14 |
+
"loss": 0.6457,
|
| 15 |
+
"step": 25
|
| 16 |
+
},
|
| 17 |
+
{
|
| 18 |
+
"epoch": 3.0,
|
| 19 |
+
"learning_rate": 5e-08,
|
| 20 |
+
"loss": 0.6364,
|
| 21 |
+
"step": 50
|
| 22 |
+
},
|
| 23 |
+
{
|
| 24 |
+
"epoch": 5.0,
|
| 25 |
+
"learning_rate": 7.5e-08,
|
| 26 |
+
"loss": 0.6418,
|
| 27 |
+
"step": 75
|
| 28 |
+
},
|
| 29 |
+
{
|
| 30 |
+
"epoch": 7.0,
|
| 31 |
+
"learning_rate": 1e-07,
|
| 32 |
+
"loss": 0.6378,
|
| 33 |
+
"step": 100
|
| 34 |
+
},
|
| 35 |
+
{
|
| 36 |
+
"epoch": 8.0,
|
| 37 |
+
"learning_rate": 1.25e-07,
|
| 38 |
+
"loss": 0.6318,
|
| 39 |
+
"step": 125
|
| 40 |
+
},
|
| 41 |
+
{
|
| 42 |
+
"epoch": 10.0,
|
| 43 |
+
"learning_rate": 1.5e-07,
|
| 44 |
+
"loss": 0.6215,
|
| 45 |
+
"step": 150
|
| 46 |
+
},
|
| 47 |
+
{
|
| 48 |
+
"epoch": 12.0,
|
| 49 |
+
"learning_rate": 1.75e-07,
|
| 50 |
+
"loss": 0.6273,
|
| 51 |
+
"step": 175
|
| 52 |
+
},
|
| 53 |
+
{
|
| 54 |
+
"epoch": 14.0,
|
| 55 |
+
"learning_rate": 2e-07,
|
| 56 |
+
"loss": 0.6081,
|
| 57 |
+
"step": 200
|
| 58 |
+
},
|
| 59 |
+
{
|
| 60 |
+
"epoch": 16.0,
|
| 61 |
+
"learning_rate": 2.25e-07,
|
| 62 |
+
"loss": 0.6053,
|
| 63 |
+
"step": 225
|
| 64 |
+
},
|
| 65 |
+
{
|
| 66 |
+
"epoch": 17.0,
|
| 67 |
+
"learning_rate": 2.5e-07,
|
| 68 |
+
"loss": 0.5989,
|
| 69 |
+
"step": 250
|
| 70 |
+
},
|
| 71 |
+
{
|
| 72 |
+
"epoch": 19.0,
|
| 73 |
+
"learning_rate": 2.75e-07,
|
| 74 |
+
"loss": 0.5812,
|
| 75 |
+
"step": 275
|
| 76 |
+
},
|
| 77 |
+
{
|
| 78 |
+
"epoch": 21.0,
|
| 79 |
+
"learning_rate": 3e-07,
|
| 80 |
+
"loss": 0.5828,
|
| 81 |
+
"step": 300
|
| 82 |
+
},
|
| 83 |
+
{
|
| 84 |
+
"epoch": 23.0,
|
| 85 |
+
"learning_rate": 3.25e-07,
|
| 86 |
+
"loss": 0.5657,
|
| 87 |
+
"step": 325
|
| 88 |
+
},
|
| 89 |
+
{
|
| 90 |
+
"epoch": 24.0,
|
| 91 |
+
"learning_rate": 3.5e-07,
|
| 92 |
+
"loss": 0.5588,
|
| 93 |
+
"step": 350
|
| 94 |
+
},
|
| 95 |
+
{
|
| 96 |
+
"epoch": 26.0,
|
| 97 |
+
"learning_rate": 3.75e-07,
|
| 98 |
+
"loss": 0.5519,
|
| 99 |
+
"step": 375
|
| 100 |
+
},
|
| 101 |
+
{
|
| 102 |
+
"epoch": 28.0,
|
| 103 |
+
"learning_rate": 4e-07,
|
| 104 |
+
"loss": 0.545,
|
| 105 |
+
"step": 400
|
| 106 |
+
},
|
| 107 |
+
{
|
| 108 |
+
"epoch": 30.0,
|
| 109 |
+
"learning_rate": 4.2499999999999995e-07,
|
| 110 |
+
"loss": 0.533,
|
| 111 |
+
"step": 425
|
| 112 |
+
},
|
| 113 |
+
{
|
| 114 |
+
"epoch": 32.0,
|
| 115 |
+
"learning_rate": 4.5e-07,
|
| 116 |
+
"loss": 0.5229,
|
| 117 |
+
"step": 450
|
| 118 |
+
},
|
| 119 |
+
{
|
| 120 |
+
"epoch": 33.0,
|
| 121 |
+
"learning_rate": 4.7499999999999995e-07,
|
| 122 |
+
"loss": 0.5166,
|
| 123 |
+
"step": 475
|
| 124 |
+
},
|
| 125 |
+
{
|
| 126 |
+
"epoch": 35.0,
|
| 127 |
+
"learning_rate": 5e-07,
|
| 128 |
+
"loss": 0.5011,
|
| 129 |
+
"step": 500
|
| 130 |
+
},
|
| 131 |
+
{
|
| 132 |
+
"epoch": 35.0,
|
| 133 |
+
"eval_loss": 0.5963125228881836,
|
| 134 |
+
"eval_runtime": 242.2717,
|
| 135 |
+
"eval_samples_per_second": 4.214,
|
| 136 |
+
"eval_steps_per_second": 0.132,
|
| 137 |
+
"eval_wer": 59.88676791247466,
|
| 138 |
+
"step": 500
|
| 139 |
+
},
|
| 140 |
+
{
|
| 141 |
+
"epoch": 37.0,
|
| 142 |
+
"learning_rate": 4.972222222222222e-07,
|
| 143 |
+
"loss": 0.4965,
|
| 144 |
+
"step": 525
|
| 145 |
+
},
|
| 146 |
+
{
|
| 147 |
+
"epoch": 39.0,
|
| 148 |
+
"learning_rate": 4.944444444444445e-07,
|
| 149 |
+
"loss": 0.4828,
|
| 150 |
+
"step": 550
|
| 151 |
+
},
|
| 152 |
+
{
|
| 153 |
+
"epoch": 41.0,
|
| 154 |
+
"learning_rate": 4.916666666666666e-07,
|
| 155 |
+
"loss": 0.4732,
|
| 156 |
+
"step": 575
|
| 157 |
+
},
|
| 158 |
+
{
|
| 159 |
+
"epoch": 42.0,
|
| 160 |
+
"learning_rate": 4.888888888888889e-07,
|
| 161 |
+
"loss": 0.4708,
|
| 162 |
+
"step": 600
|
| 163 |
+
},
|
| 164 |
+
{
|
| 165 |
+
"epoch": 44.0,
|
| 166 |
+
"learning_rate": 4.861111111111111e-07,
|
| 167 |
+
"loss": 0.4586,
|
| 168 |
+
"step": 625
|
| 169 |
+
},
|
| 170 |
+
{
|
| 171 |
+
"epoch": 46.0,
|
| 172 |
+
"learning_rate": 4.833333333333333e-07,
|
| 173 |
+
"loss": 0.4551,
|
| 174 |
+
"step": 650
|
| 175 |
+
},
|
| 176 |
+
{
|
| 177 |
+
"epoch": 48.0,
|
| 178 |
+
"learning_rate": 4.805555555555555e-07,
|
| 179 |
+
"loss": 0.4483,
|
| 180 |
+
"step": 675
|
| 181 |
+
},
|
| 182 |
+
{
|
| 183 |
+
"epoch": 49.0,
|
| 184 |
+
"learning_rate": 4.777777777777778e-07,
|
| 185 |
+
"loss": 0.437,
|
| 186 |
+
"step": 700
|
| 187 |
+
},
|
| 188 |
+
{
|
| 189 |
+
"epoch": 51.0,
|
| 190 |
+
"learning_rate": 4.7499999999999995e-07,
|
| 191 |
+
"loss": 0.4322,
|
| 192 |
+
"step": 725
|
| 193 |
+
},
|
| 194 |
+
{
|
| 195 |
+
"epoch": 53.0,
|
| 196 |
+
"learning_rate": 4.722222222222222e-07,
|
| 197 |
+
"loss": 0.4207,
|
| 198 |
+
"step": 750
|
| 199 |
+
},
|
| 200 |
+
{
|
| 201 |
+
"epoch": 55.0,
|
| 202 |
+
"learning_rate": 4.694444444444444e-07,
|
| 203 |
+
"loss": 0.4165,
|
| 204 |
+
"step": 775
|
| 205 |
+
},
|
| 206 |
+
{
|
| 207 |
+
"epoch": 57.0,
|
| 208 |
+
"learning_rate": 4.6666666666666666e-07,
|
| 209 |
+
"loss": 0.4121,
|
| 210 |
+
"step": 800
|
| 211 |
+
},
|
| 212 |
+
{
|
| 213 |
+
"epoch": 58.0,
|
| 214 |
+
"learning_rate": 4.6388888888888886e-07,
|
| 215 |
+
"loss": 0.4014,
|
| 216 |
+
"step": 825
|
| 217 |
+
},
|
| 218 |
+
{
|
| 219 |
+
"epoch": 60.0,
|
| 220 |
+
"learning_rate": 4.611111111111111e-07,
|
| 221 |
+
"loss": 0.3975,
|
| 222 |
+
"step": 850
|
| 223 |
+
},
|
| 224 |
+
{
|
| 225 |
+
"epoch": 62.0,
|
| 226 |
+
"learning_rate": 4.5833333333333327e-07,
|
| 227 |
+
"loss": 0.3938,
|
| 228 |
+
"step": 875
|
| 229 |
+
},
|
| 230 |
+
{
|
| 231 |
+
"epoch": 64.0,
|
| 232 |
+
"learning_rate": 4.555555555555555e-07,
|
| 233 |
+
"loss": 0.3806,
|
| 234 |
+
"step": 900
|
| 235 |
+
},
|
| 236 |
+
{
|
| 237 |
+
"epoch": 66.0,
|
| 238 |
+
"learning_rate": 4.527777777777778e-07,
|
| 239 |
+
"loss": 0.3835,
|
| 240 |
+
"step": 925
|
| 241 |
+
},
|
| 242 |
+
{
|
| 243 |
+
"epoch": 67.0,
|
| 244 |
+
"learning_rate": 4.5e-07,
|
| 245 |
+
"loss": 0.372,
|
| 246 |
+
"step": 950
|
| 247 |
+
},
|
| 248 |
+
{
|
| 249 |
+
"epoch": 69.0,
|
| 250 |
+
"learning_rate": 4.4722222222222223e-07,
|
| 251 |
+
"loss": 0.3652,
|
| 252 |
+
"step": 975
|
| 253 |
+
},
|
| 254 |
+
{
|
| 255 |
+
"epoch": 71.0,
|
| 256 |
+
"learning_rate": 4.444444444444444e-07,
|
| 257 |
+
"loss": 0.3648,
|
| 258 |
+
"step": 1000
|
| 259 |
+
},
|
| 260 |
+
{
|
| 261 |
+
"epoch": 71.0,
|
| 262 |
+
"eval_loss": 0.5612689256668091,
|
| 263 |
+
"eval_runtime": 219.2597,
|
| 264 |
+
"eval_samples_per_second": 4.657,
|
| 265 |
+
"eval_steps_per_second": 0.146,
|
| 266 |
+
"eval_wer": 55.95424811598638,
|
| 267 |
+
"step": 1000
|
| 268 |
+
},
|
| 269 |
+
{
|
| 270 |
+
"epoch": 73.0,
|
| 271 |
+
"learning_rate": 4.4166666666666664e-07,
|
| 272 |
+
"loss": 0.3552,
|
| 273 |
+
"step": 1025
|
| 274 |
+
},
|
| 275 |
+
{
|
| 276 |
+
"epoch": 74.0,
|
| 277 |
+
"learning_rate": 4.3888888888888884e-07,
|
| 278 |
+
"loss": 0.3504,
|
| 279 |
+
"step": 1050
|
| 280 |
+
},
|
| 281 |
+
{
|
| 282 |
+
"epoch": 76.0,
|
| 283 |
+
"learning_rate": 4.361111111111111e-07,
|
| 284 |
+
"loss": 0.3452,
|
| 285 |
+
"step": 1075
|
| 286 |
+
},
|
| 287 |
+
{
|
| 288 |
+
"epoch": 78.0,
|
| 289 |
+
"learning_rate": 4.3333333333333335e-07,
|
| 290 |
+
"loss": 0.3425,
|
| 291 |
+
"step": 1100
|
| 292 |
+
},
|
| 293 |
+
{
|
| 294 |
+
"epoch": 80.0,
|
| 295 |
+
"learning_rate": 4.3055555555555555e-07,
|
| 296 |
+
"loss": 0.3314,
|
| 297 |
+
"step": 1125
|
| 298 |
+
},
|
| 299 |
+
{
|
| 300 |
+
"epoch": 82.0,
|
| 301 |
+
"learning_rate": 4.2777777777777775e-07,
|
| 302 |
+
"loss": 0.3347,
|
| 303 |
+
"step": 1150
|
| 304 |
+
},
|
| 305 |
+
{
|
| 306 |
+
"epoch": 83.0,
|
| 307 |
+
"learning_rate": 4.2499999999999995e-07,
|
| 308 |
+
"loss": 0.3247,
|
| 309 |
+
"step": 1175
|
| 310 |
+
},
|
| 311 |
+
{
|
| 312 |
+
"epoch": 85.0,
|
| 313 |
+
"learning_rate": 4.222222222222222e-07,
|
| 314 |
+
"loss": 0.3229,
|
| 315 |
+
"step": 1200
|
| 316 |
+
},
|
| 317 |
+
{
|
| 318 |
+
"epoch": 87.0,
|
| 319 |
+
"learning_rate": 4.194444444444444e-07,
|
| 320 |
+
"loss": 0.3165,
|
| 321 |
+
"step": 1225
|
| 322 |
+
},
|
| 323 |
+
{
|
| 324 |
+
"epoch": 89.0,
|
| 325 |
+
"learning_rate": 4.1666666666666667e-07,
|
| 326 |
+
"loss": 0.3168,
|
| 327 |
+
"step": 1250
|
| 328 |
+
},
|
| 329 |
+
{
|
| 330 |
+
"epoch": 91.0,
|
| 331 |
+
"learning_rate": 4.1388888888888887e-07,
|
| 332 |
+
"loss": 0.3051,
|
| 333 |
+
"step": 1275
|
| 334 |
+
},
|
| 335 |
+
{
|
| 336 |
+
"epoch": 92.0,
|
| 337 |
+
"learning_rate": 4.1111111111111107e-07,
|
| 338 |
+
"loss": 0.3035,
|
| 339 |
+
"step": 1300
|
| 340 |
+
},
|
| 341 |
+
{
|
| 342 |
+
"epoch": 94.0,
|
| 343 |
+
"learning_rate": 4.083333333333333e-07,
|
| 344 |
+
"loss": 0.3009,
|
| 345 |
+
"step": 1325
|
| 346 |
+
},
|
| 347 |
+
{
|
| 348 |
+
"epoch": 96.0,
|
| 349 |
+
"learning_rate": 4.055555555555555e-07,
|
| 350 |
+
"loss": 0.2981,
|
| 351 |
+
"step": 1350
|
| 352 |
+
},
|
| 353 |
+
{
|
| 354 |
+
"epoch": 98.0,
|
| 355 |
+
"learning_rate": 4.027777777777778e-07,
|
| 356 |
+
"loss": 0.29,
|
| 357 |
+
"step": 1375
|
| 358 |
+
},
|
| 359 |
+
{
|
| 360 |
+
"epoch": 99.0,
|
| 361 |
+
"learning_rate": 4e-07,
|
| 362 |
+
"loss": 0.2854,
|
| 363 |
+
"step": 1400
|
| 364 |
+
},
|
| 365 |
+
{
|
| 366 |
+
"epoch": 101.0,
|
| 367 |
+
"learning_rate": 3.972222222222222e-07,
|
| 368 |
+
"loss": 0.284,
|
| 369 |
+
"step": 1425
|
| 370 |
+
},
|
| 371 |
+
{
|
| 372 |
+
"epoch": 103.0,
|
| 373 |
+
"learning_rate": 3.9444444444444444e-07,
|
| 374 |
+
"loss": 0.2805,
|
| 375 |
+
"step": 1450
|
| 376 |
+
},
|
| 377 |
+
{
|
| 378 |
+
"epoch": 105.0,
|
| 379 |
+
"learning_rate": 3.9166666666666664e-07,
|
| 380 |
+
"loss": 0.2743,
|
| 381 |
+
"step": 1475
|
| 382 |
+
},
|
| 383 |
+
{
|
| 384 |
+
"epoch": 107.0,
|
| 385 |
+
"learning_rate": 3.888888888888889e-07,
|
| 386 |
+
"loss": 0.2732,
|
| 387 |
+
"step": 1500
|
| 388 |
+
},
|
| 389 |
+
{
|
| 390 |
+
"epoch": 107.0,
|
| 391 |
+
"eval_loss": 0.5504247546195984,
|
| 392 |
+
"eval_runtime": 221.6747,
|
| 393 |
+
"eval_samples_per_second": 4.606,
|
| 394 |
+
"eval_steps_per_second": 0.144,
|
| 395 |
+
"eval_wer": 54.4585134463104,
|
| 396 |
+
"step": 1500
|
| 397 |
+
},
|
| 398 |
+
{
|
| 399 |
+
"epoch": 108.0,
|
| 400 |
+
"learning_rate": 3.861111111111111e-07,
|
| 401 |
+
"loss": 0.2677,
|
| 402 |
+
"step": 1525
|
| 403 |
+
},
|
| 404 |
+
{
|
| 405 |
+
"epoch": 110.0,
|
| 406 |
+
"learning_rate": 3.8333333333333335e-07,
|
| 407 |
+
"loss": 0.2633,
|
| 408 |
+
"step": 1550
|
| 409 |
+
},
|
| 410 |
+
{
|
| 411 |
+
"epoch": 112.0,
|
| 412 |
+
"learning_rate": 3.805555555555555e-07,
|
| 413 |
+
"loss": 0.2601,
|
| 414 |
+
"step": 1575
|
| 415 |
+
},
|
| 416 |
+
{
|
| 417 |
+
"epoch": 114.0,
|
| 418 |
+
"learning_rate": 3.7777777777777775e-07,
|
| 419 |
+
"loss": 0.2608,
|
| 420 |
+
"step": 1600
|
| 421 |
+
},
|
| 422 |
+
{
|
| 423 |
+
"epoch": 116.0,
|
| 424 |
+
"learning_rate": 3.75e-07,
|
| 425 |
+
"loss": 0.2539,
|
| 426 |
+
"step": 1625
|
| 427 |
+
},
|
| 428 |
+
{
|
| 429 |
+
"epoch": 117.0,
|
| 430 |
+
"learning_rate": 3.722222222222222e-07,
|
| 431 |
+
"loss": 0.2513,
|
| 432 |
+
"step": 1650
|
| 433 |
+
},
|
| 434 |
+
{
|
| 435 |
+
"epoch": 119.0,
|
| 436 |
+
"learning_rate": 3.6944444444444447e-07,
|
| 437 |
+
"loss": 0.2483,
|
| 438 |
+
"step": 1675
|
| 439 |
+
},
|
| 440 |
+
{
|
| 441 |
+
"epoch": 121.0,
|
| 442 |
+
"learning_rate": 3.666666666666666e-07,
|
| 443 |
+
"loss": 0.2436,
|
| 444 |
+
"step": 1700
|
| 445 |
+
},
|
| 446 |
+
{
|
| 447 |
+
"epoch": 123.0,
|
| 448 |
+
"learning_rate": 3.6388888888888887e-07,
|
| 449 |
+
"loss": 0.2394,
|
| 450 |
+
"step": 1725
|
| 451 |
+
},
|
| 452 |
+
{
|
| 453 |
+
"epoch": 124.0,
|
| 454 |
+
"learning_rate": 3.6111111111111107e-07,
|
| 455 |
+
"loss": 0.2372,
|
| 456 |
+
"step": 1750
|
| 457 |
+
},
|
| 458 |
+
{
|
| 459 |
+
"epoch": 126.0,
|
| 460 |
+
"learning_rate": 3.583333333333333e-07,
|
| 461 |
+
"loss": 0.2347,
|
| 462 |
+
"step": 1775
|
| 463 |
+
},
|
| 464 |
+
{
|
| 465 |
+
"epoch": 128.0,
|
| 466 |
+
"learning_rate": 3.5555555555555553e-07,
|
| 467 |
+
"loss": 0.2312,
|
| 468 |
+
"step": 1800
|
| 469 |
+
},
|
| 470 |
+
{
|
| 471 |
+
"epoch": 130.0,
|
| 472 |
+
"learning_rate": 3.527777777777778e-07,
|
| 473 |
+
"loss": 0.2294,
|
| 474 |
+
"step": 1825
|
| 475 |
+
},
|
| 476 |
+
{
|
| 477 |
+
"epoch": 132.0,
|
| 478 |
+
"learning_rate": 3.5e-07,
|
| 479 |
+
"loss": 0.2241,
|
| 480 |
+
"step": 1850
|
| 481 |
+
},
|
| 482 |
+
{
|
| 483 |
+
"epoch": 133.0,
|
| 484 |
+
"learning_rate": 3.472222222222222e-07,
|
| 485 |
+
"loss": 0.2249,
|
| 486 |
+
"step": 1875
|
| 487 |
+
},
|
| 488 |
+
{
|
| 489 |
+
"epoch": 135.0,
|
| 490 |
+
"learning_rate": 3.4444444444444444e-07,
|
| 491 |
+
"loss": 0.2187,
|
| 492 |
+
"step": 1900
|
| 493 |
+
},
|
| 494 |
+
{
|
| 495 |
+
"epoch": 137.0,
|
| 496 |
+
"learning_rate": 3.4166666666666664e-07,
|
| 497 |
+
"loss": 0.2171,
|
| 498 |
+
"step": 1925
|
| 499 |
+
},
|
| 500 |
+
{
|
| 501 |
+
"epoch": 139.0,
|
| 502 |
+
"learning_rate": 3.388888888888889e-07,
|
| 503 |
+
"loss": 0.2139,
|
| 504 |
+
"step": 1950
|
| 505 |
+
},
|
| 506 |
+
{
|
| 507 |
+
"epoch": 141.0,
|
| 508 |
+
"learning_rate": 3.361111111111111e-07,
|
| 509 |
+
"loss": 0.2121,
|
| 510 |
+
"step": 1975
|
| 511 |
+
},
|
| 512 |
+
{
|
| 513 |
+
"epoch": 142.0,
|
| 514 |
+
"learning_rate": 3.333333333333333e-07,
|
| 515 |
+
"loss": 0.2081,
|
| 516 |
+
"step": 2000
|
| 517 |
+
},
|
| 518 |
+
{
|
| 519 |
+
"epoch": 142.0,
|
| 520 |
+
"eval_loss": 0.5501875281333923,
|
| 521 |
+
"eval_runtime": 220.7799,
|
| 522 |
+
"eval_samples_per_second": 4.625,
|
| 523 |
+
"eval_steps_per_second": 0.145,
|
| 524 |
+
"eval_wer": 53.6704793236678,
|
| 525 |
+
"step": 2000
|
| 526 |
+
},
|
| 527 |
+
{
|
| 528 |
+
"epoch": 144.0,
|
| 529 |
+
"learning_rate": 3.3055555555555556e-07,
|
| 530 |
+
"loss": 0.2082,
|
| 531 |
+
"step": 2025
|
| 532 |
+
},
|
| 533 |
+
{
|
| 534 |
+
"epoch": 146.0,
|
| 535 |
+
"learning_rate": 3.2777777777777776e-07,
|
| 536 |
+
"loss": 0.2039,
|
| 537 |
+
"step": 2050
|
| 538 |
+
},
|
| 539 |
+
{
|
| 540 |
+
"epoch": 148.0,
|
| 541 |
+
"learning_rate": 3.25e-07,
|
| 542 |
+
"loss": 0.2022,
|
| 543 |
+
"step": 2075
|
| 544 |
+
},
|
| 545 |
+
{
|
| 546 |
+
"epoch": 149.0,
|
| 547 |
+
"learning_rate": 3.222222222222222e-07,
|
| 548 |
+
"loss": 0.1984,
|
| 549 |
+
"step": 2100
|
| 550 |
+
},
|
| 551 |
+
{
|
| 552 |
+
"epoch": 151.0,
|
| 553 |
+
"learning_rate": 3.194444444444444e-07,
|
| 554 |
+
"loss": 0.1969,
|
| 555 |
+
"step": 2125
|
| 556 |
+
},
|
| 557 |
+
{
|
| 558 |
+
"epoch": 153.0,
|
| 559 |
+
"learning_rate": 3.166666666666666e-07,
|
| 560 |
+
"loss": 0.1938,
|
| 561 |
+
"step": 2150
|
| 562 |
+
},
|
| 563 |
+
{
|
| 564 |
+
"epoch": 155.0,
|
| 565 |
+
"learning_rate": 3.1388888888888887e-07,
|
| 566 |
+
"loss": 0.193,
|
| 567 |
+
"step": 2175
|
| 568 |
+
},
|
| 569 |
+
{
|
| 570 |
+
"epoch": 157.0,
|
| 571 |
+
"learning_rate": 3.111111111111111e-07,
|
| 572 |
+
"loss": 0.1908,
|
| 573 |
+
"step": 2200
|
| 574 |
+
},
|
| 575 |
+
{
|
| 576 |
+
"epoch": 158.0,
|
| 577 |
+
"learning_rate": 3.0833333333333333e-07,
|
| 578 |
+
"loss": 0.1864,
|
| 579 |
+
"step": 2225
|
| 580 |
+
},
|
| 581 |
+
{
|
| 582 |
+
"epoch": 160.0,
|
| 583 |
+
"learning_rate": 3.055555555555556e-07,
|
| 584 |
+
"loss": 0.1846,
|
| 585 |
+
"step": 2250
|
| 586 |
+
},
|
| 587 |
+
{
|
| 588 |
+
"epoch": 162.0,
|
| 589 |
+
"learning_rate": 3.0277777777777773e-07,
|
| 590 |
+
"loss": 0.1834,
|
| 591 |
+
"step": 2275
|
| 592 |
+
},
|
| 593 |
+
{
|
| 594 |
+
"epoch": 164.0,
|
| 595 |
+
"learning_rate": 3e-07,
|
| 596 |
+
"loss": 0.1808,
|
| 597 |
+
"step": 2300
|
| 598 |
+
},
|
| 599 |
+
{
|
| 600 |
+
"epoch": 166.0,
|
| 601 |
+
"learning_rate": 2.972222222222222e-07,
|
| 602 |
+
"loss": 0.1787,
|
| 603 |
+
"step": 2325
|
| 604 |
+
},
|
| 605 |
+
{
|
| 606 |
+
"epoch": 167.0,
|
| 607 |
+
"learning_rate": 2.9444444444444444e-07,
|
| 608 |
+
"loss": 0.1769,
|
| 609 |
+
"step": 2350
|
| 610 |
+
},
|
| 611 |
+
{
|
| 612 |
+
"epoch": 169.0,
|
| 613 |
+
"learning_rate": 2.916666666666667e-07,
|
| 614 |
+
"loss": 0.174,
|
| 615 |
+
"step": 2375
|
| 616 |
+
},
|
| 617 |
+
{
|
| 618 |
+
"epoch": 171.0,
|
| 619 |
+
"learning_rate": 2.8888888888888885e-07,
|
| 620 |
+
"loss": 0.1718,
|
| 621 |
+
"step": 2400
|
| 622 |
+
},
|
| 623 |
+
{
|
| 624 |
+
"epoch": 173.0,
|
| 625 |
+
"learning_rate": 2.861111111111111e-07,
|
| 626 |
+
"loss": 0.1707,
|
| 627 |
+
"step": 2425
|
| 628 |
+
},
|
| 629 |
+
{
|
| 630 |
+
"epoch": 174.0,
|
| 631 |
+
"learning_rate": 2.833333333333333e-07,
|
| 632 |
+
"loss": 0.1684,
|
| 633 |
+
"step": 2450
|
| 634 |
+
},
|
| 635 |
+
{
|
| 636 |
+
"epoch": 176.0,
|
| 637 |
+
"learning_rate": 2.8055555555555556e-07,
|
| 638 |
+
"loss": 0.1671,
|
| 639 |
+
"step": 2475
|
| 640 |
+
},
|
| 641 |
+
{
|
| 642 |
+
"epoch": 178.0,
|
| 643 |
+
"learning_rate": 2.7777777777777776e-07,
|
| 644 |
+
"loss": 0.1627,
|
| 645 |
+
"step": 2500
|
| 646 |
+
},
|
| 647 |
+
{
|
| 648 |
+
"epoch": 178.0,
|
| 649 |
+
"eval_loss": 0.5558388233184814,
|
| 650 |
+
"eval_runtime": 219.4948,
|
| 651 |
+
"eval_samples_per_second": 4.652,
|
| 652 |
+
"eval_steps_per_second": 0.146,
|
| 653 |
+
"eval_wer": 53.82732106652385,
|
| 654 |
+
"step": 2500
|
| 655 |
+
},
|
| 656 |
+
{
|
| 657 |
+
"epoch": 180.0,
|
| 658 |
+
"learning_rate": 2.75e-07,
|
| 659 |
+
"loss": 0.1628,
|
| 660 |
+
"step": 2525
|
| 661 |
+
},
|
| 662 |
+
{
|
| 663 |
+
"epoch": 182.0,
|
| 664 |
+
"learning_rate": 2.7222222222222216e-07,
|
| 665 |
+
"loss": 0.1587,
|
| 666 |
+
"step": 2550
|
| 667 |
+
},
|
| 668 |
+
{
|
| 669 |
+
"epoch": 183.0,
|
| 670 |
+
"learning_rate": 2.694444444444444e-07,
|
| 671 |
+
"loss": 0.1598,
|
| 672 |
+
"step": 2575
|
| 673 |
+
},
|
| 674 |
+
{
|
| 675 |
+
"epoch": 185.0,
|
| 676 |
+
"learning_rate": 2.6666666666666667e-07,
|
| 677 |
+
"loss": 0.1575,
|
| 678 |
+
"step": 2600
|
| 679 |
+
},
|
| 680 |
+
{
|
| 681 |
+
"epoch": 187.0,
|
| 682 |
+
"learning_rate": 2.638888888888889e-07,
|
| 683 |
+
"loss": 0.1547,
|
| 684 |
+
"step": 2625
|
| 685 |
+
},
|
| 686 |
+
{
|
| 687 |
+
"epoch": 189.0,
|
| 688 |
+
"learning_rate": 2.6111111111111113e-07,
|
| 689 |
+
"loss": 0.1524,
|
| 690 |
+
"step": 2650
|
| 691 |
+
},
|
| 692 |
+
{
|
| 693 |
+
"epoch": 191.0,
|
| 694 |
+
"learning_rate": 2.5833333333333333e-07,
|
| 695 |
+
"loss": 0.1523,
|
| 696 |
+
"step": 2675
|
| 697 |
+
},
|
| 698 |
+
{
|
| 699 |
+
"epoch": 192.0,
|
| 700 |
+
"learning_rate": 2.5555555555555553e-07,
|
| 701 |
+
"loss": 0.1507,
|
| 702 |
+
"step": 2700
|
| 703 |
+
},
|
| 704 |
+
{
|
| 705 |
+
"epoch": 194.0,
|
| 706 |
+
"learning_rate": 2.5277777777777773e-07,
|
| 707 |
+
"loss": 0.1484,
|
| 708 |
+
"step": 2725
|
| 709 |
+
},
|
| 710 |
+
{
|
| 711 |
+
"epoch": 196.0,
|
| 712 |
+
"learning_rate": 2.5e-07,
|
| 713 |
+
"loss": 0.1468,
|
| 714 |
+
"step": 2750
|
| 715 |
+
},
|
| 716 |
+
{
|
| 717 |
+
"epoch": 198.0,
|
| 718 |
+
"learning_rate": 2.4722222222222224e-07,
|
| 719 |
+
"loss": 0.144,
|
| 720 |
+
"step": 2775
|
| 721 |
+
},
|
| 722 |
+
{
|
| 723 |
+
"epoch": 199.0,
|
| 724 |
+
"learning_rate": 2.4444444444444445e-07,
|
| 725 |
+
"loss": 0.1452,
|
| 726 |
+
"step": 2800
|
| 727 |
+
},
|
| 728 |
+
{
|
| 729 |
+
"epoch": 201.0,
|
| 730 |
+
"learning_rate": 2.4166666666666665e-07,
|
| 731 |
+
"loss": 0.1427,
|
| 732 |
+
"step": 2825
|
| 733 |
+
},
|
| 734 |
+
{
|
| 735 |
+
"epoch": 203.0,
|
| 736 |
+
"learning_rate": 2.388888888888889e-07,
|
| 737 |
+
"loss": 0.14,
|
| 738 |
+
"step": 2850
|
| 739 |
+
},
|
| 740 |
+
{
|
| 741 |
+
"epoch": 205.0,
|
| 742 |
+
"learning_rate": 2.361111111111111e-07,
|
| 743 |
+
"loss": 0.1399,
|
| 744 |
+
"step": 2875
|
| 745 |
+
},
|
| 746 |
+
{
|
| 747 |
+
"epoch": 207.0,
|
| 748 |
+
"learning_rate": 2.3333333333333333e-07,
|
| 749 |
+
"loss": 0.1384,
|
| 750 |
+
"step": 2900
|
| 751 |
+
},
|
| 752 |
+
{
|
| 753 |
+
"epoch": 208.0,
|
| 754 |
+
"learning_rate": 2.3055555555555556e-07,
|
| 755 |
+
"loss": 0.1366,
|
| 756 |
+
"step": 2925
|
| 757 |
+
},
|
| 758 |
+
{
|
| 759 |
+
"epoch": 210.0,
|
| 760 |
+
"learning_rate": 2.2777777777777776e-07,
|
| 761 |
+
"loss": 0.1363,
|
| 762 |
+
"step": 2950
|
| 763 |
+
},
|
| 764 |
+
{
|
| 765 |
+
"epoch": 212.0,
|
| 766 |
+
"learning_rate": 2.25e-07,
|
| 767 |
+
"loss": 0.1331,
|
| 768 |
+
"step": 2975
|
| 769 |
+
},
|
| 770 |
+
{
|
| 771 |
+
"epoch": 214.0,
|
| 772 |
+
"learning_rate": 2.222222222222222e-07,
|
| 773 |
+
"loss": 0.133,
|
| 774 |
+
"step": 3000
|
| 775 |
+
},
|
| 776 |
+
{
|
| 777 |
+
"epoch": 214.0,
|
| 778 |
+
"eval_loss": 0.5628067255020142,
|
| 779 |
+
"eval_runtime": 220.2103,
|
| 780 |
+
"eval_samples_per_second": 4.636,
|
| 781 |
+
"eval_steps_per_second": 0.145,
|
| 782 |
+
"eval_wer": 53.662828506943114,
|
| 783 |
+
"step": 3000
|
| 784 |
+
},
|
| 785 |
+
{
|
| 786 |
+
"epoch": 216.0,
|
| 787 |
+
"learning_rate": 2.1944444444444442e-07,
|
| 788 |
+
"loss": 0.1324,
|
| 789 |
+
"step": 3025
|
| 790 |
+
},
|
| 791 |
+
{
|
| 792 |
+
"epoch": 217.0,
|
| 793 |
+
"learning_rate": 2.1666666666666667e-07,
|
| 794 |
+
"loss": 0.1301,
|
| 795 |
+
"step": 3050
|
| 796 |
+
},
|
| 797 |
+
{
|
| 798 |
+
"epoch": 219.0,
|
| 799 |
+
"learning_rate": 2.1388888888888888e-07,
|
| 800 |
+
"loss": 0.1288,
|
| 801 |
+
"step": 3075
|
| 802 |
+
},
|
| 803 |
+
{
|
| 804 |
+
"epoch": 221.0,
|
| 805 |
+
"learning_rate": 2.111111111111111e-07,
|
| 806 |
+
"loss": 0.128,
|
| 807 |
+
"step": 3100
|
| 808 |
+
},
|
| 809 |
+
{
|
| 810 |
+
"epoch": 223.0,
|
| 811 |
+
"learning_rate": 2.0833333333333333e-07,
|
| 812 |
+
"loss": 0.1266,
|
| 813 |
+
"step": 3125
|
| 814 |
+
},
|
| 815 |
+
{
|
| 816 |
+
"epoch": 224.0,
|
| 817 |
+
"learning_rate": 2.0555555555555553e-07,
|
| 818 |
+
"loss": 0.1258,
|
| 819 |
+
"step": 3150
|
| 820 |
+
},
|
| 821 |
+
{
|
| 822 |
+
"epoch": 226.0,
|
| 823 |
+
"learning_rate": 2.0277777777777776e-07,
|
| 824 |
+
"loss": 0.1251,
|
| 825 |
+
"step": 3175
|
| 826 |
+
},
|
| 827 |
+
{
|
| 828 |
+
"epoch": 228.0,
|
| 829 |
+
"learning_rate": 2e-07,
|
| 830 |
+
"loss": 0.1219,
|
| 831 |
+
"step": 3200
|
| 832 |
+
},
|
| 833 |
+
{
|
| 834 |
+
"epoch": 230.0,
|
| 835 |
+
"learning_rate": 1.9722222222222222e-07,
|
| 836 |
+
"loss": 0.1226,
|
| 837 |
+
"step": 3225
|
| 838 |
+
},
|
| 839 |
+
{
|
| 840 |
+
"epoch": 232.0,
|
| 841 |
+
"learning_rate": 1.9444444444444445e-07,
|
| 842 |
+
"loss": 0.1202,
|
| 843 |
+
"step": 3250
|
| 844 |
+
},
|
| 845 |
+
{
|
| 846 |
+
"epoch": 233.0,
|
| 847 |
+
"learning_rate": 1.9166666666666668e-07,
|
| 848 |
+
"loss": 0.1199,
|
| 849 |
+
"step": 3275
|
| 850 |
+
},
|
| 851 |
+
{
|
| 852 |
+
"epoch": 235.0,
|
| 853 |
+
"learning_rate": 1.8888888888888888e-07,
|
| 854 |
+
"loss": 0.1186,
|
| 855 |
+
"step": 3300
|
| 856 |
+
},
|
| 857 |
+
{
|
| 858 |
+
"epoch": 237.0,
|
| 859 |
+
"learning_rate": 1.861111111111111e-07,
|
| 860 |
+
"loss": 0.1173,
|
| 861 |
+
"step": 3325
|
| 862 |
+
},
|
| 863 |
+
{
|
| 864 |
+
"epoch": 239.0,
|
| 865 |
+
"learning_rate": 1.833333333333333e-07,
|
| 866 |
+
"loss": 0.1172,
|
| 867 |
+
"step": 3350
|
| 868 |
+
},
|
| 869 |
+
{
|
| 870 |
+
"epoch": 241.0,
|
| 871 |
+
"learning_rate": 1.8055555555555554e-07,
|
| 872 |
+
"loss": 0.115,
|
| 873 |
+
"step": 3375
|
| 874 |
+
},
|
| 875 |
+
{
|
| 876 |
+
"epoch": 242.0,
|
| 877 |
+
"learning_rate": 1.7777777777777776e-07,
|
| 878 |
+
"loss": 0.1151,
|
| 879 |
+
"step": 3400
|
| 880 |
+
},
|
| 881 |
+
{
|
| 882 |
+
"epoch": 244.0,
|
| 883 |
+
"learning_rate": 1.75e-07,
|
| 884 |
+
"loss": 0.1141,
|
| 885 |
+
"step": 3425
|
| 886 |
+
},
|
| 887 |
+
{
|
| 888 |
+
"epoch": 246.0,
|
| 889 |
+
"learning_rate": 1.7222222222222222e-07,
|
| 890 |
+
"loss": 0.1137,
|
| 891 |
+
"step": 3450
|
| 892 |
+
},
|
| 893 |
+
{
|
| 894 |
+
"epoch": 248.0,
|
| 895 |
+
"learning_rate": 1.6944444444444445e-07,
|
| 896 |
+
"loss": 0.1109,
|
| 897 |
+
"step": 3475
|
| 898 |
+
},
|
| 899 |
+
{
|
| 900 |
+
"epoch": 249.0,
|
| 901 |
+
"learning_rate": 1.6666666666666665e-07,
|
| 902 |
+
"loss": 0.1112,
|
| 903 |
+
"step": 3500
|
| 904 |
+
},
|
| 905 |
+
{
|
| 906 |
+
"epoch": 249.0,
|
| 907 |
+
"eval_loss": 0.5695652365684509,
|
| 908 |
+
"eval_runtime": 216.96,
|
| 909 |
+
"eval_samples_per_second": 4.706,
|
| 910 |
+
"eval_steps_per_second": 0.147,
|
| 911 |
+
"eval_wer": 54.07979801843846,
|
| 912 |
+
"step": 3500
|
| 913 |
+
},
|
| 914 |
+
{
|
| 915 |
+
"epoch": 251.0,
|
| 916 |
+
"learning_rate": 1.6388888888888888e-07,
|
| 917 |
+
"loss": 0.1103,
|
| 918 |
+
"step": 3525
|
| 919 |
+
},
|
| 920 |
+
{
|
| 921 |
+
"epoch": 253.0,
|
| 922 |
+
"learning_rate": 1.611111111111111e-07,
|
| 923 |
+
"loss": 0.1112,
|
| 924 |
+
"step": 3550
|
| 925 |
+
},
|
| 926 |
+
{
|
| 927 |
+
"epoch": 255.0,
|
| 928 |
+
"learning_rate": 1.583333333333333e-07,
|
| 929 |
+
"loss": 0.1081,
|
| 930 |
+
"step": 3575
|
| 931 |
+
},
|
| 932 |
+
{
|
| 933 |
+
"epoch": 257.0,
|
| 934 |
+
"learning_rate": 1.5555555555555556e-07,
|
| 935 |
+
"loss": 0.108,
|
| 936 |
+
"step": 3600
|
| 937 |
+
},
|
| 938 |
+
{
|
| 939 |
+
"epoch": 258.0,
|
| 940 |
+
"learning_rate": 1.527777777777778e-07,
|
| 941 |
+
"loss": 0.107,
|
| 942 |
+
"step": 3625
|
| 943 |
+
},
|
| 944 |
+
{
|
| 945 |
+
"epoch": 260.0,
|
| 946 |
+
"learning_rate": 1.5e-07,
|
| 947 |
+
"loss": 0.1057,
|
| 948 |
+
"step": 3650
|
| 949 |
+
},
|
| 950 |
+
{
|
| 951 |
+
"epoch": 262.0,
|
| 952 |
+
"learning_rate": 1.4722222222222222e-07,
|
| 953 |
+
"loss": 0.1053,
|
| 954 |
+
"step": 3675
|
| 955 |
+
},
|
| 956 |
+
{
|
| 957 |
+
"epoch": 264.0,
|
| 958 |
+
"learning_rate": 1.4444444444444442e-07,
|
| 959 |
+
"loss": 0.1055,
|
| 960 |
+
"step": 3700
|
| 961 |
+
},
|
| 962 |
+
{
|
| 963 |
+
"epoch": 266.0,
|
| 964 |
+
"learning_rate": 1.4166666666666665e-07,
|
| 965 |
+
"loss": 0.1053,
|
| 966 |
+
"step": 3725
|
| 967 |
+
},
|
| 968 |
+
{
|
| 969 |
+
"epoch": 267.0,
|
| 970 |
+
"learning_rate": 1.3888888888888888e-07,
|
| 971 |
+
"loss": 0.1043,
|
| 972 |
+
"step": 3750
|
| 973 |
+
},
|
| 974 |
+
{
|
| 975 |
+
"epoch": 269.0,
|
| 976 |
+
"learning_rate": 1.3611111111111108e-07,
|
| 977 |
+
"loss": 0.1031,
|
| 978 |
+
"step": 3775
|
| 979 |
+
},
|
| 980 |
+
{
|
| 981 |
+
"epoch": 271.0,
|
| 982 |
+
"learning_rate": 1.3333333333333334e-07,
|
| 983 |
+
"loss": 0.1008,
|
| 984 |
+
"step": 3800
|
| 985 |
+
},
|
| 986 |
+
{
|
| 987 |
+
"epoch": 273.0,
|
| 988 |
+
"learning_rate": 1.3055555555555556e-07,
|
| 989 |
+
"loss": 0.1027,
|
| 990 |
+
"step": 3825
|
| 991 |
+
},
|
| 992 |
+
{
|
| 993 |
+
"epoch": 274.0,
|
| 994 |
+
"learning_rate": 1.2777777777777777e-07,
|
| 995 |
+
"loss": 0.0995,
|
| 996 |
+
"step": 3850
|
| 997 |
+
},
|
| 998 |
+
{
|
| 999 |
+
"epoch": 276.0,
|
| 1000 |
+
"learning_rate": 1.25e-07,
|
| 1001 |
+
"loss": 0.0997,
|
| 1002 |
+
"step": 3875
|
| 1003 |
+
},
|
| 1004 |
+
{
|
| 1005 |
+
"epoch": 278.0,
|
| 1006 |
+
"learning_rate": 1.2222222222222222e-07,
|
| 1007 |
+
"loss": 0.101,
|
| 1008 |
+
"step": 3900
|
| 1009 |
+
},
|
| 1010 |
+
{
|
| 1011 |
+
"epoch": 280.0,
|
| 1012 |
+
"learning_rate": 1.1944444444444445e-07,
|
| 1013 |
+
"loss": 0.0985,
|
| 1014 |
+
"step": 3925
|
| 1015 |
+
},
|
| 1016 |
+
{
|
| 1017 |
+
"epoch": 282.0,
|
| 1018 |
+
"learning_rate": 1.1666666666666667e-07,
|
| 1019 |
+
"loss": 0.0983,
|
| 1020 |
+
"step": 3950
|
| 1021 |
+
},
|
| 1022 |
+
{
|
| 1023 |
+
"epoch": 283.0,
|
| 1024 |
+
"learning_rate": 1.1388888888888888e-07,
|
| 1025 |
+
"loss": 0.0979,
|
| 1026 |
+
"step": 3975
|
| 1027 |
+
},
|
| 1028 |
+
{
|
| 1029 |
+
"epoch": 285.0,
|
| 1030 |
+
"learning_rate": 1.111111111111111e-07,
|
| 1031 |
+
"loss": 0.0973,
|
| 1032 |
+
"step": 4000
|
| 1033 |
+
},
|
| 1034 |
+
{
|
| 1035 |
+
"epoch": 285.0,
|
| 1036 |
+
"eval_loss": 0.5748891830444336,
|
| 1037 |
+
"eval_runtime": 217.6279,
|
| 1038 |
+
"eval_samples_per_second": 4.691,
|
| 1039 |
+
"eval_steps_per_second": 0.147,
|
| 1040 |
+
"eval_wer": 53.99946444282927,
|
| 1041 |
+
"step": 4000
|
| 1042 |
+
},
|
| 1043 |
+
{
|
| 1044 |
+
"epoch": 287.0,
|
| 1045 |
+
"learning_rate": 1.0833333333333334e-07,
|
| 1046 |
+
"loss": 0.0967,
|
| 1047 |
+
"step": 4025
|
| 1048 |
+
},
|
| 1049 |
+
{
|
| 1050 |
+
"epoch": 289.0,
|
| 1051 |
+
"learning_rate": 1.0555555555555555e-07,
|
| 1052 |
+
"loss": 0.0958,
|
| 1053 |
+
"step": 4050
|
| 1054 |
+
},
|
| 1055 |
+
{
|
| 1056 |
+
"epoch": 291.0,
|
| 1057 |
+
"learning_rate": 1.0277777777777777e-07,
|
| 1058 |
+
"loss": 0.0965,
|
| 1059 |
+
"step": 4075
|
| 1060 |
+
},
|
| 1061 |
+
{
|
| 1062 |
+
"epoch": 292.0,
|
| 1063 |
+
"learning_rate": 1e-07,
|
| 1064 |
+
"loss": 0.0965,
|
| 1065 |
+
"step": 4100
|
| 1066 |
+
},
|
| 1067 |
+
{
|
| 1068 |
+
"epoch": 294.0,
|
| 1069 |
+
"learning_rate": 9.722222222222222e-08,
|
| 1070 |
+
"loss": 0.0948,
|
| 1071 |
+
"step": 4125
|
| 1072 |
+
},
|
| 1073 |
+
{
|
| 1074 |
+
"epoch": 296.0,
|
| 1075 |
+
"learning_rate": 9.444444444444444e-08,
|
| 1076 |
+
"loss": 0.0941,
|
| 1077 |
+
"step": 4150
|
| 1078 |
+
},
|
| 1079 |
+
{
|
| 1080 |
+
"epoch": 298.0,
|
| 1081 |
+
"learning_rate": 9.166666666666665e-08,
|
| 1082 |
+
"loss": 0.0935,
|
| 1083 |
+
"step": 4175
|
| 1084 |
+
},
|
| 1085 |
+
{
|
| 1086 |
+
"epoch": 299.0,
|
| 1087 |
+
"learning_rate": 8.888888888888888e-08,
|
| 1088 |
+
"loss": 0.0937,
|
| 1089 |
+
"step": 4200
|
| 1090 |
+
},
|
| 1091 |
+
{
|
| 1092 |
+
"epoch": 301.0,
|
| 1093 |
+
"learning_rate": 8.611111111111111e-08,
|
| 1094 |
+
"loss": 0.0928,
|
| 1095 |
+
"step": 4225
|
| 1096 |
+
},
|
| 1097 |
+
{
|
| 1098 |
+
"epoch": 303.0,
|
| 1099 |
+
"learning_rate": 8.333333333333333e-08,
|
| 1100 |
+
"loss": 0.0924,
|
| 1101 |
+
"step": 4250
|
| 1102 |
+
},
|
| 1103 |
+
{
|
| 1104 |
+
"epoch": 305.0,
|
| 1105 |
+
"learning_rate": 8.055555555555555e-08,
|
| 1106 |
+
"loss": 0.0924,
|
| 1107 |
+
"step": 4275
|
| 1108 |
+
},
|
| 1109 |
+
{
|
| 1110 |
+
"epoch": 307.0,
|
| 1111 |
+
"learning_rate": 7.777777777777778e-08,
|
| 1112 |
+
"loss": 0.091,
|
| 1113 |
+
"step": 4300
|
| 1114 |
+
},
|
| 1115 |
+
{
|
| 1116 |
+
"epoch": 308.0,
|
| 1117 |
+
"learning_rate": 7.5e-08,
|
| 1118 |
+
"loss": 0.0916,
|
| 1119 |
+
"step": 4325
|
| 1120 |
+
},
|
| 1121 |
+
{
|
| 1122 |
+
"epoch": 310.0,
|
| 1123 |
+
"learning_rate": 7.222222222222221e-08,
|
| 1124 |
+
"loss": 0.0915,
|
| 1125 |
+
"step": 4350
|
| 1126 |
+
},
|
| 1127 |
+
{
|
| 1128 |
+
"epoch": 312.0,
|
| 1129 |
+
"learning_rate": 6.944444444444444e-08,
|
| 1130 |
+
"loss": 0.0907,
|
| 1131 |
+
"step": 4375
|
| 1132 |
+
},
|
| 1133 |
+
{
|
| 1134 |
+
"epoch": 314.0,
|
| 1135 |
+
"learning_rate": 6.666666666666667e-08,
|
| 1136 |
+
"loss": 0.0894,
|
| 1137 |
+
"step": 4400
|
| 1138 |
+
},
|
| 1139 |
+
{
|
| 1140 |
+
"epoch": 316.0,
|
| 1141 |
+
"learning_rate": 6.388888888888888e-08,
|
| 1142 |
+
"loss": 0.0911,
|
| 1143 |
+
"step": 4425
|
| 1144 |
+
},
|
| 1145 |
+
{
|
| 1146 |
+
"epoch": 317.0,
|
| 1147 |
+
"learning_rate": 6.111111111111111e-08,
|
| 1148 |
+
"loss": 0.0899,
|
| 1149 |
+
"step": 4450
|
| 1150 |
+
},
|
| 1151 |
+
{
|
| 1152 |
+
"epoch": 319.0,
|
| 1153 |
+
"learning_rate": 5.833333333333333e-08,
|
| 1154 |
+
"loss": 0.0892,
|
| 1155 |
+
"step": 4475
|
| 1156 |
+
},
|
| 1157 |
+
{
|
| 1158 |
+
"epoch": 321.0,
|
| 1159 |
+
"learning_rate": 5.555555555555555e-08,
|
| 1160 |
+
"loss": 0.0906,
|
| 1161 |
+
"step": 4500
|
| 1162 |
+
},
|
| 1163 |
+
{
|
| 1164 |
+
"epoch": 321.0,
|
| 1165 |
+
"eval_loss": 0.5782893300056458,
|
| 1166 |
+
"eval_runtime": 236.4641,
|
| 1167 |
+
"eval_samples_per_second": 4.318,
|
| 1168 |
+
"eval_steps_per_second": 0.135,
|
| 1169 |
+
"eval_wer": 54.14865536896064,
|
| 1170 |
+
"step": 4500
|
| 1171 |
+
},
|
| 1172 |
+
{
|
| 1173 |
+
"epoch": 323.0,
|
| 1174 |
+
"learning_rate": 5.2777777777777776e-08,
|
| 1175 |
+
"loss": 0.089,
|
| 1176 |
+
"step": 4525
|
| 1177 |
+
},
|
| 1178 |
+
{
|
| 1179 |
+
"epoch": 324.0,
|
| 1180 |
+
"learning_rate": 5e-08,
|
| 1181 |
+
"loss": 0.0891,
|
| 1182 |
+
"step": 4550
|
| 1183 |
+
},
|
| 1184 |
+
{
|
| 1185 |
+
"epoch": 326.0,
|
| 1186 |
+
"learning_rate": 4.722222222222222e-08,
|
| 1187 |
+
"loss": 0.0891,
|
| 1188 |
+
"step": 4575
|
| 1189 |
+
},
|
| 1190 |
+
{
|
| 1191 |
+
"epoch": 328.0,
|
| 1192 |
+
"learning_rate": 4.444444444444444e-08,
|
| 1193 |
+
"loss": 0.0878,
|
| 1194 |
+
"step": 4600
|
| 1195 |
+
},
|
| 1196 |
+
{
|
| 1197 |
+
"epoch": 330.0,
|
| 1198 |
+
"learning_rate": 4.166666666666666e-08,
|
| 1199 |
+
"loss": 0.0886,
|
| 1200 |
+
"step": 4625
|
| 1201 |
+
},
|
| 1202 |
+
{
|
| 1203 |
+
"epoch": 332.0,
|
| 1204 |
+
"learning_rate": 3.888888888888889e-08,
|
| 1205 |
+
"loss": 0.0877,
|
| 1206 |
+
"step": 4650
|
| 1207 |
+
},
|
| 1208 |
+
{
|
| 1209 |
+
"epoch": 333.0,
|
| 1210 |
+
"learning_rate": 3.6111111111111106e-08,
|
| 1211 |
+
"loss": 0.0882,
|
| 1212 |
+
"step": 4675
|
| 1213 |
+
},
|
| 1214 |
+
{
|
| 1215 |
+
"epoch": 335.0,
|
| 1216 |
+
"learning_rate": 3.3333333333333334e-08,
|
| 1217 |
+
"loss": 0.0871,
|
| 1218 |
+
"step": 4700
|
| 1219 |
+
},
|
| 1220 |
+
{
|
| 1221 |
+
"epoch": 337.0,
|
| 1222 |
+
"learning_rate": 3.0555555555555556e-08,
|
| 1223 |
+
"loss": 0.0886,
|
| 1224 |
+
"step": 4725
|
| 1225 |
+
},
|
| 1226 |
+
{
|
| 1227 |
+
"epoch": 339.0,
|
| 1228 |
+
"learning_rate": 2.7777777777777774e-08,
|
| 1229 |
+
"loss": 0.0882,
|
| 1230 |
+
"step": 4750
|
| 1231 |
+
},
|
| 1232 |
+
{
|
| 1233 |
+
"epoch": 341.0,
|
| 1234 |
+
"learning_rate": 2.5e-08,
|
| 1235 |
+
"loss": 0.0872,
|
| 1236 |
+
"step": 4775
|
| 1237 |
+
},
|
| 1238 |
+
{
|
| 1239 |
+
"epoch": 342.0,
|
| 1240 |
+
"learning_rate": 2.222222222222222e-08,
|
| 1241 |
+
"loss": 0.0871,
|
| 1242 |
+
"step": 4800
|
| 1243 |
+
},
|
| 1244 |
+
{
|
| 1245 |
+
"epoch": 344.0,
|
| 1246 |
+
"learning_rate": 1.9444444444444445e-08,
|
| 1247 |
+
"loss": 0.0884,
|
| 1248 |
+
"step": 4825
|
| 1249 |
+
},
|
| 1250 |
+
{
|
| 1251 |
+
"epoch": 346.0,
|
| 1252 |
+
"learning_rate": 1.6666666666666667e-08,
|
| 1253 |
+
"loss": 0.0864,
|
| 1254 |
+
"step": 4850
|
| 1255 |
+
},
|
| 1256 |
+
{
|
| 1257 |
+
"epoch": 348.0,
|
| 1258 |
+
"learning_rate": 1.3888888888888887e-08,
|
| 1259 |
+
"loss": 0.0878,
|
| 1260 |
+
"step": 4875
|
| 1261 |
+
},
|
| 1262 |
+
{
|
| 1263 |
+
"epoch": 349.0,
|
| 1264 |
+
"learning_rate": 1.111111111111111e-08,
|
| 1265 |
+
"loss": 0.0865,
|
| 1266 |
+
"step": 4900
|
| 1267 |
+
},
|
| 1268 |
+
{
|
| 1269 |
+
"epoch": 351.0,
|
| 1270 |
+
"learning_rate": 8.333333333333334e-09,
|
| 1271 |
+
"loss": 0.0873,
|
| 1272 |
+
"step": 4925
|
| 1273 |
+
},
|
| 1274 |
+
{
|
| 1275 |
+
"epoch": 353.0,
|
| 1276 |
+
"learning_rate": 5.555555555555555e-09,
|
| 1277 |
+
"loss": 0.0868,
|
| 1278 |
+
"step": 4950
|
| 1279 |
+
},
|
| 1280 |
+
{
|
| 1281 |
+
"epoch": 355.0,
|
| 1282 |
+
"learning_rate": 2.7777777777777776e-09,
|
| 1283 |
+
"loss": 0.0865,
|
| 1284 |
+
"step": 4975
|
| 1285 |
+
},
|
| 1286 |
+
{
|
| 1287 |
+
"epoch": 357.0,
|
| 1288 |
+
"learning_rate": 0.0,
|
| 1289 |
+
"loss": 0.0874,
|
| 1290 |
+
"step": 5000
|
| 1291 |
+
},
|
| 1292 |
+
{
|
| 1293 |
+
"epoch": 357.0,
|
| 1294 |
+
"eval_loss": 0.5793120861053467,
|
| 1295 |
+
"eval_runtime": 218.3558,
|
| 1296 |
+
"eval_samples_per_second": 4.676,
|
| 1297 |
+
"eval_steps_per_second": 0.147,
|
| 1298 |
+
"eval_wer": 54.228988944569835,
|
| 1299 |
+
"step": 5000
|
| 1300 |
+
},
|
| 1301 |
+
{
|
| 1302 |
+
"epoch": 357.0,
|
| 1303 |
+
"step": 5000,
|
| 1304 |
+
"total_flos": 2.024578836135936e+19,
|
| 1305 |
+
"train_loss": 0.22935251636505127,
|
| 1306 |
+
"train_runtime": 22650.8672,
|
| 1307 |
+
"train_samples_per_second": 14.127,
|
| 1308 |
+
"train_steps_per_second": 0.221
|
| 1309 |
+
}
|
| 1310 |
+
],
|
| 1311 |
+
"logging_steps": 25,
|
| 1312 |
+
"max_steps": 5000,
|
| 1313 |
+
"num_input_tokens_seen": 0,
|
| 1314 |
+
"num_train_epochs": 9223372036854775807,
|
| 1315 |
+
"save_steps": 500,
|
| 1316 |
+
"total_flos": 2.024578836135936e+19,
|
| 1317 |
+
"train_batch_size": 32,
|
| 1318 |
+
"trial_name": null,
|
| 1319 |
+
"trial_params": null
|
| 1320 |
+
}
|