End of training
Browse files- README.md +7 -2
- all_results.json +15 -0
- phase_1_train_results.json +9 -0
- runs/Apr25_23-51-20_dgx-a100/events.out.tfevents.1745717211.dgx-a100.855256.1 +3 -0
- trainer_state.json +2243 -0
- val_results.json +9 -0
README.md
CHANGED
|
@@ -1,20 +1,25 @@
|
|
| 1 |
---
|
| 2 |
library_name: transformers
|
|
|
|
|
|
|
| 3 |
license: apache-2.0
|
| 4 |
base_model: openai/whisper-tiny
|
| 5 |
tags:
|
|
|
|
|
|
|
|
|
|
| 6 |
- generated_from_trainer
|
| 7 |
metrics:
|
| 8 |
- wer
|
| 9 |
model-index:
|
| 10 |
-
- name:
|
| 11 |
results: []
|
| 12 |
---
|
| 13 |
|
| 14 |
<!-- This model card has been generated automatically according to the information the Trainer had access to. You
|
| 15 |
should probably proofread and complete it, then remove this comment. -->
|
| 16 |
|
| 17 |
-
#
|
| 18 |
|
| 19 |
This model is a fine-tuned version of [openai/whisper-tiny](https://huggingface.co/openai/whisper-tiny) on the None dataset.
|
| 20 |
It achieves the following results on the evaluation set:
|
|
|
|
| 1 |
---
|
| 2 |
library_name: transformers
|
| 3 |
+
language:
|
| 4 |
+
- jv
|
| 5 |
license: apache-2.0
|
| 6 |
base_model: openai/whisper-tiny
|
| 7 |
tags:
|
| 8 |
+
- whisper
|
| 9 |
+
- javanese
|
| 10 |
+
- asr
|
| 11 |
- generated_from_trainer
|
| 12 |
metrics:
|
| 13 |
- wer
|
| 14 |
model-index:
|
| 15 |
+
- name: Whisper-Tiny-Java-v6
|
| 16 |
results: []
|
| 17 |
---
|
| 18 |
|
| 19 |
<!-- This model card has been generated automatically according to the information the Trainer had access to. You
|
| 20 |
should probably proofread and complete it, then remove this comment. -->
|
| 21 |
|
| 22 |
+
# Whisper-Tiny-Java-v6
|
| 23 |
|
| 24 |
This model is a fine-tuned version of [openai/whisper-tiny](https://huggingface.co/openai/whisper-tiny) on the None dataset.
|
| 25 |
It achieves the following results on the evaluation set:
|
all_results.json
ADDED
|
@@ -0,0 +1,15 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"epoch": 159.7456,
|
| 3 |
+
"phase_1_val_loss": 1.0391464233398438,
|
| 4 |
+
"phase_1_val_runtime": 101.9929,
|
| 5 |
+
"phase_1_val_samples_per_second": 6.128,
|
| 6 |
+
"phase_1_val_steps_per_second": 1.539,
|
| 7 |
+
"phase_1_val_wer": 0.359779867002981,
|
| 8 |
+
"total_flos": 1.966379287412736e+19,
|
| 9 |
+
"train_loss": 0.0770821487186849,
|
| 10 |
+
"train_runtime": 117170.0508,
|
| 11 |
+
"train_samples": 5000,
|
| 12 |
+
"train_samples_per_second": 6.828,
|
| 13 |
+
"train_steps_per_second": 0.427,
|
| 14 |
+
"val_samples": 625
|
| 15 |
+
}
|
phase_1_train_results.json
ADDED
|
@@ -0,0 +1,9 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"epoch": 159.7456,
|
| 3 |
+
"total_flos": 1.966379287412736e+19,
|
| 4 |
+
"train_loss": 0.0770821487186849,
|
| 5 |
+
"train_runtime": 117170.0508,
|
| 6 |
+
"train_samples": 5000,
|
| 7 |
+
"train_samples_per_second": 6.828,
|
| 8 |
+
"train_steps_per_second": 0.427
|
| 9 |
+
}
|
runs/Apr25_23-51-20_dgx-a100/events.out.tfevents.1745717211.dgx-a100.855256.1
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:5d81344b8862919491dad649d662e5ca723de2ecaa14545c758c31971d5766cd
|
| 3 |
+
size 477
|
trainer_state.json
ADDED
|
@@ -0,0 +1,2243 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"best_global_step": 41000,
|
| 3 |
+
"best_metric": 0.359779867002981,
|
| 4 |
+
"best_model_checkpoint": "/home/cluster-dgxa100/slp01/bagas-fine-tune-whisper/whisper-tiny-javanese-openslr-v6/checkpoint-41000",
|
| 5 |
+
"epoch": 159.7456,
|
| 6 |
+
"eval_steps": 1000,
|
| 7 |
+
"global_step": 50000,
|
| 8 |
+
"is_hyper_param_search": false,
|
| 9 |
+
"is_local_process_zero": true,
|
| 10 |
+
"is_world_process_zero": true,
|
| 11 |
+
"log_history": [
|
| 12 |
+
{
|
| 13 |
+
"epoch": 0.64,
|
| 14 |
+
"grad_norm": 25.32911491394043,
|
| 15 |
+
"learning_rate": 7.88e-07,
|
| 16 |
+
"loss": 3.6275,
|
| 17 |
+
"step": 200
|
| 18 |
+
},
|
| 19 |
+
{
|
| 20 |
+
"epoch": 1.2784,
|
| 21 |
+
"grad_norm": 20.188899993896484,
|
| 22 |
+
"learning_rate": 1.588e-06,
|
| 23 |
+
"loss": 2.0941,
|
| 24 |
+
"step": 400
|
| 25 |
+
},
|
| 26 |
+
{
|
| 27 |
+
"epoch": 1.9184,
|
| 28 |
+
"grad_norm": 20.781444549560547,
|
| 29 |
+
"learning_rate": 2.3880000000000003e-06,
|
| 30 |
+
"loss": 1.6316,
|
| 31 |
+
"step": 600
|
| 32 |
+
},
|
| 33 |
+
{
|
| 34 |
+
"epoch": 2.5568,
|
| 35 |
+
"grad_norm": 18.28050994873047,
|
| 36 |
+
"learning_rate": 3.188e-06,
|
| 37 |
+
"loss": 1.3303,
|
| 38 |
+
"step": 800
|
| 39 |
+
},
|
| 40 |
+
{
|
| 41 |
+
"epoch": 3.1952,
|
| 42 |
+
"grad_norm": 19.185575485229492,
|
| 43 |
+
"learning_rate": 3.988000000000001e-06,
|
| 44 |
+
"loss": 1.1166,
|
| 45 |
+
"step": 1000
|
| 46 |
+
},
|
| 47 |
+
{
|
| 48 |
+
"epoch": 3.1952,
|
| 49 |
+
"eval_loss": 1.0200245380401611,
|
| 50 |
+
"eval_runtime": 143.6337,
|
| 51 |
+
"eval_samples_per_second": 4.351,
|
| 52 |
+
"eval_steps_per_second": 1.093,
|
| 53 |
+
"eval_wer": 0.6975464343040587,
|
| 54 |
+
"step": 1000
|
| 55 |
+
},
|
| 56 |
+
{
|
| 57 |
+
"epoch": 3.8352,
|
| 58 |
+
"grad_norm": 16.41274070739746,
|
| 59 |
+
"learning_rate": 4.7880000000000006e-06,
|
| 60 |
+
"loss": 0.9805,
|
| 61 |
+
"step": 1200
|
| 62 |
+
},
|
| 63 |
+
{
|
| 64 |
+
"epoch": 4.4736,
|
| 65 |
+
"grad_norm": 14.031892776489258,
|
| 66 |
+
"learning_rate": 5.588e-06,
|
| 67 |
+
"loss": 0.8521,
|
| 68 |
+
"step": 1400
|
| 69 |
+
},
|
| 70 |
+
{
|
| 71 |
+
"epoch": 5.112,
|
| 72 |
+
"grad_norm": 14.105754852294922,
|
| 73 |
+
"learning_rate": 6.3880000000000005e-06,
|
| 74 |
+
"loss": 0.7633,
|
| 75 |
+
"step": 1600
|
| 76 |
+
},
|
| 77 |
+
{
|
| 78 |
+
"epoch": 5.752,
|
| 79 |
+
"grad_norm": 13.716428756713867,
|
| 80 |
+
"learning_rate": 7.1880000000000005e-06,
|
| 81 |
+
"loss": 0.6676,
|
| 82 |
+
"step": 1800
|
| 83 |
+
},
|
| 84 |
+
{
|
| 85 |
+
"epoch": 6.3904,
|
| 86 |
+
"grad_norm": 12.681669235229492,
|
| 87 |
+
"learning_rate": 7.988e-06,
|
| 88 |
+
"loss": 0.5644,
|
| 89 |
+
"step": 2000
|
| 90 |
+
},
|
| 91 |
+
{
|
| 92 |
+
"epoch": 6.3904,
|
| 93 |
+
"eval_loss": 0.7341693043708801,
|
| 94 |
+
"eval_runtime": 113.3954,
|
| 95 |
+
"eval_samples_per_second": 5.512,
|
| 96 |
+
"eval_steps_per_second": 1.385,
|
| 97 |
+
"eval_wer": 0.5897729878468241,
|
| 98 |
+
"step": 2000
|
| 99 |
+
},
|
| 100 |
+
{
|
| 101 |
+
"epoch": 7.0288,
|
| 102 |
+
"grad_norm": 12.7446928024292,
|
| 103 |
+
"learning_rate": 8.788000000000001e-06,
|
| 104 |
+
"loss": 0.5326,
|
| 105 |
+
"step": 2200
|
| 106 |
+
},
|
| 107 |
+
{
|
| 108 |
+
"epoch": 7.6688,
|
| 109 |
+
"grad_norm": 12.206501007080078,
|
| 110 |
+
"learning_rate": 9.588e-06,
|
| 111 |
+
"loss": 0.4346,
|
| 112 |
+
"step": 2400
|
| 113 |
+
},
|
| 114 |
+
{
|
| 115 |
+
"epoch": 8.3072,
|
| 116 |
+
"grad_norm": 9.13988208770752,
|
| 117 |
+
"learning_rate": 1.0388e-05,
|
| 118 |
+
"loss": 0.3967,
|
| 119 |
+
"step": 2600
|
| 120 |
+
},
|
| 121 |
+
{
|
| 122 |
+
"epoch": 8.9472,
|
| 123 |
+
"grad_norm": 12.071162223815918,
|
| 124 |
+
"learning_rate": 1.1188e-05,
|
| 125 |
+
"loss": 0.3532,
|
| 126 |
+
"step": 2800
|
| 127 |
+
},
|
| 128 |
+
{
|
| 129 |
+
"epoch": 9.5856,
|
| 130 |
+
"grad_norm": 8.742654800415039,
|
| 131 |
+
"learning_rate": 1.1988000000000001e-05,
|
| 132 |
+
"loss": 0.2837,
|
| 133 |
+
"step": 3000
|
| 134 |
+
},
|
| 135 |
+
{
|
| 136 |
+
"epoch": 9.5856,
|
| 137 |
+
"eval_loss": 0.666342556476593,
|
| 138 |
+
"eval_runtime": 133.4076,
|
| 139 |
+
"eval_samples_per_second": 4.685,
|
| 140 |
+
"eval_steps_per_second": 1.177,
|
| 141 |
+
"eval_wer": 0.5700527401972024,
|
| 142 |
+
"step": 3000
|
| 143 |
+
},
|
| 144 |
+
{
|
| 145 |
+
"epoch": 10.224,
|
| 146 |
+
"grad_norm": 10.311443328857422,
|
| 147 |
+
"learning_rate": 1.2788e-05,
|
| 148 |
+
"loss": 0.2601,
|
| 149 |
+
"step": 3200
|
| 150 |
+
},
|
| 151 |
+
{
|
| 152 |
+
"epoch": 10.864,
|
| 153 |
+
"grad_norm": 10.367297172546387,
|
| 154 |
+
"learning_rate": 1.3588000000000001e-05,
|
| 155 |
+
"loss": 0.227,
|
| 156 |
+
"step": 3400
|
| 157 |
+
},
|
| 158 |
+
{
|
| 159 |
+
"epoch": 11.5024,
|
| 160 |
+
"grad_norm": 10.210521697998047,
|
| 161 |
+
"learning_rate": 1.4388000000000002e-05,
|
| 162 |
+
"loss": 0.1883,
|
| 163 |
+
"step": 3600
|
| 164 |
+
},
|
| 165 |
+
{
|
| 166 |
+
"epoch": 12.1408,
|
| 167 |
+
"grad_norm": 9.120365142822266,
|
| 168 |
+
"learning_rate": 1.5188000000000001e-05,
|
| 169 |
+
"loss": 0.1716,
|
| 170 |
+
"step": 3800
|
| 171 |
+
},
|
| 172 |
+
{
|
| 173 |
+
"epoch": 12.7808,
|
| 174 |
+
"grad_norm": 8.728864669799805,
|
| 175 |
+
"learning_rate": 1.5988e-05,
|
| 176 |
+
"loss": 0.1451,
|
| 177 |
+
"step": 4000
|
| 178 |
+
},
|
| 179 |
+
{
|
| 180 |
+
"epoch": 12.7808,
|
| 181 |
+
"eval_loss": 0.669156551361084,
|
| 182 |
+
"eval_runtime": 111.1099,
|
| 183 |
+
"eval_samples_per_second": 5.625,
|
| 184 |
+
"eval_steps_per_second": 1.413,
|
| 185 |
+
"eval_wer": 0.615684476037606,
|
| 186 |
+
"step": 4000
|
| 187 |
+
},
|
| 188 |
+
{
|
| 189 |
+
"epoch": 13.4192,
|
| 190 |
+
"grad_norm": 8.055611610412598,
|
| 191 |
+
"learning_rate": 1.6788000000000003e-05,
|
| 192 |
+
"loss": 0.1277,
|
| 193 |
+
"step": 4200
|
| 194 |
+
},
|
| 195 |
+
{
|
| 196 |
+
"epoch": 14.0576,
|
| 197 |
+
"grad_norm": 10.252028465270996,
|
| 198 |
+
"learning_rate": 1.7588e-05,
|
| 199 |
+
"loss": 0.1231,
|
| 200 |
+
"step": 4400
|
| 201 |
+
},
|
| 202 |
+
{
|
| 203 |
+
"epoch": 14.6976,
|
| 204 |
+
"grad_norm": 8.062397956848145,
|
| 205 |
+
"learning_rate": 1.8388e-05,
|
| 206 |
+
"loss": 0.1015,
|
| 207 |
+
"step": 4600
|
| 208 |
+
},
|
| 209 |
+
{
|
| 210 |
+
"epoch": 15.336,
|
| 211 |
+
"grad_norm": 5.604861736297607,
|
| 212 |
+
"learning_rate": 1.9188000000000003e-05,
|
| 213 |
+
"loss": 0.0966,
|
| 214 |
+
"step": 4800
|
| 215 |
+
},
|
| 216 |
+
{
|
| 217 |
+
"epoch": 15.975999999999999,
|
| 218 |
+
"grad_norm": 5.753713130950928,
|
| 219 |
+
"learning_rate": 1.9988000000000002e-05,
|
| 220 |
+
"loss": 0.0949,
|
| 221 |
+
"step": 5000
|
| 222 |
+
},
|
| 223 |
+
{
|
| 224 |
+
"epoch": 15.975999999999999,
|
| 225 |
+
"eval_loss": 0.6928630471229553,
|
| 226 |
+
"eval_runtime": 104.9987,
|
| 227 |
+
"eval_samples_per_second": 5.952,
|
| 228 |
+
"eval_steps_per_second": 1.495,
|
| 229 |
+
"eval_wer": 0.6140793396010089,
|
| 230 |
+
"step": 5000
|
| 231 |
+
},
|
| 232 |
+
{
|
| 233 |
+
"epoch": 16.6144,
|
| 234 |
+
"grad_norm": 7.435614109039307,
|
| 235 |
+
"learning_rate": 1.991288888888889e-05,
|
| 236 |
+
"loss": 0.0763,
|
| 237 |
+
"step": 5200
|
| 238 |
+
},
|
| 239 |
+
{
|
| 240 |
+
"epoch": 17.2528,
|
| 241 |
+
"grad_norm": 5.643408298492432,
|
| 242 |
+
"learning_rate": 1.9824000000000002e-05,
|
| 243 |
+
"loss": 0.0767,
|
| 244 |
+
"step": 5400
|
| 245 |
+
},
|
| 246 |
+
{
|
| 247 |
+
"epoch": 17.8928,
|
| 248 |
+
"grad_norm": 6.685403823852539,
|
| 249 |
+
"learning_rate": 1.9735111111111112e-05,
|
| 250 |
+
"loss": 0.0695,
|
| 251 |
+
"step": 5600
|
| 252 |
+
},
|
| 253 |
+
{
|
| 254 |
+
"epoch": 18.5312,
|
| 255 |
+
"grad_norm": 5.571532249450684,
|
| 256 |
+
"learning_rate": 1.9646222222222223e-05,
|
| 257 |
+
"loss": 0.0649,
|
| 258 |
+
"step": 5800
|
| 259 |
+
},
|
| 260 |
+
{
|
| 261 |
+
"epoch": 19.1696,
|
| 262 |
+
"grad_norm": 5.548845291137695,
|
| 263 |
+
"learning_rate": 1.9557333333333333e-05,
|
| 264 |
+
"loss": 0.0621,
|
| 265 |
+
"step": 6000
|
| 266 |
+
},
|
| 267 |
+
{
|
| 268 |
+
"epoch": 19.1696,
|
| 269 |
+
"eval_loss": 0.7081567049026489,
|
| 270 |
+
"eval_runtime": 106.1316,
|
| 271 |
+
"eval_samples_per_second": 5.889,
|
| 272 |
+
"eval_steps_per_second": 1.479,
|
| 273 |
+
"eval_wer": 0.4790185737216235,
|
| 274 |
+
"step": 6000
|
| 275 |
+
},
|
| 276 |
+
{
|
| 277 |
+
"epoch": 19.8096,
|
| 278 |
+
"grad_norm": 5.352997779846191,
|
| 279 |
+
"learning_rate": 1.9468444444444444e-05,
|
| 280 |
+
"loss": 0.0578,
|
| 281 |
+
"step": 6200
|
| 282 |
+
},
|
| 283 |
+
{
|
| 284 |
+
"epoch": 20.448,
|
| 285 |
+
"grad_norm": 4.763257026672363,
|
| 286 |
+
"learning_rate": 1.9379555555555558e-05,
|
| 287 |
+
"loss": 0.0521,
|
| 288 |
+
"step": 6400
|
| 289 |
+
},
|
| 290 |
+
{
|
| 291 |
+
"epoch": 21.0864,
|
| 292 |
+
"grad_norm": 2.98003888130188,
|
| 293 |
+
"learning_rate": 1.9290666666666668e-05,
|
| 294 |
+
"loss": 0.0527,
|
| 295 |
+
"step": 6600
|
| 296 |
+
},
|
| 297 |
+
{
|
| 298 |
+
"epoch": 21.7264,
|
| 299 |
+
"grad_norm": 12.276226043701172,
|
| 300 |
+
"learning_rate": 1.920177777777778e-05,
|
| 301 |
+
"loss": 0.0456,
|
| 302 |
+
"step": 6800
|
| 303 |
+
},
|
| 304 |
+
{
|
| 305 |
+
"epoch": 22.3648,
|
| 306 |
+
"grad_norm": 4.935642719268799,
|
| 307 |
+
"learning_rate": 1.911288888888889e-05,
|
| 308 |
+
"loss": 0.0466,
|
| 309 |
+
"step": 7000
|
| 310 |
+
},
|
| 311 |
+
{
|
| 312 |
+
"epoch": 22.3648,
|
| 313 |
+
"eval_loss": 0.7455945611000061,
|
| 314 |
+
"eval_runtime": 105.1482,
|
| 315 |
+
"eval_samples_per_second": 5.944,
|
| 316 |
+
"eval_steps_per_second": 1.493,
|
| 317 |
+
"eval_wer": 0.4469158449896813,
|
| 318 |
+
"step": 7000
|
| 319 |
+
},
|
| 320 |
+
{
|
| 321 |
+
"epoch": 23.0032,
|
| 322 |
+
"grad_norm": 4.248728275299072,
|
| 323 |
+
"learning_rate": 1.9024000000000003e-05,
|
| 324 |
+
"loss": 0.0449,
|
| 325 |
+
"step": 7200
|
| 326 |
+
},
|
| 327 |
+
{
|
| 328 |
+
"epoch": 23.6432,
|
| 329 |
+
"grad_norm": 6.321920394897461,
|
| 330 |
+
"learning_rate": 1.8935111111111113e-05,
|
| 331 |
+
"loss": 0.0388,
|
| 332 |
+
"step": 7400
|
| 333 |
+
},
|
| 334 |
+
{
|
| 335 |
+
"epoch": 24.2816,
|
| 336 |
+
"grad_norm": 4.33453893661499,
|
| 337 |
+
"learning_rate": 1.8846222222222224e-05,
|
| 338 |
+
"loss": 0.0412,
|
| 339 |
+
"step": 7600
|
| 340 |
+
},
|
| 341 |
+
{
|
| 342 |
+
"epoch": 24.9216,
|
| 343 |
+
"grad_norm": 8.20227336883545,
|
| 344 |
+
"learning_rate": 1.8757333333333334e-05,
|
| 345 |
+
"loss": 0.0386,
|
| 346 |
+
"step": 7800
|
| 347 |
+
},
|
| 348 |
+
{
|
| 349 |
+
"epoch": 25.56,
|
| 350 |
+
"grad_norm": 4.103025913238525,
|
| 351 |
+
"learning_rate": 1.8668444444444448e-05,
|
| 352 |
+
"loss": 0.0338,
|
| 353 |
+
"step": 8000
|
| 354 |
+
},
|
| 355 |
+
{
|
| 356 |
+
"epoch": 25.56,
|
| 357 |
+
"eval_loss": 0.760087251663208,
|
| 358 |
+
"eval_runtime": 103.2404,
|
| 359 |
+
"eval_samples_per_second": 6.054,
|
| 360 |
+
"eval_steps_per_second": 1.521,
|
| 361 |
+
"eval_wer": 0.43659711075441415,
|
| 362 |
+
"step": 8000
|
| 363 |
+
},
|
| 364 |
+
{
|
| 365 |
+
"epoch": 26.1984,
|
| 366 |
+
"grad_norm": 3.744155168533325,
|
| 367 |
+
"learning_rate": 1.858e-05,
|
| 368 |
+
"loss": 0.0356,
|
| 369 |
+
"step": 8200
|
| 370 |
+
},
|
| 371 |
+
{
|
| 372 |
+
"epoch": 26.8384,
|
| 373 |
+
"grad_norm": 3.403186082839966,
|
| 374 |
+
"learning_rate": 1.8491111111111112e-05,
|
| 375 |
+
"loss": 0.0334,
|
| 376 |
+
"step": 8400
|
| 377 |
+
},
|
| 378 |
+
{
|
| 379 |
+
"epoch": 27.4768,
|
| 380 |
+
"grad_norm": 4.127572536468506,
|
| 381 |
+
"learning_rate": 1.8402222222222223e-05,
|
| 382 |
+
"loss": 0.0327,
|
| 383 |
+
"step": 8600
|
| 384 |
+
},
|
| 385 |
+
{
|
| 386 |
+
"epoch": 28.1152,
|
| 387 |
+
"grad_norm": 4.966237545013428,
|
| 388 |
+
"learning_rate": 1.8313333333333333e-05,
|
| 389 |
+
"loss": 0.0309,
|
| 390 |
+
"step": 8800
|
| 391 |
+
},
|
| 392 |
+
{
|
| 393 |
+
"epoch": 28.7552,
|
| 394 |
+
"grad_norm": 6.874843120574951,
|
| 395 |
+
"learning_rate": 1.8224444444444447e-05,
|
| 396 |
+
"loss": 0.0288,
|
| 397 |
+
"step": 9000
|
| 398 |
+
},
|
| 399 |
+
{
|
| 400 |
+
"epoch": 28.7552,
|
| 401 |
+
"eval_loss": 0.7782142758369446,
|
| 402 |
+
"eval_runtime": 101.996,
|
| 403 |
+
"eval_samples_per_second": 6.128,
|
| 404 |
+
"eval_steps_per_second": 1.539,
|
| 405 |
+
"eval_wer": 0.3893602384774134,
|
| 406 |
+
"step": 9000
|
| 407 |
+
},
|
| 408 |
+
{
|
| 409 |
+
"epoch": 29.3936,
|
| 410 |
+
"grad_norm": 2.1933653354644775,
|
| 411 |
+
"learning_rate": 1.8135555555555557e-05,
|
| 412 |
+
"loss": 0.0275,
|
| 413 |
+
"step": 9200
|
| 414 |
+
},
|
| 415 |
+
{
|
| 416 |
+
"epoch": 30.032,
|
| 417 |
+
"grad_norm": 2.639747381210327,
|
| 418 |
+
"learning_rate": 1.8046666666666668e-05,
|
| 419 |
+
"loss": 0.027,
|
| 420 |
+
"step": 9400
|
| 421 |
+
},
|
| 422 |
+
{
|
| 423 |
+
"epoch": 30.672,
|
| 424 |
+
"grad_norm": 2.232715368270874,
|
| 425 |
+
"learning_rate": 1.7957777777777778e-05,
|
| 426 |
+
"loss": 0.0248,
|
| 427 |
+
"step": 9600
|
| 428 |
+
},
|
| 429 |
+
{
|
| 430 |
+
"epoch": 31.3104,
|
| 431 |
+
"grad_norm": 1.4926921129226685,
|
| 432 |
+
"learning_rate": 1.7868888888888892e-05,
|
| 433 |
+
"loss": 0.0266,
|
| 434 |
+
"step": 9800
|
| 435 |
+
},
|
| 436 |
+
{
|
| 437 |
+
"epoch": 31.9504,
|
| 438 |
+
"grad_norm": 2.4175844192504883,
|
| 439 |
+
"learning_rate": 1.7780000000000003e-05,
|
| 440 |
+
"loss": 0.0232,
|
| 441 |
+
"step": 10000
|
| 442 |
+
},
|
| 443 |
+
{
|
| 444 |
+
"epoch": 31.9504,
|
| 445 |
+
"eval_loss": 0.79770827293396,
|
| 446 |
+
"eval_runtime": 104.7759,
|
| 447 |
+
"eval_samples_per_second": 5.965,
|
| 448 |
+
"eval_steps_per_second": 1.498,
|
| 449 |
+
"eval_wer": 0.41068562256363217,
|
| 450 |
+
"step": 10000
|
| 451 |
+
},
|
| 452 |
+
{
|
| 453 |
+
"epoch": 32.5888,
|
| 454 |
+
"grad_norm": 1.8864479064941406,
|
| 455 |
+
"learning_rate": 1.7691111111111113e-05,
|
| 456 |
+
"loss": 0.0218,
|
| 457 |
+
"step": 10200
|
| 458 |
+
},
|
| 459 |
+
{
|
| 460 |
+
"epoch": 33.2272,
|
| 461 |
+
"grad_norm": 8.317153930664062,
|
| 462 |
+
"learning_rate": 1.7602222222222223e-05,
|
| 463 |
+
"loss": 0.0239,
|
| 464 |
+
"step": 10400
|
| 465 |
+
},
|
| 466 |
+
{
|
| 467 |
+
"epoch": 33.8672,
|
| 468 |
+
"grad_norm": 4.491159439086914,
|
| 469 |
+
"learning_rate": 1.7513333333333334e-05,
|
| 470 |
+
"loss": 0.023,
|
| 471 |
+
"step": 10600
|
| 472 |
+
},
|
| 473 |
+
{
|
| 474 |
+
"epoch": 34.5056,
|
| 475 |
+
"grad_norm": 2.753185272216797,
|
| 476 |
+
"learning_rate": 1.7424444444444444e-05,
|
| 477 |
+
"loss": 0.0202,
|
| 478 |
+
"step": 10800
|
| 479 |
+
},
|
| 480 |
+
{
|
| 481 |
+
"epoch": 35.144,
|
| 482 |
+
"grad_norm": 2.579869270324707,
|
| 483 |
+
"learning_rate": 1.7335555555555558e-05,
|
| 484 |
+
"loss": 0.0212,
|
| 485 |
+
"step": 11000
|
| 486 |
+
},
|
| 487 |
+
{
|
| 488 |
+
"epoch": 35.144,
|
| 489 |
+
"eval_loss": 0.797561764717102,
|
| 490 |
+
"eval_runtime": 106.2768,
|
| 491 |
+
"eval_samples_per_second": 5.881,
|
| 492 |
+
"eval_steps_per_second": 1.477,
|
| 493 |
+
"eval_wer": 0.41435450584728273,
|
| 494 |
+
"step": 11000
|
| 495 |
+
},
|
| 496 |
+
{
|
| 497 |
+
"epoch": 35.784,
|
| 498 |
+
"grad_norm": 2.2582919597625732,
|
| 499 |
+
"learning_rate": 1.724666666666667e-05,
|
| 500 |
+
"loss": 0.0193,
|
| 501 |
+
"step": 11200
|
| 502 |
+
},
|
| 503 |
+
{
|
| 504 |
+
"epoch": 36.4224,
|
| 505 |
+
"grad_norm": 2.982079029083252,
|
| 506 |
+
"learning_rate": 1.715777777777778e-05,
|
| 507 |
+
"loss": 0.0207,
|
| 508 |
+
"step": 11400
|
| 509 |
+
},
|
| 510 |
+
{
|
| 511 |
+
"epoch": 37.0608,
|
| 512 |
+
"grad_norm": 3.09155011177063,
|
| 513 |
+
"learning_rate": 1.706888888888889e-05,
|
| 514 |
+
"loss": 0.0176,
|
| 515 |
+
"step": 11600
|
| 516 |
+
},
|
| 517 |
+
{
|
| 518 |
+
"epoch": 37.7008,
|
| 519 |
+
"grad_norm": 3.805440902709961,
|
| 520 |
+
"learning_rate": 1.6980444444444447e-05,
|
| 521 |
+
"loss": 0.0177,
|
| 522 |
+
"step": 11800
|
| 523 |
+
},
|
| 524 |
+
{
|
| 525 |
+
"epoch": 38.3392,
|
| 526 |
+
"grad_norm": 2.447655439376831,
|
| 527 |
+
"learning_rate": 1.6891555555555557e-05,
|
| 528 |
+
"loss": 0.0178,
|
| 529 |
+
"step": 12000
|
| 530 |
+
},
|
| 531 |
+
{
|
| 532 |
+
"epoch": 38.3392,
|
| 533 |
+
"eval_loss": 0.8184047341346741,
|
| 534 |
+
"eval_runtime": 140.2428,
|
| 535 |
+
"eval_samples_per_second": 4.457,
|
| 536 |
+
"eval_steps_per_second": 1.119,
|
| 537 |
+
"eval_wer": 0.4010548039440495,
|
| 538 |
+
"step": 12000
|
| 539 |
+
},
|
| 540 |
+
{
|
| 541 |
+
"epoch": 38.9792,
|
| 542 |
+
"grad_norm": 6.444581031799316,
|
| 543 |
+
"learning_rate": 1.6802666666666668e-05,
|
| 544 |
+
"loss": 0.018,
|
| 545 |
+
"step": 12200
|
| 546 |
+
},
|
| 547 |
+
{
|
| 548 |
+
"epoch": 39.6176,
|
| 549 |
+
"grad_norm": 1.9832179546356201,
|
| 550 |
+
"learning_rate": 1.671377777777778e-05,
|
| 551 |
+
"loss": 0.0185,
|
| 552 |
+
"step": 12400
|
| 553 |
+
},
|
| 554 |
+
{
|
| 555 |
+
"epoch": 40.256,
|
| 556 |
+
"grad_norm": 1.374569058418274,
|
| 557 |
+
"learning_rate": 1.6624888888888892e-05,
|
| 558 |
+
"loss": 0.0159,
|
| 559 |
+
"step": 12600
|
| 560 |
+
},
|
| 561 |
+
{
|
| 562 |
+
"epoch": 40.896,
|
| 563 |
+
"grad_norm": 4.2071309089660645,
|
| 564 |
+
"learning_rate": 1.6536000000000002e-05,
|
| 565 |
+
"loss": 0.0138,
|
| 566 |
+
"step": 12800
|
| 567 |
+
},
|
| 568 |
+
{
|
| 569 |
+
"epoch": 41.5344,
|
| 570 |
+
"grad_norm": 2.3069539070129395,
|
| 571 |
+
"learning_rate": 1.6447111111111113e-05,
|
| 572 |
+
"loss": 0.0132,
|
| 573 |
+
"step": 13000
|
| 574 |
+
},
|
| 575 |
+
{
|
| 576 |
+
"epoch": 41.5344,
|
| 577 |
+
"eval_loss": 0.8310635089874268,
|
| 578 |
+
"eval_runtime": 138.029,
|
| 579 |
+
"eval_samples_per_second": 4.528,
|
| 580 |
+
"eval_steps_per_second": 1.137,
|
| 581 |
+
"eval_wer": 0.3762898417794084,
|
| 582 |
+
"step": 13000
|
| 583 |
+
},
|
| 584 |
+
{
|
| 585 |
+
"epoch": 42.1728,
|
| 586 |
+
"grad_norm": 1.6480227708816528,
|
| 587 |
+
"learning_rate": 1.6358222222222223e-05,
|
| 588 |
+
"loss": 0.016,
|
| 589 |
+
"step": 13200
|
| 590 |
+
},
|
| 591 |
+
{
|
| 592 |
+
"epoch": 42.8128,
|
| 593 |
+
"grad_norm": 4.483267784118652,
|
| 594 |
+
"learning_rate": 1.6269333333333334e-05,
|
| 595 |
+
"loss": 0.0145,
|
| 596 |
+
"step": 13400
|
| 597 |
+
},
|
| 598 |
+
{
|
| 599 |
+
"epoch": 43.4512,
|
| 600 |
+
"grad_norm": 1.1435110569000244,
|
| 601 |
+
"learning_rate": 1.6180444444444444e-05,
|
| 602 |
+
"loss": 0.0129,
|
| 603 |
+
"step": 13600
|
| 604 |
+
},
|
| 605 |
+
{
|
| 606 |
+
"epoch": 44.0896,
|
| 607 |
+
"grad_norm": 0.6056487560272217,
|
| 608 |
+
"learning_rate": 1.6091555555555555e-05,
|
| 609 |
+
"loss": 0.015,
|
| 610 |
+
"step": 13800
|
| 611 |
+
},
|
| 612 |
+
{
|
| 613 |
+
"epoch": 44.7296,
|
| 614 |
+
"grad_norm": 7.710687160491943,
|
| 615 |
+
"learning_rate": 1.600266666666667e-05,
|
| 616 |
+
"loss": 0.0145,
|
| 617 |
+
"step": 14000
|
| 618 |
+
},
|
| 619 |
+
{
|
| 620 |
+
"epoch": 44.7296,
|
| 621 |
+
"eval_loss": 0.8473762273788452,
|
| 622 |
+
"eval_runtime": 109.9727,
|
| 623 |
+
"eval_samples_per_second": 5.683,
|
| 624 |
+
"eval_steps_per_second": 1.428,
|
| 625 |
+
"eval_wer": 0.3790415042421463,
|
| 626 |
+
"step": 14000
|
| 627 |
+
},
|
| 628 |
+
{
|
| 629 |
+
"epoch": 45.368,
|
| 630 |
+
"grad_norm": 0.3126671314239502,
|
| 631 |
+
"learning_rate": 1.591377777777778e-05,
|
| 632 |
+
"loss": 0.0151,
|
| 633 |
+
"step": 14200
|
| 634 |
+
},
|
| 635 |
+
{
|
| 636 |
+
"epoch": 46.0064,
|
| 637 |
+
"grad_norm": 1.5604488849639893,
|
| 638 |
+
"learning_rate": 1.582488888888889e-05,
|
| 639 |
+
"loss": 0.0126,
|
| 640 |
+
"step": 14400
|
| 641 |
+
},
|
| 642 |
+
{
|
| 643 |
+
"epoch": 46.6464,
|
| 644 |
+
"grad_norm": 1.0523457527160645,
|
| 645 |
+
"learning_rate": 1.5736000000000003e-05,
|
| 646 |
+
"loss": 0.0124,
|
| 647 |
+
"step": 14600
|
| 648 |
+
},
|
| 649 |
+
{
|
| 650 |
+
"epoch": 47.2848,
|
| 651 |
+
"grad_norm": 2.6398849487304688,
|
| 652 |
+
"learning_rate": 1.5647111111111114e-05,
|
| 653 |
+
"loss": 0.013,
|
| 654 |
+
"step": 14800
|
| 655 |
+
},
|
| 656 |
+
{
|
| 657 |
+
"epoch": 47.9248,
|
| 658 |
+
"grad_norm": 1.1916667222976685,
|
| 659 |
+
"learning_rate": 1.5558222222222224e-05,
|
| 660 |
+
"loss": 0.0117,
|
| 661 |
+
"step": 15000
|
| 662 |
+
},
|
| 663 |
+
{
|
| 664 |
+
"epoch": 47.9248,
|
| 665 |
+
"eval_loss": 0.8624815940856934,
|
| 666 |
+
"eval_runtime": 105.5676,
|
| 667 |
+
"eval_samples_per_second": 5.92,
|
| 668 |
+
"eval_steps_per_second": 1.487,
|
| 669 |
+
"eval_wer": 0.4155010318734235,
|
| 670 |
+
"step": 15000
|
| 671 |
+
},
|
| 672 |
+
{
|
| 673 |
+
"epoch": 48.5632,
|
| 674 |
+
"grad_norm": 1.593981146812439,
|
| 675 |
+
"learning_rate": 1.5469333333333335e-05,
|
| 676 |
+
"loss": 0.0115,
|
| 677 |
+
"step": 15200
|
| 678 |
+
},
|
| 679 |
+
{
|
| 680 |
+
"epoch": 49.2016,
|
| 681 |
+
"grad_norm": 2.9825565814971924,
|
| 682 |
+
"learning_rate": 1.538044444444445e-05,
|
| 683 |
+
"loss": 0.0111,
|
| 684 |
+
"step": 15400
|
| 685 |
+
},
|
| 686 |
+
{
|
| 687 |
+
"epoch": 49.8416,
|
| 688 |
+
"grad_norm": 1.278347134590149,
|
| 689 |
+
"learning_rate": 1.529155555555556e-05,
|
| 690 |
+
"loss": 0.0109,
|
| 691 |
+
"step": 15600
|
| 692 |
+
},
|
| 693 |
+
{
|
| 694 |
+
"epoch": 50.48,
|
| 695 |
+
"grad_norm": 1.986265778541565,
|
| 696 |
+
"learning_rate": 1.5202666666666668e-05,
|
| 697 |
+
"loss": 0.0115,
|
| 698 |
+
"step": 15800
|
| 699 |
+
},
|
| 700 |
+
{
|
| 701 |
+
"epoch": 51.1184,
|
| 702 |
+
"grad_norm": 0.23057551681995392,
|
| 703 |
+
"learning_rate": 1.5114222222222223e-05,
|
| 704 |
+
"loss": 0.0101,
|
| 705 |
+
"step": 16000
|
| 706 |
+
},
|
| 707 |
+
{
|
| 708 |
+
"epoch": 51.1184,
|
| 709 |
+
"eval_loss": 0.8906803727149963,
|
| 710 |
+
"eval_runtime": 106.8162,
|
| 711 |
+
"eval_samples_per_second": 5.851,
|
| 712 |
+
"eval_steps_per_second": 1.47,
|
| 713 |
+
"eval_wer": 0.3758312313689521,
|
| 714 |
+
"step": 16000
|
| 715 |
+
},
|
| 716 |
+
{
|
| 717 |
+
"epoch": 51.7584,
|
| 718 |
+
"grad_norm": 2.8394742012023926,
|
| 719 |
+
"learning_rate": 1.5025333333333333e-05,
|
| 720 |
+
"loss": 0.01,
|
| 721 |
+
"step": 16200
|
| 722 |
+
},
|
| 723 |
+
{
|
| 724 |
+
"epoch": 52.3968,
|
| 725 |
+
"grad_norm": 3.907292366027832,
|
| 726 |
+
"learning_rate": 1.4936444444444447e-05,
|
| 727 |
+
"loss": 0.0101,
|
| 728 |
+
"step": 16400
|
| 729 |
+
},
|
| 730 |
+
{
|
| 731 |
+
"epoch": 53.0352,
|
| 732 |
+
"grad_norm": 0.3342689871788025,
|
| 733 |
+
"learning_rate": 1.4847555555555558e-05,
|
| 734 |
+
"loss": 0.0106,
|
| 735 |
+
"step": 16600
|
| 736 |
+
},
|
| 737 |
+
{
|
| 738 |
+
"epoch": 53.6752,
|
| 739 |
+
"grad_norm": 14.257401466369629,
|
| 740 |
+
"learning_rate": 1.4758666666666668e-05,
|
| 741 |
+
"loss": 0.0096,
|
| 742 |
+
"step": 16800
|
| 743 |
+
},
|
| 744 |
+
{
|
| 745 |
+
"epoch": 54.3136,
|
| 746 |
+
"grad_norm": 4.482171535491943,
|
| 747 |
+
"learning_rate": 1.4669777777777779e-05,
|
| 748 |
+
"loss": 0.0091,
|
| 749 |
+
"step": 17000
|
| 750 |
+
},
|
| 751 |
+
{
|
| 752 |
+
"epoch": 54.3136,
|
| 753 |
+
"eval_loss": 0.8973252773284912,
|
| 754 |
+
"eval_runtime": 107.1216,
|
| 755 |
+
"eval_samples_per_second": 5.834,
|
| 756 |
+
"eval_steps_per_second": 1.466,
|
| 757 |
+
"eval_wer": 0.3999082779179087,
|
| 758 |
+
"step": 17000
|
| 759 |
+
},
|
| 760 |
+
{
|
| 761 |
+
"epoch": 54.9536,
|
| 762 |
+
"grad_norm": 0.3064260184764862,
|
| 763 |
+
"learning_rate": 1.458088888888889e-05,
|
| 764 |
+
"loss": 0.0089,
|
| 765 |
+
"step": 17200
|
| 766 |
+
},
|
| 767 |
+
{
|
| 768 |
+
"epoch": 55.592,
|
| 769 |
+
"grad_norm": 3.7441632747650146,
|
| 770 |
+
"learning_rate": 1.4492000000000001e-05,
|
| 771 |
+
"loss": 0.01,
|
| 772 |
+
"step": 17400
|
| 773 |
+
},
|
| 774 |
+
{
|
| 775 |
+
"epoch": 56.2304,
|
| 776 |
+
"grad_norm": 2.9929962158203125,
|
| 777 |
+
"learning_rate": 1.4403111111111112e-05,
|
| 778 |
+
"loss": 0.0099,
|
| 779 |
+
"step": 17600
|
| 780 |
+
},
|
| 781 |
+
{
|
| 782 |
+
"epoch": 56.8704,
|
| 783 |
+
"grad_norm": 3.1295619010925293,
|
| 784 |
+
"learning_rate": 1.4314222222222222e-05,
|
| 785 |
+
"loss": 0.0083,
|
| 786 |
+
"step": 17800
|
| 787 |
+
},
|
| 788 |
+
{
|
| 789 |
+
"epoch": 57.5088,
|
| 790 |
+
"grad_norm": 3.032670736312866,
|
| 791 |
+
"learning_rate": 1.4225777777777779e-05,
|
| 792 |
+
"loss": 0.0087,
|
| 793 |
+
"step": 18000
|
| 794 |
+
},
|
| 795 |
+
{
|
| 796 |
+
"epoch": 57.5088,
|
| 797 |
+
"eval_loss": 0.9277347922325134,
|
| 798 |
+
"eval_runtime": 108.6583,
|
| 799 |
+
"eval_samples_per_second": 5.752,
|
| 800 |
+
"eval_steps_per_second": 1.445,
|
| 801 |
+
"eval_wer": 0.4182526943361614,
|
| 802 |
+
"step": 18000
|
| 803 |
+
},
|
| 804 |
+
{
|
| 805 |
+
"epoch": 58.1472,
|
| 806 |
+
"grad_norm": 0.5450573563575745,
|
| 807 |
+
"learning_rate": 1.4137333333333334e-05,
|
| 808 |
+
"loss": 0.0081,
|
| 809 |
+
"step": 18200
|
| 810 |
+
},
|
| 811 |
+
{
|
| 812 |
+
"epoch": 58.7872,
|
| 813 |
+
"grad_norm": 3.2385668754577637,
|
| 814 |
+
"learning_rate": 1.4048444444444445e-05,
|
| 815 |
+
"loss": 0.008,
|
| 816 |
+
"step": 18400
|
| 817 |
+
},
|
| 818 |
+
{
|
| 819 |
+
"epoch": 59.4256,
|
| 820 |
+
"grad_norm": 2.490182638168335,
|
| 821 |
+
"learning_rate": 1.3959555555555557e-05,
|
| 822 |
+
"loss": 0.0074,
|
| 823 |
+
"step": 18600
|
| 824 |
+
},
|
| 825 |
+
{
|
| 826 |
+
"epoch": 60.064,
|
| 827 |
+
"grad_norm": 2.916256904602051,
|
| 828 |
+
"learning_rate": 1.3870666666666667e-05,
|
| 829 |
+
"loss": 0.009,
|
| 830 |
+
"step": 18800
|
| 831 |
+
},
|
| 832 |
+
{
|
| 833 |
+
"epoch": 60.704,
|
| 834 |
+
"grad_norm": 0.9266663789749146,
|
| 835 |
+
"learning_rate": 1.3781777777777778e-05,
|
| 836 |
+
"loss": 0.0068,
|
| 837 |
+
"step": 19000
|
| 838 |
+
},
|
| 839 |
+
{
|
| 840 |
+
"epoch": 60.704,
|
| 841 |
+
"eval_loss": 0.9448923468589783,
|
| 842 |
+
"eval_runtime": 110.101,
|
| 843 |
+
"eval_samples_per_second": 5.677,
|
| 844 |
+
"eval_steps_per_second": 1.426,
|
| 845 |
+
"eval_wer": 0.43889016280669574,
|
| 846 |
+
"step": 19000
|
| 847 |
+
},
|
| 848 |
+
{
|
| 849 |
+
"epoch": 61.3424,
|
| 850 |
+
"grad_norm": 0.37736016511917114,
|
| 851 |
+
"learning_rate": 1.3692888888888892e-05,
|
| 852 |
+
"loss": 0.0079,
|
| 853 |
+
"step": 19200
|
| 854 |
+
},
|
| 855 |
+
{
|
| 856 |
+
"epoch": 61.9824,
|
| 857 |
+
"grad_norm": 2.7307522296905518,
|
| 858 |
+
"learning_rate": 1.3604000000000002e-05,
|
| 859 |
+
"loss": 0.008,
|
| 860 |
+
"step": 19400
|
| 861 |
+
},
|
| 862 |
+
{
|
| 863 |
+
"epoch": 62.6208,
|
| 864 |
+
"grad_norm": 0.1229301169514656,
|
| 865 |
+
"learning_rate": 1.3515111111111113e-05,
|
| 866 |
+
"loss": 0.0065,
|
| 867 |
+
"step": 19600
|
| 868 |
+
},
|
| 869 |
+
{
|
| 870 |
+
"epoch": 63.2592,
|
| 871 |
+
"grad_norm": 0.7501879930496216,
|
| 872 |
+
"learning_rate": 1.3426222222222223e-05,
|
| 873 |
+
"loss": 0.0074,
|
| 874 |
+
"step": 19800
|
| 875 |
+
},
|
| 876 |
+
{
|
| 877 |
+
"epoch": 63.8992,
|
| 878 |
+
"grad_norm": 0.5861854553222656,
|
| 879 |
+
"learning_rate": 1.3337333333333335e-05,
|
| 880 |
+
"loss": 0.0073,
|
| 881 |
+
"step": 20000
|
| 882 |
+
},
|
| 883 |
+
{
|
| 884 |
+
"epoch": 63.8992,
|
| 885 |
+
"eval_loss": 0.9372482895851135,
|
| 886 |
+
"eval_runtime": 139.4248,
|
| 887 |
+
"eval_samples_per_second": 4.483,
|
| 888 |
+
"eval_steps_per_second": 1.126,
|
| 889 |
+
"eval_wer": 0.3833983031414813,
|
| 890 |
+
"step": 20000
|
| 891 |
+
},
|
| 892 |
+
{
|
| 893 |
+
"epoch": 64.5376,
|
| 894 |
+
"grad_norm": 1.7161897420883179,
|
| 895 |
+
"learning_rate": 1.3248444444444446e-05,
|
| 896 |
+
"loss": 0.0072,
|
| 897 |
+
"step": 20200
|
| 898 |
+
},
|
| 899 |
+
{
|
| 900 |
+
"epoch": 65.176,
|
| 901 |
+
"grad_norm": 1.3181058168411255,
|
| 902 |
+
"learning_rate": 1.3159555555555556e-05,
|
| 903 |
+
"loss": 0.0064,
|
| 904 |
+
"step": 20400
|
| 905 |
+
},
|
| 906 |
+
{
|
| 907 |
+
"epoch": 65.816,
|
| 908 |
+
"grad_norm": 0.576824426651001,
|
| 909 |
+
"learning_rate": 1.3070666666666667e-05,
|
| 910 |
+
"loss": 0.0073,
|
| 911 |
+
"step": 20600
|
| 912 |
+
},
|
| 913 |
+
{
|
| 914 |
+
"epoch": 66.4544,
|
| 915 |
+
"grad_norm": 1.2488709688186646,
|
| 916 |
+
"learning_rate": 1.298177777777778e-05,
|
| 917 |
+
"loss": 0.0067,
|
| 918 |
+
"step": 20800
|
| 919 |
+
},
|
| 920 |
+
{
|
| 921 |
+
"epoch": 67.0928,
|
| 922 |
+
"grad_norm": 0.8621447682380676,
|
| 923 |
+
"learning_rate": 1.2892888888888891e-05,
|
| 924 |
+
"loss": 0.0066,
|
| 925 |
+
"step": 21000
|
| 926 |
+
},
|
| 927 |
+
{
|
| 928 |
+
"epoch": 67.0928,
|
| 929 |
+
"eval_loss": 0.9511934518814087,
|
| 930 |
+
"eval_runtime": 138.3581,
|
| 931 |
+
"eval_samples_per_second": 4.517,
|
| 932 |
+
"eval_steps_per_second": 1.135,
|
| 933 |
+
"eval_wer": 0.4038064664067874,
|
| 934 |
+
"step": 21000
|
| 935 |
+
},
|
| 936 |
+
{
|
| 937 |
+
"epoch": 67.7328,
|
| 938 |
+
"grad_norm": 2.959468364715576,
|
| 939 |
+
"learning_rate": 1.2804000000000001e-05,
|
| 940 |
+
"loss": 0.0063,
|
| 941 |
+
"step": 21200
|
| 942 |
+
},
|
| 943 |
+
{
|
| 944 |
+
"epoch": 68.3712,
|
| 945 |
+
"grad_norm": 4.359962463378906,
|
| 946 |
+
"learning_rate": 1.2715111111111112e-05,
|
| 947 |
+
"loss": 0.0069,
|
| 948 |
+
"step": 21400
|
| 949 |
+
},
|
| 950 |
+
{
|
| 951 |
+
"epoch": 69.0096,
|
| 952 |
+
"grad_norm": 0.8827808499336243,
|
| 953 |
+
"learning_rate": 1.2626222222222224e-05,
|
| 954 |
+
"loss": 0.0067,
|
| 955 |
+
"step": 21600
|
| 956 |
+
},
|
| 957 |
+
{
|
| 958 |
+
"epoch": 69.6496,
|
| 959 |
+
"grad_norm": 4.478594779968262,
|
| 960 |
+
"learning_rate": 1.2537333333333334e-05,
|
| 961 |
+
"loss": 0.0063,
|
| 962 |
+
"step": 21800
|
| 963 |
+
},
|
| 964 |
+
{
|
| 965 |
+
"epoch": 70.288,
|
| 966 |
+
"grad_norm": 0.3562716245651245,
|
| 967 |
+
"learning_rate": 1.2448444444444445e-05,
|
| 968 |
+
"loss": 0.0056,
|
| 969 |
+
"step": 22000
|
| 970 |
+
},
|
| 971 |
+
{
|
| 972 |
+
"epoch": 70.288,
|
| 973 |
+
"eval_loss": 0.9799156188964844,
|
| 974 |
+
"eval_runtime": 108.661,
|
| 975 |
+
"eval_samples_per_second": 5.752,
|
| 976 |
+
"eval_steps_per_second": 1.445,
|
| 977 |
+
"eval_wer": 0.4063288236642972,
|
| 978 |
+
"step": 22000
|
| 979 |
+
},
|
| 980 |
+
{
|
| 981 |
+
"epoch": 70.928,
|
| 982 |
+
"grad_norm": 0.21852029860019684,
|
| 983 |
+
"learning_rate": 1.2359555555555555e-05,
|
| 984 |
+
"loss": 0.0058,
|
| 985 |
+
"step": 22200
|
| 986 |
+
},
|
| 987 |
+
{
|
| 988 |
+
"epoch": 71.5664,
|
| 989 |
+
"grad_norm": 0.7513341307640076,
|
| 990 |
+
"learning_rate": 1.227111111111111e-05,
|
| 991 |
+
"loss": 0.0051,
|
| 992 |
+
"step": 22400
|
| 993 |
+
},
|
| 994 |
+
{
|
| 995 |
+
"epoch": 72.2048,
|
| 996 |
+
"grad_norm": 1.729013204574585,
|
| 997 |
+
"learning_rate": 1.2182222222222225e-05,
|
| 998 |
+
"loss": 0.0048,
|
| 999 |
+
"step": 22600
|
| 1000 |
+
},
|
| 1001 |
+
{
|
| 1002 |
+
"epoch": 72.8448,
|
| 1003 |
+
"grad_norm": 0.363949179649353,
|
| 1004 |
+
"learning_rate": 1.2093333333333335e-05,
|
| 1005 |
+
"loss": 0.0043,
|
| 1006 |
+
"step": 22800
|
| 1007 |
+
},
|
| 1008 |
+
{
|
| 1009 |
+
"epoch": 73.4832,
|
| 1010 |
+
"grad_norm": 0.09823896735906601,
|
| 1011 |
+
"learning_rate": 1.2004444444444445e-05,
|
| 1012 |
+
"loss": 0.0049,
|
| 1013 |
+
"step": 23000
|
| 1014 |
+
},
|
| 1015 |
+
{
|
| 1016 |
+
"epoch": 73.4832,
|
| 1017 |
+
"eval_loss": 0.9892968535423279,
|
| 1018 |
+
"eval_runtime": 127.4129,
|
| 1019 |
+
"eval_samples_per_second": 4.905,
|
| 1020 |
+
"eval_steps_per_second": 1.232,
|
| 1021 |
+
"eval_wer": 0.3845448291676221,
|
| 1022 |
+
"step": 23000
|
| 1023 |
+
},
|
| 1024 |
+
{
|
| 1025 |
+
"epoch": 74.1216,
|
| 1026 |
+
"grad_norm": 0.18766258656978607,
|
| 1027 |
+
"learning_rate": 1.1915555555555556e-05,
|
| 1028 |
+
"loss": 0.0061,
|
| 1029 |
+
"step": 23200
|
| 1030 |
+
},
|
| 1031 |
+
{
|
| 1032 |
+
"epoch": 74.7616,
|
| 1033 |
+
"grad_norm": 2.374441623687744,
|
| 1034 |
+
"learning_rate": 1.1827111111111111e-05,
|
| 1035 |
+
"loss": 0.0049,
|
| 1036 |
+
"step": 23400
|
| 1037 |
+
},
|
| 1038 |
+
{
|
| 1039 |
+
"epoch": 75.4,
|
| 1040 |
+
"grad_norm": 1.901435136795044,
|
| 1041 |
+
"learning_rate": 1.1738222222222223e-05,
|
| 1042 |
+
"loss": 0.0042,
|
| 1043 |
+
"step": 23600
|
| 1044 |
+
},
|
| 1045 |
+
{
|
| 1046 |
+
"epoch": 76.0384,
|
| 1047 |
+
"grad_norm": 1.8781205415725708,
|
| 1048 |
+
"learning_rate": 1.1649333333333336e-05,
|
| 1049 |
+
"loss": 0.0066,
|
| 1050 |
+
"step": 23800
|
| 1051 |
+
},
|
| 1052 |
+
{
|
| 1053 |
+
"epoch": 76.6784,
|
| 1054 |
+
"grad_norm": 4.333975791931152,
|
| 1055 |
+
"learning_rate": 1.1560444444444446e-05,
|
| 1056 |
+
"loss": 0.0046,
|
| 1057 |
+
"step": 24000
|
| 1058 |
+
},
|
| 1059 |
+
{
|
| 1060 |
+
"epoch": 76.6784,
|
| 1061 |
+
"eval_loss": 0.9896656274795532,
|
| 1062 |
+
"eval_runtime": 128.2479,
|
| 1063 |
+
"eval_samples_per_second": 4.873,
|
| 1064 |
+
"eval_steps_per_second": 1.224,
|
| 1065 |
+
"eval_wer": 0.38087594588397156,
|
| 1066 |
+
"step": 24000
|
| 1067 |
+
},
|
| 1068 |
+
{
|
| 1069 |
+
"epoch": 77.3168,
|
| 1070 |
+
"grad_norm": 1.9272364377975464,
|
| 1071 |
+
"learning_rate": 1.1471555555555556e-05,
|
| 1072 |
+
"loss": 0.0035,
|
| 1073 |
+
"step": 24200
|
| 1074 |
+
},
|
| 1075 |
+
{
|
| 1076 |
+
"epoch": 77.9568,
|
| 1077 |
+
"grad_norm": 0.6323594450950623,
|
| 1078 |
+
"learning_rate": 1.1382666666666669e-05,
|
| 1079 |
+
"loss": 0.0051,
|
| 1080 |
+
"step": 24400
|
| 1081 |
+
},
|
| 1082 |
+
{
|
| 1083 |
+
"epoch": 78.5952,
|
| 1084 |
+
"grad_norm": 1.1000163555145264,
|
| 1085 |
+
"learning_rate": 1.1293777777777779e-05,
|
| 1086 |
+
"loss": 0.0038,
|
| 1087 |
+
"step": 24600
|
| 1088 |
+
},
|
| 1089 |
+
{
|
| 1090 |
+
"epoch": 79.2336,
|
| 1091 |
+
"grad_norm": 0.48444664478302,
|
| 1092 |
+
"learning_rate": 1.120488888888889e-05,
|
| 1093 |
+
"loss": 0.0045,
|
| 1094 |
+
"step": 24800
|
| 1095 |
+
},
|
| 1096 |
+
{
|
| 1097 |
+
"epoch": 79.8736,
|
| 1098 |
+
"grad_norm": 0.6538843512535095,
|
| 1099 |
+
"learning_rate": 1.1116e-05,
|
| 1100 |
+
"loss": 0.0044,
|
| 1101 |
+
"step": 25000
|
| 1102 |
+
},
|
| 1103 |
+
{
|
| 1104 |
+
"epoch": 79.8736,
|
| 1105 |
+
"eval_loss": 0.9970444440841675,
|
| 1106 |
+
"eval_runtime": 131.0293,
|
| 1107 |
+
"eval_samples_per_second": 4.77,
|
| 1108 |
+
"eval_steps_per_second": 1.198,
|
| 1109 |
+
"eval_wer": 0.37491401054803947,
|
| 1110 |
+
"step": 25000
|
| 1111 |
+
},
|
| 1112 |
+
{
|
| 1113 |
+
"epoch": 80.512,
|
| 1114 |
+
"grad_norm": 3.5870893001556396,
|
| 1115 |
+
"learning_rate": 1.1027111111111112e-05,
|
| 1116 |
+
"loss": 0.0045,
|
| 1117 |
+
"step": 25200
|
| 1118 |
+
},
|
| 1119 |
+
{
|
| 1120 |
+
"epoch": 81.1504,
|
| 1121 |
+
"grad_norm": 0.1785198152065277,
|
| 1122 |
+
"learning_rate": 1.0938222222222223e-05,
|
| 1123 |
+
"loss": 0.0035,
|
| 1124 |
+
"step": 25400
|
| 1125 |
+
},
|
| 1126 |
+
{
|
| 1127 |
+
"epoch": 81.7904,
|
| 1128 |
+
"grad_norm": 0.870185375213623,
|
| 1129 |
+
"learning_rate": 1.0849333333333335e-05,
|
| 1130 |
+
"loss": 0.0039,
|
| 1131 |
+
"step": 25600
|
| 1132 |
+
},
|
| 1133 |
+
{
|
| 1134 |
+
"epoch": 82.4288,
|
| 1135 |
+
"grad_norm": 0.21784505248069763,
|
| 1136 |
+
"learning_rate": 1.0760444444444445e-05,
|
| 1137 |
+
"loss": 0.0041,
|
| 1138 |
+
"step": 25800
|
| 1139 |
+
},
|
| 1140 |
+
{
|
| 1141 |
+
"epoch": 83.0672,
|
| 1142 |
+
"grad_norm": 3.4108142852783203,
|
| 1143 |
+
"learning_rate": 1.0671555555555557e-05,
|
| 1144 |
+
"loss": 0.0039,
|
| 1145 |
+
"step": 26000
|
| 1146 |
+
},
|
| 1147 |
+
{
|
| 1148 |
+
"epoch": 83.0672,
|
| 1149 |
+
"eval_loss": 1.011253833770752,
|
| 1150 |
+
"eval_runtime": 125.5354,
|
| 1151 |
+
"eval_samples_per_second": 4.979,
|
| 1152 |
+
"eval_steps_per_second": 1.251,
|
| 1153 |
+
"eval_wer": 0.37606053657418026,
|
| 1154 |
+
"step": 26000
|
| 1155 |
+
},
|
| 1156 |
+
{
|
| 1157 |
+
"epoch": 83.7072,
|
| 1158 |
+
"grad_norm": 0.04159156233072281,
|
| 1159 |
+
"learning_rate": 1.0582666666666668e-05,
|
| 1160 |
+
"loss": 0.0031,
|
| 1161 |
+
"step": 26200
|
| 1162 |
+
},
|
| 1163 |
+
{
|
| 1164 |
+
"epoch": 84.3456,
|
| 1165 |
+
"grad_norm": 0.20573270320892334,
|
| 1166 |
+
"learning_rate": 1.0493777777777778e-05,
|
| 1167 |
+
"loss": 0.0043,
|
| 1168 |
+
"step": 26400
|
| 1169 |
+
},
|
| 1170 |
+
{
|
| 1171 |
+
"epoch": 84.9856,
|
| 1172 |
+
"grad_norm": 2.9891881942749023,
|
| 1173 |
+
"learning_rate": 1.0404888888888889e-05,
|
| 1174 |
+
"loss": 0.0041,
|
| 1175 |
+
"step": 26600
|
| 1176 |
+
},
|
| 1177 |
+
{
|
| 1178 |
+
"epoch": 85.624,
|
| 1179 |
+
"grad_norm": 2.5772018432617188,
|
| 1180 |
+
"learning_rate": 1.0316e-05,
|
| 1181 |
+
"loss": 0.0039,
|
| 1182 |
+
"step": 26800
|
| 1183 |
+
},
|
| 1184 |
+
{
|
| 1185 |
+
"epoch": 86.2624,
|
| 1186 |
+
"grad_norm": 0.12921689450740814,
|
| 1187 |
+
"learning_rate": 1.0227111111111111e-05,
|
| 1188 |
+
"loss": 0.0035,
|
| 1189 |
+
"step": 27000
|
| 1190 |
+
},
|
| 1191 |
+
{
|
| 1192 |
+
"epoch": 86.2624,
|
| 1193 |
+
"eval_loss": 1.0148670673370361,
|
| 1194 |
+
"eval_runtime": 126.4781,
|
| 1195 |
+
"eval_samples_per_second": 4.942,
|
| 1196 |
+
"eval_steps_per_second": 1.241,
|
| 1197 |
+
"eval_wer": 0.38316899793625314,
|
| 1198 |
+
"step": 27000
|
| 1199 |
+
},
|
| 1200 |
+
{
|
| 1201 |
+
"epoch": 86.9024,
|
| 1202 |
+
"grad_norm": 1.1538918018341064,
|
| 1203 |
+
"learning_rate": 1.0138222222222223e-05,
|
| 1204 |
+
"loss": 0.0041,
|
| 1205 |
+
"step": 27200
|
| 1206 |
+
},
|
| 1207 |
+
{
|
| 1208 |
+
"epoch": 87.5408,
|
| 1209 |
+
"grad_norm": 3.53802490234375,
|
| 1210 |
+
"learning_rate": 1.0049333333333334e-05,
|
| 1211 |
+
"loss": 0.0037,
|
| 1212 |
+
"step": 27400
|
| 1213 |
+
},
|
| 1214 |
+
{
|
| 1215 |
+
"epoch": 88.1792,
|
| 1216 |
+
"grad_norm": 0.03811612352728844,
|
| 1217 |
+
"learning_rate": 9.960444444444444e-06,
|
| 1218 |
+
"loss": 0.0035,
|
| 1219 |
+
"step": 27600
|
| 1220 |
+
},
|
| 1221 |
+
{
|
| 1222 |
+
"epoch": 88.8192,
|
| 1223 |
+
"grad_norm": 0.8571044206619263,
|
| 1224 |
+
"learning_rate": 9.871555555555557e-06,
|
| 1225 |
+
"loss": 0.0027,
|
| 1226 |
+
"step": 27800
|
| 1227 |
+
},
|
| 1228 |
+
{
|
| 1229 |
+
"epoch": 89.4576,
|
| 1230 |
+
"grad_norm": 1.1792042255401611,
|
| 1231 |
+
"learning_rate": 9.782666666666667e-06,
|
| 1232 |
+
"loss": 0.003,
|
| 1233 |
+
"step": 28000
|
| 1234 |
+
},
|
| 1235 |
+
{
|
| 1236 |
+
"epoch": 89.4576,
|
| 1237 |
+
"eval_loss": 1.0032455921173096,
|
| 1238 |
+
"eval_runtime": 152.9287,
|
| 1239 |
+
"eval_samples_per_second": 4.087,
|
| 1240 |
+
"eval_steps_per_second": 1.027,
|
| 1241 |
+
"eval_wer": 0.38592066039899103,
|
| 1242 |
+
"step": 28000
|
| 1243 |
+
},
|
| 1244 |
+
{
|
| 1245 |
+
"epoch": 90.096,
|
| 1246 |
+
"grad_norm": 0.32366737723350525,
|
| 1247 |
+
"learning_rate": 9.693777777777779e-06,
|
| 1248 |
+
"loss": 0.0034,
|
| 1249 |
+
"step": 28200
|
| 1250 |
+
},
|
| 1251 |
+
{
|
| 1252 |
+
"epoch": 90.736,
|
| 1253 |
+
"grad_norm": 0.4921131134033203,
|
| 1254 |
+
"learning_rate": 9.60488888888889e-06,
|
| 1255 |
+
"loss": 0.0035,
|
| 1256 |
+
"step": 28400
|
| 1257 |
+
},
|
| 1258 |
+
{
|
| 1259 |
+
"epoch": 91.3744,
|
| 1260 |
+
"grad_norm": 3.0860531330108643,
|
| 1261 |
+
"learning_rate": 9.516e-06,
|
| 1262 |
+
"loss": 0.0031,
|
| 1263 |
+
"step": 28600
|
| 1264 |
+
},
|
| 1265 |
+
{
|
| 1266 |
+
"epoch": 92.0128,
|
| 1267 |
+
"grad_norm": 3.8854892253875732,
|
| 1268 |
+
"learning_rate": 9.427111111111112e-06,
|
| 1269 |
+
"loss": 0.0034,
|
| 1270 |
+
"step": 28800
|
| 1271 |
+
},
|
| 1272 |
+
{
|
| 1273 |
+
"epoch": 92.6528,
|
| 1274 |
+
"grad_norm": 0.3374783396720886,
|
| 1275 |
+
"learning_rate": 9.338222222222223e-06,
|
| 1276 |
+
"loss": 0.0025,
|
| 1277 |
+
"step": 29000
|
| 1278 |
+
},
|
| 1279 |
+
{
|
| 1280 |
+
"epoch": 92.6528,
|
| 1281 |
+
"eval_loss": 1.0094186067581177,
|
| 1282 |
+
"eval_runtime": 135.2467,
|
| 1283 |
+
"eval_samples_per_second": 4.621,
|
| 1284 |
+
"eval_steps_per_second": 1.161,
|
| 1285 |
+
"eval_wer": 0.3856913551937629,
|
| 1286 |
+
"step": 29000
|
| 1287 |
+
},
|
| 1288 |
+
{
|
| 1289 |
+
"epoch": 93.2912,
|
| 1290 |
+
"grad_norm": 0.9511881470680237,
|
| 1291 |
+
"learning_rate": 9.249333333333335e-06,
|
| 1292 |
+
"loss": 0.0023,
|
| 1293 |
+
"step": 29200
|
| 1294 |
+
},
|
| 1295 |
+
{
|
| 1296 |
+
"epoch": 93.9312,
|
| 1297 |
+
"grad_norm": 0.5302645564079285,
|
| 1298 |
+
"learning_rate": 9.160888888888888e-06,
|
| 1299 |
+
"loss": 0.0025,
|
| 1300 |
+
"step": 29400
|
| 1301 |
+
},
|
| 1302 |
+
{
|
| 1303 |
+
"epoch": 94.5696,
|
| 1304 |
+
"grad_norm": 0.22430481016635895,
|
| 1305 |
+
"learning_rate": 9.072444444444445e-06,
|
| 1306 |
+
"loss": 0.003,
|
| 1307 |
+
"step": 29600
|
| 1308 |
+
},
|
| 1309 |
+
{
|
| 1310 |
+
"epoch": 95.208,
|
| 1311 |
+
"grad_norm": 0.18836861848831177,
|
| 1312 |
+
"learning_rate": 8.983555555555556e-06,
|
| 1313 |
+
"loss": 0.0025,
|
| 1314 |
+
"step": 29800
|
| 1315 |
+
},
|
| 1316 |
+
{
|
| 1317 |
+
"epoch": 95.848,
|
| 1318 |
+
"grad_norm": 0.5240734815597534,
|
| 1319 |
+
"learning_rate": 8.894666666666666e-06,
|
| 1320 |
+
"loss": 0.0034,
|
| 1321 |
+
"step": 30000
|
| 1322 |
+
},
|
| 1323 |
+
{
|
| 1324 |
+
"epoch": 95.848,
|
| 1325 |
+
"eval_loss": 1.02016282081604,
|
| 1326 |
+
"eval_runtime": 122.5101,
|
| 1327 |
+
"eval_samples_per_second": 5.102,
|
| 1328 |
+
"eval_steps_per_second": 1.282,
|
| 1329 |
+
"eval_wer": 0.3733088741114423,
|
| 1330 |
+
"step": 30000
|
| 1331 |
+
},
|
| 1332 |
+
{
|
| 1333 |
+
"epoch": 96.4864,
|
| 1334 |
+
"grad_norm": 0.10021472722291946,
|
| 1335 |
+
"learning_rate": 8.805777777777778e-06,
|
| 1336 |
+
"loss": 0.0029,
|
| 1337 |
+
"step": 30200
|
| 1338 |
+
},
|
| 1339 |
+
{
|
| 1340 |
+
"epoch": 97.1248,
|
| 1341 |
+
"grad_norm": 0.07506231218576431,
|
| 1342 |
+
"learning_rate": 8.716888888888889e-06,
|
| 1343 |
+
"loss": 0.0028,
|
| 1344 |
+
"step": 30400
|
| 1345 |
+
},
|
| 1346 |
+
{
|
| 1347 |
+
"epoch": 97.7648,
|
| 1348 |
+
"grad_norm": 0.6782508492469788,
|
| 1349 |
+
"learning_rate": 8.628000000000001e-06,
|
| 1350 |
+
"loss": 0.0031,
|
| 1351 |
+
"step": 30600
|
| 1352 |
+
},
|
| 1353 |
+
{
|
| 1354 |
+
"epoch": 98.4032,
|
| 1355 |
+
"grad_norm": 0.1539822518825531,
|
| 1356 |
+
"learning_rate": 8.539111111111112e-06,
|
| 1357 |
+
"loss": 0.0032,
|
| 1358 |
+
"step": 30800
|
| 1359 |
+
},
|
| 1360 |
+
{
|
| 1361 |
+
"epoch": 99.0416,
|
| 1362 |
+
"grad_norm": 0.3094239830970764,
|
| 1363 |
+
"learning_rate": 8.450222222222224e-06,
|
| 1364 |
+
"loss": 0.0027,
|
| 1365 |
+
"step": 31000
|
| 1366 |
+
},
|
| 1367 |
+
{
|
| 1368 |
+
"epoch": 99.0416,
|
| 1369 |
+
"eval_loss": 1.011278510093689,
|
| 1370 |
+
"eval_runtime": 106.3459,
|
| 1371 |
+
"eval_samples_per_second": 5.877,
|
| 1372 |
+
"eval_steps_per_second": 1.476,
|
| 1373 |
+
"eval_wer": 0.36551249713368494,
|
| 1374 |
+
"step": 31000
|
| 1375 |
+
},
|
| 1376 |
+
{
|
| 1377 |
+
"epoch": 99.6816,
|
| 1378 |
+
"grad_norm": 0.32241737842559814,
|
| 1379 |
+
"learning_rate": 8.361333333333334e-06,
|
| 1380 |
+
"loss": 0.0025,
|
| 1381 |
+
"step": 31200
|
| 1382 |
+
},
|
| 1383 |
+
{
|
| 1384 |
+
"epoch": 100.32,
|
| 1385 |
+
"grad_norm": 0.7617099285125732,
|
| 1386 |
+
"learning_rate": 8.272444444444445e-06,
|
| 1387 |
+
"loss": 0.0024,
|
| 1388 |
+
"step": 31400
|
| 1389 |
+
},
|
| 1390 |
+
{
|
| 1391 |
+
"epoch": 100.96,
|
| 1392 |
+
"grad_norm": 0.16760019958019257,
|
| 1393 |
+
"learning_rate": 8.183555555555555e-06,
|
| 1394 |
+
"loss": 0.0022,
|
| 1395 |
+
"step": 31600
|
| 1396 |
+
},
|
| 1397 |
+
{
|
| 1398 |
+
"epoch": 101.5984,
|
| 1399 |
+
"grad_norm": 0.05670633167028427,
|
| 1400 |
+
"learning_rate": 8.095111111111112e-06,
|
| 1401 |
+
"loss": 0.0029,
|
| 1402 |
+
"step": 31800
|
| 1403 |
+
},
|
| 1404 |
+
{
|
| 1405 |
+
"epoch": 102.2368,
|
| 1406 |
+
"grad_norm": 1.7071377038955688,
|
| 1407 |
+
"learning_rate": 8.006222222222223e-06,
|
| 1408 |
+
"loss": 0.0023,
|
| 1409 |
+
"step": 32000
|
| 1410 |
+
},
|
| 1411 |
+
{
|
| 1412 |
+
"epoch": 102.2368,
|
| 1413 |
+
"eval_loss": 1.0178241729736328,
|
| 1414 |
+
"eval_runtime": 104.5181,
|
| 1415 |
+
"eval_samples_per_second": 5.98,
|
| 1416 |
+
"eval_steps_per_second": 1.502,
|
| 1417 |
+
"eval_wer": 0.3767484521898647,
|
| 1418 |
+
"step": 32000
|
| 1419 |
+
},
|
| 1420 |
+
{
|
| 1421 |
+
"epoch": 102.8768,
|
| 1422 |
+
"grad_norm": 0.09089858084917068,
|
| 1423 |
+
"learning_rate": 7.917333333333333e-06,
|
| 1424 |
+
"loss": 0.0021,
|
| 1425 |
+
"step": 32200
|
| 1426 |
+
},
|
| 1427 |
+
{
|
| 1428 |
+
"epoch": 103.5152,
|
| 1429 |
+
"grad_norm": 0.31245774030685425,
|
| 1430 |
+
"learning_rate": 7.828444444444445e-06,
|
| 1431 |
+
"loss": 0.0027,
|
| 1432 |
+
"step": 32400
|
| 1433 |
+
},
|
| 1434 |
+
{
|
| 1435 |
+
"epoch": 104.1536,
|
| 1436 |
+
"grad_norm": 0.11016162484884262,
|
| 1437 |
+
"learning_rate": 7.739555555555556e-06,
|
| 1438 |
+
"loss": 0.0024,
|
| 1439 |
+
"step": 32600
|
| 1440 |
+
},
|
| 1441 |
+
{
|
| 1442 |
+
"epoch": 104.7936,
|
| 1443 |
+
"grad_norm": 1.1260924339294434,
|
| 1444 |
+
"learning_rate": 7.650666666666668e-06,
|
| 1445 |
+
"loss": 0.002,
|
| 1446 |
+
"step": 32800
|
| 1447 |
+
},
|
| 1448 |
+
{
|
| 1449 |
+
"epoch": 105.432,
|
| 1450 |
+
"grad_norm": 0.3586418330669403,
|
| 1451 |
+
"learning_rate": 7.561777777777778e-06,
|
| 1452 |
+
"loss": 0.002,
|
| 1453 |
+
"step": 33000
|
| 1454 |
+
},
|
| 1455 |
+
{
|
| 1456 |
+
"epoch": 105.432,
|
| 1457 |
+
"eval_loss": 1.0109715461730957,
|
| 1458 |
+
"eval_runtime": 110.8526,
|
| 1459 |
+
"eval_samples_per_second": 5.638,
|
| 1460 |
+
"eval_steps_per_second": 1.416,
|
| 1461 |
+
"eval_wer": 0.36711763357028204,
|
| 1462 |
+
"step": 33000
|
| 1463 |
+
},
|
| 1464 |
+
{
|
| 1465 |
+
"epoch": 106.0704,
|
| 1466 |
+
"grad_norm": 0.06334754824638367,
|
| 1467 |
+
"learning_rate": 7.4728888888888895e-06,
|
| 1468 |
+
"loss": 0.0021,
|
| 1469 |
+
"step": 33200
|
| 1470 |
+
},
|
| 1471 |
+
{
|
| 1472 |
+
"epoch": 106.7104,
|
| 1473 |
+
"grad_norm": 0.018475733697414398,
|
| 1474 |
+
"learning_rate": 7.384e-06,
|
| 1475 |
+
"loss": 0.002,
|
| 1476 |
+
"step": 33400
|
| 1477 |
+
},
|
| 1478 |
+
{
|
| 1479 |
+
"epoch": 107.3488,
|
| 1480 |
+
"grad_norm": 0.0939849466085434,
|
| 1481 |
+
"learning_rate": 7.295111111111112e-06,
|
| 1482 |
+
"loss": 0.0018,
|
| 1483 |
+
"step": 33600
|
| 1484 |
+
},
|
| 1485 |
+
{
|
| 1486 |
+
"epoch": 107.9888,
|
| 1487 |
+
"grad_norm": 0.6405438780784607,
|
| 1488 |
+
"learning_rate": 7.206222222222223e-06,
|
| 1489 |
+
"loss": 0.0025,
|
| 1490 |
+
"step": 33800
|
| 1491 |
+
},
|
| 1492 |
+
{
|
| 1493 |
+
"epoch": 108.6272,
|
| 1494 |
+
"grad_norm": 2.7534918785095215,
|
| 1495 |
+
"learning_rate": 7.117333333333334e-06,
|
| 1496 |
+
"loss": 0.002,
|
| 1497 |
+
"step": 34000
|
| 1498 |
+
},
|
| 1499 |
+
{
|
| 1500 |
+
"epoch": 108.6272,
|
| 1501 |
+
"eval_loss": 1.0300931930541992,
|
| 1502 |
+
"eval_runtime": 115.653,
|
| 1503 |
+
"eval_samples_per_second": 5.404,
|
| 1504 |
+
"eval_steps_per_second": 1.358,
|
| 1505 |
+
"eval_wer": 0.3733088741114423,
|
| 1506 |
+
"step": 34000
|
| 1507 |
+
},
|
| 1508 |
+
{
|
| 1509 |
+
"epoch": 109.2656,
|
| 1510 |
+
"grad_norm": 0.15731623768806458,
|
| 1511 |
+
"learning_rate": 7.028444444444445e-06,
|
| 1512 |
+
"loss": 0.0018,
|
| 1513 |
+
"step": 34200
|
| 1514 |
+
},
|
| 1515 |
+
{
|
| 1516 |
+
"epoch": 109.9056,
|
| 1517 |
+
"grad_norm": 1.9066952466964722,
|
| 1518 |
+
"learning_rate": 6.9395555555555565e-06,
|
| 1519 |
+
"loss": 0.0025,
|
| 1520 |
+
"step": 34400
|
| 1521 |
+
},
|
| 1522 |
+
{
|
| 1523 |
+
"epoch": 110.544,
|
| 1524 |
+
"grad_norm": 0.7084272503852844,
|
| 1525 |
+
"learning_rate": 6.850666666666668e-06,
|
| 1526 |
+
"loss": 0.002,
|
| 1527 |
+
"step": 34600
|
| 1528 |
+
},
|
| 1529 |
+
{
|
| 1530 |
+
"epoch": 111.1824,
|
| 1531 |
+
"grad_norm": 0.11165124177932739,
|
| 1532 |
+
"learning_rate": 6.761777777777778e-06,
|
| 1533 |
+
"loss": 0.0017,
|
| 1534 |
+
"step": 34800
|
| 1535 |
+
},
|
| 1536 |
+
{
|
| 1537 |
+
"epoch": 111.8224,
|
| 1538 |
+
"grad_norm": 0.16031847894191742,
|
| 1539 |
+
"learning_rate": 6.6733333333333335e-06,
|
| 1540 |
+
"loss": 0.0018,
|
| 1541 |
+
"step": 35000
|
| 1542 |
+
},
|
| 1543 |
+
{
|
| 1544 |
+
"epoch": 111.8224,
|
| 1545 |
+
"eval_loss": 1.047860026359558,
|
| 1546 |
+
"eval_runtime": 128.5964,
|
| 1547 |
+
"eval_samples_per_second": 4.86,
|
| 1548 |
+
"eval_steps_per_second": 1.221,
|
| 1549 |
+
"eval_wer": 0.3850034395780784,
|
| 1550 |
+
"step": 35000
|
| 1551 |
+
},
|
| 1552 |
+
{
|
| 1553 |
+
"epoch": 112.4608,
|
| 1554 |
+
"grad_norm": 0.2045450359582901,
|
| 1555 |
+
"learning_rate": 6.584444444444446e-06,
|
| 1556 |
+
"loss": 0.0023,
|
| 1557 |
+
"step": 35200
|
| 1558 |
+
},
|
| 1559 |
+
{
|
| 1560 |
+
"epoch": 113.0992,
|
| 1561 |
+
"grad_norm": 0.3708573281764984,
|
| 1562 |
+
"learning_rate": 6.495555555555556e-06,
|
| 1563 |
+
"loss": 0.002,
|
| 1564 |
+
"step": 35400
|
| 1565 |
+
},
|
| 1566 |
+
{
|
| 1567 |
+
"epoch": 113.7392,
|
| 1568 |
+
"grad_norm": 0.015015633776783943,
|
| 1569 |
+
"learning_rate": 6.4066666666666674e-06,
|
| 1570 |
+
"loss": 0.002,
|
| 1571 |
+
"step": 35600
|
| 1572 |
+
},
|
| 1573 |
+
{
|
| 1574 |
+
"epoch": 114.3776,
|
| 1575 |
+
"grad_norm": 0.04302428662776947,
|
| 1576 |
+
"learning_rate": 6.317777777777778e-06,
|
| 1577 |
+
"loss": 0.002,
|
| 1578 |
+
"step": 35800
|
| 1579 |
+
},
|
| 1580 |
+
{
|
| 1581 |
+
"epoch": 115.016,
|
| 1582 |
+
"grad_norm": 0.14831684529781342,
|
| 1583 |
+
"learning_rate": 6.22888888888889e-06,
|
| 1584 |
+
"loss": 0.002,
|
| 1585 |
+
"step": 36000
|
| 1586 |
+
},
|
| 1587 |
+
{
|
| 1588 |
+
"epoch": 115.016,
|
| 1589 |
+
"eval_loss": 1.0215903520584106,
|
| 1590 |
+
"eval_runtime": 125.0936,
|
| 1591 |
+
"eval_samples_per_second": 4.996,
|
| 1592 |
+
"eval_steps_per_second": 1.255,
|
| 1593 |
+
"eval_wer": 0.36253152946571887,
|
| 1594 |
+
"step": 36000
|
| 1595 |
+
},
|
| 1596 |
+
{
|
| 1597 |
+
"epoch": 115.656,
|
| 1598 |
+
"grad_norm": 0.06726387143135071,
|
| 1599 |
+
"learning_rate": 6.1400000000000005e-06,
|
| 1600 |
+
"loss": 0.0017,
|
| 1601 |
+
"step": 36200
|
| 1602 |
+
},
|
| 1603 |
+
{
|
| 1604 |
+
"epoch": 116.2944,
|
| 1605 |
+
"grad_norm": 0.059277065098285675,
|
| 1606 |
+
"learning_rate": 6.051111111111112e-06,
|
| 1607 |
+
"loss": 0.0016,
|
| 1608 |
+
"step": 36400
|
| 1609 |
+
},
|
| 1610 |
+
{
|
| 1611 |
+
"epoch": 116.9344,
|
| 1612 |
+
"grad_norm": 1.2333521842956543,
|
| 1613 |
+
"learning_rate": 5.962222222222222e-06,
|
| 1614 |
+
"loss": 0.0015,
|
| 1615 |
+
"step": 36600
|
| 1616 |
+
},
|
| 1617 |
+
{
|
| 1618 |
+
"epoch": 117.5728,
|
| 1619 |
+
"grad_norm": 0.1064881980419159,
|
| 1620 |
+
"learning_rate": 5.873333333333334e-06,
|
| 1621 |
+
"loss": 0.0013,
|
| 1622 |
+
"step": 36800
|
| 1623 |
+
},
|
| 1624 |
+
{
|
| 1625 |
+
"epoch": 118.2112,
|
| 1626 |
+
"grad_norm": 0.025924814864993095,
|
| 1627 |
+
"learning_rate": 5.784444444444445e-06,
|
| 1628 |
+
"loss": 0.002,
|
| 1629 |
+
"step": 37000
|
| 1630 |
+
},
|
| 1631 |
+
{
|
| 1632 |
+
"epoch": 118.2112,
|
| 1633 |
+
"eval_loss": 1.040561556816101,
|
| 1634 |
+
"eval_runtime": 127.336,
|
| 1635 |
+
"eval_samples_per_second": 4.908,
|
| 1636 |
+
"eval_steps_per_second": 1.233,
|
| 1637 |
+
"eval_wer": 0.36872277000687914,
|
| 1638 |
+
"step": 37000
|
| 1639 |
+
},
|
| 1640 |
+
{
|
| 1641 |
+
"epoch": 118.8512,
|
| 1642 |
+
"grad_norm": 0.05642084404826164,
|
| 1643 |
+
"learning_rate": 5.695555555555556e-06,
|
| 1644 |
+
"loss": 0.0017,
|
| 1645 |
+
"step": 37200
|
| 1646 |
+
},
|
| 1647 |
+
{
|
| 1648 |
+
"epoch": 119.4896,
|
| 1649 |
+
"grad_norm": 0.045298777520656586,
|
| 1650 |
+
"learning_rate": 5.606666666666667e-06,
|
| 1651 |
+
"loss": 0.0013,
|
| 1652 |
+
"step": 37400
|
| 1653 |
+
},
|
| 1654 |
+
{
|
| 1655 |
+
"epoch": 120.128,
|
| 1656 |
+
"grad_norm": 0.4968818724155426,
|
| 1657 |
+
"learning_rate": 5.517777777777779e-06,
|
| 1658 |
+
"loss": 0.0018,
|
| 1659 |
+
"step": 37600
|
| 1660 |
+
},
|
| 1661 |
+
{
|
| 1662 |
+
"epoch": 120.768,
|
| 1663 |
+
"grad_norm": 0.10297541320323944,
|
| 1664 |
+
"learning_rate": 5.428888888888889e-06,
|
| 1665 |
+
"loss": 0.0013,
|
| 1666 |
+
"step": 37800
|
| 1667 |
+
},
|
| 1668 |
+
{
|
| 1669 |
+
"epoch": 121.4064,
|
| 1670 |
+
"grad_norm": 0.022102700546383858,
|
| 1671 |
+
"learning_rate": 5.3400000000000005e-06,
|
| 1672 |
+
"loss": 0.0011,
|
| 1673 |
+
"step": 38000
|
| 1674 |
+
},
|
| 1675 |
+
{
|
| 1676 |
+
"epoch": 121.4064,
|
| 1677 |
+
"eval_loss": 1.0520411729812622,
|
| 1678 |
+
"eval_runtime": 132.307,
|
| 1679 |
+
"eval_samples_per_second": 4.724,
|
| 1680 |
+
"eval_steps_per_second": 1.187,
|
| 1681 |
+
"eval_wer": 0.41297867461591375,
|
| 1682 |
+
"step": 38000
|
| 1683 |
+
},
|
| 1684 |
+
{
|
| 1685 |
+
"epoch": 122.0448,
|
| 1686 |
+
"grad_norm": 0.13088802993297577,
|
| 1687 |
+
"learning_rate": 5.251555555555557e-06,
|
| 1688 |
+
"loss": 0.0018,
|
| 1689 |
+
"step": 38200
|
| 1690 |
+
},
|
| 1691 |
+
{
|
| 1692 |
+
"epoch": 122.6848,
|
| 1693 |
+
"grad_norm": 0.1453741490840912,
|
| 1694 |
+
"learning_rate": 5.162666666666667e-06,
|
| 1695 |
+
"loss": 0.0016,
|
| 1696 |
+
"step": 38400
|
| 1697 |
+
},
|
| 1698 |
+
{
|
| 1699 |
+
"epoch": 123.3232,
|
| 1700 |
+
"grad_norm": 0.0916297510266304,
|
| 1701 |
+
"learning_rate": 5.0737777777777785e-06,
|
| 1702 |
+
"loss": 0.0014,
|
| 1703 |
+
"step": 38600
|
| 1704 |
+
},
|
| 1705 |
+
{
|
| 1706 |
+
"epoch": 123.9632,
|
| 1707 |
+
"grad_norm": 3.053272008895874,
|
| 1708 |
+
"learning_rate": 4.984888888888889e-06,
|
| 1709 |
+
"loss": 0.0014,
|
| 1710 |
+
"step": 38800
|
| 1711 |
+
},
|
| 1712 |
+
{
|
| 1713 |
+
"epoch": 124.6016,
|
| 1714 |
+
"grad_norm": 0.06683260202407837,
|
| 1715 |
+
"learning_rate": 4.896e-06,
|
| 1716 |
+
"loss": 0.0014,
|
| 1717 |
+
"step": 39000
|
| 1718 |
+
},
|
| 1719 |
+
{
|
| 1720 |
+
"epoch": 124.6016,
|
| 1721 |
+
"eval_loss": 1.0563929080963135,
|
| 1722 |
+
"eval_runtime": 129.9946,
|
| 1723 |
+
"eval_samples_per_second": 4.808,
|
| 1724 |
+
"eval_steps_per_second": 1.208,
|
| 1725 |
+
"eval_wer": 0.36620041274936943,
|
| 1726 |
+
"step": 39000
|
| 1727 |
+
},
|
| 1728 |
+
{
|
| 1729 |
+
"epoch": 125.24,
|
| 1730 |
+
"grad_norm": 0.12738560140132904,
|
| 1731 |
+
"learning_rate": 4.8071111111111115e-06,
|
| 1732 |
+
"loss": 0.001,
|
| 1733 |
+
"step": 39200
|
| 1734 |
+
},
|
| 1735 |
+
{
|
| 1736 |
+
"epoch": 125.88,
|
| 1737 |
+
"grad_norm": 0.2464442402124405,
|
| 1738 |
+
"learning_rate": 4.718222222222222e-06,
|
| 1739 |
+
"loss": 0.0016,
|
| 1740 |
+
"step": 39400
|
| 1741 |
+
},
|
| 1742 |
+
{
|
| 1743 |
+
"epoch": 126.5184,
|
| 1744 |
+
"grad_norm": 0.026401793584227562,
|
| 1745 |
+
"learning_rate": 4.629333333333333e-06,
|
| 1746 |
+
"loss": 0.0015,
|
| 1747 |
+
"step": 39600
|
| 1748 |
+
},
|
| 1749 |
+
{
|
| 1750 |
+
"epoch": 127.1568,
|
| 1751 |
+
"grad_norm": 0.2598608732223511,
|
| 1752 |
+
"learning_rate": 4.540444444444445e-06,
|
| 1753 |
+
"loss": 0.0017,
|
| 1754 |
+
"step": 39800
|
| 1755 |
+
},
|
| 1756 |
+
{
|
| 1757 |
+
"epoch": 127.7968,
|
| 1758 |
+
"grad_norm": 0.11029820889234543,
|
| 1759 |
+
"learning_rate": 4.451555555555556e-06,
|
| 1760 |
+
"loss": 0.0009,
|
| 1761 |
+
"step": 40000
|
| 1762 |
+
},
|
| 1763 |
+
{
|
| 1764 |
+
"epoch": 127.7968,
|
| 1765 |
+
"eval_loss": 1.0525336265563965,
|
| 1766 |
+
"eval_runtime": 130.5048,
|
| 1767 |
+
"eval_samples_per_second": 4.789,
|
| 1768 |
+
"eval_steps_per_second": 1.203,
|
| 1769 |
+
"eval_wer": 0.37881219903691815,
|
| 1770 |
+
"step": 40000
|
| 1771 |
+
},
|
| 1772 |
+
{
|
| 1773 |
+
"epoch": 128.4352,
|
| 1774 |
+
"grad_norm": 0.03331665322184563,
|
| 1775 |
+
"learning_rate": 4.362666666666667e-06,
|
| 1776 |
+
"loss": 0.001,
|
| 1777 |
+
"step": 40200
|
| 1778 |
+
},
|
| 1779 |
+
{
|
| 1780 |
+
"epoch": 129.0736,
|
| 1781 |
+
"grad_norm": 0.12977726757526398,
|
| 1782 |
+
"learning_rate": 4.2737777777777785e-06,
|
| 1783 |
+
"loss": 0.0015,
|
| 1784 |
+
"step": 40400
|
| 1785 |
+
},
|
| 1786 |
+
{
|
| 1787 |
+
"epoch": 129.7136,
|
| 1788 |
+
"grad_norm": 0.031082535162568092,
|
| 1789 |
+
"learning_rate": 4.18488888888889e-06,
|
| 1790 |
+
"loss": 0.0011,
|
| 1791 |
+
"step": 40600
|
| 1792 |
+
},
|
| 1793 |
+
{
|
| 1794 |
+
"epoch": 130.352,
|
| 1795 |
+
"grad_norm": 2.1375820636749268,
|
| 1796 |
+
"learning_rate": 4.096e-06,
|
| 1797 |
+
"loss": 0.001,
|
| 1798 |
+
"step": 40800
|
| 1799 |
+
},
|
| 1800 |
+
{
|
| 1801 |
+
"epoch": 130.992,
|
| 1802 |
+
"grad_norm": 1.0629844665527344,
|
| 1803 |
+
"learning_rate": 4.0071111111111116e-06,
|
| 1804 |
+
"loss": 0.0009,
|
| 1805 |
+
"step": 41000
|
| 1806 |
+
},
|
| 1807 |
+
{
|
| 1808 |
+
"epoch": 130.992,
|
| 1809 |
+
"eval_loss": 1.0391464233398438,
|
| 1810 |
+
"eval_runtime": 131.1426,
|
| 1811 |
+
"eval_samples_per_second": 4.766,
|
| 1812 |
+
"eval_steps_per_second": 1.197,
|
| 1813 |
+
"eval_wer": 0.359779867002981,
|
| 1814 |
+
"step": 41000
|
| 1815 |
+
},
|
| 1816 |
+
{
|
| 1817 |
+
"epoch": 131.6304,
|
| 1818 |
+
"grad_norm": 0.0641772672533989,
|
| 1819 |
+
"learning_rate": 3.918222222222223e-06,
|
| 1820 |
+
"loss": 0.0012,
|
| 1821 |
+
"step": 41200
|
| 1822 |
+
},
|
| 1823 |
+
{
|
| 1824 |
+
"epoch": 132.2688,
|
| 1825 |
+
"grad_norm": 0.018780462443828583,
|
| 1826 |
+
"learning_rate": 3.829333333333334e-06,
|
| 1827 |
+
"loss": 0.0012,
|
| 1828 |
+
"step": 41400
|
| 1829 |
+
},
|
| 1830 |
+
{
|
| 1831 |
+
"epoch": 132.9088,
|
| 1832 |
+
"grad_norm": 0.021932663396000862,
|
| 1833 |
+
"learning_rate": 3.7408888888888895e-06,
|
| 1834 |
+
"loss": 0.0013,
|
| 1835 |
+
"step": 41600
|
| 1836 |
+
},
|
| 1837 |
+
{
|
| 1838 |
+
"epoch": 133.5472,
|
| 1839 |
+
"grad_norm": 0.40411534905433655,
|
| 1840 |
+
"learning_rate": 3.6520000000000004e-06,
|
| 1841 |
+
"loss": 0.0013,
|
| 1842 |
+
"step": 41800
|
| 1843 |
+
},
|
| 1844 |
+
{
|
| 1845 |
+
"epoch": 134.1856,
|
| 1846 |
+
"grad_norm": 2.085688352584839,
|
| 1847 |
+
"learning_rate": 3.5631111111111117e-06,
|
| 1848 |
+
"loss": 0.0011,
|
| 1849 |
+
"step": 42000
|
| 1850 |
+
},
|
| 1851 |
+
{
|
| 1852 |
+
"epoch": 134.1856,
|
| 1853 |
+
"eval_loss": 1.04826819896698,
|
| 1854 |
+
"eval_runtime": 125.1858,
|
| 1855 |
+
"eval_samples_per_second": 4.993,
|
| 1856 |
+
"eval_steps_per_second": 1.254,
|
| 1857 |
+
"eval_wer": 0.3762898417794084,
|
| 1858 |
+
"step": 42000
|
| 1859 |
+
},
|
| 1860 |
+
{
|
| 1861 |
+
"epoch": 134.8256,
|
| 1862 |
+
"grad_norm": 0.1101042777299881,
|
| 1863 |
+
"learning_rate": 3.4742222222222225e-06,
|
| 1864 |
+
"loss": 0.001,
|
| 1865 |
+
"step": 42200
|
| 1866 |
+
},
|
| 1867 |
+
{
|
| 1868 |
+
"epoch": 135.464,
|
| 1869 |
+
"grad_norm": 0.5308594107627869,
|
| 1870 |
+
"learning_rate": 3.385333333333334e-06,
|
| 1871 |
+
"loss": 0.0012,
|
| 1872 |
+
"step": 42400
|
| 1873 |
+
},
|
| 1874 |
+
{
|
| 1875 |
+
"epoch": 136.1024,
|
| 1876 |
+
"grad_norm": 0.10180327296257019,
|
| 1877 |
+
"learning_rate": 3.2964444444444447e-06,
|
| 1878 |
+
"loss": 0.001,
|
| 1879 |
+
"step": 42600
|
| 1880 |
+
},
|
| 1881 |
+
{
|
| 1882 |
+
"epoch": 136.7424,
|
| 1883 |
+
"grad_norm": 0.029545770958065987,
|
| 1884 |
+
"learning_rate": 3.207555555555556e-06,
|
| 1885 |
+
"loss": 0.0009,
|
| 1886 |
+
"step": 42800
|
| 1887 |
+
},
|
| 1888 |
+
{
|
| 1889 |
+
"epoch": 137.3808,
|
| 1890 |
+
"grad_norm": 0.022081894800066948,
|
| 1891 |
+
"learning_rate": 3.118666666666667e-06,
|
| 1892 |
+
"loss": 0.001,
|
| 1893 |
+
"step": 43000
|
| 1894 |
+
},
|
| 1895 |
+
{
|
| 1896 |
+
"epoch": 137.3808,
|
| 1897 |
+
"eval_loss": 1.0376399755477905,
|
| 1898 |
+
"eval_runtime": 129.9317,
|
| 1899 |
+
"eval_samples_per_second": 4.81,
|
| 1900 |
+
"eval_steps_per_second": 1.208,
|
| 1901 |
+
"eval_wer": 0.4168768631047925,
|
| 1902 |
+
"step": 43000
|
| 1903 |
+
},
|
| 1904 |
+
{
|
| 1905 |
+
"epoch": 138.0192,
|
| 1906 |
+
"grad_norm": 0.7658660411834717,
|
| 1907 |
+
"learning_rate": 3.029777777777778e-06,
|
| 1908 |
+
"loss": 0.0011,
|
| 1909 |
+
"step": 43200
|
| 1910 |
+
},
|
| 1911 |
+
{
|
| 1912 |
+
"epoch": 138.6592,
|
| 1913 |
+
"grad_norm": 0.027192605659365654,
|
| 1914 |
+
"learning_rate": 2.940888888888889e-06,
|
| 1915 |
+
"loss": 0.0011,
|
| 1916 |
+
"step": 43400
|
| 1917 |
+
},
|
| 1918 |
+
{
|
| 1919 |
+
"epoch": 139.2976,
|
| 1920 |
+
"grad_norm": 0.03390691056847572,
|
| 1921 |
+
"learning_rate": 2.8520000000000004e-06,
|
| 1922 |
+
"loss": 0.001,
|
| 1923 |
+
"step": 43600
|
| 1924 |
+
},
|
| 1925 |
+
{
|
| 1926 |
+
"epoch": 139.9376,
|
| 1927 |
+
"grad_norm": 0.08181264251470566,
|
| 1928 |
+
"learning_rate": 2.7631111111111113e-06,
|
| 1929 |
+
"loss": 0.0011,
|
| 1930 |
+
"step": 43800
|
| 1931 |
+
},
|
| 1932 |
+
{
|
| 1933 |
+
"epoch": 140.576,
|
| 1934 |
+
"grad_norm": 0.3259316086769104,
|
| 1935 |
+
"learning_rate": 2.6742222222222226e-06,
|
| 1936 |
+
"loss": 0.0008,
|
| 1937 |
+
"step": 44000
|
| 1938 |
+
},
|
| 1939 |
+
{
|
| 1940 |
+
"epoch": 140.576,
|
| 1941 |
+
"eval_loss": 1.0424144268035889,
|
| 1942 |
+
"eval_runtime": 104.1123,
|
| 1943 |
+
"eval_samples_per_second": 6.003,
|
| 1944 |
+
"eval_steps_per_second": 1.508,
|
| 1945 |
+
"eval_wer": 0.36321944508140336,
|
| 1946 |
+
"step": 44000
|
| 1947 |
+
},
|
| 1948 |
+
{
|
| 1949 |
+
"epoch": 141.2144,
|
| 1950 |
+
"grad_norm": 0.02337617427110672,
|
| 1951 |
+
"learning_rate": 2.5853333333333335e-06,
|
| 1952 |
+
"loss": 0.0005,
|
| 1953 |
+
"step": 44200
|
| 1954 |
+
},
|
| 1955 |
+
{
|
| 1956 |
+
"epoch": 141.8544,
|
| 1957 |
+
"grad_norm": 0.01961325667798519,
|
| 1958 |
+
"learning_rate": 2.4964444444444448e-06,
|
| 1959 |
+
"loss": 0.001,
|
| 1960 |
+
"step": 44400
|
| 1961 |
+
},
|
| 1962 |
+
{
|
| 1963 |
+
"epoch": 142.4928,
|
| 1964 |
+
"grad_norm": 0.0774613469839096,
|
| 1965 |
+
"learning_rate": 2.4075555555555556e-06,
|
| 1966 |
+
"loss": 0.0008,
|
| 1967 |
+
"step": 44600
|
| 1968 |
+
},
|
| 1969 |
+
{
|
| 1970 |
+
"epoch": 143.1312,
|
| 1971 |
+
"grad_norm": 0.4658966660499573,
|
| 1972 |
+
"learning_rate": 2.318666666666667e-06,
|
| 1973 |
+
"loss": 0.0007,
|
| 1974 |
+
"step": 44800
|
| 1975 |
+
},
|
| 1976 |
+
{
|
| 1977 |
+
"epoch": 143.7712,
|
| 1978 |
+
"grad_norm": 0.39419323205947876,
|
| 1979 |
+
"learning_rate": 2.229777777777778e-06,
|
| 1980 |
+
"loss": 0.0007,
|
| 1981 |
+
"step": 45000
|
| 1982 |
+
},
|
| 1983 |
+
{
|
| 1984 |
+
"epoch": 143.7712,
|
| 1985 |
+
"eval_loss": 1.0494757890701294,
|
| 1986 |
+
"eval_runtime": 104.6831,
|
| 1987 |
+
"eval_samples_per_second": 5.97,
|
| 1988 |
+
"eval_steps_per_second": 1.5,
|
| 1989 |
+
"eval_wer": 0.4159596422838798,
|
| 1990 |
+
"step": 45000
|
| 1991 |
+
},
|
| 1992 |
+
{
|
| 1993 |
+
"epoch": 144.4096,
|
| 1994 |
+
"grad_norm": 1.093592882156372,
|
| 1995 |
+
"learning_rate": 2.140888888888889e-06,
|
| 1996 |
+
"loss": 0.0008,
|
| 1997 |
+
"step": 45200
|
| 1998 |
+
},
|
| 1999 |
+
{
|
| 2000 |
+
"epoch": 145.048,
|
| 2001 |
+
"grad_norm": 0.011096829548478127,
|
| 2002 |
+
"learning_rate": 2.052e-06,
|
| 2003 |
+
"loss": 0.0008,
|
| 2004 |
+
"step": 45400
|
| 2005 |
+
},
|
| 2006 |
+
{
|
| 2007 |
+
"epoch": 145.688,
|
| 2008 |
+
"grad_norm": 0.18569226562976837,
|
| 2009 |
+
"learning_rate": 1.9631111111111113e-06,
|
| 2010 |
+
"loss": 0.0009,
|
| 2011 |
+
"step": 45600
|
| 2012 |
+
},
|
| 2013 |
+
{
|
| 2014 |
+
"epoch": 146.3264,
|
| 2015 |
+
"grad_norm": 0.17811138927936554,
|
| 2016 |
+
"learning_rate": 1.8742222222222222e-06,
|
| 2017 |
+
"loss": 0.0008,
|
| 2018 |
+
"step": 45800
|
| 2019 |
+
},
|
| 2020 |
+
{
|
| 2021 |
+
"epoch": 146.9664,
|
| 2022 |
+
"grad_norm": 0.01262740883976221,
|
| 2023 |
+
"learning_rate": 1.7853333333333333e-06,
|
| 2024 |
+
"loss": 0.0009,
|
| 2025 |
+
"step": 46000
|
| 2026 |
+
},
|
| 2027 |
+
{
|
| 2028 |
+
"epoch": 146.9664,
|
| 2029 |
+
"eval_loss": 1.054679274559021,
|
| 2030 |
+
"eval_runtime": 104.7175,
|
| 2031 |
+
"eval_samples_per_second": 5.968,
|
| 2032 |
+
"eval_steps_per_second": 1.499,
|
| 2033 |
+
"eval_wer": 0.36872277000687914,
|
| 2034 |
+
"step": 46000
|
| 2035 |
+
},
|
| 2036 |
+
{
|
| 2037 |
+
"epoch": 147.6048,
|
| 2038 |
+
"grad_norm": 7.735331058502197,
|
| 2039 |
+
"learning_rate": 1.6964444444444444e-06,
|
| 2040 |
+
"loss": 0.0008,
|
| 2041 |
+
"step": 46200
|
| 2042 |
+
},
|
| 2043 |
+
{
|
| 2044 |
+
"epoch": 148.2432,
|
| 2045 |
+
"grad_norm": 0.2187137007713318,
|
| 2046 |
+
"learning_rate": 1.608e-06,
|
| 2047 |
+
"loss": 0.0009,
|
| 2048 |
+
"step": 46400
|
| 2049 |
+
},
|
| 2050 |
+
{
|
| 2051 |
+
"epoch": 148.8832,
|
| 2052 |
+
"grad_norm": 0.15815149247646332,
|
| 2053 |
+
"learning_rate": 1.5191111111111112e-06,
|
| 2054 |
+
"loss": 0.0008,
|
| 2055 |
+
"step": 46600
|
| 2056 |
+
},
|
| 2057 |
+
{
|
| 2058 |
+
"epoch": 149.5216,
|
| 2059 |
+
"grad_norm": 0.3170558214187622,
|
| 2060 |
+
"learning_rate": 1.430666666666667e-06,
|
| 2061 |
+
"loss": 0.0005,
|
| 2062 |
+
"step": 46800
|
| 2063 |
+
},
|
| 2064 |
+
{
|
| 2065 |
+
"epoch": 150.16,
|
| 2066 |
+
"grad_norm": 0.09482572972774506,
|
| 2067 |
+
"learning_rate": 1.341777777777778e-06,
|
| 2068 |
+
"loss": 0.0005,
|
| 2069 |
+
"step": 47000
|
| 2070 |
+
},
|
| 2071 |
+
{
|
| 2072 |
+
"epoch": 150.16,
|
| 2073 |
+
"eval_loss": 1.058286428451538,
|
| 2074 |
+
"eval_runtime": 103.6331,
|
| 2075 |
+
"eval_samples_per_second": 6.031,
|
| 2076 |
+
"eval_steps_per_second": 1.515,
|
| 2077 |
+
"eval_wer": 0.3689520752121073,
|
| 2078 |
+
"step": 47000
|
| 2079 |
+
},
|
| 2080 |
+
{
|
| 2081 |
+
"epoch": 150.8,
|
| 2082 |
+
"grad_norm": 0.012617244385182858,
|
| 2083 |
+
"learning_rate": 1.2528888888888891e-06,
|
| 2084 |
+
"loss": 0.0006,
|
| 2085 |
+
"step": 47200
|
| 2086 |
+
},
|
| 2087 |
+
{
|
| 2088 |
+
"epoch": 151.4384,
|
| 2089 |
+
"grad_norm": 0.032891806215047836,
|
| 2090 |
+
"learning_rate": 1.1640000000000002e-06,
|
| 2091 |
+
"loss": 0.0009,
|
| 2092 |
+
"step": 47400
|
| 2093 |
+
},
|
| 2094 |
+
{
|
| 2095 |
+
"epoch": 152.0768,
|
| 2096 |
+
"grad_norm": 1.6969341039657593,
|
| 2097 |
+
"learning_rate": 1.0751111111111113e-06,
|
| 2098 |
+
"loss": 0.0006,
|
| 2099 |
+
"step": 47600
|
| 2100 |
+
},
|
| 2101 |
+
{
|
| 2102 |
+
"epoch": 152.7168,
|
| 2103 |
+
"grad_norm": 0.2790795862674713,
|
| 2104 |
+
"learning_rate": 9.862222222222224e-07,
|
| 2105 |
+
"loss": 0.0006,
|
| 2106 |
+
"step": 47800
|
| 2107 |
+
},
|
| 2108 |
+
{
|
| 2109 |
+
"epoch": 153.3552,
|
| 2110 |
+
"grad_norm": 0.33090606331825256,
|
| 2111 |
+
"learning_rate": 8.973333333333334e-07,
|
| 2112 |
+
"loss": 0.0007,
|
| 2113 |
+
"step": 48000
|
| 2114 |
+
},
|
| 2115 |
+
{
|
| 2116 |
+
"epoch": 153.3552,
|
| 2117 |
+
"eval_loss": 1.0462735891342163,
|
| 2118 |
+
"eval_runtime": 105.9059,
|
| 2119 |
+
"eval_samples_per_second": 5.901,
|
| 2120 |
+
"eval_steps_per_second": 1.482,
|
| 2121 |
+
"eval_wer": 0.4122907590002293,
|
| 2122 |
+
"step": 48000
|
| 2123 |
+
},
|
| 2124 |
+
{
|
| 2125 |
+
"epoch": 153.9952,
|
| 2126 |
+
"grad_norm": 0.005133169237524271,
|
| 2127 |
+
"learning_rate": 8.084444444444445e-07,
|
| 2128 |
+
"loss": 0.0008,
|
| 2129 |
+
"step": 48200
|
| 2130 |
+
},
|
| 2131 |
+
{
|
| 2132 |
+
"epoch": 154.6336,
|
| 2133 |
+
"grad_norm": 0.020690323784947395,
|
| 2134 |
+
"learning_rate": 7.195555555555556e-07,
|
| 2135 |
+
"loss": 0.001,
|
| 2136 |
+
"step": 48400
|
| 2137 |
+
},
|
| 2138 |
+
{
|
| 2139 |
+
"epoch": 155.272,
|
| 2140 |
+
"grad_norm": 2.341998338699341,
|
| 2141 |
+
"learning_rate": 6.306666666666666e-07,
|
| 2142 |
+
"loss": 0.0007,
|
| 2143 |
+
"step": 48600
|
| 2144 |
+
},
|
| 2145 |
+
{
|
| 2146 |
+
"epoch": 155.912,
|
| 2147 |
+
"grad_norm": 0.0469956174492836,
|
| 2148 |
+
"learning_rate": 5.417777777777778e-07,
|
| 2149 |
+
"loss": 0.0006,
|
| 2150 |
+
"step": 48800
|
| 2151 |
+
},
|
| 2152 |
+
{
|
| 2153 |
+
"epoch": 156.5504,
|
| 2154 |
+
"grad_norm": 0.006763980723917484,
|
| 2155 |
+
"learning_rate": 4.528888888888889e-07,
|
| 2156 |
+
"loss": 0.0007,
|
| 2157 |
+
"step": 49000
|
| 2158 |
+
},
|
| 2159 |
+
{
|
| 2160 |
+
"epoch": 156.5504,
|
| 2161 |
+
"eval_loss": 1.0463974475860596,
|
| 2162 |
+
"eval_runtime": 104.646,
|
| 2163 |
+
"eval_samples_per_second": 5.973,
|
| 2164 |
+
"eval_steps_per_second": 1.5,
|
| 2165 |
+
"eval_wer": 0.3600091722082091,
|
| 2166 |
+
"step": 49000
|
| 2167 |
+
},
|
| 2168 |
+
{
|
| 2169 |
+
"epoch": 157.1888,
|
| 2170 |
+
"grad_norm": 0.01508264522999525,
|
| 2171 |
+
"learning_rate": 3.6400000000000003e-07,
|
| 2172 |
+
"loss": 0.0008,
|
| 2173 |
+
"step": 49200
|
| 2174 |
+
},
|
| 2175 |
+
{
|
| 2176 |
+
"epoch": 157.8288,
|
| 2177 |
+
"grad_norm": 0.008547438308596611,
|
| 2178 |
+
"learning_rate": 2.751111111111111e-07,
|
| 2179 |
+
"loss": 0.0013,
|
| 2180 |
+
"step": 49400
|
| 2181 |
+
},
|
| 2182 |
+
{
|
| 2183 |
+
"epoch": 158.4672,
|
| 2184 |
+
"grad_norm": 0.03822452574968338,
|
| 2185 |
+
"learning_rate": 1.8622222222222221e-07,
|
| 2186 |
+
"loss": 0.0005,
|
| 2187 |
+
"step": 49600
|
| 2188 |
+
},
|
| 2189 |
+
{
|
| 2190 |
+
"epoch": 159.1056,
|
| 2191 |
+
"grad_norm": 0.07067080587148666,
|
| 2192 |
+
"learning_rate": 9.733333333333334e-08,
|
| 2193 |
+
"loss": 0.0006,
|
| 2194 |
+
"step": 49800
|
| 2195 |
+
},
|
| 2196 |
+
{
|
| 2197 |
+
"epoch": 159.7456,
|
| 2198 |
+
"grad_norm": 0.007019846700131893,
|
| 2199 |
+
"learning_rate": 8.444444444444445e-09,
|
| 2200 |
+
"loss": 0.0005,
|
| 2201 |
+
"step": 50000
|
| 2202 |
+
},
|
| 2203 |
+
{
|
| 2204 |
+
"epoch": 159.7456,
|
| 2205 |
+
"eval_loss": 1.0454859733581543,
|
| 2206 |
+
"eval_runtime": 104.5679,
|
| 2207 |
+
"eval_samples_per_second": 5.977,
|
| 2208 |
+
"eval_steps_per_second": 1.501,
|
| 2209 |
+
"eval_wer": 0.3609263930291218,
|
| 2210 |
+
"step": 50000
|
| 2211 |
+
},
|
| 2212 |
+
{
|
| 2213 |
+
"epoch": 159.7456,
|
| 2214 |
+
"step": 50000,
|
| 2215 |
+
"total_flos": 1.966379287412736e+19,
|
| 2216 |
+
"train_loss": 0.0770821487186849,
|
| 2217 |
+
"train_runtime": 117170.0508,
|
| 2218 |
+
"train_samples_per_second": 6.828,
|
| 2219 |
+
"train_steps_per_second": 0.427
|
| 2220 |
+
}
|
| 2221 |
+
],
|
| 2222 |
+
"logging_steps": 200,
|
| 2223 |
+
"max_steps": 50000,
|
| 2224 |
+
"num_input_tokens_seen": 0,
|
| 2225 |
+
"num_train_epochs": 161,
|
| 2226 |
+
"save_steps": 1000,
|
| 2227 |
+
"stateful_callbacks": {
|
| 2228 |
+
"TrainerControl": {
|
| 2229 |
+
"args": {
|
| 2230 |
+
"should_epoch_stop": false,
|
| 2231 |
+
"should_evaluate": false,
|
| 2232 |
+
"should_log": false,
|
| 2233 |
+
"should_save": true,
|
| 2234 |
+
"should_training_stop": true
|
| 2235 |
+
},
|
| 2236 |
+
"attributes": {}
|
| 2237 |
+
}
|
| 2238 |
+
},
|
| 2239 |
+
"total_flos": 1.966379287412736e+19,
|
| 2240 |
+
"train_batch_size": 8,
|
| 2241 |
+
"trial_name": null,
|
| 2242 |
+
"trial_params": null
|
| 2243 |
+
}
|
val_results.json
ADDED
|
@@ -0,0 +1,9 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"epoch": 159.7456,
|
| 3 |
+
"phase_1_val_loss": 1.0391464233398438,
|
| 4 |
+
"phase_1_val_runtime": 101.9929,
|
| 5 |
+
"phase_1_val_samples_per_second": 6.128,
|
| 6 |
+
"phase_1_val_steps_per_second": 1.539,
|
| 7 |
+
"phase_1_val_wer": 0.359779867002981,
|
| 8 |
+
"val_samples": 625
|
| 9 |
+
}
|