End of training
Browse files- README.md +7 -2
- all_results.json +15 -0
- phase_1_train_results.json +9 -0
- runs/Apr29_22-22-00_dgx-a100/events.out.tfevents.1746043134.dgx-a100.2626384.1 +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-v7
|
| 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-v7
|
| 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": 400.0,
|
| 3 |
+
"phase_1_val_loss": 1.27247154712677,
|
| 4 |
+
"phase_1_val_runtime": 41.2147,
|
| 5 |
+
"phase_1_val_samples_per_second": 6.066,
|
| 6 |
+
"phase_1_val_steps_per_second": 1.529,
|
| 7 |
+
"phase_1_val_wer": 0.44776962168266515,
|
| 8 |
+
"total_flos": 1.9695108096e+19,
|
| 9 |
+
"train_loss": 0.05668430684482679,
|
| 10 |
+
"train_runtime": 102928.0936,
|
| 11 |
+
"train_samples": 2000,
|
| 12 |
+
"train_samples_per_second": 7.772,
|
| 13 |
+
"train_steps_per_second": 0.486,
|
| 14 |
+
"val_samples": 250
|
| 15 |
+
}
|
phase_1_train_results.json
ADDED
|
@@ -0,0 +1,9 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"epoch": 400.0,
|
| 3 |
+
"total_flos": 1.9695108096e+19,
|
| 4 |
+
"train_loss": 0.05668430684482679,
|
| 5 |
+
"train_runtime": 102928.0936,
|
| 6 |
+
"train_samples": 2000,
|
| 7 |
+
"train_samples_per_second": 7.772,
|
| 8 |
+
"train_steps_per_second": 0.486
|
| 9 |
+
}
|
runs/Apr29_22-22-00_dgx-a100/events.out.tfevents.1746043134.dgx-a100.2626384.1
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:43c68ec6797bb3a328b1c0ae3f4ac0f9053b567511fee75e30781fe7cadaa8fe
|
| 3 |
+
size 477
|
trainer_state.json
ADDED
|
@@ -0,0 +1,2243 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"best_global_step": 13000,
|
| 3 |
+
"best_metric": 0.44776962168266515,
|
| 4 |
+
"best_model_checkpoint": "/home/cluster-dgxa100/slp01/bagas-fine-tune-whisper/whisper-tiny-javanese-openslr-v7/checkpoint-13000",
|
| 5 |
+
"epoch": 400.0,
|
| 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": 1.6,
|
| 14 |
+
"grad_norm": 28.316560745239258,
|
| 15 |
+
"learning_rate": 7.88e-07,
|
| 16 |
+
"loss": 3.6257,
|
| 17 |
+
"step": 200
|
| 18 |
+
},
|
| 19 |
+
{
|
| 20 |
+
"epoch": 3.2,
|
| 21 |
+
"grad_norm": 19.273317337036133,
|
| 22 |
+
"learning_rate": 1.588e-06,
|
| 23 |
+
"loss": 2.0745,
|
| 24 |
+
"step": 400
|
| 25 |
+
},
|
| 26 |
+
{
|
| 27 |
+
"epoch": 4.8,
|
| 28 |
+
"grad_norm": 20.253143310546875,
|
| 29 |
+
"learning_rate": 2.3880000000000003e-06,
|
| 30 |
+
"loss": 1.5564,
|
| 31 |
+
"step": 600
|
| 32 |
+
},
|
| 33 |
+
{
|
| 34 |
+
"epoch": 6.4,
|
| 35 |
+
"grad_norm": 18.321447372436523,
|
| 36 |
+
"learning_rate": 3.188e-06,
|
| 37 |
+
"loss": 1.2332,
|
| 38 |
+
"step": 800
|
| 39 |
+
},
|
| 40 |
+
{
|
| 41 |
+
"epoch": 8.0,
|
| 42 |
+
"grad_norm": 15.124577522277832,
|
| 43 |
+
"learning_rate": 3.988000000000001e-06,
|
| 44 |
+
"loss": 0.9973,
|
| 45 |
+
"step": 1000
|
| 46 |
+
},
|
| 47 |
+
{
|
| 48 |
+
"epoch": 8.0,
|
| 49 |
+
"eval_loss": 1.0427539348602295,
|
| 50 |
+
"eval_runtime": 42.8057,
|
| 51 |
+
"eval_samples_per_second": 5.84,
|
| 52 |
+
"eval_steps_per_second": 1.472,
|
| 53 |
+
"eval_wer": 0.6821005081874647,
|
| 54 |
+
"step": 1000
|
| 55 |
+
},
|
| 56 |
+
{
|
| 57 |
+
"epoch": 9.6,
|
| 58 |
+
"grad_norm": 17.642391204833984,
|
| 59 |
+
"learning_rate": 4.7880000000000006e-06,
|
| 60 |
+
"loss": 0.7932,
|
| 61 |
+
"step": 1200
|
| 62 |
+
},
|
| 63 |
+
{
|
| 64 |
+
"epoch": 11.2,
|
| 65 |
+
"grad_norm": 12.182209014892578,
|
| 66 |
+
"learning_rate": 5.588e-06,
|
| 67 |
+
"loss": 0.6471,
|
| 68 |
+
"step": 1400
|
| 69 |
+
},
|
| 70 |
+
{
|
| 71 |
+
"epoch": 12.8,
|
| 72 |
+
"grad_norm": 14.519808769226074,
|
| 73 |
+
"learning_rate": 6.3880000000000005e-06,
|
| 74 |
+
"loss": 0.5151,
|
| 75 |
+
"step": 1600
|
| 76 |
+
},
|
| 77 |
+
{
|
| 78 |
+
"epoch": 14.4,
|
| 79 |
+
"grad_norm": 13.092401504516602,
|
| 80 |
+
"learning_rate": 7.1880000000000005e-06,
|
| 81 |
+
"loss": 0.3948,
|
| 82 |
+
"step": 1800
|
| 83 |
+
},
|
| 84 |
+
{
|
| 85 |
+
"epoch": 16.0,
|
| 86 |
+
"grad_norm": 11.094600677490234,
|
| 87 |
+
"learning_rate": 7.988e-06,
|
| 88 |
+
"loss": 0.3139,
|
| 89 |
+
"step": 2000
|
| 90 |
+
},
|
| 91 |
+
{
|
| 92 |
+
"epoch": 16.0,
|
| 93 |
+
"eval_loss": 0.8733022212982178,
|
| 94 |
+
"eval_runtime": 46.221,
|
| 95 |
+
"eval_samples_per_second": 5.409,
|
| 96 |
+
"eval_steps_per_second": 1.363,
|
| 97 |
+
"eval_wer": 0.6233766233766234,
|
| 98 |
+
"step": 2000
|
| 99 |
+
},
|
| 100 |
+
{
|
| 101 |
+
"epoch": 17.6,
|
| 102 |
+
"grad_norm": 13.814520835876465,
|
| 103 |
+
"learning_rate": 8.788000000000001e-06,
|
| 104 |
+
"loss": 0.2389,
|
| 105 |
+
"step": 2200
|
| 106 |
+
},
|
| 107 |
+
{
|
| 108 |
+
"epoch": 19.2,
|
| 109 |
+
"grad_norm": 10.147465705871582,
|
| 110 |
+
"learning_rate": 9.588e-06,
|
| 111 |
+
"loss": 0.1868,
|
| 112 |
+
"step": 2400
|
| 113 |
+
},
|
| 114 |
+
{
|
| 115 |
+
"epoch": 20.8,
|
| 116 |
+
"grad_norm": 6.512983798980713,
|
| 117 |
+
"learning_rate": 1.0388e-05,
|
| 118 |
+
"loss": 0.1454,
|
| 119 |
+
"step": 2600
|
| 120 |
+
},
|
| 121 |
+
{
|
| 122 |
+
"epoch": 22.4,
|
| 123 |
+
"grad_norm": 7.759674549102783,
|
| 124 |
+
"learning_rate": 1.1188e-05,
|
| 125 |
+
"loss": 0.1186,
|
| 126 |
+
"step": 2800
|
| 127 |
+
},
|
| 128 |
+
{
|
| 129 |
+
"epoch": 24.0,
|
| 130 |
+
"grad_norm": 8.339158058166504,
|
| 131 |
+
"learning_rate": 1.1988000000000001e-05,
|
| 132 |
+
"loss": 0.0988,
|
| 133 |
+
"step": 3000
|
| 134 |
+
},
|
| 135 |
+
{
|
| 136 |
+
"epoch": 24.0,
|
| 137 |
+
"eval_loss": 0.8925266861915588,
|
| 138 |
+
"eval_runtime": 44.0477,
|
| 139 |
+
"eval_samples_per_second": 5.676,
|
| 140 |
+
"eval_steps_per_second": 1.43,
|
| 141 |
+
"eval_wer": 0.6346696781479391,
|
| 142 |
+
"step": 3000
|
| 143 |
+
},
|
| 144 |
+
{
|
| 145 |
+
"epoch": 25.6,
|
| 146 |
+
"grad_norm": 4.570150375366211,
|
| 147 |
+
"learning_rate": 1.2788e-05,
|
| 148 |
+
"loss": 0.0821,
|
| 149 |
+
"step": 3200
|
| 150 |
+
},
|
| 151 |
+
{
|
| 152 |
+
"epoch": 27.2,
|
| 153 |
+
"grad_norm": 6.510232925415039,
|
| 154 |
+
"learning_rate": 1.3588000000000001e-05,
|
| 155 |
+
"loss": 0.0745,
|
| 156 |
+
"step": 3400
|
| 157 |
+
},
|
| 158 |
+
{
|
| 159 |
+
"epoch": 28.8,
|
| 160 |
+
"grad_norm": 6.720197677612305,
|
| 161 |
+
"learning_rate": 1.4388000000000002e-05,
|
| 162 |
+
"loss": 0.0664,
|
| 163 |
+
"step": 3600
|
| 164 |
+
},
|
| 165 |
+
{
|
| 166 |
+
"epoch": 30.4,
|
| 167 |
+
"grad_norm": 6.113977432250977,
|
| 168 |
+
"learning_rate": 1.5188000000000001e-05,
|
| 169 |
+
"loss": 0.0579,
|
| 170 |
+
"step": 3800
|
| 171 |
+
},
|
| 172 |
+
{
|
| 173 |
+
"epoch": 32.0,
|
| 174 |
+
"grad_norm": 4.528390884399414,
|
| 175 |
+
"learning_rate": 1.5988e-05,
|
| 176 |
+
"loss": 0.0499,
|
| 177 |
+
"step": 4000
|
| 178 |
+
},
|
| 179 |
+
{
|
| 180 |
+
"epoch": 32.0,
|
| 181 |
+
"eval_loss": 0.9389314651489258,
|
| 182 |
+
"eval_runtime": 43.1155,
|
| 183 |
+
"eval_samples_per_second": 5.798,
|
| 184 |
+
"eval_steps_per_second": 1.461,
|
| 185 |
+
"eval_wer": 0.6149068322981367,
|
| 186 |
+
"step": 4000
|
| 187 |
+
},
|
| 188 |
+
{
|
| 189 |
+
"epoch": 33.6,
|
| 190 |
+
"grad_norm": 8.023102760314941,
|
| 191 |
+
"learning_rate": 1.6784e-05,
|
| 192 |
+
"loss": 0.0488,
|
| 193 |
+
"step": 4200
|
| 194 |
+
},
|
| 195 |
+
{
|
| 196 |
+
"epoch": 35.2,
|
| 197 |
+
"grad_norm": 5.378859043121338,
|
| 198 |
+
"learning_rate": 1.7584e-05,
|
| 199 |
+
"loss": 0.0443,
|
| 200 |
+
"step": 4400
|
| 201 |
+
},
|
| 202 |
+
{
|
| 203 |
+
"epoch": 36.8,
|
| 204 |
+
"grad_norm": 6.565428256988525,
|
| 205 |
+
"learning_rate": 1.8380000000000004e-05,
|
| 206 |
+
"loss": 0.0415,
|
| 207 |
+
"step": 4600
|
| 208 |
+
},
|
| 209 |
+
{
|
| 210 |
+
"epoch": 38.4,
|
| 211 |
+
"grad_norm": 3.81632137298584,
|
| 212 |
+
"learning_rate": 1.918e-05,
|
| 213 |
+
"loss": 0.0387,
|
| 214 |
+
"step": 4800
|
| 215 |
+
},
|
| 216 |
+
{
|
| 217 |
+
"epoch": 40.0,
|
| 218 |
+
"grad_norm": 1.75031316280365,
|
| 219 |
+
"learning_rate": 1.9980000000000002e-05,
|
| 220 |
+
"loss": 0.0349,
|
| 221 |
+
"step": 5000
|
| 222 |
+
},
|
| 223 |
+
{
|
| 224 |
+
"epoch": 40.0,
|
| 225 |
+
"eval_loss": 0.9878742098808289,
|
| 226 |
+
"eval_runtime": 41.8583,
|
| 227 |
+
"eval_samples_per_second": 5.973,
|
| 228 |
+
"eval_steps_per_second": 1.505,
|
| 229 |
+
"eval_wer": 0.6205533596837944,
|
| 230 |
+
"step": 5000
|
| 231 |
+
},
|
| 232 |
+
{
|
| 233 |
+
"epoch": 41.6,
|
| 234 |
+
"grad_norm": 1.5769927501678467,
|
| 235 |
+
"learning_rate": 1.9913333333333335e-05,
|
| 236 |
+
"loss": 0.0335,
|
| 237 |
+
"step": 5200
|
| 238 |
+
},
|
| 239 |
+
{
|
| 240 |
+
"epoch": 43.2,
|
| 241 |
+
"grad_norm": 4.837329387664795,
|
| 242 |
+
"learning_rate": 1.9824444444444445e-05,
|
| 243 |
+
"loss": 0.0303,
|
| 244 |
+
"step": 5400
|
| 245 |
+
},
|
| 246 |
+
{
|
| 247 |
+
"epoch": 44.8,
|
| 248 |
+
"grad_norm": 4.526083469390869,
|
| 249 |
+
"learning_rate": 1.9735555555555556e-05,
|
| 250 |
+
"loss": 0.0282,
|
| 251 |
+
"step": 5600
|
| 252 |
+
},
|
| 253 |
+
{
|
| 254 |
+
"epoch": 46.4,
|
| 255 |
+
"grad_norm": 4.3648176193237305,
|
| 256 |
+
"learning_rate": 1.9646666666666666e-05,
|
| 257 |
+
"loss": 0.0285,
|
| 258 |
+
"step": 5800
|
| 259 |
+
},
|
| 260 |
+
{
|
| 261 |
+
"epoch": 48.0,
|
| 262 |
+
"grad_norm": 4.617649078369141,
|
| 263 |
+
"learning_rate": 1.955777777777778e-05,
|
| 264 |
+
"loss": 0.0246,
|
| 265 |
+
"step": 6000
|
| 266 |
+
},
|
| 267 |
+
{
|
| 268 |
+
"epoch": 48.0,
|
| 269 |
+
"eval_loss": 1.02446711063385,
|
| 270 |
+
"eval_runtime": 41.3069,
|
| 271 |
+
"eval_samples_per_second": 6.052,
|
| 272 |
+
"eval_steps_per_second": 1.525,
|
| 273 |
+
"eval_wer": 0.46809712027103334,
|
| 274 |
+
"step": 6000
|
| 275 |
+
},
|
| 276 |
+
{
|
| 277 |
+
"epoch": 49.6,
|
| 278 |
+
"grad_norm": 1.5408967733383179,
|
| 279 |
+
"learning_rate": 1.946888888888889e-05,
|
| 280 |
+
"loss": 0.0235,
|
| 281 |
+
"step": 6200
|
| 282 |
+
},
|
| 283 |
+
{
|
| 284 |
+
"epoch": 51.2,
|
| 285 |
+
"grad_norm": 4.78511381149292,
|
| 286 |
+
"learning_rate": 1.938e-05,
|
| 287 |
+
"loss": 0.0212,
|
| 288 |
+
"step": 6400
|
| 289 |
+
},
|
| 290 |
+
{
|
| 291 |
+
"epoch": 52.8,
|
| 292 |
+
"grad_norm": 0.8843356370925903,
|
| 293 |
+
"learning_rate": 1.9291111111111115e-05,
|
| 294 |
+
"loss": 0.0193,
|
| 295 |
+
"step": 6600
|
| 296 |
+
},
|
| 297 |
+
{
|
| 298 |
+
"epoch": 54.4,
|
| 299 |
+
"grad_norm": 2.4114930629730225,
|
| 300 |
+
"learning_rate": 1.9202222222222225e-05,
|
| 301 |
+
"loss": 0.0193,
|
| 302 |
+
"step": 6800
|
| 303 |
+
},
|
| 304 |
+
{
|
| 305 |
+
"epoch": 56.0,
|
| 306 |
+
"grad_norm": 3.1046841144561768,
|
| 307 |
+
"learning_rate": 1.9113333333333336e-05,
|
| 308 |
+
"loss": 0.0164,
|
| 309 |
+
"step": 7000
|
| 310 |
+
},
|
| 311 |
+
{
|
| 312 |
+
"epoch": 56.0,
|
| 313 |
+
"eval_loss": 1.103171706199646,
|
| 314 |
+
"eval_runtime": 41.2052,
|
| 315 |
+
"eval_samples_per_second": 6.067,
|
| 316 |
+
"eval_steps_per_second": 1.529,
|
| 317 |
+
"eval_wer": 0.47148503670242803,
|
| 318 |
+
"step": 7000
|
| 319 |
+
},
|
| 320 |
+
{
|
| 321 |
+
"epoch": 57.6,
|
| 322 |
+
"grad_norm": 2.163294792175293,
|
| 323 |
+
"learning_rate": 1.902488888888889e-05,
|
| 324 |
+
"loss": 0.0157,
|
| 325 |
+
"step": 7200
|
| 326 |
+
},
|
| 327 |
+
{
|
| 328 |
+
"epoch": 59.2,
|
| 329 |
+
"grad_norm": 4.016397953033447,
|
| 330 |
+
"learning_rate": 1.8936e-05,
|
| 331 |
+
"loss": 0.015,
|
| 332 |
+
"step": 7400
|
| 333 |
+
},
|
| 334 |
+
{
|
| 335 |
+
"epoch": 60.8,
|
| 336 |
+
"grad_norm": 3.4655325412750244,
|
| 337 |
+
"learning_rate": 1.884711111111111e-05,
|
| 338 |
+
"loss": 0.0124,
|
| 339 |
+
"step": 7600
|
| 340 |
+
},
|
| 341 |
+
{
|
| 342 |
+
"epoch": 62.4,
|
| 343 |
+
"grad_norm": 2.6923351287841797,
|
| 344 |
+
"learning_rate": 1.8758222222222224e-05,
|
| 345 |
+
"loss": 0.0121,
|
| 346 |
+
"step": 7800
|
| 347 |
+
},
|
| 348 |
+
{
|
| 349 |
+
"epoch": 64.0,
|
| 350 |
+
"grad_norm": 3.1834566593170166,
|
| 351 |
+
"learning_rate": 1.8669333333333334e-05,
|
| 352 |
+
"loss": 0.0118,
|
| 353 |
+
"step": 8000
|
| 354 |
+
},
|
| 355 |
+
{
|
| 356 |
+
"epoch": 64.0,
|
| 357 |
+
"eval_loss": 1.1389034986495972,
|
| 358 |
+
"eval_runtime": 42.003,
|
| 359 |
+
"eval_samples_per_second": 5.952,
|
| 360 |
+
"eval_steps_per_second": 1.5,
|
| 361 |
+
"eval_wer": 0.598531902879729,
|
| 362 |
+
"step": 8000
|
| 363 |
+
},
|
| 364 |
+
{
|
| 365 |
+
"epoch": 65.6,
|
| 366 |
+
"grad_norm": 1.3980759382247925,
|
| 367 |
+
"learning_rate": 1.8580444444444445e-05,
|
| 368 |
+
"loss": 0.0126,
|
| 369 |
+
"step": 8200
|
| 370 |
+
},
|
| 371 |
+
{
|
| 372 |
+
"epoch": 67.2,
|
| 373 |
+
"grad_norm": 3.1268885135650635,
|
| 374 |
+
"learning_rate": 1.849155555555556e-05,
|
| 375 |
+
"loss": 0.0109,
|
| 376 |
+
"step": 8400
|
| 377 |
+
},
|
| 378 |
+
{
|
| 379 |
+
"epoch": 68.8,
|
| 380 |
+
"grad_norm": 2.908175468444824,
|
| 381 |
+
"learning_rate": 1.840266666666667e-05,
|
| 382 |
+
"loss": 0.0126,
|
| 383 |
+
"step": 8600
|
| 384 |
+
},
|
| 385 |
+
{
|
| 386 |
+
"epoch": 70.4,
|
| 387 |
+
"grad_norm": 3.613409996032715,
|
| 388 |
+
"learning_rate": 1.831377777777778e-05,
|
| 389 |
+
"loss": 0.0105,
|
| 390 |
+
"step": 8800
|
| 391 |
+
},
|
| 392 |
+
{
|
| 393 |
+
"epoch": 72.0,
|
| 394 |
+
"grad_norm": 2.127668619155884,
|
| 395 |
+
"learning_rate": 1.822488888888889e-05,
|
| 396 |
+
"loss": 0.0094,
|
| 397 |
+
"step": 9000
|
| 398 |
+
},
|
| 399 |
+
{
|
| 400 |
+
"epoch": 72.0,
|
| 401 |
+
"eval_loss": 1.145694375038147,
|
| 402 |
+
"eval_runtime": 41.9604,
|
| 403 |
+
"eval_samples_per_second": 5.958,
|
| 404 |
+
"eval_steps_per_second": 1.501,
|
| 405 |
+
"eval_wer": 0.5793337097684924,
|
| 406 |
+
"step": 9000
|
| 407 |
+
},
|
| 408 |
+
{
|
| 409 |
+
"epoch": 73.6,
|
| 410 |
+
"grad_norm": 3.8510990142822266,
|
| 411 |
+
"learning_rate": 1.8136000000000004e-05,
|
| 412 |
+
"loss": 0.0106,
|
| 413 |
+
"step": 9200
|
| 414 |
+
},
|
| 415 |
+
{
|
| 416 |
+
"epoch": 75.2,
|
| 417 |
+
"grad_norm": 0.6357870697975159,
|
| 418 |
+
"learning_rate": 1.8047111111111114e-05,
|
| 419 |
+
"loss": 0.0077,
|
| 420 |
+
"step": 9400
|
| 421 |
+
},
|
| 422 |
+
{
|
| 423 |
+
"epoch": 76.8,
|
| 424 |
+
"grad_norm": 0.7974900603294373,
|
| 425 |
+
"learning_rate": 1.7958222222222225e-05,
|
| 426 |
+
"loss": 0.0092,
|
| 427 |
+
"step": 9600
|
| 428 |
+
},
|
| 429 |
+
{
|
| 430 |
+
"epoch": 78.4,
|
| 431 |
+
"grad_norm": 3.332366943359375,
|
| 432 |
+
"learning_rate": 1.7869333333333335e-05,
|
| 433 |
+
"loss": 0.0097,
|
| 434 |
+
"step": 9800
|
| 435 |
+
},
|
| 436 |
+
{
|
| 437 |
+
"epoch": 80.0,
|
| 438 |
+
"grad_norm": 3.0786008834838867,
|
| 439 |
+
"learning_rate": 1.7780444444444446e-05,
|
| 440 |
+
"loss": 0.0086,
|
| 441 |
+
"step": 10000
|
| 442 |
+
},
|
| 443 |
+
{
|
| 444 |
+
"epoch": 80.0,
|
| 445 |
+
"eval_loss": 1.1853593587875366,
|
| 446 |
+
"eval_runtime": 42.1314,
|
| 447 |
+
"eval_samples_per_second": 5.934,
|
| 448 |
+
"eval_steps_per_second": 1.495,
|
| 449 |
+
"eval_wer": 0.5375494071146245,
|
| 450 |
+
"step": 10000
|
| 451 |
+
},
|
| 452 |
+
{
|
| 453 |
+
"epoch": 81.6,
|
| 454 |
+
"grad_norm": 1.0725177526474,
|
| 455 |
+
"learning_rate": 1.7691555555555556e-05,
|
| 456 |
+
"loss": 0.0072,
|
| 457 |
+
"step": 10200
|
| 458 |
+
},
|
| 459 |
+
{
|
| 460 |
+
"epoch": 83.2,
|
| 461 |
+
"grad_norm": 0.27933818101882935,
|
| 462 |
+
"learning_rate": 1.7602666666666667e-05,
|
| 463 |
+
"loss": 0.0074,
|
| 464 |
+
"step": 10400
|
| 465 |
+
},
|
| 466 |
+
{
|
| 467 |
+
"epoch": 84.8,
|
| 468 |
+
"grad_norm": 1.2738243341445923,
|
| 469 |
+
"learning_rate": 1.7513777777777777e-05,
|
| 470 |
+
"loss": 0.0061,
|
| 471 |
+
"step": 10600
|
| 472 |
+
},
|
| 473 |
+
{
|
| 474 |
+
"epoch": 86.4,
|
| 475 |
+
"grad_norm": 6.480189323425293,
|
| 476 |
+
"learning_rate": 1.742488888888889e-05,
|
| 477 |
+
"loss": 0.0065,
|
| 478 |
+
"step": 10800
|
| 479 |
+
},
|
| 480 |
+
{
|
| 481 |
+
"epoch": 88.0,
|
| 482 |
+
"grad_norm": 0.33281826972961426,
|
| 483 |
+
"learning_rate": 1.7336e-05,
|
| 484 |
+
"loss": 0.0067,
|
| 485 |
+
"step": 11000
|
| 486 |
+
},
|
| 487 |
+
{
|
| 488 |
+
"epoch": 88.0,
|
| 489 |
+
"eval_loss": 1.237168788909912,
|
| 490 |
+
"eval_runtime": 40.9166,
|
| 491 |
+
"eval_samples_per_second": 6.11,
|
| 492 |
+
"eval_steps_per_second": 1.54,
|
| 493 |
+
"eval_wer": 0.466403162055336,
|
| 494 |
+
"step": 11000
|
| 495 |
+
},
|
| 496 |
+
{
|
| 497 |
+
"epoch": 89.6,
|
| 498 |
+
"grad_norm": 0.395781934261322,
|
| 499 |
+
"learning_rate": 1.7247111111111112e-05,
|
| 500 |
+
"loss": 0.0055,
|
| 501 |
+
"step": 11200
|
| 502 |
+
},
|
| 503 |
+
{
|
| 504 |
+
"epoch": 91.2,
|
| 505 |
+
"grad_norm": 2.044802188873291,
|
| 506 |
+
"learning_rate": 1.7158222222222222e-05,
|
| 507 |
+
"loss": 0.0071,
|
| 508 |
+
"step": 11400
|
| 509 |
+
},
|
| 510 |
+
{
|
| 511 |
+
"epoch": 92.8,
|
| 512 |
+
"grad_norm": 0.7053709030151367,
|
| 513 |
+
"learning_rate": 1.7069333333333336e-05,
|
| 514 |
+
"loss": 0.0069,
|
| 515 |
+
"step": 11600
|
| 516 |
+
},
|
| 517 |
+
{
|
| 518 |
+
"epoch": 94.4,
|
| 519 |
+
"grad_norm": 2.6809892654418945,
|
| 520 |
+
"learning_rate": 1.6980444444444447e-05,
|
| 521 |
+
"loss": 0.0059,
|
| 522 |
+
"step": 11800
|
| 523 |
+
},
|
| 524 |
+
{
|
| 525 |
+
"epoch": 96.0,
|
| 526 |
+
"grad_norm": 0.43073830008506775,
|
| 527 |
+
"learning_rate": 1.6891555555555557e-05,
|
| 528 |
+
"loss": 0.005,
|
| 529 |
+
"step": 12000
|
| 530 |
+
},
|
| 531 |
+
{
|
| 532 |
+
"epoch": 96.0,
|
| 533 |
+
"eval_loss": 1.2564021348953247,
|
| 534 |
+
"eval_runtime": 43.0999,
|
| 535 |
+
"eval_samples_per_second": 5.8,
|
| 536 |
+
"eval_steps_per_second": 1.462,
|
| 537 |
+
"eval_wer": 0.4991530208921513,
|
| 538 |
+
"step": 12000
|
| 539 |
+
},
|
| 540 |
+
{
|
| 541 |
+
"epoch": 97.6,
|
| 542 |
+
"grad_norm": 0.7439399361610413,
|
| 543 |
+
"learning_rate": 1.6802666666666668e-05,
|
| 544 |
+
"loss": 0.0052,
|
| 545 |
+
"step": 12200
|
| 546 |
+
},
|
| 547 |
+
{
|
| 548 |
+
"epoch": 99.2,
|
| 549 |
+
"grad_norm": 0.18939250707626343,
|
| 550 |
+
"learning_rate": 1.671377777777778e-05,
|
| 551 |
+
"loss": 0.005,
|
| 552 |
+
"step": 12400
|
| 553 |
+
},
|
| 554 |
+
{
|
| 555 |
+
"epoch": 100.8,
|
| 556 |
+
"grad_norm": 2.827566385269165,
|
| 557 |
+
"learning_rate": 1.6624888888888892e-05,
|
| 558 |
+
"loss": 0.0052,
|
| 559 |
+
"step": 12600
|
| 560 |
+
},
|
| 561 |
+
{
|
| 562 |
+
"epoch": 102.4,
|
| 563 |
+
"grad_norm": 3.124176502227783,
|
| 564 |
+
"learning_rate": 1.6536000000000002e-05,
|
| 565 |
+
"loss": 0.0041,
|
| 566 |
+
"step": 12800
|
| 567 |
+
},
|
| 568 |
+
{
|
| 569 |
+
"epoch": 104.0,
|
| 570 |
+
"grad_norm": 1.91826593875885,
|
| 571 |
+
"learning_rate": 1.6447111111111113e-05,
|
| 572 |
+
"loss": 0.0043,
|
| 573 |
+
"step": 13000
|
| 574 |
+
},
|
| 575 |
+
{
|
| 576 |
+
"epoch": 104.0,
|
| 577 |
+
"eval_loss": 1.27247154712677,
|
| 578 |
+
"eval_runtime": 41.9766,
|
| 579 |
+
"eval_samples_per_second": 5.956,
|
| 580 |
+
"eval_steps_per_second": 1.501,
|
| 581 |
+
"eval_wer": 0.44776962168266515,
|
| 582 |
+
"step": 13000
|
| 583 |
+
},
|
| 584 |
+
{
|
| 585 |
+
"epoch": 105.6,
|
| 586 |
+
"grad_norm": 0.8713739514350891,
|
| 587 |
+
"learning_rate": 1.6358222222222223e-05,
|
| 588 |
+
"loss": 0.0032,
|
| 589 |
+
"step": 13200
|
| 590 |
+
},
|
| 591 |
+
{
|
| 592 |
+
"epoch": 107.2,
|
| 593 |
+
"grad_norm": 0.05066121742129326,
|
| 594 |
+
"learning_rate": 1.6269333333333334e-05,
|
| 595 |
+
"loss": 0.0042,
|
| 596 |
+
"step": 13400
|
| 597 |
+
},
|
| 598 |
+
{
|
| 599 |
+
"epoch": 108.8,
|
| 600 |
+
"grad_norm": 1.1562632322311401,
|
| 601 |
+
"learning_rate": 1.6180444444444444e-05,
|
| 602 |
+
"loss": 0.0032,
|
| 603 |
+
"step": 13600
|
| 604 |
+
},
|
| 605 |
+
{
|
| 606 |
+
"epoch": 110.4,
|
| 607 |
+
"grad_norm": 0.07406218349933624,
|
| 608 |
+
"learning_rate": 1.6091555555555555e-05,
|
| 609 |
+
"loss": 0.0037,
|
| 610 |
+
"step": 13800
|
| 611 |
+
},
|
| 612 |
+
{
|
| 613 |
+
"epoch": 112.0,
|
| 614 |
+
"grad_norm": 0.997751772403717,
|
| 615 |
+
"learning_rate": 1.600266666666667e-05,
|
| 616 |
+
"loss": 0.0039,
|
| 617 |
+
"step": 14000
|
| 618 |
+
},
|
| 619 |
+
{
|
| 620 |
+
"epoch": 112.0,
|
| 621 |
+
"eval_loss": 1.3331384658813477,
|
| 622 |
+
"eval_runtime": 43.8205,
|
| 623 |
+
"eval_samples_per_second": 5.705,
|
| 624 |
+
"eval_steps_per_second": 1.438,
|
| 625 |
+
"eval_wer": 0.5307735742518351,
|
| 626 |
+
"step": 14000
|
| 627 |
+
},
|
| 628 |
+
{
|
| 629 |
+
"epoch": 113.6,
|
| 630 |
+
"grad_norm": 0.19617311656475067,
|
| 631 |
+
"learning_rate": 1.591377777777778e-05,
|
| 632 |
+
"loss": 0.0043,
|
| 633 |
+
"step": 14200
|
| 634 |
+
},
|
| 635 |
+
{
|
| 636 |
+
"epoch": 115.2,
|
| 637 |
+
"grad_norm": 3.2143056392669678,
|
| 638 |
+
"learning_rate": 1.5825333333333336e-05,
|
| 639 |
+
"loss": 0.0051,
|
| 640 |
+
"step": 14400
|
| 641 |
+
},
|
| 642 |
+
{
|
| 643 |
+
"epoch": 116.8,
|
| 644 |
+
"grad_norm": 0.6758298873901367,
|
| 645 |
+
"learning_rate": 1.5736444444444446e-05,
|
| 646 |
+
"loss": 0.0033,
|
| 647 |
+
"step": 14600
|
| 648 |
+
},
|
| 649 |
+
{
|
| 650 |
+
"epoch": 118.4,
|
| 651 |
+
"grad_norm": 2.0771877765655518,
|
| 652 |
+
"learning_rate": 1.5647555555555557e-05,
|
| 653 |
+
"loss": 0.0037,
|
| 654 |
+
"step": 14800
|
| 655 |
+
},
|
| 656 |
+
{
|
| 657 |
+
"epoch": 120.0,
|
| 658 |
+
"grad_norm": 2.100074529647827,
|
| 659 |
+
"learning_rate": 1.5558666666666667e-05,
|
| 660 |
+
"loss": 0.0036,
|
| 661 |
+
"step": 15000
|
| 662 |
+
},
|
| 663 |
+
{
|
| 664 |
+
"epoch": 120.0,
|
| 665 |
+
"eval_loss": 1.3492019176483154,
|
| 666 |
+
"eval_runtime": 43.5776,
|
| 667 |
+
"eval_samples_per_second": 5.737,
|
| 668 |
+
"eval_steps_per_second": 1.446,
|
| 669 |
+
"eval_wer": 0.603613777526821,
|
| 670 |
+
"step": 15000
|
| 671 |
+
},
|
| 672 |
+
{
|
| 673 |
+
"epoch": 121.6,
|
| 674 |
+
"grad_norm": 0.3508625030517578,
|
| 675 |
+
"learning_rate": 1.5469777777777778e-05,
|
| 676 |
+
"loss": 0.0026,
|
| 677 |
+
"step": 15200
|
| 678 |
+
},
|
| 679 |
+
{
|
| 680 |
+
"epoch": 123.2,
|
| 681 |
+
"grad_norm": 0.0235364381223917,
|
| 682 |
+
"learning_rate": 1.5380888888888888e-05,
|
| 683 |
+
"loss": 0.0033,
|
| 684 |
+
"step": 15400
|
| 685 |
+
},
|
| 686 |
+
{
|
| 687 |
+
"epoch": 124.8,
|
| 688 |
+
"grad_norm": 1.745226263999939,
|
| 689 |
+
"learning_rate": 1.5292e-05,
|
| 690 |
+
"loss": 0.0024,
|
| 691 |
+
"step": 15600
|
| 692 |
+
},
|
| 693 |
+
{
|
| 694 |
+
"epoch": 126.4,
|
| 695 |
+
"grad_norm": 1.7244856357574463,
|
| 696 |
+
"learning_rate": 1.5203111111111112e-05,
|
| 697 |
+
"loss": 0.0027,
|
| 698 |
+
"step": 15800
|
| 699 |
+
},
|
| 700 |
+
{
|
| 701 |
+
"epoch": 128.0,
|
| 702 |
+
"grad_norm": 2.978816509246826,
|
| 703 |
+
"learning_rate": 1.5114222222222223e-05,
|
| 704 |
+
"loss": 0.0028,
|
| 705 |
+
"step": 16000
|
| 706 |
+
},
|
| 707 |
+
{
|
| 708 |
+
"epoch": 128.0,
|
| 709 |
+
"eval_loss": 1.3909789323806763,
|
| 710 |
+
"eval_runtime": 41.3413,
|
| 711 |
+
"eval_samples_per_second": 6.047,
|
| 712 |
+
"eval_steps_per_second": 1.524,
|
| 713 |
+
"eval_wer": 0.45511010728402035,
|
| 714 |
+
"step": 16000
|
| 715 |
+
},
|
| 716 |
+
{
|
| 717 |
+
"epoch": 129.6,
|
| 718 |
+
"grad_norm": 0.11152245104312897,
|
| 719 |
+
"learning_rate": 1.5025333333333333e-05,
|
| 720 |
+
"loss": 0.0023,
|
| 721 |
+
"step": 16200
|
| 722 |
+
},
|
| 723 |
+
{
|
| 724 |
+
"epoch": 131.2,
|
| 725 |
+
"grad_norm": 0.12898045778274536,
|
| 726 |
+
"learning_rate": 1.4936444444444447e-05,
|
| 727 |
+
"loss": 0.0029,
|
| 728 |
+
"step": 16400
|
| 729 |
+
},
|
| 730 |
+
{
|
| 731 |
+
"epoch": 132.8,
|
| 732 |
+
"grad_norm": 0.8840957880020142,
|
| 733 |
+
"learning_rate": 1.4847555555555558e-05,
|
| 734 |
+
"loss": 0.0026,
|
| 735 |
+
"step": 16600
|
| 736 |
+
},
|
| 737 |
+
{
|
| 738 |
+
"epoch": 134.4,
|
| 739 |
+
"grad_norm": 0.07675749063491821,
|
| 740 |
+
"learning_rate": 1.4758666666666668e-05,
|
| 741 |
+
"loss": 0.0027,
|
| 742 |
+
"step": 16800
|
| 743 |
+
},
|
| 744 |
+
{
|
| 745 |
+
"epoch": 136.0,
|
| 746 |
+
"grad_norm": 0.5362949371337891,
|
| 747 |
+
"learning_rate": 1.4669777777777779e-05,
|
| 748 |
+
"loss": 0.0034,
|
| 749 |
+
"step": 17000
|
| 750 |
+
},
|
| 751 |
+
{
|
| 752 |
+
"epoch": 136.0,
|
| 753 |
+
"eval_loss": 1.436603307723999,
|
| 754 |
+
"eval_runtime": 41.9502,
|
| 755 |
+
"eval_samples_per_second": 5.959,
|
| 756 |
+
"eval_steps_per_second": 1.502,
|
| 757 |
+
"eval_wer": 0.4906832298136646,
|
| 758 |
+
"step": 17000
|
| 759 |
+
},
|
| 760 |
+
{
|
| 761 |
+
"epoch": 137.6,
|
| 762 |
+
"grad_norm": 1.4152299165725708,
|
| 763 |
+
"learning_rate": 1.458088888888889e-05,
|
| 764 |
+
"loss": 0.0024,
|
| 765 |
+
"step": 17200
|
| 766 |
+
},
|
| 767 |
+
{
|
| 768 |
+
"epoch": 139.2,
|
| 769 |
+
"grad_norm": 0.1923665702342987,
|
| 770 |
+
"learning_rate": 1.4492000000000001e-05,
|
| 771 |
+
"loss": 0.0023,
|
| 772 |
+
"step": 17400
|
| 773 |
+
},
|
| 774 |
+
{
|
| 775 |
+
"epoch": 140.8,
|
| 776 |
+
"grad_norm": 0.9098421931266785,
|
| 777 |
+
"learning_rate": 1.4403111111111112e-05,
|
| 778 |
+
"loss": 0.0018,
|
| 779 |
+
"step": 17600
|
| 780 |
+
},
|
| 781 |
+
{
|
| 782 |
+
"epoch": 142.4,
|
| 783 |
+
"grad_norm": 0.06011312082409859,
|
| 784 |
+
"learning_rate": 1.4314222222222222e-05,
|
| 785 |
+
"loss": 0.0023,
|
| 786 |
+
"step": 17800
|
| 787 |
+
},
|
| 788 |
+
{
|
| 789 |
+
"epoch": 144.0,
|
| 790 |
+
"grad_norm": 0.4919435381889343,
|
| 791 |
+
"learning_rate": 1.4225333333333336e-05,
|
| 792 |
+
"loss": 0.0024,
|
| 793 |
+
"step": 18000
|
| 794 |
+
},
|
| 795 |
+
{
|
| 796 |
+
"epoch": 144.0,
|
| 797 |
+
"eval_loss": 1.3797271251678467,
|
| 798 |
+
"eval_runtime": 46.4998,
|
| 799 |
+
"eval_samples_per_second": 5.376,
|
| 800 |
+
"eval_steps_per_second": 1.355,
|
| 801 |
+
"eval_wer": 0.6126482213438735,
|
| 802 |
+
"step": 18000
|
| 803 |
+
},
|
| 804 |
+
{
|
| 805 |
+
"epoch": 145.6,
|
| 806 |
+
"grad_norm": 1.19382905960083,
|
| 807 |
+
"learning_rate": 1.4136444444444446e-05,
|
| 808 |
+
"loss": 0.0026,
|
| 809 |
+
"step": 18200
|
| 810 |
+
},
|
| 811 |
+
{
|
| 812 |
+
"epoch": 147.2,
|
| 813 |
+
"grad_norm": 0.1949184685945511,
|
| 814 |
+
"learning_rate": 1.4047555555555557e-05,
|
| 815 |
+
"loss": 0.0017,
|
| 816 |
+
"step": 18400
|
| 817 |
+
},
|
| 818 |
+
{
|
| 819 |
+
"epoch": 148.8,
|
| 820 |
+
"grad_norm": 0.030353059992194176,
|
| 821 |
+
"learning_rate": 1.3959111111111112e-05,
|
| 822 |
+
"loss": 0.002,
|
| 823 |
+
"step": 18600
|
| 824 |
+
},
|
| 825 |
+
{
|
| 826 |
+
"epoch": 150.4,
|
| 827 |
+
"grad_norm": 0.8148756623268127,
|
| 828 |
+
"learning_rate": 1.3870222222222223e-05,
|
| 829 |
+
"loss": 0.0016,
|
| 830 |
+
"step": 18800
|
| 831 |
+
},
|
| 832 |
+
{
|
| 833 |
+
"epoch": 152.0,
|
| 834 |
+
"grad_norm": 0.1468518227338791,
|
| 835 |
+
"learning_rate": 1.3781333333333335e-05,
|
| 836 |
+
"loss": 0.0019,
|
| 837 |
+
"step": 19000
|
| 838 |
+
},
|
| 839 |
+
{
|
| 840 |
+
"epoch": 152.0,
|
| 841 |
+
"eval_loss": 1.4368208646774292,
|
| 842 |
+
"eval_runtime": 42.0557,
|
| 843 |
+
"eval_samples_per_second": 5.945,
|
| 844 |
+
"eval_steps_per_second": 1.498,
|
| 845 |
+
"eval_wer": 0.5228684359119141,
|
| 846 |
+
"step": 19000
|
| 847 |
+
},
|
| 848 |
+
{
|
| 849 |
+
"epoch": 153.6,
|
| 850 |
+
"grad_norm": 3.330853223800659,
|
| 851 |
+
"learning_rate": 1.3692888888888892e-05,
|
| 852 |
+
"loss": 0.0024,
|
| 853 |
+
"step": 19200
|
| 854 |
+
},
|
| 855 |
+
{
|
| 856 |
+
"epoch": 155.2,
|
| 857 |
+
"grad_norm": 2.411620616912842,
|
| 858 |
+
"learning_rate": 1.3604000000000002e-05,
|
| 859 |
+
"loss": 0.0018,
|
| 860 |
+
"step": 19400
|
| 861 |
+
},
|
| 862 |
+
{
|
| 863 |
+
"epoch": 156.8,
|
| 864 |
+
"grad_norm": 1.7824522256851196,
|
| 865 |
+
"learning_rate": 1.3515111111111113e-05,
|
| 866 |
+
"loss": 0.0019,
|
| 867 |
+
"step": 19600
|
| 868 |
+
},
|
| 869 |
+
{
|
| 870 |
+
"epoch": 158.4,
|
| 871 |
+
"grad_norm": 0.05631573125720024,
|
| 872 |
+
"learning_rate": 1.3426222222222223e-05,
|
| 873 |
+
"loss": 0.0018,
|
| 874 |
+
"step": 19800
|
| 875 |
+
},
|
| 876 |
+
{
|
| 877 |
+
"epoch": 160.0,
|
| 878 |
+
"grad_norm": 0.20461511611938477,
|
| 879 |
+
"learning_rate": 1.3337333333333335e-05,
|
| 880 |
+
"loss": 0.0017,
|
| 881 |
+
"step": 20000
|
| 882 |
+
},
|
| 883 |
+
{
|
| 884 |
+
"epoch": 160.0,
|
| 885 |
+
"eval_loss": 1.4499236345291138,
|
| 886 |
+
"eval_runtime": 43.1323,
|
| 887 |
+
"eval_samples_per_second": 5.796,
|
| 888 |
+
"eval_steps_per_second": 1.461,
|
| 889 |
+
"eval_wer": 0.5979672501411631,
|
| 890 |
+
"step": 20000
|
| 891 |
+
},
|
| 892 |
+
{
|
| 893 |
+
"epoch": 161.6,
|
| 894 |
+
"grad_norm": 0.01897699199616909,
|
| 895 |
+
"learning_rate": 1.3248444444444446e-05,
|
| 896 |
+
"loss": 0.0028,
|
| 897 |
+
"step": 20200
|
| 898 |
+
},
|
| 899 |
+
{
|
| 900 |
+
"epoch": 163.2,
|
| 901 |
+
"grad_norm": 0.9474685192108154,
|
| 902 |
+
"learning_rate": 1.3160000000000001e-05,
|
| 903 |
+
"loss": 0.0027,
|
| 904 |
+
"step": 20400
|
| 905 |
+
},
|
| 906 |
+
{
|
| 907 |
+
"epoch": 164.8,
|
| 908 |
+
"grad_norm": 0.10914404690265656,
|
| 909 |
+
"learning_rate": 1.3071111111111112e-05,
|
| 910 |
+
"loss": 0.0026,
|
| 911 |
+
"step": 20600
|
| 912 |
+
},
|
| 913 |
+
{
|
| 914 |
+
"epoch": 166.4,
|
| 915 |
+
"grad_norm": 0.6951639652252197,
|
| 916 |
+
"learning_rate": 1.2982222222222222e-05,
|
| 917 |
+
"loss": 0.0015,
|
| 918 |
+
"step": 20800
|
| 919 |
+
},
|
| 920 |
+
{
|
| 921 |
+
"epoch": 168.0,
|
| 922 |
+
"grad_norm": 0.591296911239624,
|
| 923 |
+
"learning_rate": 1.2893333333333336e-05,
|
| 924 |
+
"loss": 0.0014,
|
| 925 |
+
"step": 21000
|
| 926 |
+
},
|
| 927 |
+
{
|
| 928 |
+
"epoch": 168.0,
|
| 929 |
+
"eval_loss": 1.4369308948516846,
|
| 930 |
+
"eval_runtime": 43.7269,
|
| 931 |
+
"eval_samples_per_second": 5.717,
|
| 932 |
+
"eval_steps_per_second": 1.441,
|
| 933 |
+
"eval_wer": 0.5940146809712027,
|
| 934 |
+
"step": 21000
|
| 935 |
+
},
|
| 936 |
+
{
|
| 937 |
+
"epoch": 169.6,
|
| 938 |
+
"grad_norm": 0.07852908968925476,
|
| 939 |
+
"learning_rate": 1.2804444444444446e-05,
|
| 940 |
+
"loss": 0.0018,
|
| 941 |
+
"step": 21200
|
| 942 |
+
},
|
| 943 |
+
{
|
| 944 |
+
"epoch": 171.2,
|
| 945 |
+
"grad_norm": 0.4466964304447174,
|
| 946 |
+
"learning_rate": 1.2715555555555557e-05,
|
| 947 |
+
"loss": 0.0016,
|
| 948 |
+
"step": 21400
|
| 949 |
+
},
|
| 950 |
+
{
|
| 951 |
+
"epoch": 172.8,
|
| 952 |
+
"grad_norm": 0.6805703043937683,
|
| 953 |
+
"learning_rate": 1.2626666666666667e-05,
|
| 954 |
+
"loss": 0.001,
|
| 955 |
+
"step": 21600
|
| 956 |
+
},
|
| 957 |
+
{
|
| 958 |
+
"epoch": 174.4,
|
| 959 |
+
"grad_norm": 0.14335931837558746,
|
| 960 |
+
"learning_rate": 1.253777777777778e-05,
|
| 961 |
+
"loss": 0.001,
|
| 962 |
+
"step": 21800
|
| 963 |
+
},
|
| 964 |
+
{
|
| 965 |
+
"epoch": 176.0,
|
| 966 |
+
"grad_norm": 1.0165632963180542,
|
| 967 |
+
"learning_rate": 1.244888888888889e-05,
|
| 968 |
+
"loss": 0.0015,
|
| 969 |
+
"step": 22000
|
| 970 |
+
},
|
| 971 |
+
{
|
| 972 |
+
"epoch": 176.0,
|
| 973 |
+
"eval_loss": 1.4614824056625366,
|
| 974 |
+
"eval_runtime": 42.489,
|
| 975 |
+
"eval_samples_per_second": 5.884,
|
| 976 |
+
"eval_steps_per_second": 1.483,
|
| 977 |
+
"eval_wer": 0.5307735742518351,
|
| 978 |
+
"step": 22000
|
| 979 |
+
},
|
| 980 |
+
{
|
| 981 |
+
"epoch": 177.6,
|
| 982 |
+
"grad_norm": 0.4234725832939148,
|
| 983 |
+
"learning_rate": 1.236e-05,
|
| 984 |
+
"loss": 0.0019,
|
| 985 |
+
"step": 22200
|
| 986 |
+
},
|
| 987 |
+
{
|
| 988 |
+
"epoch": 179.2,
|
| 989 |
+
"grad_norm": 4.320431232452393,
|
| 990 |
+
"learning_rate": 1.227111111111111e-05,
|
| 991 |
+
"loss": 0.0027,
|
| 992 |
+
"step": 22400
|
| 993 |
+
},
|
| 994 |
+
{
|
| 995 |
+
"epoch": 180.8,
|
| 996 |
+
"grad_norm": 0.04678142070770264,
|
| 997 |
+
"learning_rate": 1.2182222222222225e-05,
|
| 998 |
+
"loss": 0.0018,
|
| 999 |
+
"step": 22600
|
| 1000 |
+
},
|
| 1001 |
+
{
|
| 1002 |
+
"epoch": 182.4,
|
| 1003 |
+
"grad_norm": 0.19874203205108643,
|
| 1004 |
+
"learning_rate": 1.2093333333333335e-05,
|
| 1005 |
+
"loss": 0.0013,
|
| 1006 |
+
"step": 22800
|
| 1007 |
+
},
|
| 1008 |
+
{
|
| 1009 |
+
"epoch": 184.0,
|
| 1010 |
+
"grad_norm": 0.636381208896637,
|
| 1011 |
+
"learning_rate": 1.2004444444444445e-05,
|
| 1012 |
+
"loss": 0.001,
|
| 1013 |
+
"step": 23000
|
| 1014 |
+
},
|
| 1015 |
+
{
|
| 1016 |
+
"epoch": 184.0,
|
| 1017 |
+
"eval_loss": 1.445263147354126,
|
| 1018 |
+
"eval_runtime": 41.9644,
|
| 1019 |
+
"eval_samples_per_second": 5.957,
|
| 1020 |
+
"eval_steps_per_second": 1.501,
|
| 1021 |
+
"eval_wer": 0.49858836815358554,
|
| 1022 |
+
"step": 23000
|
| 1023 |
+
},
|
| 1024 |
+
{
|
| 1025 |
+
"epoch": 185.6,
|
| 1026 |
+
"grad_norm": 0.021843163296580315,
|
| 1027 |
+
"learning_rate": 1.1915555555555556e-05,
|
| 1028 |
+
"loss": 0.0017,
|
| 1029 |
+
"step": 23200
|
| 1030 |
+
},
|
| 1031 |
+
{
|
| 1032 |
+
"epoch": 187.2,
|
| 1033 |
+
"grad_norm": 0.0671222060918808,
|
| 1034 |
+
"learning_rate": 1.1826666666666668e-05,
|
| 1035 |
+
"loss": 0.0012,
|
| 1036 |
+
"step": 23400
|
| 1037 |
+
},
|
| 1038 |
+
{
|
| 1039 |
+
"epoch": 188.8,
|
| 1040 |
+
"grad_norm": 0.04560863971710205,
|
| 1041 |
+
"learning_rate": 1.1737777777777779e-05,
|
| 1042 |
+
"loss": 0.0019,
|
| 1043 |
+
"step": 23600
|
| 1044 |
+
},
|
| 1045 |
+
{
|
| 1046 |
+
"epoch": 190.4,
|
| 1047 |
+
"grad_norm": 0.17737437784671783,
|
| 1048 |
+
"learning_rate": 1.1648888888888889e-05,
|
| 1049 |
+
"loss": 0.0008,
|
| 1050 |
+
"step": 23800
|
| 1051 |
+
},
|
| 1052 |
+
{
|
| 1053 |
+
"epoch": 192.0,
|
| 1054 |
+
"grad_norm": 0.05606027692556381,
|
| 1055 |
+
"learning_rate": 1.156e-05,
|
| 1056 |
+
"loss": 0.0009,
|
| 1057 |
+
"step": 24000
|
| 1058 |
+
},
|
| 1059 |
+
{
|
| 1060 |
+
"epoch": 192.0,
|
| 1061 |
+
"eval_loss": 1.4906095266342163,
|
| 1062 |
+
"eval_runtime": 43.2268,
|
| 1063 |
+
"eval_samples_per_second": 5.783,
|
| 1064 |
+
"eval_steps_per_second": 1.457,
|
| 1065 |
+
"eval_wer": 0.5471485036702428,
|
| 1066 |
+
"step": 24000
|
| 1067 |
+
},
|
| 1068 |
+
{
|
| 1069 |
+
"epoch": 193.6,
|
| 1070 |
+
"grad_norm": 2.6839418411254883,
|
| 1071 |
+
"learning_rate": 1.1471111111111113e-05,
|
| 1072 |
+
"loss": 0.0016,
|
| 1073 |
+
"step": 24200
|
| 1074 |
+
},
|
| 1075 |
+
{
|
| 1076 |
+
"epoch": 195.2,
|
| 1077 |
+
"grad_norm": 0.05472835153341293,
|
| 1078 |
+
"learning_rate": 1.1382222222222224e-05,
|
| 1079 |
+
"loss": 0.0014,
|
| 1080 |
+
"step": 24400
|
| 1081 |
+
},
|
| 1082 |
+
{
|
| 1083 |
+
"epoch": 196.8,
|
| 1084 |
+
"grad_norm": 0.4221028685569763,
|
| 1085 |
+
"learning_rate": 1.1293333333333334e-05,
|
| 1086 |
+
"loss": 0.001,
|
| 1087 |
+
"step": 24600
|
| 1088 |
+
},
|
| 1089 |
+
{
|
| 1090 |
+
"epoch": 198.4,
|
| 1091 |
+
"grad_norm": 0.025390515103936195,
|
| 1092 |
+
"learning_rate": 1.120488888888889e-05,
|
| 1093 |
+
"loss": 0.0017,
|
| 1094 |
+
"step": 24800
|
| 1095 |
+
},
|
| 1096 |
+
{
|
| 1097 |
+
"epoch": 200.0,
|
| 1098 |
+
"grad_norm": 0.02648915909230709,
|
| 1099 |
+
"learning_rate": 1.1116e-05,
|
| 1100 |
+
"loss": 0.0007,
|
| 1101 |
+
"step": 25000
|
| 1102 |
+
},
|
| 1103 |
+
{
|
| 1104 |
+
"epoch": 200.0,
|
| 1105 |
+
"eval_loss": 1.4574248790740967,
|
| 1106 |
+
"eval_runtime": 42.0773,
|
| 1107 |
+
"eval_samples_per_second": 5.941,
|
| 1108 |
+
"eval_steps_per_second": 1.497,
|
| 1109 |
+
"eval_wer": 0.49407114624505927,
|
| 1110 |
+
"step": 25000
|
| 1111 |
+
},
|
| 1112 |
+
{
|
| 1113 |
+
"epoch": 201.6,
|
| 1114 |
+
"grad_norm": 0.05941193923354149,
|
| 1115 |
+
"learning_rate": 1.1027111111111112e-05,
|
| 1116 |
+
"loss": 0.0011,
|
| 1117 |
+
"step": 25200
|
| 1118 |
+
},
|
| 1119 |
+
{
|
| 1120 |
+
"epoch": 203.2,
|
| 1121 |
+
"grad_norm": 0.02862560749053955,
|
| 1122 |
+
"learning_rate": 1.0938222222222223e-05,
|
| 1123 |
+
"loss": 0.0012,
|
| 1124 |
+
"step": 25400
|
| 1125 |
+
},
|
| 1126 |
+
{
|
| 1127 |
+
"epoch": 204.8,
|
| 1128 |
+
"grad_norm": 0.024127991870045662,
|
| 1129 |
+
"learning_rate": 1.0849333333333335e-05,
|
| 1130 |
+
"loss": 0.0011,
|
| 1131 |
+
"step": 25600
|
| 1132 |
+
},
|
| 1133 |
+
{
|
| 1134 |
+
"epoch": 206.4,
|
| 1135 |
+
"grad_norm": 0.08150752633810043,
|
| 1136 |
+
"learning_rate": 1.0760444444444445e-05,
|
| 1137 |
+
"loss": 0.0008,
|
| 1138 |
+
"step": 25800
|
| 1139 |
+
},
|
| 1140 |
+
{
|
| 1141 |
+
"epoch": 208.0,
|
| 1142 |
+
"grad_norm": 0.05071726813912392,
|
| 1143 |
+
"learning_rate": 1.0671555555555557e-05,
|
| 1144 |
+
"loss": 0.0011,
|
| 1145 |
+
"step": 26000
|
| 1146 |
+
},
|
| 1147 |
+
{
|
| 1148 |
+
"epoch": 208.0,
|
| 1149 |
+
"eval_loss": 1.4995194673538208,
|
| 1150 |
+
"eval_runtime": 41.3368,
|
| 1151 |
+
"eval_samples_per_second": 6.048,
|
| 1152 |
+
"eval_steps_per_second": 1.524,
|
| 1153 |
+
"eval_wer": 0.4647092038396386,
|
| 1154 |
+
"step": 26000
|
| 1155 |
+
},
|
| 1156 |
+
{
|
| 1157 |
+
"epoch": 209.6,
|
| 1158 |
+
"grad_norm": 8.064842224121094,
|
| 1159 |
+
"learning_rate": 1.0582666666666668e-05,
|
| 1160 |
+
"loss": 0.001,
|
| 1161 |
+
"step": 26200
|
| 1162 |
+
},
|
| 1163 |
+
{
|
| 1164 |
+
"epoch": 211.2,
|
| 1165 |
+
"grad_norm": 0.02605404146015644,
|
| 1166 |
+
"learning_rate": 1.0493777777777778e-05,
|
| 1167 |
+
"loss": 0.001,
|
| 1168 |
+
"step": 26400
|
| 1169 |
+
},
|
| 1170 |
+
{
|
| 1171 |
+
"epoch": 212.8,
|
| 1172 |
+
"grad_norm": 0.06135524809360504,
|
| 1173 |
+
"learning_rate": 1.0404888888888889e-05,
|
| 1174 |
+
"loss": 0.0011,
|
| 1175 |
+
"step": 26600
|
| 1176 |
+
},
|
| 1177 |
+
{
|
| 1178 |
+
"epoch": 214.4,
|
| 1179 |
+
"grad_norm": 0.014893967658281326,
|
| 1180 |
+
"learning_rate": 1.0316e-05,
|
| 1181 |
+
"loss": 0.0008,
|
| 1182 |
+
"step": 26800
|
| 1183 |
+
},
|
| 1184 |
+
{
|
| 1185 |
+
"epoch": 216.0,
|
| 1186 |
+
"grad_norm": 0.009221619926393032,
|
| 1187 |
+
"learning_rate": 1.0227111111111111e-05,
|
| 1188 |
+
"loss": 0.0007,
|
| 1189 |
+
"step": 27000
|
| 1190 |
+
},
|
| 1191 |
+
{
|
| 1192 |
+
"epoch": 216.0,
|
| 1193 |
+
"eval_loss": 1.5195382833480835,
|
| 1194 |
+
"eval_runtime": 42.2229,
|
| 1195 |
+
"eval_samples_per_second": 5.921,
|
| 1196 |
+
"eval_steps_per_second": 1.492,
|
| 1197 |
+
"eval_wer": 0.515527950310559,
|
| 1198 |
+
"step": 27000
|
| 1199 |
+
},
|
| 1200 |
+
{
|
| 1201 |
+
"epoch": 217.6,
|
| 1202 |
+
"grad_norm": 0.3534085154533386,
|
| 1203 |
+
"learning_rate": 1.0138222222222223e-05,
|
| 1204 |
+
"loss": 0.0012,
|
| 1205 |
+
"step": 27200
|
| 1206 |
+
},
|
| 1207 |
+
{
|
| 1208 |
+
"epoch": 219.2,
|
| 1209 |
+
"grad_norm": 0.09608615189790726,
|
| 1210 |
+
"learning_rate": 1.0049333333333334e-05,
|
| 1211 |
+
"loss": 0.0007,
|
| 1212 |
+
"step": 27400
|
| 1213 |
+
},
|
| 1214 |
+
{
|
| 1215 |
+
"epoch": 220.8,
|
| 1216 |
+
"grad_norm": 0.03040822595357895,
|
| 1217 |
+
"learning_rate": 9.960444444444444e-06,
|
| 1218 |
+
"loss": 0.0008,
|
| 1219 |
+
"step": 27600
|
| 1220 |
+
},
|
| 1221 |
+
{
|
| 1222 |
+
"epoch": 222.4,
|
| 1223 |
+
"grad_norm": 0.030991466715931892,
|
| 1224 |
+
"learning_rate": 9.871555555555557e-06,
|
| 1225 |
+
"loss": 0.0008,
|
| 1226 |
+
"step": 27800
|
| 1227 |
+
},
|
| 1228 |
+
{
|
| 1229 |
+
"epoch": 224.0,
|
| 1230 |
+
"grad_norm": 0.270130455493927,
|
| 1231 |
+
"learning_rate": 9.783111111111112e-06,
|
| 1232 |
+
"loss": 0.0011,
|
| 1233 |
+
"step": 28000
|
| 1234 |
+
},
|
| 1235 |
+
{
|
| 1236 |
+
"epoch": 224.0,
|
| 1237 |
+
"eval_loss": 1.4928404092788696,
|
| 1238 |
+
"eval_runtime": 42.9045,
|
| 1239 |
+
"eval_samples_per_second": 5.827,
|
| 1240 |
+
"eval_steps_per_second": 1.468,
|
| 1241 |
+
"eval_wer": 0.5482778091473743,
|
| 1242 |
+
"step": 28000
|
| 1243 |
+
},
|
| 1244 |
+
{
|
| 1245 |
+
"epoch": 225.6,
|
| 1246 |
+
"grad_norm": 0.059342194348573685,
|
| 1247 |
+
"learning_rate": 9.694222222222222e-06,
|
| 1248 |
+
"loss": 0.0008,
|
| 1249 |
+
"step": 28200
|
| 1250 |
+
},
|
| 1251 |
+
{
|
| 1252 |
+
"epoch": 227.2,
|
| 1253 |
+
"grad_norm": 0.09850233793258667,
|
| 1254 |
+
"learning_rate": 9.605333333333334e-06,
|
| 1255 |
+
"loss": 0.001,
|
| 1256 |
+
"step": 28400
|
| 1257 |
+
},
|
| 1258 |
+
{
|
| 1259 |
+
"epoch": 228.8,
|
| 1260 |
+
"grad_norm": 0.012932940386235714,
|
| 1261 |
+
"learning_rate": 9.516444444444445e-06,
|
| 1262 |
+
"loss": 0.0007,
|
| 1263 |
+
"step": 28600
|
| 1264 |
+
},
|
| 1265 |
+
{
|
| 1266 |
+
"epoch": 230.4,
|
| 1267 |
+
"grad_norm": 0.4113960862159729,
|
| 1268 |
+
"learning_rate": 9.427555555555557e-06,
|
| 1269 |
+
"loss": 0.0008,
|
| 1270 |
+
"step": 28800
|
| 1271 |
+
},
|
| 1272 |
+
{
|
| 1273 |
+
"epoch": 232.0,
|
| 1274 |
+
"grad_norm": 11.596172332763672,
|
| 1275 |
+
"learning_rate": 9.338666666666667e-06,
|
| 1276 |
+
"loss": 0.0011,
|
| 1277 |
+
"step": 29000
|
| 1278 |
+
},
|
| 1279 |
+
{
|
| 1280 |
+
"epoch": 232.0,
|
| 1281 |
+
"eval_loss": 1.5243357419967651,
|
| 1282 |
+
"eval_runtime": 42.099,
|
| 1283 |
+
"eval_samples_per_second": 5.938,
|
| 1284 |
+
"eval_steps_per_second": 1.496,
|
| 1285 |
+
"eval_wer": 0.5143986448334275,
|
| 1286 |
+
"step": 29000
|
| 1287 |
+
},
|
| 1288 |
+
{
|
| 1289 |
+
"epoch": 233.6,
|
| 1290 |
+
"grad_norm": 0.14402435719966888,
|
| 1291 |
+
"learning_rate": 9.249777777777778e-06,
|
| 1292 |
+
"loss": 0.0005,
|
| 1293 |
+
"step": 29200
|
| 1294 |
+
},
|
| 1295 |
+
{
|
| 1296 |
+
"epoch": 235.2,
|
| 1297 |
+
"grad_norm": 0.037319615483284,
|
| 1298 |
+
"learning_rate": 9.160888888888888e-06,
|
| 1299 |
+
"loss": 0.0006,
|
| 1300 |
+
"step": 29400
|
| 1301 |
+
},
|
| 1302 |
+
{
|
| 1303 |
+
"epoch": 236.8,
|
| 1304 |
+
"grad_norm": 0.05602966248989105,
|
| 1305 |
+
"learning_rate": 9.072e-06,
|
| 1306 |
+
"loss": 0.0009,
|
| 1307 |
+
"step": 29600
|
| 1308 |
+
},
|
| 1309 |
+
{
|
| 1310 |
+
"epoch": 238.4,
|
| 1311 |
+
"grad_norm": 0.018490877002477646,
|
| 1312 |
+
"learning_rate": 8.983111111111111e-06,
|
| 1313 |
+
"loss": 0.0007,
|
| 1314 |
+
"step": 29800
|
| 1315 |
+
},
|
| 1316 |
+
{
|
| 1317 |
+
"epoch": 240.0,
|
| 1318 |
+
"grad_norm": 3.7446510791778564,
|
| 1319 |
+
"learning_rate": 8.894222222222223e-06,
|
| 1320 |
+
"loss": 0.0007,
|
| 1321 |
+
"step": 30000
|
| 1322 |
+
},
|
| 1323 |
+
{
|
| 1324 |
+
"epoch": 240.0,
|
| 1325 |
+
"eval_loss": 1.5804709196090698,
|
| 1326 |
+
"eval_runtime": 41.044,
|
| 1327 |
+
"eval_samples_per_second": 6.091,
|
| 1328 |
+
"eval_steps_per_second": 1.535,
|
| 1329 |
+
"eval_wer": 0.48842461885940147,
|
| 1330 |
+
"step": 30000
|
| 1331 |
+
},
|
| 1332 |
+
{
|
| 1333 |
+
"epoch": 241.6,
|
| 1334 |
+
"grad_norm": 0.09505221247673035,
|
| 1335 |
+
"learning_rate": 8.805333333333334e-06,
|
| 1336 |
+
"loss": 0.0005,
|
| 1337 |
+
"step": 30200
|
| 1338 |
+
},
|
| 1339 |
+
{
|
| 1340 |
+
"epoch": 243.2,
|
| 1341 |
+
"grad_norm": 0.031215248629450798,
|
| 1342 |
+
"learning_rate": 8.716444444444446e-06,
|
| 1343 |
+
"loss": 0.0007,
|
| 1344 |
+
"step": 30400
|
| 1345 |
+
},
|
| 1346 |
+
{
|
| 1347 |
+
"epoch": 244.8,
|
| 1348 |
+
"grad_norm": 0.3624882996082306,
|
| 1349 |
+
"learning_rate": 8.627555555555556e-06,
|
| 1350 |
+
"loss": 0.0004,
|
| 1351 |
+
"step": 30600
|
| 1352 |
+
},
|
| 1353 |
+
{
|
| 1354 |
+
"epoch": 246.4,
|
| 1355 |
+
"grad_norm": 0.025587618350982666,
|
| 1356 |
+
"learning_rate": 8.538666666666667e-06,
|
| 1357 |
+
"loss": 0.0007,
|
| 1358 |
+
"step": 30800
|
| 1359 |
+
},
|
| 1360 |
+
{
|
| 1361 |
+
"epoch": 248.0,
|
| 1362 |
+
"grad_norm": 0.8972709774971008,
|
| 1363 |
+
"learning_rate": 8.450222222222224e-06,
|
| 1364 |
+
"loss": 0.0005,
|
| 1365 |
+
"step": 31000
|
| 1366 |
+
},
|
| 1367 |
+
{
|
| 1368 |
+
"epoch": 248.0,
|
| 1369 |
+
"eval_loss": 1.5293784141540527,
|
| 1370 |
+
"eval_runtime": 42.2688,
|
| 1371 |
+
"eval_samples_per_second": 5.915,
|
| 1372 |
+
"eval_steps_per_second": 1.49,
|
| 1373 |
+
"eval_wer": 0.5115753811405985,
|
| 1374 |
+
"step": 31000
|
| 1375 |
+
},
|
| 1376 |
+
{
|
| 1377 |
+
"epoch": 249.6,
|
| 1378 |
+
"grad_norm": 0.02251521684229374,
|
| 1379 |
+
"learning_rate": 8.361333333333334e-06,
|
| 1380 |
+
"loss": 0.0007,
|
| 1381 |
+
"step": 31200
|
| 1382 |
+
},
|
| 1383 |
+
{
|
| 1384 |
+
"epoch": 251.2,
|
| 1385 |
+
"grad_norm": 0.007881904020905495,
|
| 1386 |
+
"learning_rate": 8.272444444444445e-06,
|
| 1387 |
+
"loss": 0.0004,
|
| 1388 |
+
"step": 31400
|
| 1389 |
+
},
|
| 1390 |
+
{
|
| 1391 |
+
"epoch": 252.8,
|
| 1392 |
+
"grad_norm": 0.0084000863134861,
|
| 1393 |
+
"learning_rate": 8.183555555555555e-06,
|
| 1394 |
+
"loss": 0.0005,
|
| 1395 |
+
"step": 31600
|
| 1396 |
+
},
|
| 1397 |
+
{
|
| 1398 |
+
"epoch": 254.4,
|
| 1399 |
+
"grad_norm": 0.44422510266304016,
|
| 1400 |
+
"learning_rate": 8.094666666666667e-06,
|
| 1401 |
+
"loss": 0.0004,
|
| 1402 |
+
"step": 31800
|
| 1403 |
+
},
|
| 1404 |
+
{
|
| 1405 |
+
"epoch": 256.0,
|
| 1406 |
+
"grad_norm": 0.024682149291038513,
|
| 1407 |
+
"learning_rate": 8.00577777777778e-06,
|
| 1408 |
+
"loss": 0.0005,
|
| 1409 |
+
"step": 32000
|
| 1410 |
+
},
|
| 1411 |
+
{
|
| 1412 |
+
"epoch": 256.0,
|
| 1413 |
+
"eval_loss": 1.5940375328063965,
|
| 1414 |
+
"eval_runtime": 42.3505,
|
| 1415 |
+
"eval_samples_per_second": 5.903,
|
| 1416 |
+
"eval_steps_per_second": 1.488,
|
| 1417 |
+
"eval_wer": 0.49745906267645396,
|
| 1418 |
+
"step": 32000
|
| 1419 |
+
},
|
| 1420 |
+
{
|
| 1421 |
+
"epoch": 257.6,
|
| 1422 |
+
"grad_norm": 0.019801704213023186,
|
| 1423 |
+
"learning_rate": 7.91688888888889e-06,
|
| 1424 |
+
"loss": 0.0005,
|
| 1425 |
+
"step": 32200
|
| 1426 |
+
},
|
| 1427 |
+
{
|
| 1428 |
+
"epoch": 259.2,
|
| 1429 |
+
"grad_norm": 0.006728316657245159,
|
| 1430 |
+
"learning_rate": 7.828000000000002e-06,
|
| 1431 |
+
"loss": 0.0003,
|
| 1432 |
+
"step": 32400
|
| 1433 |
+
},
|
| 1434 |
+
{
|
| 1435 |
+
"epoch": 260.8,
|
| 1436 |
+
"grad_norm": 0.1475154608488083,
|
| 1437 |
+
"learning_rate": 7.739111111111112e-06,
|
| 1438 |
+
"loss": 0.0005,
|
| 1439 |
+
"step": 32600
|
| 1440 |
+
},
|
| 1441 |
+
{
|
| 1442 |
+
"epoch": 262.4,
|
| 1443 |
+
"grad_norm": 0.038090407848358154,
|
| 1444 |
+
"learning_rate": 7.650222222222223e-06,
|
| 1445 |
+
"loss": 0.0005,
|
| 1446 |
+
"step": 32800
|
| 1447 |
+
},
|
| 1448 |
+
{
|
| 1449 |
+
"epoch": 264.0,
|
| 1450 |
+
"grad_norm": 1.3189762830734253,
|
| 1451 |
+
"learning_rate": 7.561333333333334e-06,
|
| 1452 |
+
"loss": 0.0003,
|
| 1453 |
+
"step": 33000
|
| 1454 |
+
},
|
| 1455 |
+
{
|
| 1456 |
+
"epoch": 264.0,
|
| 1457 |
+
"eval_loss": 1.5760223865509033,
|
| 1458 |
+
"eval_runtime": 41.6476,
|
| 1459 |
+
"eval_samples_per_second": 6.003,
|
| 1460 |
+
"eval_steps_per_second": 1.513,
|
| 1461 |
+
"eval_wer": 0.5002823263692829,
|
| 1462 |
+
"step": 33000
|
| 1463 |
+
},
|
| 1464 |
+
{
|
| 1465 |
+
"epoch": 265.6,
|
| 1466 |
+
"grad_norm": 0.005491426680237055,
|
| 1467 |
+
"learning_rate": 7.4728888888888895e-06,
|
| 1468 |
+
"loss": 0.0007,
|
| 1469 |
+
"step": 33200
|
| 1470 |
+
},
|
| 1471 |
+
{
|
| 1472 |
+
"epoch": 267.2,
|
| 1473 |
+
"grad_norm": 0.010416628792881966,
|
| 1474 |
+
"learning_rate": 7.384e-06,
|
| 1475 |
+
"loss": 0.0008,
|
| 1476 |
+
"step": 33400
|
| 1477 |
+
},
|
| 1478 |
+
{
|
| 1479 |
+
"epoch": 268.8,
|
| 1480 |
+
"grad_norm": 0.004440742079168558,
|
| 1481 |
+
"learning_rate": 7.295111111111112e-06,
|
| 1482 |
+
"loss": 0.0006,
|
| 1483 |
+
"step": 33600
|
| 1484 |
+
},
|
| 1485 |
+
{
|
| 1486 |
+
"epoch": 270.4,
|
| 1487 |
+
"grad_norm": 0.010887986980378628,
|
| 1488 |
+
"learning_rate": 7.206222222222223e-06,
|
| 1489 |
+
"loss": 0.0007,
|
| 1490 |
+
"step": 33800
|
| 1491 |
+
},
|
| 1492 |
+
{
|
| 1493 |
+
"epoch": 272.0,
|
| 1494 |
+
"grad_norm": 0.02552446536719799,
|
| 1495 |
+
"learning_rate": 7.117333333333334e-06,
|
| 1496 |
+
"loss": 0.0004,
|
| 1497 |
+
"step": 34000
|
| 1498 |
+
},
|
| 1499 |
+
{
|
| 1500 |
+
"epoch": 272.0,
|
| 1501 |
+
"eval_loss": 1.5940096378326416,
|
| 1502 |
+
"eval_runtime": 41.5805,
|
| 1503 |
+
"eval_samples_per_second": 6.012,
|
| 1504 |
+
"eval_steps_per_second": 1.515,
|
| 1505 |
+
"eval_wer": 0.4872953133822699,
|
| 1506 |
+
"step": 34000
|
| 1507 |
+
},
|
| 1508 |
+
{
|
| 1509 |
+
"epoch": 273.6,
|
| 1510 |
+
"grad_norm": 0.0572277270257473,
|
| 1511 |
+
"learning_rate": 7.028444444444445e-06,
|
| 1512 |
+
"loss": 0.0006,
|
| 1513 |
+
"step": 34200
|
| 1514 |
+
},
|
| 1515 |
+
{
|
| 1516 |
+
"epoch": 275.2,
|
| 1517 |
+
"grad_norm": 0.10096795111894608,
|
| 1518 |
+
"learning_rate": 6.9395555555555565e-06,
|
| 1519 |
+
"loss": 0.0009,
|
| 1520 |
+
"step": 34400
|
| 1521 |
+
},
|
| 1522 |
+
{
|
| 1523 |
+
"epoch": 276.8,
|
| 1524 |
+
"grad_norm": 0.005929913371801376,
|
| 1525 |
+
"learning_rate": 6.850666666666668e-06,
|
| 1526 |
+
"loss": 0.0005,
|
| 1527 |
+
"step": 34600
|
| 1528 |
+
},
|
| 1529 |
+
{
|
| 1530 |
+
"epoch": 278.4,
|
| 1531 |
+
"grad_norm": 0.005341435316950083,
|
| 1532 |
+
"learning_rate": 6.761777777777778e-06,
|
| 1533 |
+
"loss": 0.0005,
|
| 1534 |
+
"step": 34800
|
| 1535 |
+
},
|
| 1536 |
+
{
|
| 1537 |
+
"epoch": 280.0,
|
| 1538 |
+
"grad_norm": 0.021666768938302994,
|
| 1539 |
+
"learning_rate": 6.6728888888888895e-06,
|
| 1540 |
+
"loss": 0.0003,
|
| 1541 |
+
"step": 35000
|
| 1542 |
+
},
|
| 1543 |
+
{
|
| 1544 |
+
"epoch": 280.0,
|
| 1545 |
+
"eval_loss": 1.6009721755981445,
|
| 1546 |
+
"eval_runtime": 41.0006,
|
| 1547 |
+
"eval_samples_per_second": 6.097,
|
| 1548 |
+
"eval_steps_per_second": 1.537,
|
| 1549 |
+
"eval_wer": 0.46809712027103334,
|
| 1550 |
+
"step": 35000
|
| 1551 |
+
},
|
| 1552 |
+
{
|
| 1553 |
+
"epoch": 281.6,
|
| 1554 |
+
"grad_norm": 0.019466301426291466,
|
| 1555 |
+
"learning_rate": 6.584444444444446e-06,
|
| 1556 |
+
"loss": 0.0005,
|
| 1557 |
+
"step": 35200
|
| 1558 |
+
},
|
| 1559 |
+
{
|
| 1560 |
+
"epoch": 283.2,
|
| 1561 |
+
"grad_norm": 0.007609171327203512,
|
| 1562 |
+
"learning_rate": 6.495555555555556e-06,
|
| 1563 |
+
"loss": 0.0006,
|
| 1564 |
+
"step": 35400
|
| 1565 |
+
},
|
| 1566 |
+
{
|
| 1567 |
+
"epoch": 284.8,
|
| 1568 |
+
"grad_norm": 0.006095775868743658,
|
| 1569 |
+
"learning_rate": 6.4066666666666674e-06,
|
| 1570 |
+
"loss": 0.0004,
|
| 1571 |
+
"step": 35600
|
| 1572 |
+
},
|
| 1573 |
+
{
|
| 1574 |
+
"epoch": 286.4,
|
| 1575 |
+
"grad_norm": 0.008476600050926208,
|
| 1576 |
+
"learning_rate": 6.317777777777778e-06,
|
| 1577 |
+
"loss": 0.0002,
|
| 1578 |
+
"step": 35800
|
| 1579 |
+
},
|
| 1580 |
+
{
|
| 1581 |
+
"epoch": 288.0,
|
| 1582 |
+
"grad_norm": 0.06994141638278961,
|
| 1583 |
+
"learning_rate": 6.22888888888889e-06,
|
| 1584 |
+
"loss": 0.0004,
|
| 1585 |
+
"step": 36000
|
| 1586 |
+
},
|
| 1587 |
+
{
|
| 1588 |
+
"epoch": 288.0,
|
| 1589 |
+
"eval_loss": 1.5837336778640747,
|
| 1590 |
+
"eval_runtime": 42.3396,
|
| 1591 |
+
"eval_samples_per_second": 5.905,
|
| 1592 |
+
"eval_steps_per_second": 1.488,
|
| 1593 |
+
"eval_wer": 0.484472049689441,
|
| 1594 |
+
"step": 36000
|
| 1595 |
+
},
|
| 1596 |
+
{
|
| 1597 |
+
"epoch": 289.6,
|
| 1598 |
+
"grad_norm": 0.9534348845481873,
|
| 1599 |
+
"learning_rate": 6.1400000000000005e-06,
|
| 1600 |
+
"loss": 0.0003,
|
| 1601 |
+
"step": 36200
|
| 1602 |
+
},
|
| 1603 |
+
{
|
| 1604 |
+
"epoch": 291.2,
|
| 1605 |
+
"grad_norm": 0.00798807479441166,
|
| 1606 |
+
"learning_rate": 6.051111111111112e-06,
|
| 1607 |
+
"loss": 0.0003,
|
| 1608 |
+
"step": 36400
|
| 1609 |
+
},
|
| 1610 |
+
{
|
| 1611 |
+
"epoch": 292.8,
|
| 1612 |
+
"grad_norm": 0.4499492347240448,
|
| 1613 |
+
"learning_rate": 5.962222222222222e-06,
|
| 1614 |
+
"loss": 0.0004,
|
| 1615 |
+
"step": 36600
|
| 1616 |
+
},
|
| 1617 |
+
{
|
| 1618 |
+
"epoch": 294.4,
|
| 1619 |
+
"grad_norm": 0.00747877499088645,
|
| 1620 |
+
"learning_rate": 5.873333333333334e-06,
|
| 1621 |
+
"loss": 0.0006,
|
| 1622 |
+
"step": 36800
|
| 1623 |
+
},
|
| 1624 |
+
{
|
| 1625 |
+
"epoch": 296.0,
|
| 1626 |
+
"grad_norm": 0.007594792637974024,
|
| 1627 |
+
"learning_rate": 5.784444444444445e-06,
|
| 1628 |
+
"loss": 0.0006,
|
| 1629 |
+
"step": 37000
|
| 1630 |
+
},
|
| 1631 |
+
{
|
| 1632 |
+
"epoch": 296.0,
|
| 1633 |
+
"eval_loss": 1.5838834047317505,
|
| 1634 |
+
"eval_runtime": 41.0456,
|
| 1635 |
+
"eval_samples_per_second": 6.091,
|
| 1636 |
+
"eval_steps_per_second": 1.535,
|
| 1637 |
+
"eval_wer": 0.47939017504234893,
|
| 1638 |
+
"step": 37000
|
| 1639 |
+
},
|
| 1640 |
+
{
|
| 1641 |
+
"epoch": 297.6,
|
| 1642 |
+
"grad_norm": 0.0062100524082779884,
|
| 1643 |
+
"learning_rate": 5.695555555555556e-06,
|
| 1644 |
+
"loss": 0.0003,
|
| 1645 |
+
"step": 37200
|
| 1646 |
+
},
|
| 1647 |
+
{
|
| 1648 |
+
"epoch": 299.2,
|
| 1649 |
+
"grad_norm": 1.5888539552688599,
|
| 1650 |
+
"learning_rate": 5.606666666666667e-06,
|
| 1651 |
+
"loss": 0.0006,
|
| 1652 |
+
"step": 37400
|
| 1653 |
+
},
|
| 1654 |
+
{
|
| 1655 |
+
"epoch": 300.8,
|
| 1656 |
+
"grad_norm": 0.004372656811028719,
|
| 1657 |
+
"learning_rate": 5.517777777777779e-06,
|
| 1658 |
+
"loss": 0.0003,
|
| 1659 |
+
"step": 37600
|
| 1660 |
+
},
|
| 1661 |
+
{
|
| 1662 |
+
"epoch": 302.4,
|
| 1663 |
+
"grad_norm": 0.004411238245666027,
|
| 1664 |
+
"learning_rate": 5.428888888888889e-06,
|
| 1665 |
+
"loss": 0.0004,
|
| 1666 |
+
"step": 37800
|
| 1667 |
+
},
|
| 1668 |
+
{
|
| 1669 |
+
"epoch": 304.0,
|
| 1670 |
+
"grad_norm": 0.007507936097681522,
|
| 1671 |
+
"learning_rate": 5.3400000000000005e-06,
|
| 1672 |
+
"loss": 0.0002,
|
| 1673 |
+
"step": 38000
|
| 1674 |
+
},
|
| 1675 |
+
{
|
| 1676 |
+
"epoch": 304.0,
|
| 1677 |
+
"eval_loss": 1.5652457475662231,
|
| 1678 |
+
"eval_runtime": 41.4549,
|
| 1679 |
+
"eval_samples_per_second": 6.031,
|
| 1680 |
+
"eval_steps_per_second": 1.52,
|
| 1681 |
+
"eval_wer": 0.4754376058723885,
|
| 1682 |
+
"step": 38000
|
| 1683 |
+
},
|
| 1684 |
+
{
|
| 1685 |
+
"epoch": 305.6,
|
| 1686 |
+
"grad_norm": 0.005323062185198069,
|
| 1687 |
+
"learning_rate": 5.251555555555557e-06,
|
| 1688 |
+
"loss": 0.0005,
|
| 1689 |
+
"step": 38200
|
| 1690 |
+
},
|
| 1691 |
+
{
|
| 1692 |
+
"epoch": 307.2,
|
| 1693 |
+
"grad_norm": 0.0021106544882059097,
|
| 1694 |
+
"learning_rate": 5.162666666666667e-06,
|
| 1695 |
+
"loss": 0.0005,
|
| 1696 |
+
"step": 38400
|
| 1697 |
+
},
|
| 1698 |
+
{
|
| 1699 |
+
"epoch": 308.8,
|
| 1700 |
+
"grad_norm": 0.0051486073061823845,
|
| 1701 |
+
"learning_rate": 5.0737777777777785e-06,
|
| 1702 |
+
"loss": 0.0007,
|
| 1703 |
+
"step": 38600
|
| 1704 |
+
},
|
| 1705 |
+
{
|
| 1706 |
+
"epoch": 310.4,
|
| 1707 |
+
"grad_norm": 0.06730670481920242,
|
| 1708 |
+
"learning_rate": 4.984888888888889e-06,
|
| 1709 |
+
"loss": 0.0004,
|
| 1710 |
+
"step": 38800
|
| 1711 |
+
},
|
| 1712 |
+
{
|
| 1713 |
+
"epoch": 312.0,
|
| 1714 |
+
"grad_norm": 0.005158218555152416,
|
| 1715 |
+
"learning_rate": 4.896e-06,
|
| 1716 |
+
"loss": 0.0003,
|
| 1717 |
+
"step": 39000
|
| 1718 |
+
},
|
| 1719 |
+
{
|
| 1720 |
+
"epoch": 312.0,
|
| 1721 |
+
"eval_loss": 1.6083264350891113,
|
| 1722 |
+
"eval_runtime": 41.1796,
|
| 1723 |
+
"eval_samples_per_second": 6.071,
|
| 1724 |
+
"eval_steps_per_second": 1.53,
|
| 1725 |
+
"eval_wer": 0.48334274421230944,
|
| 1726 |
+
"step": 39000
|
| 1727 |
+
},
|
| 1728 |
+
{
|
| 1729 |
+
"epoch": 313.6,
|
| 1730 |
+
"grad_norm": 0.007136846426874399,
|
| 1731 |
+
"learning_rate": 4.8071111111111115e-06,
|
| 1732 |
+
"loss": 0.0002,
|
| 1733 |
+
"step": 39200
|
| 1734 |
+
},
|
| 1735 |
+
{
|
| 1736 |
+
"epoch": 315.2,
|
| 1737 |
+
"grad_norm": 0.0384252667427063,
|
| 1738 |
+
"learning_rate": 4.718222222222222e-06,
|
| 1739 |
+
"loss": 0.0005,
|
| 1740 |
+
"step": 39400
|
| 1741 |
+
},
|
| 1742 |
+
{
|
| 1743 |
+
"epoch": 316.8,
|
| 1744 |
+
"grad_norm": 0.03144264966249466,
|
| 1745 |
+
"learning_rate": 4.629333333333333e-06,
|
| 1746 |
+
"loss": 0.0003,
|
| 1747 |
+
"step": 39600
|
| 1748 |
+
},
|
| 1749 |
+
{
|
| 1750 |
+
"epoch": 318.4,
|
| 1751 |
+
"grad_norm": 0.04305962473154068,
|
| 1752 |
+
"learning_rate": 4.540444444444445e-06,
|
| 1753 |
+
"loss": 0.0002,
|
| 1754 |
+
"step": 39800
|
| 1755 |
+
},
|
| 1756 |
+
{
|
| 1757 |
+
"epoch": 320.0,
|
| 1758 |
+
"grad_norm": 0.011067772284150124,
|
| 1759 |
+
"learning_rate": 4.451555555555556e-06,
|
| 1760 |
+
"loss": 0.0002,
|
| 1761 |
+
"step": 40000
|
| 1762 |
+
},
|
| 1763 |
+
{
|
| 1764 |
+
"epoch": 320.0,
|
| 1765 |
+
"eval_loss": 1.575042724609375,
|
| 1766 |
+
"eval_runtime": 42.1507,
|
| 1767 |
+
"eval_samples_per_second": 5.931,
|
| 1768 |
+
"eval_steps_per_second": 1.495,
|
| 1769 |
+
"eval_wer": 0.5189158667419537,
|
| 1770 |
+
"step": 40000
|
| 1771 |
+
},
|
| 1772 |
+
{
|
| 1773 |
+
"epoch": 321.6,
|
| 1774 |
+
"grad_norm": 0.002925801556557417,
|
| 1775 |
+
"learning_rate": 4.362666666666667e-06,
|
| 1776 |
+
"loss": 0.0004,
|
| 1777 |
+
"step": 40200
|
| 1778 |
+
},
|
| 1779 |
+
{
|
| 1780 |
+
"epoch": 323.2,
|
| 1781 |
+
"grad_norm": 0.03247452154755592,
|
| 1782 |
+
"learning_rate": 4.2737777777777785e-06,
|
| 1783 |
+
"loss": 0.0002,
|
| 1784 |
+
"step": 40400
|
| 1785 |
+
},
|
| 1786 |
+
{
|
| 1787 |
+
"epoch": 324.8,
|
| 1788 |
+
"grad_norm": 0.017400383949279785,
|
| 1789 |
+
"learning_rate": 4.18488888888889e-06,
|
| 1790 |
+
"loss": 0.0002,
|
| 1791 |
+
"step": 40600
|
| 1792 |
+
},
|
| 1793 |
+
{
|
| 1794 |
+
"epoch": 326.4,
|
| 1795 |
+
"grad_norm": 0.0033240150660276413,
|
| 1796 |
+
"learning_rate": 4.096e-06,
|
| 1797 |
+
"loss": 0.0001,
|
| 1798 |
+
"step": 40800
|
| 1799 |
+
},
|
| 1800 |
+
{
|
| 1801 |
+
"epoch": 328.0,
|
| 1802 |
+
"grad_norm": 0.009032553061842918,
|
| 1803 |
+
"learning_rate": 4.0071111111111116e-06,
|
| 1804 |
+
"loss": 0.0004,
|
| 1805 |
+
"step": 41000
|
| 1806 |
+
},
|
| 1807 |
+
{
|
| 1808 |
+
"epoch": 328.0,
|
| 1809 |
+
"eval_loss": 1.619897484779358,
|
| 1810 |
+
"eval_runtime": 44.7114,
|
| 1811 |
+
"eval_samples_per_second": 5.591,
|
| 1812 |
+
"eval_steps_per_second": 1.409,
|
| 1813 |
+
"eval_wer": 0.5979672501411631,
|
| 1814 |
+
"step": 41000
|
| 1815 |
+
},
|
| 1816 |
+
{
|
| 1817 |
+
"epoch": 329.6,
|
| 1818 |
+
"grad_norm": 0.0036150990054011345,
|
| 1819 |
+
"learning_rate": 3.918222222222223e-06,
|
| 1820 |
+
"loss": 0.0002,
|
| 1821 |
+
"step": 41200
|
| 1822 |
+
},
|
| 1823 |
+
{
|
| 1824 |
+
"epoch": 331.2,
|
| 1825 |
+
"grad_norm": 0.029861323535442352,
|
| 1826 |
+
"learning_rate": 3.829333333333334e-06,
|
| 1827 |
+
"loss": 0.0002,
|
| 1828 |
+
"step": 41400
|
| 1829 |
+
},
|
| 1830 |
+
{
|
| 1831 |
+
"epoch": 332.8,
|
| 1832 |
+
"grad_norm": 0.02380683459341526,
|
| 1833 |
+
"learning_rate": 3.740444444444445e-06,
|
| 1834 |
+
"loss": 0.0001,
|
| 1835 |
+
"step": 41600
|
| 1836 |
+
},
|
| 1837 |
+
{
|
| 1838 |
+
"epoch": 334.4,
|
| 1839 |
+
"grad_norm": 0.031281813979148865,
|
| 1840 |
+
"learning_rate": 3.651555555555556e-06,
|
| 1841 |
+
"loss": 0.0002,
|
| 1842 |
+
"step": 41800
|
| 1843 |
+
},
|
| 1844 |
+
{
|
| 1845 |
+
"epoch": 336.0,
|
| 1846 |
+
"grad_norm": 0.0028394584078341722,
|
| 1847 |
+
"learning_rate": 3.5626666666666672e-06,
|
| 1848 |
+
"loss": 0.0001,
|
| 1849 |
+
"step": 42000
|
| 1850 |
+
},
|
| 1851 |
+
{
|
| 1852 |
+
"epoch": 336.0,
|
| 1853 |
+
"eval_loss": 1.5782560110092163,
|
| 1854 |
+
"eval_runtime": 42.7351,
|
| 1855 |
+
"eval_samples_per_second": 5.85,
|
| 1856 |
+
"eval_steps_per_second": 1.474,
|
| 1857 |
+
"eval_wer": 0.5352907961603613,
|
| 1858 |
+
"step": 42000
|
| 1859 |
+
},
|
| 1860 |
+
{
|
| 1861 |
+
"epoch": 337.6,
|
| 1862 |
+
"grad_norm": 0.11405123770236969,
|
| 1863 |
+
"learning_rate": 3.473777777777778e-06,
|
| 1864 |
+
"loss": 0.0002,
|
| 1865 |
+
"step": 42200
|
| 1866 |
+
},
|
| 1867 |
+
{
|
| 1868 |
+
"epoch": 339.2,
|
| 1869 |
+
"grad_norm": 0.022922609001398087,
|
| 1870 |
+
"learning_rate": 3.3848888888888894e-06,
|
| 1871 |
+
"loss": 0.0003,
|
| 1872 |
+
"step": 42400
|
| 1873 |
+
},
|
| 1874 |
+
{
|
| 1875 |
+
"epoch": 340.8,
|
| 1876 |
+
"grad_norm": 0.0035330464597791433,
|
| 1877 |
+
"learning_rate": 3.2960000000000003e-06,
|
| 1878 |
+
"loss": 0.0002,
|
| 1879 |
+
"step": 42600
|
| 1880 |
+
},
|
| 1881 |
+
{
|
| 1882 |
+
"epoch": 342.4,
|
| 1883 |
+
"grad_norm": 0.0057142325676977634,
|
| 1884 |
+
"learning_rate": 3.2071111111111116e-06,
|
| 1885 |
+
"loss": 0.0003,
|
| 1886 |
+
"step": 42800
|
| 1887 |
+
},
|
| 1888 |
+
{
|
| 1889 |
+
"epoch": 344.0,
|
| 1890 |
+
"grad_norm": 0.009726474992930889,
|
| 1891 |
+
"learning_rate": 3.1182222222222225e-06,
|
| 1892 |
+
"loss": 0.0001,
|
| 1893 |
+
"step": 43000
|
| 1894 |
+
},
|
| 1895 |
+
{
|
| 1896 |
+
"epoch": 344.0,
|
| 1897 |
+
"eval_loss": 1.5897696018218994,
|
| 1898 |
+
"eval_runtime": 42.1045,
|
| 1899 |
+
"eval_samples_per_second": 5.938,
|
| 1900 |
+
"eval_steps_per_second": 1.496,
|
| 1901 |
+
"eval_wer": 0.5098814229249012,
|
| 1902 |
+
"step": 43000
|
| 1903 |
+
},
|
| 1904 |
+
{
|
| 1905 |
+
"epoch": 345.6,
|
| 1906 |
+
"grad_norm": 0.002822835696861148,
|
| 1907 |
+
"learning_rate": 3.0293333333333338e-06,
|
| 1908 |
+
"loss": 0.0001,
|
| 1909 |
+
"step": 43200
|
| 1910 |
+
},
|
| 1911 |
+
{
|
| 1912 |
+
"epoch": 347.2,
|
| 1913 |
+
"grad_norm": 0.18968430161476135,
|
| 1914 |
+
"learning_rate": 2.9404444444444447e-06,
|
| 1915 |
+
"loss": 0.0003,
|
| 1916 |
+
"step": 43400
|
| 1917 |
+
},
|
| 1918 |
+
{
|
| 1919 |
+
"epoch": 348.8,
|
| 1920 |
+
"grad_norm": 0.01890464872121811,
|
| 1921 |
+
"learning_rate": 2.8515555555555555e-06,
|
| 1922 |
+
"loss": 0.0004,
|
| 1923 |
+
"step": 43600
|
| 1924 |
+
},
|
| 1925 |
+
{
|
| 1926 |
+
"epoch": 350.4,
|
| 1927 |
+
"grad_norm": 0.01565505564212799,
|
| 1928 |
+
"learning_rate": 2.7631111111111113e-06,
|
| 1929 |
+
"loss": 0.0006,
|
| 1930 |
+
"step": 43800
|
| 1931 |
+
},
|
| 1932 |
+
{
|
| 1933 |
+
"epoch": 352.0,
|
| 1934 |
+
"grad_norm": 0.031101664528250694,
|
| 1935 |
+
"learning_rate": 2.6742222222222226e-06,
|
| 1936 |
+
"loss": 0.0005,
|
| 1937 |
+
"step": 44000
|
| 1938 |
+
},
|
| 1939 |
+
{
|
| 1940 |
+
"epoch": 352.0,
|
| 1941 |
+
"eval_loss": 1.6004997491836548,
|
| 1942 |
+
"eval_runtime": 43.7134,
|
| 1943 |
+
"eval_samples_per_second": 5.719,
|
| 1944 |
+
"eval_steps_per_second": 1.441,
|
| 1945 |
+
"eval_wer": 0.5832862789384529,
|
| 1946 |
+
"step": 44000
|
| 1947 |
+
},
|
| 1948 |
+
{
|
| 1949 |
+
"epoch": 353.6,
|
| 1950 |
+
"grad_norm": 0.0026319867465645075,
|
| 1951 |
+
"learning_rate": 2.5853333333333335e-06,
|
| 1952 |
+
"loss": 0.0002,
|
| 1953 |
+
"step": 44200
|
| 1954 |
+
},
|
| 1955 |
+
{
|
| 1956 |
+
"epoch": 355.2,
|
| 1957 |
+
"grad_norm": 0.0039254105649888515,
|
| 1958 |
+
"learning_rate": 2.4964444444444448e-06,
|
| 1959 |
+
"loss": 0.0002,
|
| 1960 |
+
"step": 44400
|
| 1961 |
+
},
|
| 1962 |
+
{
|
| 1963 |
+
"epoch": 356.8,
|
| 1964 |
+
"grad_norm": 0.10020512342453003,
|
| 1965 |
+
"learning_rate": 2.4075555555555556e-06,
|
| 1966 |
+
"loss": 0.0002,
|
| 1967 |
+
"step": 44600
|
| 1968 |
+
},
|
| 1969 |
+
{
|
| 1970 |
+
"epoch": 358.4,
|
| 1971 |
+
"grad_norm": 0.0036830666940659285,
|
| 1972 |
+
"learning_rate": 2.318666666666667e-06,
|
| 1973 |
+
"loss": 0.0001,
|
| 1974 |
+
"step": 44800
|
| 1975 |
+
},
|
| 1976 |
+
{
|
| 1977 |
+
"epoch": 360.0,
|
| 1978 |
+
"grad_norm": 0.008613605052232742,
|
| 1979 |
+
"learning_rate": 2.229777777777778e-06,
|
| 1980 |
+
"loss": 0.0002,
|
| 1981 |
+
"step": 45000
|
| 1982 |
+
},
|
| 1983 |
+
{
|
| 1984 |
+
"epoch": 360.0,
|
| 1985 |
+
"eval_loss": 1.5903429985046387,
|
| 1986 |
+
"eval_runtime": 41.6423,
|
| 1987 |
+
"eval_samples_per_second": 6.004,
|
| 1988 |
+
"eval_steps_per_second": 1.513,
|
| 1989 |
+
"eval_wer": 0.4872953133822699,
|
| 1990 |
+
"step": 45000
|
| 1991 |
+
},
|
| 1992 |
+
{
|
| 1993 |
+
"epoch": 361.6,
|
| 1994 |
+
"grad_norm": 0.0025289321783930063,
|
| 1995 |
+
"learning_rate": 2.140888888888889e-06,
|
| 1996 |
+
"loss": 0.0003,
|
| 1997 |
+
"step": 45200
|
| 1998 |
+
},
|
| 1999 |
+
{
|
| 2000 |
+
"epoch": 363.2,
|
| 2001 |
+
"grad_norm": 0.004556621424853802,
|
| 2002 |
+
"learning_rate": 2.052e-06,
|
| 2003 |
+
"loss": 0.0002,
|
| 2004 |
+
"step": 45400
|
| 2005 |
+
},
|
| 2006 |
+
{
|
| 2007 |
+
"epoch": 364.8,
|
| 2008 |
+
"grad_norm": 0.06288538128137589,
|
| 2009 |
+
"learning_rate": 1.9631111111111113e-06,
|
| 2010 |
+
"loss": 0.0002,
|
| 2011 |
+
"step": 45600
|
| 2012 |
+
},
|
| 2013 |
+
{
|
| 2014 |
+
"epoch": 366.4,
|
| 2015 |
+
"grad_norm": 0.002961724065244198,
|
| 2016 |
+
"learning_rate": 1.8742222222222222e-06,
|
| 2017 |
+
"loss": 0.0001,
|
| 2018 |
+
"step": 45800
|
| 2019 |
+
},
|
| 2020 |
+
{
|
| 2021 |
+
"epoch": 368.0,
|
| 2022 |
+
"grad_norm": 0.0020394865423440933,
|
| 2023 |
+
"learning_rate": 1.7853333333333333e-06,
|
| 2024 |
+
"loss": 0.0002,
|
| 2025 |
+
"step": 46000
|
| 2026 |
+
},
|
| 2027 |
+
{
|
| 2028 |
+
"epoch": 368.0,
|
| 2029 |
+
"eval_loss": 1.619638442993164,
|
| 2030 |
+
"eval_runtime": 42.1006,
|
| 2031 |
+
"eval_samples_per_second": 5.938,
|
| 2032 |
+
"eval_steps_per_second": 1.496,
|
| 2033 |
+
"eval_wer": 0.5149632975719932,
|
| 2034 |
+
"step": 46000
|
| 2035 |
+
},
|
| 2036 |
+
{
|
| 2037 |
+
"epoch": 369.6,
|
| 2038 |
+
"grad_norm": 0.0018333194311708212,
|
| 2039 |
+
"learning_rate": 1.6964444444444444e-06,
|
| 2040 |
+
"loss": 0.0001,
|
| 2041 |
+
"step": 46200
|
| 2042 |
+
},
|
| 2043 |
+
{
|
| 2044 |
+
"epoch": 371.2,
|
| 2045 |
+
"grad_norm": 0.00915595330297947,
|
| 2046 |
+
"learning_rate": 1.6075555555555559e-06,
|
| 2047 |
+
"loss": 0.0001,
|
| 2048 |
+
"step": 46400
|
| 2049 |
+
},
|
| 2050 |
+
{
|
| 2051 |
+
"epoch": 372.8,
|
| 2052 |
+
"grad_norm": 0.006657126825302839,
|
| 2053 |
+
"learning_rate": 1.518666666666667e-06,
|
| 2054 |
+
"loss": 0.0001,
|
| 2055 |
+
"step": 46600
|
| 2056 |
+
},
|
| 2057 |
+
{
|
| 2058 |
+
"epoch": 374.4,
|
| 2059 |
+
"grad_norm": 0.1112598404288292,
|
| 2060 |
+
"learning_rate": 1.429777777777778e-06,
|
| 2061 |
+
"loss": 0.0001,
|
| 2062 |
+
"step": 46800
|
| 2063 |
+
},
|
| 2064 |
+
{
|
| 2065 |
+
"epoch": 376.0,
|
| 2066 |
+
"grad_norm": 0.004303216468542814,
|
| 2067 |
+
"learning_rate": 1.3408888888888892e-06,
|
| 2068 |
+
"loss": 0.0001,
|
| 2069 |
+
"step": 47000
|
| 2070 |
+
},
|
| 2071 |
+
{
|
| 2072 |
+
"epoch": 376.0,
|
| 2073 |
+
"eval_loss": 1.6212029457092285,
|
| 2074 |
+
"eval_runtime": 42.453,
|
| 2075 |
+
"eval_samples_per_second": 5.889,
|
| 2076 |
+
"eval_steps_per_second": 1.484,
|
| 2077 |
+
"eval_wer": 0.5251270468661773,
|
| 2078 |
+
"step": 47000
|
| 2079 |
+
},
|
| 2080 |
+
{
|
| 2081 |
+
"epoch": 377.6,
|
| 2082 |
+
"grad_norm": 0.003241022815927863,
|
| 2083 |
+
"learning_rate": 1.2520000000000003e-06,
|
| 2084 |
+
"loss": 0.0001,
|
| 2085 |
+
"step": 47200
|
| 2086 |
+
},
|
| 2087 |
+
{
|
| 2088 |
+
"epoch": 379.2,
|
| 2089 |
+
"grad_norm": 0.008798100054264069,
|
| 2090 |
+
"learning_rate": 1.1631111111111113e-06,
|
| 2091 |
+
"loss": 0.0001,
|
| 2092 |
+
"step": 47400
|
| 2093 |
+
},
|
| 2094 |
+
{
|
| 2095 |
+
"epoch": 380.8,
|
| 2096 |
+
"grad_norm": 0.013746335171163082,
|
| 2097 |
+
"learning_rate": 1.0742222222222224e-06,
|
| 2098 |
+
"loss": 0.0001,
|
| 2099 |
+
"step": 47600
|
| 2100 |
+
},
|
| 2101 |
+
{
|
| 2102 |
+
"epoch": 382.4,
|
| 2103 |
+
"grad_norm": 0.0022988717537373304,
|
| 2104 |
+
"learning_rate": 9.853333333333333e-07,
|
| 2105 |
+
"loss": 0.0001,
|
| 2106 |
+
"step": 47800
|
| 2107 |
+
},
|
| 2108 |
+
{
|
| 2109 |
+
"epoch": 384.0,
|
| 2110 |
+
"grad_norm": 0.0472625233232975,
|
| 2111 |
+
"learning_rate": 8.964444444444445e-07,
|
| 2112 |
+
"loss": 0.0002,
|
| 2113 |
+
"step": 48000
|
| 2114 |
+
},
|
| 2115 |
+
{
|
| 2116 |
+
"epoch": 384.0,
|
| 2117 |
+
"eval_loss": 1.618031620979309,
|
| 2118 |
+
"eval_runtime": 43.7587,
|
| 2119 |
+
"eval_samples_per_second": 5.713,
|
| 2120 |
+
"eval_steps_per_second": 1.44,
|
| 2121 |
+
"eval_wer": 0.5539243365330322,
|
| 2122 |
+
"step": 48000
|
| 2123 |
+
},
|
| 2124 |
+
{
|
| 2125 |
+
"epoch": 385.6,
|
| 2126 |
+
"grad_norm": 0.0067636617459356785,
|
| 2127 |
+
"learning_rate": 8.075555555555556e-07,
|
| 2128 |
+
"loss": 0.0001,
|
| 2129 |
+
"step": 48200
|
| 2130 |
+
},
|
| 2131 |
+
{
|
| 2132 |
+
"epoch": 387.2,
|
| 2133 |
+
"grad_norm": 0.0014746059896424413,
|
| 2134 |
+
"learning_rate": 7.186666666666667e-07,
|
| 2135 |
+
"loss": 0.0001,
|
| 2136 |
+
"step": 48400
|
| 2137 |
+
},
|
| 2138 |
+
{
|
| 2139 |
+
"epoch": 388.8,
|
| 2140 |
+
"grad_norm": 0.03201240673661232,
|
| 2141 |
+
"learning_rate": 6.297777777777778e-07,
|
| 2142 |
+
"loss": 0.0001,
|
| 2143 |
+
"step": 48600
|
| 2144 |
+
},
|
| 2145 |
+
{
|
| 2146 |
+
"epoch": 390.4,
|
| 2147 |
+
"grad_norm": 0.0024084942415356636,
|
| 2148 |
+
"learning_rate": 5.408888888888889e-07,
|
| 2149 |
+
"loss": 0.0001,
|
| 2150 |
+
"step": 48800
|
| 2151 |
+
},
|
| 2152 |
+
{
|
| 2153 |
+
"epoch": 392.0,
|
| 2154 |
+
"grad_norm": 0.0010368505027145147,
|
| 2155 |
+
"learning_rate": 4.5244444444444445e-07,
|
| 2156 |
+
"loss": 0.0001,
|
| 2157 |
+
"step": 49000
|
| 2158 |
+
},
|
| 2159 |
+
{
|
| 2160 |
+
"epoch": 392.0,
|
| 2161 |
+
"eval_loss": 1.6104307174682617,
|
| 2162 |
+
"eval_runtime": 41.3895,
|
| 2163 |
+
"eval_samples_per_second": 6.04,
|
| 2164 |
+
"eval_steps_per_second": 1.522,
|
| 2165 |
+
"eval_wer": 0.49632975719932243,
|
| 2166 |
+
"step": 49000
|
| 2167 |
+
},
|
| 2168 |
+
{
|
| 2169 |
+
"epoch": 393.6,
|
| 2170 |
+
"grad_norm": 0.6977916359901428,
|
| 2171 |
+
"learning_rate": 3.635555555555556e-07,
|
| 2172 |
+
"loss": 0.0001,
|
| 2173 |
+
"step": 49200
|
| 2174 |
+
},
|
| 2175 |
+
{
|
| 2176 |
+
"epoch": 395.2,
|
| 2177 |
+
"grad_norm": 0.0007687339093536139,
|
| 2178 |
+
"learning_rate": 2.746666666666667e-07,
|
| 2179 |
+
"loss": 0.0001,
|
| 2180 |
+
"step": 49400
|
| 2181 |
+
},
|
| 2182 |
+
{
|
| 2183 |
+
"epoch": 396.8,
|
| 2184 |
+
"grad_norm": 0.009420102462172508,
|
| 2185 |
+
"learning_rate": 1.8577777777777778e-07,
|
| 2186 |
+
"loss": 0.0001,
|
| 2187 |
+
"step": 49600
|
| 2188 |
+
},
|
| 2189 |
+
{
|
| 2190 |
+
"epoch": 398.4,
|
| 2191 |
+
"grad_norm": 0.0071104601956903934,
|
| 2192 |
+
"learning_rate": 9.68888888888889e-08,
|
| 2193 |
+
"loss": 0.0001,
|
| 2194 |
+
"step": 49800
|
| 2195 |
+
},
|
| 2196 |
+
{
|
| 2197 |
+
"epoch": 400.0,
|
| 2198 |
+
"grad_norm": 0.0014255736023187637,
|
| 2199 |
+
"learning_rate": 8e-09,
|
| 2200 |
+
"loss": 0.0001,
|
| 2201 |
+
"step": 50000
|
| 2202 |
+
},
|
| 2203 |
+
{
|
| 2204 |
+
"epoch": 400.0,
|
| 2205 |
+
"eval_loss": 1.6090388298034668,
|
| 2206 |
+
"eval_runtime": 42.8761,
|
| 2207 |
+
"eval_samples_per_second": 5.831,
|
| 2208 |
+
"eval_steps_per_second": 1.469,
|
| 2209 |
+
"eval_wer": 0.510446075663467,
|
| 2210 |
+
"step": 50000
|
| 2211 |
+
},
|
| 2212 |
+
{
|
| 2213 |
+
"epoch": 400.0,
|
| 2214 |
+
"step": 50000,
|
| 2215 |
+
"total_flos": 1.9695108096e+19,
|
| 2216 |
+
"train_loss": 0.05668430684482679,
|
| 2217 |
+
"train_runtime": 102928.0936,
|
| 2218 |
+
"train_samples_per_second": 7.772,
|
| 2219 |
+
"train_steps_per_second": 0.486
|
| 2220 |
+
}
|
| 2221 |
+
],
|
| 2222 |
+
"logging_steps": 200,
|
| 2223 |
+
"max_steps": 50000,
|
| 2224 |
+
"num_input_tokens_seen": 0,
|
| 2225 |
+
"num_train_epochs": 400,
|
| 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": 1.9695108096e+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": 400.0,
|
| 3 |
+
"phase_1_val_loss": 1.27247154712677,
|
| 4 |
+
"phase_1_val_runtime": 41.2147,
|
| 5 |
+
"phase_1_val_samples_per_second": 6.066,
|
| 6 |
+
"phase_1_val_steps_per_second": 1.529,
|
| 7 |
+
"phase_1_val_wer": 0.44776962168266515,
|
| 8 |
+
"val_samples": 250
|
| 9 |
+
}
|