End of training
Browse files- README.md +16 -4
- all_results.json +15 -0
- eval_results.json +9 -0
- runs/Feb09_22-40-00_srvrocgpu011.uct.ac.za/events.out.tfevents.1739142666.srvrocgpu011.uct.ac.za +3 -0
- train_results.json +9 -0
- trainer_state.json +701 -0
README.md
CHANGED
|
@@ -4,11 +4,23 @@ license: apache-2.0
|
|
| 4 |
base_model: openai/whisper-medium
|
| 5 |
tags:
|
| 6 |
- generated_from_trainer
|
|
|
|
|
|
|
| 7 |
metrics:
|
| 8 |
- wer
|
| 9 |
model-index:
|
| 10 |
- name: whisper-medium-bemgen-100f50m-model
|
| 11 |
-
results:
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 12 |
---
|
| 13 |
|
| 14 |
<!-- This model card has been generated automatically according to the information the Trainer had access to. You
|
|
@@ -16,10 +28,10 @@ should probably proofread and complete it, then remove this comment. -->
|
|
| 16 |
|
| 17 |
# whisper-medium-bemgen-100f50m-model
|
| 18 |
|
| 19 |
-
This model is a fine-tuned version of [openai/whisper-medium](https://huggingface.co/openai/whisper-medium) on
|
| 20 |
It achieves the following results on the evaluation set:
|
| 21 |
-
- Loss: 0.
|
| 22 |
-
- Wer: 0.
|
| 23 |
|
| 24 |
## Model description
|
| 25 |
|
|
|
|
| 4 |
base_model: openai/whisper-medium
|
| 5 |
tags:
|
| 6 |
- generated_from_trainer
|
| 7 |
+
datasets:
|
| 8 |
+
- bemgen
|
| 9 |
metrics:
|
| 10 |
- wer
|
| 11 |
model-index:
|
| 12 |
- name: whisper-medium-bemgen-100f50m-model
|
| 13 |
+
results:
|
| 14 |
+
- task:
|
| 15 |
+
name: Automatic Speech Recognition
|
| 16 |
+
type: automatic-speech-recognition
|
| 17 |
+
dataset:
|
| 18 |
+
name: bemgen
|
| 19 |
+
type: bemgen
|
| 20 |
+
metrics:
|
| 21 |
+
- name: Wer
|
| 22 |
+
type: wer
|
| 23 |
+
value: 0.38982326111744586
|
| 24 |
---
|
| 25 |
|
| 26 |
<!-- This model card has been generated automatically according to the information the Trainer had access to. You
|
|
|
|
| 28 |
|
| 29 |
# whisper-medium-bemgen-100f50m-model
|
| 30 |
|
| 31 |
+
This model is a fine-tuned version of [openai/whisper-medium](https://huggingface.co/openai/whisper-medium) on the bemgen dataset.
|
| 32 |
It achieves the following results on the evaluation set:
|
| 33 |
+
- Loss: 0.4354
|
| 34 |
+
- Wer: 0.3898
|
| 35 |
|
| 36 |
## Model description
|
| 37 |
|
all_results.json
ADDED
|
@@ -0,0 +1,15 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"epoch": 2.6391548365797295,
|
| 3 |
+
"eval_loss": 0.4354272186756134,
|
| 4 |
+
"eval_runtime": 430.7377,
|
| 5 |
+
"eval_samples": 758,
|
| 6 |
+
"eval_samples_per_second": 1.76,
|
| 7 |
+
"eval_steps_per_second": 0.88,
|
| 8 |
+
"eval_wer": 0.38982326111744586,
|
| 9 |
+
"total_flos": 1.631539005161472e+19,
|
| 10 |
+
"train_loss": 1.9525959782600404,
|
| 11 |
+
"train_runtime": 8300.7195,
|
| 12 |
+
"train_samples": 6057,
|
| 13 |
+
"train_samples_per_second": 4.819,
|
| 14 |
+
"train_steps_per_second": 0.602
|
| 15 |
+
}
|
eval_results.json
ADDED
|
@@ -0,0 +1,9 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"epoch": 2.6391548365797295,
|
| 3 |
+
"eval_loss": 0.4354272186756134,
|
| 4 |
+
"eval_runtime": 430.7377,
|
| 5 |
+
"eval_samples": 758,
|
| 6 |
+
"eval_samples_per_second": 1.76,
|
| 7 |
+
"eval_steps_per_second": 0.88,
|
| 8 |
+
"eval_wer": 0.38982326111744586
|
| 9 |
+
}
|
runs/Feb09_22-40-00_srvrocgpu011.uct.ac.za/events.out.tfevents.1739142666.srvrocgpu011.uct.ac.za
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:a22e27e8bad18174cd13f00f691383c6768d40f2f07181c7c9d08436633f30dc
|
| 3 |
+
size 40
|
train_results.json
ADDED
|
@@ -0,0 +1,9 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"epoch": 2.6391548365797295,
|
| 3 |
+
"total_flos": 1.631539005161472e+19,
|
| 4 |
+
"train_loss": 1.9525959782600404,
|
| 5 |
+
"train_runtime": 8300.7195,
|
| 6 |
+
"train_samples": 6057,
|
| 7 |
+
"train_samples_per_second": 4.819,
|
| 8 |
+
"train_steps_per_second": 0.602
|
| 9 |
+
}
|
trainer_state.json
ADDED
|
@@ -0,0 +1,701 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"best_metric": 0.4354272186756134,
|
| 3 |
+
"best_model_checkpoint": "/scratch/skscla001/speech/results/whisper-medium-bemgen-100f50m-model/checkpoint-1400",
|
| 4 |
+
"epoch": 2.6391548365797295,
|
| 5 |
+
"eval_steps": 200,
|
| 6 |
+
"global_step": 2000,
|
| 7 |
+
"is_hyper_param_search": false,
|
| 8 |
+
"is_local_process_zero": true,
|
| 9 |
+
"is_world_process_zero": true,
|
| 10 |
+
"log_history": [
|
| 11 |
+
{
|
| 12 |
+
"epoch": 0.03301419610432486,
|
| 13 |
+
"grad_norm": 117.47059631347656,
|
| 14 |
+
"learning_rate": 3.8e-07,
|
| 15 |
+
"loss": 10.9949,
|
| 16 |
+
"step": 25
|
| 17 |
+
},
|
| 18 |
+
{
|
| 19 |
+
"epoch": 0.06602839220864971,
|
| 20 |
+
"grad_norm": 102.7651138305664,
|
| 21 |
+
"learning_rate": 8.8e-07,
|
| 22 |
+
"loss": 9.4482,
|
| 23 |
+
"step": 50
|
| 24 |
+
},
|
| 25 |
+
{
|
| 26 |
+
"epoch": 0.09904258831297458,
|
| 27 |
+
"grad_norm": 64.81011199951172,
|
| 28 |
+
"learning_rate": 1.3800000000000001e-06,
|
| 29 |
+
"loss": 7.2641,
|
| 30 |
+
"step": 75
|
| 31 |
+
},
|
| 32 |
+
{
|
| 33 |
+
"epoch": 0.13205678441729943,
|
| 34 |
+
"grad_norm": 81.74808502197266,
|
| 35 |
+
"learning_rate": 1.8800000000000002e-06,
|
| 36 |
+
"loss": 5.5236,
|
| 37 |
+
"step": 100
|
| 38 |
+
},
|
| 39 |
+
{
|
| 40 |
+
"epoch": 0.1650709805216243,
|
| 41 |
+
"grad_norm": 93.43115234375,
|
| 42 |
+
"learning_rate": 2.38e-06,
|
| 43 |
+
"loss": 4.7369,
|
| 44 |
+
"step": 125
|
| 45 |
+
},
|
| 46 |
+
{
|
| 47 |
+
"epoch": 0.19808517662594916,
|
| 48 |
+
"grad_norm": 89.20719909667969,
|
| 49 |
+
"learning_rate": 2.88e-06,
|
| 50 |
+
"loss": 3.8232,
|
| 51 |
+
"step": 150
|
| 52 |
+
},
|
| 53 |
+
{
|
| 54 |
+
"epoch": 0.231099372730274,
|
| 55 |
+
"grad_norm": 64.15678405761719,
|
| 56 |
+
"learning_rate": 3.3800000000000007e-06,
|
| 57 |
+
"loss": 3.806,
|
| 58 |
+
"step": 175
|
| 59 |
+
},
|
| 60 |
+
{
|
| 61 |
+
"epoch": 0.26411356883459886,
|
| 62 |
+
"grad_norm": 64.91326904296875,
|
| 63 |
+
"learning_rate": 3.88e-06,
|
| 64 |
+
"loss": 3.1512,
|
| 65 |
+
"step": 200
|
| 66 |
+
},
|
| 67 |
+
{
|
| 68 |
+
"epoch": 0.26411356883459886,
|
| 69 |
+
"eval_loss": 0.8282185792922974,
|
| 70 |
+
"eval_runtime": 423.1335,
|
| 71 |
+
"eval_samples_per_second": 1.791,
|
| 72 |
+
"eval_steps_per_second": 0.896,
|
| 73 |
+
"eval_wer": 0.6533637400228051,
|
| 74 |
+
"step": 200
|
| 75 |
+
},
|
| 76 |
+
{
|
| 77 |
+
"epoch": 0.29712776493892373,
|
| 78 |
+
"grad_norm": 75.13513946533203,
|
| 79 |
+
"learning_rate": 4.38e-06,
|
| 80 |
+
"loss": 3.3773,
|
| 81 |
+
"step": 225
|
| 82 |
+
},
|
| 83 |
+
{
|
| 84 |
+
"epoch": 0.3301419610432486,
|
| 85 |
+
"grad_norm": 62.38654708862305,
|
| 86 |
+
"learning_rate": 4.880000000000001e-06,
|
| 87 |
+
"loss": 3.3841,
|
| 88 |
+
"step": 250
|
| 89 |
+
},
|
| 90 |
+
{
|
| 91 |
+
"epoch": 0.36315615714757343,
|
| 92 |
+
"grad_norm": 49.35546875,
|
| 93 |
+
"learning_rate": 5.380000000000001e-06,
|
| 94 |
+
"loss": 2.8367,
|
| 95 |
+
"step": 275
|
| 96 |
+
},
|
| 97 |
+
{
|
| 98 |
+
"epoch": 0.3961703532518983,
|
| 99 |
+
"grad_norm": 77.38804626464844,
|
| 100 |
+
"learning_rate": 5.8800000000000005e-06,
|
| 101 |
+
"loss": 2.9174,
|
| 102 |
+
"step": 300
|
| 103 |
+
},
|
| 104 |
+
{
|
| 105 |
+
"epoch": 0.4291845493562232,
|
| 106 |
+
"grad_norm": 78.50695037841797,
|
| 107 |
+
"learning_rate": 6.380000000000001e-06,
|
| 108 |
+
"loss": 2.6359,
|
| 109 |
+
"step": 325
|
| 110 |
+
},
|
| 111 |
+
{
|
| 112 |
+
"epoch": 0.462198745460548,
|
| 113 |
+
"grad_norm": 55.40019607543945,
|
| 114 |
+
"learning_rate": 6.88e-06,
|
| 115 |
+
"loss": 2.4471,
|
| 116 |
+
"step": 350
|
| 117 |
+
},
|
| 118 |
+
{
|
| 119 |
+
"epoch": 0.4952129415648729,
|
| 120 |
+
"grad_norm": 55.860809326171875,
|
| 121 |
+
"learning_rate": 7.3800000000000005e-06,
|
| 122 |
+
"loss": 2.6426,
|
| 123 |
+
"step": 375
|
| 124 |
+
},
|
| 125 |
+
{
|
| 126 |
+
"epoch": 0.5282271376691977,
|
| 127 |
+
"grad_norm": 51.24520492553711,
|
| 128 |
+
"learning_rate": 7.88e-06,
|
| 129 |
+
"loss": 2.4008,
|
| 130 |
+
"step": 400
|
| 131 |
+
},
|
| 132 |
+
{
|
| 133 |
+
"epoch": 0.5282271376691977,
|
| 134 |
+
"eval_loss": 0.6202582716941833,
|
| 135 |
+
"eval_runtime": 422.1868,
|
| 136 |
+
"eval_samples_per_second": 1.795,
|
| 137 |
+
"eval_steps_per_second": 0.898,
|
| 138 |
+
"eval_wer": 0.5096921322690992,
|
| 139 |
+
"step": 400
|
| 140 |
+
},
|
| 141 |
+
{
|
| 142 |
+
"epoch": 0.5612413337735226,
|
| 143 |
+
"grad_norm": 70.62708282470703,
|
| 144 |
+
"learning_rate": 8.380000000000001e-06,
|
| 145 |
+
"loss": 2.4166,
|
| 146 |
+
"step": 425
|
| 147 |
+
},
|
| 148 |
+
{
|
| 149 |
+
"epoch": 0.5942555298778475,
|
| 150 |
+
"grad_norm": 54.51411437988281,
|
| 151 |
+
"learning_rate": 8.880000000000001e-06,
|
| 152 |
+
"loss": 2.7424,
|
| 153 |
+
"step": 450
|
| 154 |
+
},
|
| 155 |
+
{
|
| 156 |
+
"epoch": 0.6272697259821723,
|
| 157 |
+
"grad_norm": 51.03986740112305,
|
| 158 |
+
"learning_rate": 9.38e-06,
|
| 159 |
+
"loss": 2.2441,
|
| 160 |
+
"step": 475
|
| 161 |
+
},
|
| 162 |
+
{
|
| 163 |
+
"epoch": 0.6602839220864972,
|
| 164 |
+
"grad_norm": 64.73434448242188,
|
| 165 |
+
"learning_rate": 9.88e-06,
|
| 166 |
+
"loss": 2.1702,
|
| 167 |
+
"step": 500
|
| 168 |
+
},
|
| 169 |
+
{
|
| 170 |
+
"epoch": 0.6932981181908221,
|
| 171 |
+
"grad_norm": 58.62995147705078,
|
| 172 |
+
"learning_rate": 9.957777777777779e-06,
|
| 173 |
+
"loss": 2.3904,
|
| 174 |
+
"step": 525
|
| 175 |
+
},
|
| 176 |
+
{
|
| 177 |
+
"epoch": 0.7263123142951469,
|
| 178 |
+
"grad_norm": 61.52521514892578,
|
| 179 |
+
"learning_rate": 9.902222222222223e-06,
|
| 180 |
+
"loss": 2.2456,
|
| 181 |
+
"step": 550
|
| 182 |
+
},
|
| 183 |
+
{
|
| 184 |
+
"epoch": 0.7593265103994717,
|
| 185 |
+
"grad_norm": 45.40699768066406,
|
| 186 |
+
"learning_rate": 9.846666666666668e-06,
|
| 187 |
+
"loss": 2.3398,
|
| 188 |
+
"step": 575
|
| 189 |
+
},
|
| 190 |
+
{
|
| 191 |
+
"epoch": 0.7923407065037966,
|
| 192 |
+
"grad_norm": 52.031795501708984,
|
| 193 |
+
"learning_rate": 9.791111111111112e-06,
|
| 194 |
+
"loss": 2.2459,
|
| 195 |
+
"step": 600
|
| 196 |
+
},
|
| 197 |
+
{
|
| 198 |
+
"epoch": 0.7923407065037966,
|
| 199 |
+
"eval_loss": 0.5506373643875122,
|
| 200 |
+
"eval_runtime": 424.3675,
|
| 201 |
+
"eval_samples_per_second": 1.786,
|
| 202 |
+
"eval_steps_per_second": 0.893,
|
| 203 |
+
"eval_wer": 0.4643671607753706,
|
| 204 |
+
"step": 600
|
| 205 |
+
},
|
| 206 |
+
{
|
| 207 |
+
"epoch": 0.8253549026081215,
|
| 208 |
+
"grad_norm": 51.51828384399414,
|
| 209 |
+
"learning_rate": 9.735555555555556e-06,
|
| 210 |
+
"loss": 2.1322,
|
| 211 |
+
"step": 625
|
| 212 |
+
},
|
| 213 |
+
{
|
| 214 |
+
"epoch": 0.8583690987124464,
|
| 215 |
+
"grad_norm": 47.97977066040039,
|
| 216 |
+
"learning_rate": 9.68e-06,
|
| 217 |
+
"loss": 2.3046,
|
| 218 |
+
"step": 650
|
| 219 |
+
},
|
| 220 |
+
{
|
| 221 |
+
"epoch": 0.8913832948167713,
|
| 222 |
+
"grad_norm": 47.63246154785156,
|
| 223 |
+
"learning_rate": 9.624444444444445e-06,
|
| 224 |
+
"loss": 2.0285,
|
| 225 |
+
"step": 675
|
| 226 |
+
},
|
| 227 |
+
{
|
| 228 |
+
"epoch": 0.924397490921096,
|
| 229 |
+
"grad_norm": 36.78068161010742,
|
| 230 |
+
"learning_rate": 9.56888888888889e-06,
|
| 231 |
+
"loss": 1.9158,
|
| 232 |
+
"step": 700
|
| 233 |
+
},
|
| 234 |
+
{
|
| 235 |
+
"epoch": 0.9574116870254209,
|
| 236 |
+
"grad_norm": 57.10271453857422,
|
| 237 |
+
"learning_rate": 9.513333333333334e-06,
|
| 238 |
+
"loss": 2.034,
|
| 239 |
+
"step": 725
|
| 240 |
+
},
|
| 241 |
+
{
|
| 242 |
+
"epoch": 0.9904258831297458,
|
| 243 |
+
"grad_norm": 35.37094497680664,
|
| 244 |
+
"learning_rate": 9.457777777777778e-06,
|
| 245 |
+
"loss": 1.92,
|
| 246 |
+
"step": 750
|
| 247 |
+
},
|
| 248 |
+
{
|
| 249 |
+
"epoch": 1.0224496533509408,
|
| 250 |
+
"grad_norm": 59.17182159423828,
|
| 251 |
+
"learning_rate": 9.402222222222222e-06,
|
| 252 |
+
"loss": 1.5202,
|
| 253 |
+
"step": 775
|
| 254 |
+
},
|
| 255 |
+
{
|
| 256 |
+
"epoch": 1.0554638494552657,
|
| 257 |
+
"grad_norm": 48.809059143066406,
|
| 258 |
+
"learning_rate": 9.346666666666666e-06,
|
| 259 |
+
"loss": 1.3319,
|
| 260 |
+
"step": 800
|
| 261 |
+
},
|
| 262 |
+
{
|
| 263 |
+
"epoch": 1.0554638494552657,
|
| 264 |
+
"eval_loss": 0.5028851628303528,
|
| 265 |
+
"eval_runtime": 423.0433,
|
| 266 |
+
"eval_samples_per_second": 1.792,
|
| 267 |
+
"eval_steps_per_second": 0.896,
|
| 268 |
+
"eval_wer": 0.4016533637400228,
|
| 269 |
+
"step": 800
|
| 270 |
+
},
|
| 271 |
+
{
|
| 272 |
+
"epoch": 1.0884780455595906,
|
| 273 |
+
"grad_norm": 59.115562438964844,
|
| 274 |
+
"learning_rate": 9.291111111111112e-06,
|
| 275 |
+
"loss": 1.2718,
|
| 276 |
+
"step": 825
|
| 277 |
+
},
|
| 278 |
+
{
|
| 279 |
+
"epoch": 1.1214922416639155,
|
| 280 |
+
"grad_norm": 22.62116050720215,
|
| 281 |
+
"learning_rate": 9.235555555555556e-06,
|
| 282 |
+
"loss": 1.4598,
|
| 283 |
+
"step": 850
|
| 284 |
+
},
|
| 285 |
+
{
|
| 286 |
+
"epoch": 1.1545064377682404,
|
| 287 |
+
"grad_norm": 37.58869934082031,
|
| 288 |
+
"learning_rate": 9.180000000000002e-06,
|
| 289 |
+
"loss": 1.33,
|
| 290 |
+
"step": 875
|
| 291 |
+
},
|
| 292 |
+
{
|
| 293 |
+
"epoch": 1.1875206338725652,
|
| 294 |
+
"grad_norm": 41.18476867675781,
|
| 295 |
+
"learning_rate": 9.124444444444444e-06,
|
| 296 |
+
"loss": 1.4802,
|
| 297 |
+
"step": 900
|
| 298 |
+
},
|
| 299 |
+
{
|
| 300 |
+
"epoch": 1.2205348299768901,
|
| 301 |
+
"grad_norm": 34.091915130615234,
|
| 302 |
+
"learning_rate": 9.06888888888889e-06,
|
| 303 |
+
"loss": 1.325,
|
| 304 |
+
"step": 925
|
| 305 |
+
},
|
| 306 |
+
{
|
| 307 |
+
"epoch": 1.253549026081215,
|
| 308 |
+
"grad_norm": 43.94548416137695,
|
| 309 |
+
"learning_rate": 9.013333333333334e-06,
|
| 310 |
+
"loss": 1.3758,
|
| 311 |
+
"step": 950
|
| 312 |
+
},
|
| 313 |
+
{
|
| 314 |
+
"epoch": 1.2865632221855399,
|
| 315 |
+
"grad_norm": 50.9454231262207,
|
| 316 |
+
"learning_rate": 8.957777777777778e-06,
|
| 317 |
+
"loss": 1.4197,
|
| 318 |
+
"step": 975
|
| 319 |
+
},
|
| 320 |
+
{
|
| 321 |
+
"epoch": 1.3195774182898647,
|
| 322 |
+
"grad_norm": 39.400634765625,
|
| 323 |
+
"learning_rate": 8.902222222222224e-06,
|
| 324 |
+
"loss": 1.5588,
|
| 325 |
+
"step": 1000
|
| 326 |
+
},
|
| 327 |
+
{
|
| 328 |
+
"epoch": 1.3195774182898647,
|
| 329 |
+
"eval_loss": 0.4675042927265167,
|
| 330 |
+
"eval_runtime": 426.8283,
|
| 331 |
+
"eval_samples_per_second": 1.776,
|
| 332 |
+
"eval_steps_per_second": 0.888,
|
| 333 |
+
"eval_wer": 0.3896807297605473,
|
| 334 |
+
"step": 1000
|
| 335 |
+
},
|
| 336 |
+
{
|
| 337 |
+
"epoch": 1.3525916143941896,
|
| 338 |
+
"grad_norm": 64.05392456054688,
|
| 339 |
+
"learning_rate": 8.846666666666668e-06,
|
| 340 |
+
"loss": 1.4914,
|
| 341 |
+
"step": 1025
|
| 342 |
+
},
|
| 343 |
+
{
|
| 344 |
+
"epoch": 1.3856058104985143,
|
| 345 |
+
"grad_norm": 38.36800765991211,
|
| 346 |
+
"learning_rate": 8.791111111111112e-06,
|
| 347 |
+
"loss": 1.3193,
|
| 348 |
+
"step": 1050
|
| 349 |
+
},
|
| 350 |
+
{
|
| 351 |
+
"epoch": 1.4186200066028392,
|
| 352 |
+
"grad_norm": 56.90019226074219,
|
| 353 |
+
"learning_rate": 8.735555555555556e-06,
|
| 354 |
+
"loss": 1.3192,
|
| 355 |
+
"step": 1075
|
| 356 |
+
},
|
| 357 |
+
{
|
| 358 |
+
"epoch": 1.451634202707164,
|
| 359 |
+
"grad_norm": 41.730682373046875,
|
| 360 |
+
"learning_rate": 8.68e-06,
|
| 361 |
+
"loss": 1.4852,
|
| 362 |
+
"step": 1100
|
| 363 |
+
},
|
| 364 |
+
{
|
| 365 |
+
"epoch": 1.484648398811489,
|
| 366 |
+
"grad_norm": 37.495849609375,
|
| 367 |
+
"learning_rate": 8.624444444444446e-06,
|
| 368 |
+
"loss": 1.244,
|
| 369 |
+
"step": 1125
|
| 370 |
+
},
|
| 371 |
+
{
|
| 372 |
+
"epoch": 1.5176625949158138,
|
| 373 |
+
"grad_norm": 34.61660385131836,
|
| 374 |
+
"learning_rate": 8.56888888888889e-06,
|
| 375 |
+
"loss": 1.3055,
|
| 376 |
+
"step": 1150
|
| 377 |
+
},
|
| 378 |
+
{
|
| 379 |
+
"epoch": 1.5506767910201387,
|
| 380 |
+
"grad_norm": 32.87508773803711,
|
| 381 |
+
"learning_rate": 8.513333333333335e-06,
|
| 382 |
+
"loss": 1.4434,
|
| 383 |
+
"step": 1175
|
| 384 |
+
},
|
| 385 |
+
{
|
| 386 |
+
"epoch": 1.5836909871244635,
|
| 387 |
+
"grad_norm": 43.57270812988281,
|
| 388 |
+
"learning_rate": 8.457777777777778e-06,
|
| 389 |
+
"loss": 1.2908,
|
| 390 |
+
"step": 1200
|
| 391 |
+
},
|
| 392 |
+
{
|
| 393 |
+
"epoch": 1.5836909871244635,
|
| 394 |
+
"eval_loss": 0.45341047644615173,
|
| 395 |
+
"eval_runtime": 422.1999,
|
| 396 |
+
"eval_samples_per_second": 1.795,
|
| 397 |
+
"eval_steps_per_second": 0.898,
|
| 398 |
+
"eval_wer": 0.3727194982896237,
|
| 399 |
+
"step": 1200
|
| 400 |
+
},
|
| 401 |
+
{
|
| 402 |
+
"epoch": 1.6167051832287884,
|
| 403 |
+
"grad_norm": 46.83863830566406,
|
| 404 |
+
"learning_rate": 8.402222222222223e-06,
|
| 405 |
+
"loss": 1.3848,
|
| 406 |
+
"step": 1225
|
| 407 |
+
},
|
| 408 |
+
{
|
| 409 |
+
"epoch": 1.649719379333113,
|
| 410 |
+
"grad_norm": 46.93707275390625,
|
| 411 |
+
"learning_rate": 8.346666666666668e-06,
|
| 412 |
+
"loss": 1.2536,
|
| 413 |
+
"step": 1250
|
| 414 |
+
},
|
| 415 |
+
{
|
| 416 |
+
"epoch": 1.682733575437438,
|
| 417 |
+
"grad_norm": 44.47553253173828,
|
| 418 |
+
"learning_rate": 8.291111111111112e-06,
|
| 419 |
+
"loss": 1.3358,
|
| 420 |
+
"step": 1275
|
| 421 |
+
},
|
| 422 |
+
{
|
| 423 |
+
"epoch": 1.7157477715417628,
|
| 424 |
+
"grad_norm": 35.15113830566406,
|
| 425 |
+
"learning_rate": 8.235555555555557e-06,
|
| 426 |
+
"loss": 1.1214,
|
| 427 |
+
"step": 1300
|
| 428 |
+
},
|
| 429 |
+
{
|
| 430 |
+
"epoch": 1.7487619676460877,
|
| 431 |
+
"grad_norm": 27.538331985473633,
|
| 432 |
+
"learning_rate": 8.18e-06,
|
| 433 |
+
"loss": 1.097,
|
| 434 |
+
"step": 1325
|
| 435 |
+
},
|
| 436 |
+
{
|
| 437 |
+
"epoch": 1.7817761637504126,
|
| 438 |
+
"grad_norm": 35.05278015136719,
|
| 439 |
+
"learning_rate": 8.124444444444445e-06,
|
| 440 |
+
"loss": 1.1746,
|
| 441 |
+
"step": 1350
|
| 442 |
+
},
|
| 443 |
+
{
|
| 444 |
+
"epoch": 1.8147903598547375,
|
| 445 |
+
"grad_norm": 50.673095703125,
|
| 446 |
+
"learning_rate": 8.06888888888889e-06,
|
| 447 |
+
"loss": 1.397,
|
| 448 |
+
"step": 1375
|
| 449 |
+
},
|
| 450 |
+
{
|
| 451 |
+
"epoch": 1.8478045559590623,
|
| 452 |
+
"grad_norm": 38.40850067138672,
|
| 453 |
+
"learning_rate": 8.013333333333333e-06,
|
| 454 |
+
"loss": 1.4258,
|
| 455 |
+
"step": 1400
|
| 456 |
+
},
|
| 457 |
+
{
|
| 458 |
+
"epoch": 1.8478045559590623,
|
| 459 |
+
"eval_loss": 0.4354272186756134,
|
| 460 |
+
"eval_runtime": 431.6561,
|
| 461 |
+
"eval_samples_per_second": 1.756,
|
| 462 |
+
"eval_steps_per_second": 0.878,
|
| 463 |
+
"eval_wer": 0.38982326111744586,
|
| 464 |
+
"step": 1400
|
| 465 |
+
},
|
| 466 |
+
{
|
| 467 |
+
"epoch": 1.8808187520633872,
|
| 468 |
+
"grad_norm": 27.05866050720215,
|
| 469 |
+
"learning_rate": 7.957777777777779e-06,
|
| 470 |
+
"loss": 1.4042,
|
| 471 |
+
"step": 1425
|
| 472 |
+
},
|
| 473 |
+
{
|
| 474 |
+
"epoch": 1.913832948167712,
|
| 475 |
+
"grad_norm": 36.17206573486328,
|
| 476 |
+
"learning_rate": 7.902222222222223e-06,
|
| 477 |
+
"loss": 1.0702,
|
| 478 |
+
"step": 1450
|
| 479 |
+
},
|
| 480 |
+
{
|
| 481 |
+
"epoch": 1.946847144272037,
|
| 482 |
+
"grad_norm": 40.80569076538086,
|
| 483 |
+
"learning_rate": 7.846666666666667e-06,
|
| 484 |
+
"loss": 1.2552,
|
| 485 |
+
"step": 1475
|
| 486 |
+
},
|
| 487 |
+
{
|
| 488 |
+
"epoch": 1.9798613403763619,
|
| 489 |
+
"grad_norm": 34.74424362182617,
|
| 490 |
+
"learning_rate": 7.791111111111111e-06,
|
| 491 |
+
"loss": 1.349,
|
| 492 |
+
"step": 1500
|
| 493 |
+
},
|
| 494 |
+
{
|
| 495 |
+
"epoch": 2.011885110597557,
|
| 496 |
+
"grad_norm": 18.330013275146484,
|
| 497 |
+
"learning_rate": 7.735555555555557e-06,
|
| 498 |
+
"loss": 1.0107,
|
| 499 |
+
"step": 1525
|
| 500 |
+
},
|
| 501 |
+
{
|
| 502 |
+
"epoch": 2.0448993067018817,
|
| 503 |
+
"grad_norm": 20.316329956054688,
|
| 504 |
+
"learning_rate": 7.680000000000001e-06,
|
| 505 |
+
"loss": 0.6744,
|
| 506 |
+
"step": 1550
|
| 507 |
+
},
|
| 508 |
+
{
|
| 509 |
+
"epoch": 2.0779135028062066,
|
| 510 |
+
"grad_norm": 56.3316764831543,
|
| 511 |
+
"learning_rate": 7.624444444444445e-06,
|
| 512 |
+
"loss": 0.6948,
|
| 513 |
+
"step": 1575
|
| 514 |
+
},
|
| 515 |
+
{
|
| 516 |
+
"epoch": 2.1109276989105314,
|
| 517 |
+
"grad_norm": 28.286052703857422,
|
| 518 |
+
"learning_rate": 7.56888888888889e-06,
|
| 519 |
+
"loss": 0.6383,
|
| 520 |
+
"step": 1600
|
| 521 |
+
},
|
| 522 |
+
{
|
| 523 |
+
"epoch": 2.1109276989105314,
|
| 524 |
+
"eval_loss": 0.44802892208099365,
|
| 525 |
+
"eval_runtime": 429.0328,
|
| 526 |
+
"eval_samples_per_second": 1.767,
|
| 527 |
+
"eval_steps_per_second": 0.883,
|
| 528 |
+
"eval_wer": 0.3600342075256556,
|
| 529 |
+
"step": 1600
|
| 530 |
+
},
|
| 531 |
+
{
|
| 532 |
+
"epoch": 2.1439418950148563,
|
| 533 |
+
"grad_norm": 33.71493148803711,
|
| 534 |
+
"learning_rate": 7.513333333333334e-06,
|
| 535 |
+
"loss": 0.5558,
|
| 536 |
+
"step": 1625
|
| 537 |
+
},
|
| 538 |
+
{
|
| 539 |
+
"epoch": 2.176956091119181,
|
| 540 |
+
"grad_norm": 29.385910034179688,
|
| 541 |
+
"learning_rate": 7.457777777777778e-06,
|
| 542 |
+
"loss": 0.6777,
|
| 543 |
+
"step": 1650
|
| 544 |
+
},
|
| 545 |
+
{
|
| 546 |
+
"epoch": 2.209970287223506,
|
| 547 |
+
"grad_norm": 20.755077362060547,
|
| 548 |
+
"learning_rate": 7.402222222222223e-06,
|
| 549 |
+
"loss": 0.6332,
|
| 550 |
+
"step": 1675
|
| 551 |
+
},
|
| 552 |
+
{
|
| 553 |
+
"epoch": 2.242984483327831,
|
| 554 |
+
"grad_norm": 37.17219543457031,
|
| 555 |
+
"learning_rate": 7.346666666666668e-06,
|
| 556 |
+
"loss": 0.6242,
|
| 557 |
+
"step": 1700
|
| 558 |
+
},
|
| 559 |
+
{
|
| 560 |
+
"epoch": 2.275998679432156,
|
| 561 |
+
"grad_norm": 37.963409423828125,
|
| 562 |
+
"learning_rate": 7.291111111111112e-06,
|
| 563 |
+
"loss": 0.7322,
|
| 564 |
+
"step": 1725
|
| 565 |
+
},
|
| 566 |
+
{
|
| 567 |
+
"epoch": 2.3090128755364807,
|
| 568 |
+
"grad_norm": 25.863115310668945,
|
| 569 |
+
"learning_rate": 7.235555555555556e-06,
|
| 570 |
+
"loss": 0.697,
|
| 571 |
+
"step": 1750
|
| 572 |
+
},
|
| 573 |
+
{
|
| 574 |
+
"epoch": 2.3420270716408056,
|
| 575 |
+
"grad_norm": 25.60062026977539,
|
| 576 |
+
"learning_rate": 7.180000000000001e-06,
|
| 577 |
+
"loss": 0.6266,
|
| 578 |
+
"step": 1775
|
| 579 |
+
},
|
| 580 |
+
{
|
| 581 |
+
"epoch": 2.3750412677451305,
|
| 582 |
+
"grad_norm": 20.695463180541992,
|
| 583 |
+
"learning_rate": 7.124444444444445e-06,
|
| 584 |
+
"loss": 0.6079,
|
| 585 |
+
"step": 1800
|
| 586 |
+
},
|
| 587 |
+
{
|
| 588 |
+
"epoch": 2.3750412677451305,
|
| 589 |
+
"eval_loss": 0.4444292187690735,
|
| 590 |
+
"eval_runtime": 421.917,
|
| 591 |
+
"eval_samples_per_second": 1.797,
|
| 592 |
+
"eval_steps_per_second": 0.898,
|
| 593 |
+
"eval_wer": 0.3482041049030787,
|
| 594 |
+
"step": 1800
|
| 595 |
+
},
|
| 596 |
+
{
|
| 597 |
+
"epoch": 2.4080554638494553,
|
| 598 |
+
"grad_norm": 26.363622665405273,
|
| 599 |
+
"learning_rate": 7.06888888888889e-06,
|
| 600 |
+
"loss": 0.663,
|
| 601 |
+
"step": 1825
|
| 602 |
+
},
|
| 603 |
+
{
|
| 604 |
+
"epoch": 2.4410696599537802,
|
| 605 |
+
"grad_norm": 31.212791442871094,
|
| 606 |
+
"learning_rate": 7.0133333333333345e-06,
|
| 607 |
+
"loss": 0.655,
|
| 608 |
+
"step": 1850
|
| 609 |
+
},
|
| 610 |
+
{
|
| 611 |
+
"epoch": 2.474083856058105,
|
| 612 |
+
"grad_norm": 39.38019561767578,
|
| 613 |
+
"learning_rate": 6.9577777777777785e-06,
|
| 614 |
+
"loss": 0.7364,
|
| 615 |
+
"step": 1875
|
| 616 |
+
},
|
| 617 |
+
{
|
| 618 |
+
"epoch": 2.50709805216243,
|
| 619 |
+
"grad_norm": 26.713680267333984,
|
| 620 |
+
"learning_rate": 6.902222222222223e-06,
|
| 621 |
+
"loss": 0.647,
|
| 622 |
+
"step": 1900
|
| 623 |
+
},
|
| 624 |
+
{
|
| 625 |
+
"epoch": 2.540112248266755,
|
| 626 |
+
"grad_norm": 24.708005905151367,
|
| 627 |
+
"learning_rate": 6.846666666666667e-06,
|
| 628 |
+
"loss": 0.6818,
|
| 629 |
+
"step": 1925
|
| 630 |
+
},
|
| 631 |
+
{
|
| 632 |
+
"epoch": 2.5731264443710797,
|
| 633 |
+
"grad_norm": 22.732967376708984,
|
| 634 |
+
"learning_rate": 6.7911111111111115e-06,
|
| 635 |
+
"loss": 0.6711,
|
| 636 |
+
"step": 1950
|
| 637 |
+
},
|
| 638 |
+
{
|
| 639 |
+
"epoch": 2.606140640475404,
|
| 640 |
+
"grad_norm": 23.73015022277832,
|
| 641 |
+
"learning_rate": 6.735555555555556e-06,
|
| 642 |
+
"loss": 0.6492,
|
| 643 |
+
"step": 1975
|
| 644 |
+
},
|
| 645 |
+
{
|
| 646 |
+
"epoch": 2.6391548365797295,
|
| 647 |
+
"grad_norm": 30.428665161132812,
|
| 648 |
+
"learning_rate": 6.680000000000001e-06,
|
| 649 |
+
"loss": 0.5709,
|
| 650 |
+
"step": 2000
|
| 651 |
+
},
|
| 652 |
+
{
|
| 653 |
+
"epoch": 2.6391548365797295,
|
| 654 |
+
"eval_loss": 0.4367344081401825,
|
| 655 |
+
"eval_runtime": 423.1943,
|
| 656 |
+
"eval_samples_per_second": 1.791,
|
| 657 |
+
"eval_steps_per_second": 0.896,
|
| 658 |
+
"eval_wer": 0.3405074116305587,
|
| 659 |
+
"step": 2000
|
| 660 |
+
},
|
| 661 |
+
{
|
| 662 |
+
"epoch": 2.6391548365797295,
|
| 663 |
+
"step": 2000,
|
| 664 |
+
"total_flos": 1.631539005161472e+19,
|
| 665 |
+
"train_loss": 1.9525959782600404,
|
| 666 |
+
"train_runtime": 8300.7195,
|
| 667 |
+
"train_samples_per_second": 4.819,
|
| 668 |
+
"train_steps_per_second": 0.602
|
| 669 |
+
}
|
| 670 |
+
],
|
| 671 |
+
"logging_steps": 25,
|
| 672 |
+
"max_steps": 5000,
|
| 673 |
+
"num_input_tokens_seen": 0,
|
| 674 |
+
"num_train_epochs": 7,
|
| 675 |
+
"save_steps": 200,
|
| 676 |
+
"stateful_callbacks": {
|
| 677 |
+
"EarlyStoppingCallback": {
|
| 678 |
+
"args": {
|
| 679 |
+
"early_stopping_patience": 3,
|
| 680 |
+
"early_stopping_threshold": 0.0
|
| 681 |
+
},
|
| 682 |
+
"attributes": {
|
| 683 |
+
"early_stopping_patience_counter": 3
|
| 684 |
+
}
|
| 685 |
+
},
|
| 686 |
+
"TrainerControl": {
|
| 687 |
+
"args": {
|
| 688 |
+
"should_epoch_stop": false,
|
| 689 |
+
"should_evaluate": false,
|
| 690 |
+
"should_log": false,
|
| 691 |
+
"should_save": true,
|
| 692 |
+
"should_training_stop": true
|
| 693 |
+
},
|
| 694 |
+
"attributes": {}
|
| 695 |
+
}
|
| 696 |
+
},
|
| 697 |
+
"total_flos": 1.631539005161472e+19,
|
| 698 |
+
"train_batch_size": 2,
|
| 699 |
+
"trial_name": null,
|
| 700 |
+
"trial_params": null
|
| 701 |
+
}
|