End of training
Browse files- README.md +6 -3
- adapter.default.safetensors +3 -0
- all_results.json +15 -0
- eval_results.json +9 -0
- runs/Aug30_13-54-36_a6763744c18f/events.out.tfevents.1725039891.a6763744c18f.11754.1 +3 -0
- train_results.json +9 -0
- trainer_state.json +573 -0
README.md
CHANGED
|
@@ -3,6 +3,9 @@ library_name: transformers
|
|
| 3 |
license: cc-by-nc-4.0
|
| 4 |
base_model: mms-meta/mms-zeroshot-300m
|
| 5 |
tags:
|
|
|
|
|
|
|
|
|
|
| 6 |
- generated_from_trainer
|
| 7 |
metrics:
|
| 8 |
- wer
|
|
@@ -16,10 +19,10 @@ should probably proofread and complete it, then remove this comment. -->
|
|
| 16 |
|
| 17 |
# mms-zeroshot-bem-sv-male
|
| 18 |
|
| 19 |
-
This model is a fine-tuned version of [mms-meta/mms-zeroshot-300m](https://huggingface.co/mms-meta/mms-zeroshot-300m) on
|
| 20 |
It achieves the following results on the evaluation set:
|
| 21 |
-
- Loss: 0.
|
| 22 |
-
- Wer: 0.
|
| 23 |
|
| 24 |
## Model description
|
| 25 |
|
|
|
|
| 3 |
license: cc-by-nc-4.0
|
| 4 |
base_model: mms-meta/mms-zeroshot-300m
|
| 5 |
tags:
|
| 6 |
+
- automatic-speech-recognition
|
| 7 |
+
- BembaSpeech
|
| 8 |
+
- mms
|
| 9 |
- generated_from_trainer
|
| 10 |
metrics:
|
| 11 |
- wer
|
|
|
|
| 19 |
|
| 20 |
# mms-zeroshot-bem-sv-male
|
| 21 |
|
| 22 |
+
This model is a fine-tuned version of [mms-meta/mms-zeroshot-300m](https://huggingface.co/mms-meta/mms-zeroshot-300m) on the BEMBASPEECH - BEM dataset.
|
| 23 |
It achieves the following results on the evaluation set:
|
| 24 |
+
- Loss: 0.1874
|
| 25 |
+
- Wer: 0.3949
|
| 26 |
|
| 27 |
## Model description
|
| 28 |
|
adapter.default.safetensors
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:939070ea2fcc5a030458d9c6352e327223c626a06fc0f614a61ddcfc143978cd
|
| 3 |
+
size 3586804
|
all_results.json
ADDED
|
@@ -0,0 +1,15 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"epoch": 10.0,
|
| 3 |
+
"eval_loss": 0.18741655349731445,
|
| 4 |
+
"eval_runtime": 76.5228,
|
| 5 |
+
"eval_samples": 1017,
|
| 6 |
+
"eval_samples_per_second": 13.29,
|
| 7 |
+
"eval_steps_per_second": 1.673,
|
| 8 |
+
"eval_wer": 0.394916142557652,
|
| 9 |
+
"total_flos": 1.4294624233263563e+19,
|
| 10 |
+
"train_loss": 0.6386700151268572,
|
| 11 |
+
"train_runtime": 13601.4767,
|
| 12 |
+
"train_samples": 7327,
|
| 13 |
+
"train_samples_per_second": 5.387,
|
| 14 |
+
"train_steps_per_second": 0.673
|
| 15 |
+
}
|
eval_results.json
ADDED
|
@@ -0,0 +1,9 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"epoch": 10.0,
|
| 3 |
+
"eval_loss": 0.18741655349731445,
|
| 4 |
+
"eval_runtime": 76.5228,
|
| 5 |
+
"eval_samples": 1017,
|
| 6 |
+
"eval_samples_per_second": 13.29,
|
| 7 |
+
"eval_steps_per_second": 1.673,
|
| 8 |
+
"eval_wer": 0.394916142557652
|
| 9 |
+
}
|
runs/Aug30_13-54-36_a6763744c18f/events.out.tfevents.1725039891.a6763744c18f.11754.1
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:f053b33244d77d16cae729ef418865282c3689db3d3e13e355811ba4826d9f5c
|
| 3 |
+
size 406
|
train_results.json
ADDED
|
@@ -0,0 +1,9 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"epoch": 10.0,
|
| 3 |
+
"total_flos": 1.4294624233263563e+19,
|
| 4 |
+
"train_loss": 0.6386700151268572,
|
| 5 |
+
"train_runtime": 13601.4767,
|
| 6 |
+
"train_samples": 7327,
|
| 7 |
+
"train_samples_per_second": 5.387,
|
| 8 |
+
"train_steps_per_second": 0.673
|
| 9 |
+
}
|
trainer_state.json
ADDED
|
@@ -0,0 +1,573 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"best_metric": null,
|
| 3 |
+
"best_model_checkpoint": null,
|
| 4 |
+
"epoch": 10.0,
|
| 5 |
+
"eval_steps": 200,
|
| 6 |
+
"global_step": 9160,
|
| 7 |
+
"is_hyper_param_search": false,
|
| 8 |
+
"is_local_process_zero": true,
|
| 9 |
+
"is_world_process_zero": true,
|
| 10 |
+
"log_history": [
|
| 11 |
+
{
|
| 12 |
+
"epoch": 0.2183406113537118,
|
| 13 |
+
"eval_loss": 2.382188320159912,
|
| 14 |
+
"eval_runtime": 74.685,
|
| 15 |
+
"eval_samples_per_second": 13.617,
|
| 16 |
+
"eval_steps_per_second": 1.714,
|
| 17 |
+
"eval_wer": 1.0,
|
| 18 |
+
"step": 200
|
| 19 |
+
},
|
| 20 |
+
{
|
| 21 |
+
"epoch": 0.4366812227074236,
|
| 22 |
+
"eval_loss": 0.27150478959083557,
|
| 23 |
+
"eval_runtime": 75.1491,
|
| 24 |
+
"eval_samples_per_second": 13.533,
|
| 25 |
+
"eval_steps_per_second": 1.703,
|
| 26 |
+
"eval_wer": 0.5093029350104822,
|
| 27 |
+
"step": 400
|
| 28 |
+
},
|
| 29 |
+
{
|
| 30 |
+
"epoch": 0.5458515283842795,
|
| 31 |
+
"grad_norm": 0.9615474939346313,
|
| 32 |
+
"learning_rate": 0.0002868211920529801,
|
| 33 |
+
"loss": 2.7769,
|
| 34 |
+
"step": 500
|
| 35 |
+
},
|
| 36 |
+
{
|
| 37 |
+
"epoch": 0.6550218340611353,
|
| 38 |
+
"eval_loss": 0.24891583621501923,
|
| 39 |
+
"eval_runtime": 75.3374,
|
| 40 |
+
"eval_samples_per_second": 13.499,
|
| 41 |
+
"eval_steps_per_second": 1.699,
|
| 42 |
+
"eval_wer": 0.4820492662473795,
|
| 43 |
+
"step": 600
|
| 44 |
+
},
|
| 45 |
+
{
|
| 46 |
+
"epoch": 0.8733624454148472,
|
| 47 |
+
"eval_loss": 0.2296326756477356,
|
| 48 |
+
"eval_runtime": 74.8314,
|
| 49 |
+
"eval_samples_per_second": 13.591,
|
| 50 |
+
"eval_steps_per_second": 1.711,
|
| 51 |
+
"eval_wer": 0.46947064989517817,
|
| 52 |
+
"step": 800
|
| 53 |
+
},
|
| 54 |
+
{
|
| 55 |
+
"epoch": 1.091703056768559,
|
| 56 |
+
"grad_norm": 0.819220244884491,
|
| 57 |
+
"learning_rate": 0.00027026490066225163,
|
| 58 |
+
"loss": 0.6809,
|
| 59 |
+
"step": 1000
|
| 60 |
+
},
|
| 61 |
+
{
|
| 62 |
+
"epoch": 1.091703056768559,
|
| 63 |
+
"eval_loss": 0.22090712189674377,
|
| 64 |
+
"eval_runtime": 75.0858,
|
| 65 |
+
"eval_samples_per_second": 13.544,
|
| 66 |
+
"eval_steps_per_second": 1.705,
|
| 67 |
+
"eval_wer": 0.4638364779874214,
|
| 68 |
+
"step": 1000
|
| 69 |
+
},
|
| 70 |
+
{
|
| 71 |
+
"epoch": 1.3100436681222707,
|
| 72 |
+
"eval_loss": 0.21630799770355225,
|
| 73 |
+
"eval_runtime": 75.4504,
|
| 74 |
+
"eval_samples_per_second": 13.479,
|
| 75 |
+
"eval_steps_per_second": 1.696,
|
| 76 |
+
"eval_wer": 0.44693396226415094,
|
| 77 |
+
"step": 1200
|
| 78 |
+
},
|
| 79 |
+
{
|
| 80 |
+
"epoch": 1.5283842794759825,
|
| 81 |
+
"eval_loss": 0.20916949212551117,
|
| 82 |
+
"eval_runtime": 75.3429,
|
| 83 |
+
"eval_samples_per_second": 13.498,
|
| 84 |
+
"eval_steps_per_second": 1.699,
|
| 85 |
+
"eval_wer": 0.4399895178197065,
|
| 86 |
+
"step": 1400
|
| 87 |
+
},
|
| 88 |
+
{
|
| 89 |
+
"epoch": 1.6375545851528384,
|
| 90 |
+
"grad_norm": 0.9643642902374268,
|
| 91 |
+
"learning_rate": 0.00025370860927152316,
|
| 92 |
+
"loss": 0.6113,
|
| 93 |
+
"step": 1500
|
| 94 |
+
},
|
| 95 |
+
{
|
| 96 |
+
"epoch": 1.7467248908296944,
|
| 97 |
+
"eval_loss": 0.20470178127288818,
|
| 98 |
+
"eval_runtime": 74.8687,
|
| 99 |
+
"eval_samples_per_second": 13.584,
|
| 100 |
+
"eval_steps_per_second": 1.71,
|
| 101 |
+
"eval_wer": 0.4346174004192872,
|
| 102 |
+
"step": 1600
|
| 103 |
+
},
|
| 104 |
+
{
|
| 105 |
+
"epoch": 1.965065502183406,
|
| 106 |
+
"eval_loss": 0.20738434791564941,
|
| 107 |
+
"eval_runtime": 75.3186,
|
| 108 |
+
"eval_samples_per_second": 13.503,
|
| 109 |
+
"eval_steps_per_second": 1.699,
|
| 110 |
+
"eval_wer": 0.4466719077568134,
|
| 111 |
+
"step": 1800
|
| 112 |
+
},
|
| 113 |
+
{
|
| 114 |
+
"epoch": 2.183406113537118,
|
| 115 |
+
"grad_norm": 0.6018229126930237,
|
| 116 |
+
"learning_rate": 0.00023715231788079468,
|
| 117 |
+
"loss": 0.5974,
|
| 118 |
+
"step": 2000
|
| 119 |
+
},
|
| 120 |
+
{
|
| 121 |
+
"epoch": 2.183406113537118,
|
| 122 |
+
"eval_loss": 0.20408853888511658,
|
| 123 |
+
"eval_runtime": 75.0651,
|
| 124 |
+
"eval_samples_per_second": 13.548,
|
| 125 |
+
"eval_steps_per_second": 1.705,
|
| 126 |
+
"eval_wer": 0.43042452830188677,
|
| 127 |
+
"step": 2000
|
| 128 |
+
},
|
| 129 |
+
{
|
| 130 |
+
"epoch": 2.4017467248908297,
|
| 131 |
+
"eval_loss": 0.20539002120494843,
|
| 132 |
+
"eval_runtime": 74.9069,
|
| 133 |
+
"eval_samples_per_second": 13.577,
|
| 134 |
+
"eval_steps_per_second": 1.709,
|
| 135 |
+
"eval_wer": 0.4317348008385744,
|
| 136 |
+
"step": 2200
|
| 137 |
+
},
|
| 138 |
+
{
|
| 139 |
+
"epoch": 2.6200873362445414,
|
| 140 |
+
"eval_loss": 0.19870109856128693,
|
| 141 |
+
"eval_runtime": 75.2079,
|
| 142 |
+
"eval_samples_per_second": 13.523,
|
| 143 |
+
"eval_steps_per_second": 1.702,
|
| 144 |
+
"eval_wer": 0.4240041928721174,
|
| 145 |
+
"step": 2400
|
| 146 |
+
},
|
| 147 |
+
{
|
| 148 |
+
"epoch": 2.7292576419213974,
|
| 149 |
+
"grad_norm": 0.7469688653945923,
|
| 150 |
+
"learning_rate": 0.0002205960264900662,
|
| 151 |
+
"loss": 0.5636,
|
| 152 |
+
"step": 2500
|
| 153 |
+
},
|
| 154 |
+
{
|
| 155 |
+
"epoch": 2.8384279475982535,
|
| 156 |
+
"eval_loss": 0.20032186806201935,
|
| 157 |
+
"eval_runtime": 75.3206,
|
| 158 |
+
"eval_samples_per_second": 13.502,
|
| 159 |
+
"eval_steps_per_second": 1.699,
|
| 160 |
+
"eval_wer": 0.42518343815513626,
|
| 161 |
+
"step": 2600
|
| 162 |
+
},
|
| 163 |
+
{
|
| 164 |
+
"epoch": 3.056768558951965,
|
| 165 |
+
"eval_loss": 0.19965218007564545,
|
| 166 |
+
"eval_runtime": 74.7464,
|
| 167 |
+
"eval_samples_per_second": 13.606,
|
| 168 |
+
"eval_steps_per_second": 1.712,
|
| 169 |
+
"eval_wer": 0.42872117400419285,
|
| 170 |
+
"step": 2800
|
| 171 |
+
},
|
| 172 |
+
{
|
| 173 |
+
"epoch": 3.2751091703056767,
|
| 174 |
+
"grad_norm": 0.8557049036026001,
|
| 175 |
+
"learning_rate": 0.00020403973509933773,
|
| 176 |
+
"loss": 0.5398,
|
| 177 |
+
"step": 3000
|
| 178 |
+
},
|
| 179 |
+
{
|
| 180 |
+
"epoch": 3.2751091703056767,
|
| 181 |
+
"eval_loss": 0.20972934365272522,
|
| 182 |
+
"eval_runtime": 74.9379,
|
| 183 |
+
"eval_samples_per_second": 13.571,
|
| 184 |
+
"eval_steps_per_second": 1.708,
|
| 185 |
+
"eval_wer": 0.4399895178197065,
|
| 186 |
+
"step": 3000
|
| 187 |
+
},
|
| 188 |
+
{
|
| 189 |
+
"epoch": 3.493449781659389,
|
| 190 |
+
"eval_loss": 0.19680900871753693,
|
| 191 |
+
"eval_runtime": 74.7045,
|
| 192 |
+
"eval_samples_per_second": 13.614,
|
| 193 |
+
"eval_steps_per_second": 1.713,
|
| 194 |
+
"eval_wer": 0.4165356394129979,
|
| 195 |
+
"step": 3200
|
| 196 |
+
},
|
| 197 |
+
{
|
| 198 |
+
"epoch": 3.7117903930131004,
|
| 199 |
+
"eval_loss": 0.20125041902065277,
|
| 200 |
+
"eval_runtime": 74.9279,
|
| 201 |
+
"eval_samples_per_second": 13.573,
|
| 202 |
+
"eval_steps_per_second": 1.708,
|
| 203 |
+
"eval_wer": 0.42177672955974843,
|
| 204 |
+
"step": 3400
|
| 205 |
+
},
|
| 206 |
+
{
|
| 207 |
+
"epoch": 3.8209606986899565,
|
| 208 |
+
"grad_norm": 0.7604547739028931,
|
| 209 |
+
"learning_rate": 0.00018748344370860925,
|
| 210 |
+
"loss": 0.5334,
|
| 211 |
+
"step": 3500
|
| 212 |
+
},
|
| 213 |
+
{
|
| 214 |
+
"epoch": 3.930131004366812,
|
| 215 |
+
"eval_loss": 0.20029041171073914,
|
| 216 |
+
"eval_runtime": 74.9149,
|
| 217 |
+
"eval_samples_per_second": 13.575,
|
| 218 |
+
"eval_steps_per_second": 1.709,
|
| 219 |
+
"eval_wer": 0.4229559748427673,
|
| 220 |
+
"step": 3600
|
| 221 |
+
},
|
| 222 |
+
{
|
| 223 |
+
"epoch": 4.148471615720524,
|
| 224 |
+
"eval_loss": 0.19763976335525513,
|
| 225 |
+
"eval_runtime": 75.2231,
|
| 226 |
+
"eval_samples_per_second": 13.52,
|
| 227 |
+
"eval_steps_per_second": 1.702,
|
| 228 |
+
"eval_wer": 0.4226939203354298,
|
| 229 |
+
"step": 3800
|
| 230 |
+
},
|
| 231 |
+
{
|
| 232 |
+
"epoch": 4.366812227074236,
|
| 233 |
+
"grad_norm": 1.0875004529953003,
|
| 234 |
+
"learning_rate": 0.00017092715231788077,
|
| 235 |
+
"loss": 0.5123,
|
| 236 |
+
"step": 4000
|
| 237 |
+
},
|
| 238 |
+
{
|
| 239 |
+
"epoch": 4.366812227074236,
|
| 240 |
+
"eval_loss": 0.19775792956352234,
|
| 241 |
+
"eval_runtime": 75.5854,
|
| 242 |
+
"eval_samples_per_second": 13.455,
|
| 243 |
+
"eval_steps_per_second": 1.693,
|
| 244 |
+
"eval_wer": 0.419811320754717,
|
| 245 |
+
"step": 4000
|
| 246 |
+
},
|
| 247 |
+
{
|
| 248 |
+
"epoch": 4.585152838427947,
|
| 249 |
+
"eval_loss": 0.201936736702919,
|
| 250 |
+
"eval_runtime": 74.8299,
|
| 251 |
+
"eval_samples_per_second": 13.591,
|
| 252 |
+
"eval_steps_per_second": 1.711,
|
| 253 |
+
"eval_wer": 0.429769392033543,
|
| 254 |
+
"step": 4200
|
| 255 |
+
},
|
| 256 |
+
{
|
| 257 |
+
"epoch": 4.8034934497816595,
|
| 258 |
+
"eval_loss": 0.19385558366775513,
|
| 259 |
+
"eval_runtime": 75.489,
|
| 260 |
+
"eval_samples_per_second": 13.472,
|
| 261 |
+
"eval_steps_per_second": 1.696,
|
| 262 |
+
"eval_wer": 0.41457023060796644,
|
| 263 |
+
"step": 4400
|
| 264 |
+
},
|
| 265 |
+
{
|
| 266 |
+
"epoch": 4.9126637554585155,
|
| 267 |
+
"grad_norm": 1.0764209032058716,
|
| 268 |
+
"learning_rate": 0.0001543708609271523,
|
| 269 |
+
"loss": 0.5119,
|
| 270 |
+
"step": 4500
|
| 271 |
+
},
|
| 272 |
+
{
|
| 273 |
+
"epoch": 5.021834061135372,
|
| 274 |
+
"eval_loss": 0.1989189237356186,
|
| 275 |
+
"eval_runtime": 75.1857,
|
| 276 |
+
"eval_samples_per_second": 13.527,
|
| 277 |
+
"eval_steps_per_second": 1.702,
|
| 278 |
+
"eval_wer": 0.4161425576519916,
|
| 279 |
+
"step": 4600
|
| 280 |
+
},
|
| 281 |
+
{
|
| 282 |
+
"epoch": 5.240174672489083,
|
| 283 |
+
"eval_loss": 0.1901673972606659,
|
| 284 |
+
"eval_runtime": 75.3755,
|
| 285 |
+
"eval_samples_per_second": 13.492,
|
| 286 |
+
"eval_steps_per_second": 1.698,
|
| 287 |
+
"eval_wer": 0.407625786163522,
|
| 288 |
+
"step": 4800
|
| 289 |
+
},
|
| 290 |
+
{
|
| 291 |
+
"epoch": 5.458515283842795,
|
| 292 |
+
"grad_norm": 0.5494298338890076,
|
| 293 |
+
"learning_rate": 0.00013781456953642382,
|
| 294 |
+
"loss": 0.4929,
|
| 295 |
+
"step": 5000
|
| 296 |
+
},
|
| 297 |
+
{
|
| 298 |
+
"epoch": 5.458515283842795,
|
| 299 |
+
"eval_loss": 0.1928589940071106,
|
| 300 |
+
"eval_runtime": 75.5607,
|
| 301 |
+
"eval_samples_per_second": 13.459,
|
| 302 |
+
"eval_steps_per_second": 1.694,
|
| 303 |
+
"eval_wer": 0.4115566037735849,
|
| 304 |
+
"step": 5000
|
| 305 |
+
},
|
| 306 |
+
{
|
| 307 |
+
"epoch": 5.676855895196507,
|
| 308 |
+
"eval_loss": 0.19432277977466583,
|
| 309 |
+
"eval_runtime": 75.57,
|
| 310 |
+
"eval_samples_per_second": 13.458,
|
| 311 |
+
"eval_steps_per_second": 1.694,
|
| 312 |
+
"eval_wer": 0.4144392033542977,
|
| 313 |
+
"step": 5200
|
| 314 |
+
},
|
| 315 |
+
{
|
| 316 |
+
"epoch": 5.895196506550218,
|
| 317 |
+
"eval_loss": 0.19218797981739044,
|
| 318 |
+
"eval_runtime": 75.4593,
|
| 319 |
+
"eval_samples_per_second": 13.477,
|
| 320 |
+
"eval_steps_per_second": 1.696,
|
| 321 |
+
"eval_wer": 0.4106394129979036,
|
| 322 |
+
"step": 5400
|
| 323 |
+
},
|
| 324 |
+
{
|
| 325 |
+
"epoch": 6.004366812227074,
|
| 326 |
+
"grad_norm": 0.40776267647743225,
|
| 327 |
+
"learning_rate": 0.00012125827814569536,
|
| 328 |
+
"loss": 0.4878,
|
| 329 |
+
"step": 5500
|
| 330 |
+
},
|
| 331 |
+
{
|
| 332 |
+
"epoch": 6.11353711790393,
|
| 333 |
+
"eval_loss": 0.19328303635120392,
|
| 334 |
+
"eval_runtime": 76.1865,
|
| 335 |
+
"eval_samples_per_second": 13.349,
|
| 336 |
+
"eval_steps_per_second": 1.68,
|
| 337 |
+
"eval_wer": 0.41365303983228513,
|
| 338 |
+
"step": 5600
|
| 339 |
+
},
|
| 340 |
+
{
|
| 341 |
+
"epoch": 6.331877729257642,
|
| 342 |
+
"eval_loss": 0.19196276366710663,
|
| 343 |
+
"eval_runtime": 75.3609,
|
| 344 |
+
"eval_samples_per_second": 13.495,
|
| 345 |
+
"eval_steps_per_second": 1.698,
|
| 346 |
+
"eval_wer": 0.40579140461215935,
|
| 347 |
+
"step": 5800
|
| 348 |
+
},
|
| 349 |
+
{
|
| 350 |
+
"epoch": 6.550218340611353,
|
| 351 |
+
"grad_norm": 0.5017375946044922,
|
| 352 |
+
"learning_rate": 0.00010470198675496688,
|
| 353 |
+
"loss": 0.4755,
|
| 354 |
+
"step": 6000
|
| 355 |
+
},
|
| 356 |
+
{
|
| 357 |
+
"epoch": 6.550218340611353,
|
| 358 |
+
"eval_loss": 0.19270355999469757,
|
| 359 |
+
"eval_runtime": 75.2917,
|
| 360 |
+
"eval_samples_per_second": 13.507,
|
| 361 |
+
"eval_steps_per_second": 1.7,
|
| 362 |
+
"eval_wer": 0.41705974842767296,
|
| 363 |
+
"step": 6000
|
| 364 |
+
},
|
| 365 |
+
{
|
| 366 |
+
"epoch": 6.7685589519650655,
|
| 367 |
+
"eval_loss": 0.19197432696819305,
|
| 368 |
+
"eval_runtime": 75.4413,
|
| 369 |
+
"eval_samples_per_second": 13.481,
|
| 370 |
+
"eval_steps_per_second": 1.697,
|
| 371 |
+
"eval_wer": 0.41273584905660377,
|
| 372 |
+
"step": 6200
|
| 373 |
+
},
|
| 374 |
+
{
|
| 375 |
+
"epoch": 6.986899563318778,
|
| 376 |
+
"eval_loss": 0.19249393045902252,
|
| 377 |
+
"eval_runtime": 75.5242,
|
| 378 |
+
"eval_samples_per_second": 13.466,
|
| 379 |
+
"eval_steps_per_second": 1.695,
|
| 380 |
+
"eval_wer": 0.40605345911949686,
|
| 381 |
+
"step": 6400
|
| 382 |
+
},
|
| 383 |
+
{
|
| 384 |
+
"epoch": 7.096069868995633,
|
| 385 |
+
"grad_norm": 0.8388169407844543,
|
| 386 |
+
"learning_rate": 8.81456953642384e-05,
|
| 387 |
+
"loss": 0.475,
|
| 388 |
+
"step": 6500
|
| 389 |
+
},
|
| 390 |
+
{
|
| 391 |
+
"epoch": 7.205240174672489,
|
| 392 |
+
"eval_loss": 0.18842807412147522,
|
| 393 |
+
"eval_runtime": 75.7212,
|
| 394 |
+
"eval_samples_per_second": 13.431,
|
| 395 |
+
"eval_steps_per_second": 1.69,
|
| 396 |
+
"eval_wer": 0.40579140461215935,
|
| 397 |
+
"step": 6600
|
| 398 |
+
},
|
| 399 |
+
{
|
| 400 |
+
"epoch": 7.423580786026201,
|
| 401 |
+
"eval_loss": 0.19032706320285797,
|
| 402 |
+
"eval_runtime": 75.2911,
|
| 403 |
+
"eval_samples_per_second": 13.508,
|
| 404 |
+
"eval_steps_per_second": 1.7,
|
| 405 |
+
"eval_wer": 0.40697064989517817,
|
| 406 |
+
"step": 6800
|
| 407 |
+
},
|
| 408 |
+
{
|
| 409 |
+
"epoch": 7.641921397379913,
|
| 410 |
+
"grad_norm": 1.298710584640503,
|
| 411 |
+
"learning_rate": 7.158940397350993e-05,
|
| 412 |
+
"loss": 0.4715,
|
| 413 |
+
"step": 7000
|
| 414 |
+
},
|
| 415 |
+
{
|
| 416 |
+
"epoch": 7.641921397379913,
|
| 417 |
+
"eval_loss": 0.18821676075458527,
|
| 418 |
+
"eval_runtime": 75.1357,
|
| 419 |
+
"eval_samples_per_second": 13.536,
|
| 420 |
+
"eval_steps_per_second": 1.704,
|
| 421 |
+
"eval_wer": 0.3996331236897275,
|
| 422 |
+
"step": 7000
|
| 423 |
+
},
|
| 424 |
+
{
|
| 425 |
+
"epoch": 7.860262008733624,
|
| 426 |
+
"eval_loss": 0.1881488859653473,
|
| 427 |
+
"eval_runtime": 75.7296,
|
| 428 |
+
"eval_samples_per_second": 13.429,
|
| 429 |
+
"eval_steps_per_second": 1.69,
|
| 430 |
+
"eval_wer": 0.4033018867924528,
|
| 431 |
+
"step": 7200
|
| 432 |
+
},
|
| 433 |
+
{
|
| 434 |
+
"epoch": 8.078602620087336,
|
| 435 |
+
"eval_loss": 0.1884731650352478,
|
| 436 |
+
"eval_runtime": 75.2978,
|
| 437 |
+
"eval_samples_per_second": 13.506,
|
| 438 |
+
"eval_steps_per_second": 1.7,
|
| 439 |
+
"eval_wer": 0.40068134171907754,
|
| 440 |
+
"step": 7400
|
| 441 |
+
},
|
| 442 |
+
{
|
| 443 |
+
"epoch": 8.187772925764191,
|
| 444 |
+
"grad_norm": 0.7714540362358093,
|
| 445 |
+
"learning_rate": 5.5033112582781456e-05,
|
| 446 |
+
"loss": 0.4575,
|
| 447 |
+
"step": 7500
|
| 448 |
+
},
|
| 449 |
+
{
|
| 450 |
+
"epoch": 8.296943231441048,
|
| 451 |
+
"eval_loss": 0.1885315328836441,
|
| 452 |
+
"eval_runtime": 76.1765,
|
| 453 |
+
"eval_samples_per_second": 13.351,
|
| 454 |
+
"eval_steps_per_second": 1.68,
|
| 455 |
+
"eval_wer": 0.4015985324947589,
|
| 456 |
+
"step": 7600
|
| 457 |
+
},
|
| 458 |
+
{
|
| 459 |
+
"epoch": 8.51528384279476,
|
| 460 |
+
"eval_loss": 0.18875299394130707,
|
| 461 |
+
"eval_runtime": 75.8312,
|
| 462 |
+
"eval_samples_per_second": 13.411,
|
| 463 |
+
"eval_steps_per_second": 1.688,
|
| 464 |
+
"eval_wer": 0.40500524109014674,
|
| 465 |
+
"step": 7800
|
| 466 |
+
},
|
| 467 |
+
{
|
| 468 |
+
"epoch": 8.733624454148472,
|
| 469 |
+
"grad_norm": 0.7762609124183655,
|
| 470 |
+
"learning_rate": 3.850993377483443e-05,
|
| 471 |
+
"loss": 0.4611,
|
| 472 |
+
"step": 8000
|
| 473 |
+
},
|
| 474 |
+
{
|
| 475 |
+
"epoch": 8.733624454148472,
|
| 476 |
+
"eval_loss": 0.18837310373783112,
|
| 477 |
+
"eval_runtime": 76.1442,
|
| 478 |
+
"eval_samples_per_second": 13.356,
|
| 479 |
+
"eval_steps_per_second": 1.681,
|
| 480 |
+
"eval_wer": 0.40461215932914046,
|
| 481 |
+
"step": 8000
|
| 482 |
+
},
|
| 483 |
+
{
|
| 484 |
+
"epoch": 8.951965065502183,
|
| 485 |
+
"eval_loss": 0.1881391853094101,
|
| 486 |
+
"eval_runtime": 76.0237,
|
| 487 |
+
"eval_samples_per_second": 13.377,
|
| 488 |
+
"eval_steps_per_second": 1.684,
|
| 489 |
+
"eval_wer": 0.39740566037735847,
|
| 490 |
+
"step": 8200
|
| 491 |
+
},
|
| 492 |
+
{
|
| 493 |
+
"epoch": 9.170305676855895,
|
| 494 |
+
"eval_loss": 0.1864539086818695,
|
| 495 |
+
"eval_runtime": 75.5164,
|
| 496 |
+
"eval_samples_per_second": 13.467,
|
| 497 |
+
"eval_steps_per_second": 1.695,
|
| 498 |
+
"eval_wer": 0.3955712788259958,
|
| 499 |
+
"step": 8400
|
| 500 |
+
},
|
| 501 |
+
{
|
| 502 |
+
"epoch": 9.279475982532752,
|
| 503 |
+
"grad_norm": 0.7670572400093079,
|
| 504 |
+
"learning_rate": 2.1953642384105956e-05,
|
| 505 |
+
"loss": 0.4559,
|
| 506 |
+
"step": 8500
|
| 507 |
+
},
|
| 508 |
+
{
|
| 509 |
+
"epoch": 9.388646288209607,
|
| 510 |
+
"eval_loss": 0.18750794231891632,
|
| 511 |
+
"eval_runtime": 75.8042,
|
| 512 |
+
"eval_samples_per_second": 13.416,
|
| 513 |
+
"eval_steps_per_second": 1.689,
|
| 514 |
+
"eval_wer": 0.39740566037735847,
|
| 515 |
+
"step": 8600
|
| 516 |
+
},
|
| 517 |
+
{
|
| 518 |
+
"epoch": 9.606986899563319,
|
| 519 |
+
"eval_loss": 0.1871640682220459,
|
| 520 |
+
"eval_runtime": 75.5158,
|
| 521 |
+
"eval_samples_per_second": 13.467,
|
| 522 |
+
"eval_steps_per_second": 1.695,
|
| 523 |
+
"eval_wer": 0.3996331236897275,
|
| 524 |
+
"step": 8800
|
| 525 |
+
},
|
| 526 |
+
{
|
| 527 |
+
"epoch": 9.825327510917031,
|
| 528 |
+
"grad_norm": 0.47464126348495483,
|
| 529 |
+
"learning_rate": 5.3973509933774825e-06,
|
| 530 |
+
"loss": 0.4536,
|
| 531 |
+
"step": 9000
|
| 532 |
+
},
|
| 533 |
+
{
|
| 534 |
+
"epoch": 9.825327510917031,
|
| 535 |
+
"eval_loss": 0.18756870925426483,
|
| 536 |
+
"eval_runtime": 75.5452,
|
| 537 |
+
"eval_samples_per_second": 13.462,
|
| 538 |
+
"eval_steps_per_second": 1.694,
|
| 539 |
+
"eval_wer": 0.3953092243186583,
|
| 540 |
+
"step": 9000
|
| 541 |
+
},
|
| 542 |
+
{
|
| 543 |
+
"epoch": 10.0,
|
| 544 |
+
"step": 9160,
|
| 545 |
+
"total_flos": 1.4294624233263563e+19,
|
| 546 |
+
"train_loss": 0.6386700151268572,
|
| 547 |
+
"train_runtime": 13601.4767,
|
| 548 |
+
"train_samples_per_second": 5.387,
|
| 549 |
+
"train_steps_per_second": 0.673
|
| 550 |
+
}
|
| 551 |
+
],
|
| 552 |
+
"logging_steps": 500,
|
| 553 |
+
"max_steps": 9160,
|
| 554 |
+
"num_input_tokens_seen": 0,
|
| 555 |
+
"num_train_epochs": 10,
|
| 556 |
+
"save_steps": 200,
|
| 557 |
+
"stateful_callbacks": {
|
| 558 |
+
"TrainerControl": {
|
| 559 |
+
"args": {
|
| 560 |
+
"should_epoch_stop": false,
|
| 561 |
+
"should_evaluate": false,
|
| 562 |
+
"should_log": false,
|
| 563 |
+
"should_save": true,
|
| 564 |
+
"should_training_stop": true
|
| 565 |
+
},
|
| 566 |
+
"attributes": {}
|
| 567 |
+
}
|
| 568 |
+
},
|
| 569 |
+
"total_flos": 1.4294624233263563e+19,
|
| 570 |
+
"train_batch_size": 8,
|
| 571 |
+
"trial_name": null,
|
| 572 |
+
"trial_params": null
|
| 573 |
+
}
|