End of training
Browse files- README.md +2 -0
- all_results.json +15 -0
- eval_results.json +9 -0
- runs/Oct04_07-33-11_a3cadab5f73a/events.out.tfevents.1728037286.a3cadab5f73a.29.1 +3 -0
- train_results.json +9 -0
- trainer_state.json +632 -0
README.md
CHANGED
|
@@ -2,6 +2,8 @@
|
|
| 2 |
license: apache-2.0
|
| 3 |
base_model: DewiBrynJones/wav2vec2-xlsr-53-ft-btb-cv-cy
|
| 4 |
tags:
|
|
|
|
|
|
|
| 5 |
- generated_from_trainer
|
| 6 |
metrics:
|
| 7 |
- wer
|
|
|
|
| 2 |
license: apache-2.0
|
| 3 |
base_model: DewiBrynJones/wav2vec2-xlsr-53-ft-btb-cv-cy
|
| 4 |
tags:
|
| 5 |
+
- automatic-speech-recognition
|
| 6 |
+
- ./data-configs/cv.json
|
| 7 |
- generated_from_trainer
|
| 8 |
metrics:
|
| 9 |
- wer
|
all_results.json
ADDED
|
@@ -0,0 +1,15 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"epoch": 5.020080321285141,
|
| 3 |
+
"eval_loss": 0.2515573501586914,
|
| 4 |
+
"eval_runtime": 200.0113,
|
| 5 |
+
"eval_samples": 5381,
|
| 6 |
+
"eval_samples_per_second": 26.903,
|
| 7 |
+
"eval_steps_per_second": 0.425,
|
| 8 |
+
"eval_wer": 0.24033088747042097,
|
| 9 |
+
"total_flos": 6.356146932571761e+18,
|
| 10 |
+
"train_loss": 0.3559208065032959,
|
| 11 |
+
"train_runtime": 12994.9936,
|
| 12 |
+
"train_samples": 7968,
|
| 13 |
+
"train_samples_per_second": 3.078,
|
| 14 |
+
"train_steps_per_second": 0.77
|
| 15 |
+
}
|
eval_results.json
ADDED
|
@@ -0,0 +1,9 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"epoch": 5.020080321285141,
|
| 3 |
+
"eval_loss": 0.2515573501586914,
|
| 4 |
+
"eval_runtime": 200.0113,
|
| 5 |
+
"eval_samples": 5381,
|
| 6 |
+
"eval_samples_per_second": 26.903,
|
| 7 |
+
"eval_steps_per_second": 0.425,
|
| 8 |
+
"eval_wer": 0.24033088747042097
|
| 9 |
+
}
|
runs/Oct04_07-33-11_a3cadab5f73a/events.out.tfevents.1728037286.a3cadab5f73a.29.1
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:b300327309affd6dd65c72ca0a53b29a3c009fa29d4b469e4cee0ba5e2c9d898
|
| 3 |
+
size 406
|
train_results.json
ADDED
|
@@ -0,0 +1,9 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"epoch": 5.020080321285141,
|
| 3 |
+
"total_flos": 6.356146932571761e+18,
|
| 4 |
+
"train_loss": 0.3559208065032959,
|
| 5 |
+
"train_runtime": 12994.9936,
|
| 6 |
+
"train_samples": 7968,
|
| 7 |
+
"train_samples_per_second": 3.078,
|
| 8 |
+
"train_steps_per_second": 0.77
|
| 9 |
+
}
|
trainer_state.json
ADDED
|
@@ -0,0 +1,632 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"best_metric": null,
|
| 3 |
+
"best_model_checkpoint": null,
|
| 4 |
+
"epoch": 5.020080321285141,
|
| 5 |
+
"eval_steps": 200,
|
| 6 |
+
"global_step": 10000,
|
| 7 |
+
"is_hyper_param_search": false,
|
| 8 |
+
"is_local_process_zero": true,
|
| 9 |
+
"is_world_process_zero": true,
|
| 10 |
+
"log_history": [
|
| 11 |
+
{
|
| 12 |
+
"epoch": 0.10040160642570281,
|
| 13 |
+
"eval_loss": 0.3807084858417511,
|
| 14 |
+
"eval_runtime": 206.6442,
|
| 15 |
+
"eval_samples_per_second": 26.04,
|
| 16 |
+
"eval_steps_per_second": 0.411,
|
| 17 |
+
"eval_wer": 0.2514466384298753,
|
| 18 |
+
"step": 200
|
| 19 |
+
},
|
| 20 |
+
{
|
| 21 |
+
"epoch": 0.20080321285140562,
|
| 22 |
+
"eval_loss": 0.253967821598053,
|
| 23 |
+
"eval_runtime": 206.4967,
|
| 24 |
+
"eval_samples_per_second": 26.059,
|
| 25 |
+
"eval_steps_per_second": 0.412,
|
| 26 |
+
"eval_wer": 0.2642923899858816,
|
| 27 |
+
"step": 400
|
| 28 |
+
},
|
| 29 |
+
{
|
| 30 |
+
"epoch": 0.25100401606425704,
|
| 31 |
+
"grad_norm": 1.857898235321045,
|
| 32 |
+
"learning_rate": 0.0001491,
|
| 33 |
+
"loss": 2.4874,
|
| 34 |
+
"step": 500
|
| 35 |
+
},
|
| 36 |
+
{
|
| 37 |
+
"epoch": 0.30120481927710846,
|
| 38 |
+
"eval_loss": 0.2642447352409363,
|
| 39 |
+
"eval_runtime": 205.9594,
|
| 40 |
+
"eval_samples_per_second": 26.127,
|
| 41 |
+
"eval_steps_per_second": 0.413,
|
| 42 |
+
"eval_wer": 0.30376424267732505,
|
| 43 |
+
"step": 600
|
| 44 |
+
},
|
| 45 |
+
{
|
| 46 |
+
"epoch": 0.40160642570281124,
|
| 47 |
+
"eval_loss": 0.3125462234020233,
|
| 48 |
+
"eval_runtime": 205.1387,
|
| 49 |
+
"eval_samples_per_second": 26.231,
|
| 50 |
+
"eval_steps_per_second": 0.414,
|
| 51 |
+
"eval_wer": 0.39048300821253157,
|
| 52 |
+
"step": 800
|
| 53 |
+
},
|
| 54 |
+
{
|
| 55 |
+
"epoch": 0.5020080321285141,
|
| 56 |
+
"grad_norm": 3.634869337081909,
|
| 57 |
+
"learning_rate": 0.00029909999999999995,
|
| 58 |
+
"loss": 0.3991,
|
| 59 |
+
"step": 1000
|
| 60 |
+
},
|
| 61 |
+
{
|
| 62 |
+
"epoch": 0.5020080321285141,
|
| 63 |
+
"eval_loss": 0.3531426191329956,
|
| 64 |
+
"eval_runtime": 205.1455,
|
| 65 |
+
"eval_samples_per_second": 26.23,
|
| 66 |
+
"eval_steps_per_second": 0.414,
|
| 67 |
+
"eval_wer": 0.3939430094056354,
|
| 68 |
+
"step": 1000
|
| 69 |
+
},
|
| 70 |
+
{
|
| 71 |
+
"epoch": 0.6024096385542169,
|
| 72 |
+
"eval_loss": 0.3571958839893341,
|
| 73 |
+
"eval_runtime": 209.3856,
|
| 74 |
+
"eval_samples_per_second": 25.699,
|
| 75 |
+
"eval_steps_per_second": 0.406,
|
| 76 |
+
"eval_wer": 0.40390542663405515,
|
| 77 |
+
"step": 1200
|
| 78 |
+
},
|
| 79 |
+
{
|
| 80 |
+
"epoch": 0.7028112449799196,
|
| 81 |
+
"eval_loss": 0.36791086196899414,
|
| 82 |
+
"eval_runtime": 206.8164,
|
| 83 |
+
"eval_samples_per_second": 26.018,
|
| 84 |
+
"eval_steps_per_second": 0.411,
|
| 85 |
+
"eval_wer": 0.4052576110083716,
|
| 86 |
+
"step": 1400
|
| 87 |
+
},
|
| 88 |
+
{
|
| 89 |
+
"epoch": 0.7530120481927711,
|
| 90 |
+
"grad_norm": 3.5483558177948,
|
| 91 |
+
"learning_rate": 0.0002834333333333333,
|
| 92 |
+
"loss": 0.4512,
|
| 93 |
+
"step": 1500
|
| 94 |
+
},
|
| 95 |
+
{
|
| 96 |
+
"epoch": 0.8032128514056225,
|
| 97 |
+
"eval_loss": 0.35897189378738403,
|
| 98 |
+
"eval_runtime": 207.7252,
|
| 99 |
+
"eval_samples_per_second": 25.904,
|
| 100 |
+
"eval_steps_per_second": 0.409,
|
| 101 |
+
"eval_wer": 0.38767921414225776,
|
| 102 |
+
"step": 1600
|
| 103 |
+
},
|
| 104 |
+
{
|
| 105 |
+
"epoch": 0.9036144578313253,
|
| 106 |
+
"eval_loss": 0.3732704222202301,
|
| 107 |
+
"eval_runtime": 205.6494,
|
| 108 |
+
"eval_samples_per_second": 26.166,
|
| 109 |
+
"eval_steps_per_second": 0.413,
|
| 110 |
+
"eval_wer": 0.4006840462128895,
|
| 111 |
+
"step": 1800
|
| 112 |
+
},
|
| 113 |
+
{
|
| 114 |
+
"epoch": 1.0040160642570282,
|
| 115 |
+
"grad_norm": 2.2905380725860596,
|
| 116 |
+
"learning_rate": 0.00026676666666666663,
|
| 117 |
+
"loss": 0.4333,
|
| 118 |
+
"step": 2000
|
| 119 |
+
},
|
| 120 |
+
{
|
| 121 |
+
"epoch": 1.0040160642570282,
|
| 122 |
+
"eval_loss": 0.377088725566864,
|
| 123 |
+
"eval_runtime": 204.613,
|
| 124 |
+
"eval_samples_per_second": 26.298,
|
| 125 |
+
"eval_steps_per_second": 0.415,
|
| 126 |
+
"eval_wer": 0.4243273876990992,
|
| 127 |
+
"step": 2000
|
| 128 |
+
},
|
| 129 |
+
{
|
| 130 |
+
"epoch": 1.104417670682731,
|
| 131 |
+
"eval_loss": 0.3604430556297302,
|
| 132 |
+
"eval_runtime": 208.0965,
|
| 133 |
+
"eval_samples_per_second": 25.858,
|
| 134 |
+
"eval_steps_per_second": 0.408,
|
| 135 |
+
"eval_wer": 0.3867048459901768,
|
| 136 |
+
"step": 2200
|
| 137 |
+
},
|
| 138 |
+
{
|
| 139 |
+
"epoch": 1.2048192771084336,
|
| 140 |
+
"eval_loss": 0.3431110978126526,
|
| 141 |
+
"eval_runtime": 206.2637,
|
| 142 |
+
"eval_samples_per_second": 26.088,
|
| 143 |
+
"eval_steps_per_second": 0.412,
|
| 144 |
+
"eval_wer": 0.38137564875022373,
|
| 145 |
+
"step": 2400
|
| 146 |
+
},
|
| 147 |
+
{
|
| 148 |
+
"epoch": 1.2550200803212852,
|
| 149 |
+
"grad_norm": 1.8699342012405396,
|
| 150 |
+
"learning_rate": 0.00025009999999999995,
|
| 151 |
+
"loss": 0.3468,
|
| 152 |
+
"step": 2500
|
| 153 |
+
},
|
| 154 |
+
{
|
| 155 |
+
"epoch": 1.3052208835341366,
|
| 156 |
+
"eval_loss": 0.32902058959007263,
|
| 157 |
+
"eval_runtime": 205.4777,
|
| 158 |
+
"eval_samples_per_second": 26.188,
|
| 159 |
+
"eval_steps_per_second": 0.414,
|
| 160 |
+
"eval_wer": 0.3778559923641353,
|
| 161 |
+
"step": 2600
|
| 162 |
+
},
|
| 163 |
+
{
|
| 164 |
+
"epoch": 1.4056224899598393,
|
| 165 |
+
"eval_loss": 0.33407700061798096,
|
| 166 |
+
"eval_runtime": 205.6359,
|
| 167 |
+
"eval_samples_per_second": 26.168,
|
| 168 |
+
"eval_steps_per_second": 0.413,
|
| 169 |
+
"eval_wer": 0.3647119648432063,
|
| 170 |
+
"step": 2800
|
| 171 |
+
},
|
| 172 |
+
{
|
| 173 |
+
"epoch": 1.5060240963855422,
|
| 174 |
+
"grad_norm": 2.062389373779297,
|
| 175 |
+
"learning_rate": 0.0002334333333333333,
|
| 176 |
+
"loss": 0.3503,
|
| 177 |
+
"step": 3000
|
| 178 |
+
},
|
| 179 |
+
{
|
| 180 |
+
"epoch": 1.5060240963855422,
|
| 181 |
+
"eval_loss": 0.3247535228729248,
|
| 182 |
+
"eval_runtime": 206.3116,
|
| 183 |
+
"eval_samples_per_second": 26.082,
|
| 184 |
+
"eval_steps_per_second": 0.412,
|
| 185 |
+
"eval_wer": 0.3614706993577124,
|
| 186 |
+
"step": 3000
|
| 187 |
+
},
|
| 188 |
+
{
|
| 189 |
+
"epoch": 1.606425702811245,
|
| 190 |
+
"eval_loss": 0.33116209506988525,
|
| 191 |
+
"eval_runtime": 203.9912,
|
| 192 |
+
"eval_samples_per_second": 26.379,
|
| 193 |
+
"eval_steps_per_second": 0.417,
|
| 194 |
+
"eval_wer": 0.35512736383702204,
|
| 195 |
+
"step": 3200
|
| 196 |
+
},
|
| 197 |
+
{
|
| 198 |
+
"epoch": 1.7068273092369477,
|
| 199 |
+
"eval_loss": 0.3410908281803131,
|
| 200 |
+
"eval_runtime": 204.5054,
|
| 201 |
+
"eval_samples_per_second": 26.312,
|
| 202 |
+
"eval_steps_per_second": 0.416,
|
| 203 |
+
"eval_wer": 0.3836226610193084,
|
| 204 |
+
"step": 3400
|
| 205 |
+
},
|
| 206 |
+
{
|
| 207 |
+
"epoch": 1.7570281124497993,
|
| 208 |
+
"grad_norm": 0.9907544255256653,
|
| 209 |
+
"learning_rate": 0.00021679999999999998,
|
| 210 |
+
"loss": 0.3418,
|
| 211 |
+
"step": 3500
|
| 212 |
+
},
|
| 213 |
+
{
|
| 214 |
+
"epoch": 1.8072289156626506,
|
| 215 |
+
"eval_loss": 0.3116574287414551,
|
| 216 |
+
"eval_runtime": 205.0392,
|
| 217 |
+
"eval_samples_per_second": 26.244,
|
| 218 |
+
"eval_steps_per_second": 0.415,
|
| 219 |
+
"eval_wer": 0.33752908190658,
|
| 220 |
+
"step": 3600
|
| 221 |
+
},
|
| 222 |
+
{
|
| 223 |
+
"epoch": 1.9076305220883534,
|
| 224 |
+
"eval_loss": 0.3196774423122406,
|
| 225 |
+
"eval_runtime": 206.2716,
|
| 226 |
+
"eval_samples_per_second": 26.087,
|
| 227 |
+
"eval_steps_per_second": 0.412,
|
| 228 |
+
"eval_wer": 0.34317644017578397,
|
| 229 |
+
"step": 3800
|
| 230 |
+
},
|
| 231 |
+
{
|
| 232 |
+
"epoch": 2.0080321285140563,
|
| 233 |
+
"grad_norm": 1.0384626388549805,
|
| 234 |
+
"learning_rate": 0.0002001333333333333,
|
| 235 |
+
"loss": 0.3181,
|
| 236 |
+
"step": 4000
|
| 237 |
+
},
|
| 238 |
+
{
|
| 239 |
+
"epoch": 2.0080321285140563,
|
| 240 |
+
"eval_loss": 0.30675315856933594,
|
| 241 |
+
"eval_runtime": 206.0737,
|
| 242 |
+
"eval_samples_per_second": 26.112,
|
| 243 |
+
"eval_steps_per_second": 0.412,
|
| 244 |
+
"eval_wer": 0.3339696553918352,
|
| 245 |
+
"step": 4000
|
| 246 |
+
},
|
| 247 |
+
{
|
| 248 |
+
"epoch": 2.108433734939759,
|
| 249 |
+
"eval_loss": 0.31376445293426514,
|
| 250 |
+
"eval_runtime": 209.2791,
|
| 251 |
+
"eval_samples_per_second": 25.712,
|
| 252 |
+
"eval_steps_per_second": 0.406,
|
| 253 |
+
"eval_wer": 0.3358388514386844,
|
| 254 |
+
"step": 4200
|
| 255 |
+
},
|
| 256 |
+
{
|
| 257 |
+
"epoch": 2.208835341365462,
|
| 258 |
+
"eval_loss": 0.31388720870018005,
|
| 259 |
+
"eval_runtime": 204.9118,
|
| 260 |
+
"eval_samples_per_second": 26.26,
|
| 261 |
+
"eval_steps_per_second": 0.415,
|
| 262 |
+
"eval_wer": 0.3333731034619897,
|
| 263 |
+
"step": 4400
|
| 264 |
+
},
|
| 265 |
+
{
|
| 266 |
+
"epoch": 2.2590361445783134,
|
| 267 |
+
"grad_norm": 0.5868389010429382,
|
| 268 |
+
"learning_rate": 0.00018346666666666664,
|
| 269 |
+
"loss": 0.2423,
|
| 270 |
+
"step": 4500
|
| 271 |
+
},
|
| 272 |
+
{
|
| 273 |
+
"epoch": 2.3092369477911645,
|
| 274 |
+
"eval_loss": 0.3191888928413391,
|
| 275 |
+
"eval_runtime": 204.834,
|
| 276 |
+
"eval_samples_per_second": 26.27,
|
| 277 |
+
"eval_steps_per_second": 0.415,
|
| 278 |
+
"eval_wer": 0.32848137763725666,
|
| 279 |
+
"step": 4600
|
| 280 |
+
},
|
| 281 |
+
{
|
| 282 |
+
"epoch": 2.4096385542168672,
|
| 283 |
+
"eval_loss": 0.2928995192050934,
|
| 284 |
+
"eval_runtime": 204.43,
|
| 285 |
+
"eval_samples_per_second": 26.322,
|
| 286 |
+
"eval_steps_per_second": 0.416,
|
| 287 |
+
"eval_wer": 0.31682872994094136,
|
| 288 |
+
"step": 4800
|
| 289 |
+
},
|
| 290 |
+
{
|
| 291 |
+
"epoch": 2.5100401606425704,
|
| 292 |
+
"grad_norm": 1.3247759342193604,
|
| 293 |
+
"learning_rate": 0.0001668,
|
| 294 |
+
"loss": 0.2327,
|
| 295 |
+
"step": 5000
|
| 296 |
+
},
|
| 297 |
+
{
|
| 298 |
+
"epoch": 2.5100401606425704,
|
| 299 |
+
"eval_loss": 0.29208171367645264,
|
| 300 |
+
"eval_runtime": 206.3612,
|
| 301 |
+
"eval_samples_per_second": 26.076,
|
| 302 |
+
"eval_steps_per_second": 0.412,
|
| 303 |
+
"eval_wer": 0.3103064288412973,
|
| 304 |
+
"step": 5000
|
| 305 |
+
},
|
| 306 |
+
{
|
| 307 |
+
"epoch": 2.610441767068273,
|
| 308 |
+
"eval_loss": 0.2801830470561981,
|
| 309 |
+
"eval_runtime": 204.3678,
|
| 310 |
+
"eval_samples_per_second": 26.33,
|
| 311 |
+
"eval_steps_per_second": 0.416,
|
| 312 |
+
"eval_wer": 0.3037443576129969,
|
| 313 |
+
"step": 5200
|
| 314 |
+
},
|
| 315 |
+
{
|
| 316 |
+
"epoch": 2.710843373493976,
|
| 317 |
+
"eval_loss": 0.2811721861362457,
|
| 318 |
+
"eval_runtime": 204.4403,
|
| 319 |
+
"eval_samples_per_second": 26.321,
|
| 320 |
+
"eval_steps_per_second": 0.416,
|
| 321 |
+
"eval_wer": 0.29624768836127185,
|
| 322 |
+
"step": 5400
|
| 323 |
+
},
|
| 324 |
+
{
|
| 325 |
+
"epoch": 2.7610441767068274,
|
| 326 |
+
"grad_norm": 1.381541132926941,
|
| 327 |
+
"learning_rate": 0.00015013333333333331,
|
| 328 |
+
"loss": 0.2374,
|
| 329 |
+
"step": 5500
|
| 330 |
+
},
|
| 331 |
+
{
|
| 332 |
+
"epoch": 2.8112449799196786,
|
| 333 |
+
"eval_loss": 0.28872984647750854,
|
| 334 |
+
"eval_runtime": 204.3069,
|
| 335 |
+
"eval_samples_per_second": 26.338,
|
| 336 |
+
"eval_steps_per_second": 0.416,
|
| 337 |
+
"eval_wer": 0.30422159915687325,
|
| 338 |
+
"step": 5600
|
| 339 |
+
},
|
| 340 |
+
{
|
| 341 |
+
"epoch": 2.9116465863453813,
|
| 342 |
+
"eval_loss": 0.27397701144218445,
|
| 343 |
+
"eval_runtime": 204.1464,
|
| 344 |
+
"eval_samples_per_second": 26.359,
|
| 345 |
+
"eval_steps_per_second": 0.416,
|
| 346 |
+
"eval_wer": 0.2927081469108553,
|
| 347 |
+
"step": 5800
|
| 348 |
+
},
|
| 349 |
+
{
|
| 350 |
+
"epoch": 3.0120481927710845,
|
| 351 |
+
"grad_norm": 1.4617916345596313,
|
| 352 |
+
"learning_rate": 0.00013346666666666667,
|
| 353 |
+
"loss": 0.2136,
|
| 354 |
+
"step": 6000
|
| 355 |
+
},
|
| 356 |
+
{
|
| 357 |
+
"epoch": 3.0120481927710845,
|
| 358 |
+
"eval_loss": 0.2662462592124939,
|
| 359 |
+
"eval_runtime": 203.8941,
|
| 360 |
+
"eval_samples_per_second": 26.391,
|
| 361 |
+
"eval_steps_per_second": 0.417,
|
| 362 |
+
"eval_wer": 0.28296446539004555,
|
| 363 |
+
"step": 6000
|
| 364 |
+
},
|
| 365 |
+
{
|
| 366 |
+
"epoch": 3.112449799196787,
|
| 367 |
+
"eval_loss": 0.28285130858421326,
|
| 368 |
+
"eval_runtime": 206.1704,
|
| 369 |
+
"eval_samples_per_second": 26.1,
|
| 370 |
+
"eval_steps_per_second": 0.412,
|
| 371 |
+
"eval_wer": 0.2890294100101414,
|
| 372 |
+
"step": 6200
|
| 373 |
+
},
|
| 374 |
+
{
|
| 375 |
+
"epoch": 3.21285140562249,
|
| 376 |
+
"eval_loss": 0.2729070484638214,
|
| 377 |
+
"eval_runtime": 206.2438,
|
| 378 |
+
"eval_samples_per_second": 26.09,
|
| 379 |
+
"eval_steps_per_second": 0.412,
|
| 380 |
+
"eval_wer": 0.28692159319135396,
|
| 381 |
+
"step": 6400
|
| 382 |
+
},
|
| 383 |
+
{
|
| 384 |
+
"epoch": 3.2630522088353415,
|
| 385 |
+
"grad_norm": 0.8870707750320435,
|
| 386 |
+
"learning_rate": 0.00011679999999999998,
|
| 387 |
+
"loss": 0.167,
|
| 388 |
+
"step": 6500
|
| 389 |
+
},
|
| 390 |
+
{
|
| 391 |
+
"epoch": 3.3132530120481927,
|
| 392 |
+
"eval_loss": 0.2776893675327301,
|
| 393 |
+
"eval_runtime": 204.2022,
|
| 394 |
+
"eval_samples_per_second": 26.351,
|
| 395 |
+
"eval_steps_per_second": 0.416,
|
| 396 |
+
"eval_wer": 0.28892998468850045,
|
| 397 |
+
"step": 6600
|
| 398 |
+
},
|
| 399 |
+
{
|
| 400 |
+
"epoch": 3.4136546184738954,
|
| 401 |
+
"eval_loss": 0.2711654603481293,
|
| 402 |
+
"eval_runtime": 203.7376,
|
| 403 |
+
"eval_samples_per_second": 26.411,
|
| 404 |
+
"eval_steps_per_second": 0.417,
|
| 405 |
+
"eval_wer": 0.28095607389289906,
|
| 406 |
+
"step": 6800
|
| 407 |
+
},
|
| 408 |
+
{
|
| 409 |
+
"epoch": 3.5140562248995986,
|
| 410 |
+
"grad_norm": 1.0165985822677612,
|
| 411 |
+
"learning_rate": 0.00010013333333333333,
|
| 412 |
+
"loss": 0.1614,
|
| 413 |
+
"step": 7000
|
| 414 |
+
},
|
| 415 |
+
{
|
| 416 |
+
"epoch": 3.5140562248995986,
|
| 417 |
+
"eval_loss": 0.2688385844230652,
|
| 418 |
+
"eval_runtime": 204.1623,
|
| 419 |
+
"eval_samples_per_second": 26.356,
|
| 420 |
+
"eval_steps_per_second": 0.416,
|
| 421 |
+
"eval_wer": 0.27091411640716656,
|
| 422 |
+
"step": 7000
|
| 423 |
+
},
|
| 424 |
+
{
|
| 425 |
+
"epoch": 3.6144578313253013,
|
| 426 |
+
"eval_loss": 0.2589295208454132,
|
| 427 |
+
"eval_runtime": 205.2749,
|
| 428 |
+
"eval_samples_per_second": 26.214,
|
| 429 |
+
"eval_steps_per_second": 0.414,
|
| 430 |
+
"eval_wer": 0.26626101135437175,
|
| 431 |
+
"step": 7200
|
| 432 |
+
},
|
| 433 |
+
{
|
| 434 |
+
"epoch": 3.714859437751004,
|
| 435 |
+
"eval_loss": 0.26514673233032227,
|
| 436 |
+
"eval_runtime": 204.2135,
|
| 437 |
+
"eval_samples_per_second": 26.35,
|
| 438 |
+
"eval_steps_per_second": 0.416,
|
| 439 |
+
"eval_wer": 0.2669768736701863,
|
| 440 |
+
"step": 7400
|
| 441 |
+
},
|
| 442 |
+
{
|
| 443 |
+
"epoch": 3.765060240963855,
|
| 444 |
+
"grad_norm": 0.7397546172142029,
|
| 445 |
+
"learning_rate": 8.346666666666666e-05,
|
| 446 |
+
"loss": 0.1529,
|
| 447 |
+
"step": 7500
|
| 448 |
+
},
|
| 449 |
+
{
|
| 450 |
+
"epoch": 3.8152610441767068,
|
| 451 |
+
"eval_loss": 0.25074735283851624,
|
| 452 |
+
"eval_runtime": 204.2336,
|
| 453 |
+
"eval_samples_per_second": 26.347,
|
| 454 |
+
"eval_steps_per_second": 0.416,
|
| 455 |
+
"eval_wer": 0.2637157231203643,
|
| 456 |
+
"step": 7600
|
| 457 |
+
},
|
| 458 |
+
{
|
| 459 |
+
"epoch": 3.9156626506024095,
|
| 460 |
+
"eval_loss": 0.2493942528963089,
|
| 461 |
+
"eval_runtime": 206.072,
|
| 462 |
+
"eval_samples_per_second": 26.112,
|
| 463 |
+
"eval_steps_per_second": 0.412,
|
| 464 |
+
"eval_wer": 0.2567957207341566,
|
| 465 |
+
"step": 7800
|
| 466 |
+
},
|
| 467 |
+
{
|
| 468 |
+
"epoch": 4.016064257028113,
|
| 469 |
+
"grad_norm": 0.785851776599884,
|
| 470 |
+
"learning_rate": 6.68e-05,
|
| 471 |
+
"loss": 0.1496,
|
| 472 |
+
"step": 8000
|
| 473 |
+
},
|
| 474 |
+
{
|
| 475 |
+
"epoch": 4.016064257028113,
|
| 476 |
+
"eval_loss": 0.25821030139923096,
|
| 477 |
+
"eval_runtime": 204.9558,
|
| 478 |
+
"eval_samples_per_second": 26.254,
|
| 479 |
+
"eval_steps_per_second": 0.415,
|
| 480 |
+
"eval_wer": 0.2580484797868321,
|
| 481 |
+
"step": 8000
|
| 482 |
+
},
|
| 483 |
+
{
|
| 484 |
+
"epoch": 4.116465863453815,
|
| 485 |
+
"eval_loss": 0.2650238871574402,
|
| 486 |
+
"eval_runtime": 204.7852,
|
| 487 |
+
"eval_samples_per_second": 26.276,
|
| 488 |
+
"eval_steps_per_second": 0.415,
|
| 489 |
+
"eval_wer": 0.25753146811429933,
|
| 490 |
+
"step": 8200
|
| 491 |
+
},
|
| 492 |
+
{
|
| 493 |
+
"epoch": 4.216867469879518,
|
| 494 |
+
"eval_loss": 0.26561084389686584,
|
| 495 |
+
"eval_runtime": 210.1403,
|
| 496 |
+
"eval_samples_per_second": 25.607,
|
| 497 |
+
"eval_steps_per_second": 0.404,
|
| 498 |
+
"eval_wer": 0.25598043309670104,
|
| 499 |
+
"step": 8400
|
| 500 |
+
},
|
| 501 |
+
{
|
| 502 |
+
"epoch": 4.267068273092369,
|
| 503 |
+
"grad_norm": 0.34119465947151184,
|
| 504 |
+
"learning_rate": 5.013333333333332e-05,
|
| 505 |
+
"loss": 0.1128,
|
| 506 |
+
"step": 8500
|
| 507 |
+
},
|
| 508 |
+
{
|
| 509 |
+
"epoch": 4.317269076305221,
|
| 510 |
+
"eval_loss": 0.25430822372436523,
|
| 511 |
+
"eval_runtime": 203.852,
|
| 512 |
+
"eval_samples_per_second": 26.397,
|
| 513 |
+
"eval_steps_per_second": 0.417,
|
| 514 |
+
"eval_wer": 0.25118813259360895,
|
| 515 |
+
"step": 8600
|
| 516 |
+
},
|
| 517 |
+
{
|
| 518 |
+
"epoch": 4.417670682730924,
|
| 519 |
+
"eval_loss": 0.2586837112903595,
|
| 520 |
+
"eval_runtime": 202.7677,
|
| 521 |
+
"eval_samples_per_second": 26.538,
|
| 522 |
+
"eval_steps_per_second": 0.419,
|
| 523 |
+
"eval_wer": 0.24987571834794886,
|
| 524 |
+
"step": 8800
|
| 525 |
+
},
|
| 526 |
+
{
|
| 527 |
+
"epoch": 4.518072289156627,
|
| 528 |
+
"grad_norm": 0.5558347105979919,
|
| 529 |
+
"learning_rate": 3.346666666666666e-05,
|
| 530 |
+
"loss": 0.1109,
|
| 531 |
+
"step": 9000
|
| 532 |
+
},
|
| 533 |
+
{
|
| 534 |
+
"epoch": 4.518072289156627,
|
| 535 |
+
"eval_loss": 0.2540307939052582,
|
| 536 |
+
"eval_runtime": 202.8954,
|
| 537 |
+
"eval_samples_per_second": 26.521,
|
| 538 |
+
"eval_steps_per_second": 0.419,
|
| 539 |
+
"eval_wer": 0.24599813080395316,
|
| 540 |
+
"step": 9000
|
| 541 |
+
},
|
| 542 |
+
{
|
| 543 |
+
"epoch": 4.618473895582329,
|
| 544 |
+
"eval_loss": 0.2546459436416626,
|
| 545 |
+
"eval_runtime": 208.4343,
|
| 546 |
+
"eval_samples_per_second": 25.816,
|
| 547 |
+
"eval_steps_per_second": 0.408,
|
| 548 |
+
"eval_wer": 0.24245858935353656,
|
| 549 |
+
"step": 9200
|
| 550 |
+
},
|
| 551 |
+
{
|
| 552 |
+
"epoch": 4.718875502008032,
|
| 553 |
+
"eval_loss": 0.25800344347953796,
|
| 554 |
+
"eval_runtime": 205.7304,
|
| 555 |
+
"eval_samples_per_second": 26.156,
|
| 556 |
+
"eval_steps_per_second": 0.413,
|
| 557 |
+
"eval_wer": 0.24198134780966016,
|
| 558 |
+
"step": 9400
|
| 559 |
+
},
|
| 560 |
+
{
|
| 561 |
+
"epoch": 4.769076305220883,
|
| 562 |
+
"grad_norm": 0.9226210117340088,
|
| 563 |
+
"learning_rate": 1.68e-05,
|
| 564 |
+
"loss": 0.1028,
|
| 565 |
+
"step": 9500
|
| 566 |
+
},
|
| 567 |
+
{
|
| 568 |
+
"epoch": 4.8192771084337345,
|
| 569 |
+
"eval_loss": 0.25135332345962524,
|
| 570 |
+
"eval_runtime": 203.1799,
|
| 571 |
+
"eval_samples_per_second": 26.484,
|
| 572 |
+
"eval_steps_per_second": 0.418,
|
| 573 |
+
"eval_wer": 0.24035077253474915,
|
| 574 |
+
"step": 9600
|
| 575 |
+
},
|
| 576 |
+
{
|
| 577 |
+
"epoch": 4.919678714859438,
|
| 578 |
+
"eval_loss": 0.2509777843952179,
|
| 579 |
+
"eval_runtime": 203.2596,
|
| 580 |
+
"eval_samples_per_second": 26.474,
|
| 581 |
+
"eval_steps_per_second": 0.418,
|
| 582 |
+
"eval_wer": 0.24025134721310823,
|
| 583 |
+
"step": 9800
|
| 584 |
+
},
|
| 585 |
+
{
|
| 586 |
+
"epoch": 5.020080321285141,
|
| 587 |
+
"grad_norm": 4.224822521209717,
|
| 588 |
+
"learning_rate": 1.3333333333333334e-07,
|
| 589 |
+
"loss": 0.1069,
|
| 590 |
+
"step": 10000
|
| 591 |
+
},
|
| 592 |
+
{
|
| 593 |
+
"epoch": 5.020080321285141,
|
| 594 |
+
"eval_loss": 0.2515573501586914,
|
| 595 |
+
"eval_runtime": 209.1262,
|
| 596 |
+
"eval_samples_per_second": 25.731,
|
| 597 |
+
"eval_steps_per_second": 0.406,
|
| 598 |
+
"eval_wer": 0.24033088747042097,
|
| 599 |
+
"step": 10000
|
| 600 |
+
},
|
| 601 |
+
{
|
| 602 |
+
"epoch": 5.020080321285141,
|
| 603 |
+
"step": 10000,
|
| 604 |
+
"total_flos": 6.356146932571761e+18,
|
| 605 |
+
"train_loss": 0.3559208065032959,
|
| 606 |
+
"train_runtime": 12994.9936,
|
| 607 |
+
"train_samples_per_second": 3.078,
|
| 608 |
+
"train_steps_per_second": 0.77
|
| 609 |
+
}
|
| 610 |
+
],
|
| 611 |
+
"logging_steps": 500,
|
| 612 |
+
"max_steps": 10000,
|
| 613 |
+
"num_input_tokens_seen": 0,
|
| 614 |
+
"num_train_epochs": 6,
|
| 615 |
+
"save_steps": 500,
|
| 616 |
+
"stateful_callbacks": {
|
| 617 |
+
"TrainerControl": {
|
| 618 |
+
"args": {
|
| 619 |
+
"should_epoch_stop": false,
|
| 620 |
+
"should_evaluate": false,
|
| 621 |
+
"should_log": false,
|
| 622 |
+
"should_save": true,
|
| 623 |
+
"should_training_stop": true
|
| 624 |
+
},
|
| 625 |
+
"attributes": {}
|
| 626 |
+
}
|
| 627 |
+
},
|
| 628 |
+
"total_flos": 6.356146932571761e+18,
|
| 629 |
+
"train_batch_size": 4,
|
| 630 |
+
"trial_name": null,
|
| 631 |
+
"trial_params": null
|
| 632 |
+
}
|