Commit
•
5319be6
1
Parent(s):
a1a5c4f
End of training
Browse files- README.md +1 -1
- all_results.json +15 -0
- eval_results.json +9 -0
- runs/Nov08_14-18-49_b7d2bf728bae/events.out.tfevents.1731083926.b7d2bf728bae.786.1 +3 -0
- train_results.json +9 -0
- trainer_state.json +620 -0
README.md
CHANGED
|
@@ -16,7 +16,7 @@ should probably proofread and complete it, then remove this comment. -->
|
|
| 16 |
|
| 17 |
# whisper-base-ft-cv-cy
|
| 18 |
|
| 19 |
-
This model is a fine-tuned version of [openai/whisper-base](https://huggingface.co/openai/whisper-base) on
|
| 20 |
It achieves the following results on the evaluation set:
|
| 21 |
- Loss: 0.5415
|
| 22 |
- Wer: 0.4117
|
|
|
|
| 16 |
|
| 17 |
# whisper-base-ft-cv-cy
|
| 18 |
|
| 19 |
+
This model is a fine-tuned version of [openai/whisper-base](https://huggingface.co/openai/whisper-base) on the DewiBrynJones/commonvoice_18_0_cy train_all+other_with_excluded main dataset.
|
| 20 |
It achieves the following results on the evaluation set:
|
| 21 |
- Loss: 0.5415
|
| 22 |
- Wer: 0.4117
|
all_results.json
ADDED
|
@@ -0,0 +1,15 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"epoch": 2.828854314002829,
|
| 3 |
+
"eval_loss": 0.5415101647377014,
|
| 4 |
+
"eval_runtime": 593.0299,
|
| 5 |
+
"eval_samples": 5381,
|
| 6 |
+
"eval_samples_per_second": 9.074,
|
| 7 |
+
"eval_steps_per_second": 0.568,
|
| 8 |
+
"eval_wer": 0.4117459938764961,
|
| 9 |
+
"total_flos": 4.14935339433984e+18,
|
| 10 |
+
"train_loss": 0.729869236946106,
|
| 11 |
+
"train_runtime": 7273.2114,
|
| 12 |
+
"train_samples": 22611,
|
| 13 |
+
"train_samples_per_second": 8.799,
|
| 14 |
+
"train_steps_per_second": 0.275
|
| 15 |
+
}
|
eval_results.json
ADDED
|
@@ -0,0 +1,9 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"epoch": 2.828854314002829,
|
| 3 |
+
"eval_loss": 0.5415101647377014,
|
| 4 |
+
"eval_runtime": 593.0299,
|
| 5 |
+
"eval_samples": 5381,
|
| 6 |
+
"eval_samples_per_second": 9.074,
|
| 7 |
+
"eval_steps_per_second": 0.568,
|
| 8 |
+
"eval_wer": 0.4117459938764961
|
| 9 |
+
}
|
runs/Nov08_14-18-49_b7d2bf728bae/events.out.tfevents.1731083926.b7d2bf728bae.786.1
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:51caaafcd84aa4c6c244e42e2fe269f293244cc0cfa1dd37e78f152c04cbe6c2
|
| 3 |
+
size 406
|
train_results.json
ADDED
|
@@ -0,0 +1,9 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"epoch": 2.828854314002829,
|
| 3 |
+
"total_flos": 4.14935339433984e+18,
|
| 4 |
+
"train_loss": 0.729869236946106,
|
| 5 |
+
"train_runtime": 7273.2114,
|
| 6 |
+
"train_samples": 22611,
|
| 7 |
+
"train_samples_per_second": 8.799,
|
| 8 |
+
"train_steps_per_second": 0.275
|
| 9 |
+
}
|
trainer_state.json
ADDED
|
@@ -0,0 +1,620 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"best_metric": null,
|
| 3 |
+
"best_model_checkpoint": null,
|
| 4 |
+
"epoch": 2.828854314002829,
|
| 5 |
+
"eval_steps": 1000,
|
| 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": 0.03536067892503536,
|
| 13 |
+
"grad_norm": 30.994422912597656,
|
| 14 |
+
"learning_rate": 4.6000000000000004e-07,
|
| 15 |
+
"loss": 2.5424,
|
| 16 |
+
"step": 25
|
| 17 |
+
},
|
| 18 |
+
{
|
| 19 |
+
"epoch": 0.07072135785007072,
|
| 20 |
+
"grad_norm": 28.130775451660156,
|
| 21 |
+
"learning_rate": 9.600000000000001e-07,
|
| 22 |
+
"loss": 2.3297,
|
| 23 |
+
"step": 50
|
| 24 |
+
},
|
| 25 |
+
{
|
| 26 |
+
"epoch": 0.10608203677510608,
|
| 27 |
+
"grad_norm": 23.87168312072754,
|
| 28 |
+
"learning_rate": 1.46e-06,
|
| 29 |
+
"loss": 2.0791,
|
| 30 |
+
"step": 75
|
| 31 |
+
},
|
| 32 |
+
{
|
| 33 |
+
"epoch": 0.14144271570014144,
|
| 34 |
+
"grad_norm": 22.275480270385742,
|
| 35 |
+
"learning_rate": 1.9600000000000003e-06,
|
| 36 |
+
"loss": 1.8166,
|
| 37 |
+
"step": 100
|
| 38 |
+
},
|
| 39 |
+
{
|
| 40 |
+
"epoch": 0.1768033946251768,
|
| 41 |
+
"grad_norm": 20.288251876831055,
|
| 42 |
+
"learning_rate": 2.46e-06,
|
| 43 |
+
"loss": 1.6861,
|
| 44 |
+
"step": 125
|
| 45 |
+
},
|
| 46 |
+
{
|
| 47 |
+
"epoch": 0.21216407355021216,
|
| 48 |
+
"grad_norm": 19.130552291870117,
|
| 49 |
+
"learning_rate": 2.96e-06,
|
| 50 |
+
"loss": 1.5617,
|
| 51 |
+
"step": 150
|
| 52 |
+
},
|
| 53 |
+
{
|
| 54 |
+
"epoch": 0.24752475247524752,
|
| 55 |
+
"grad_norm": 19.718809127807617,
|
| 56 |
+
"learning_rate": 3.46e-06,
|
| 57 |
+
"loss": 1.4696,
|
| 58 |
+
"step": 175
|
| 59 |
+
},
|
| 60 |
+
{
|
| 61 |
+
"epoch": 0.2828854314002829,
|
| 62 |
+
"grad_norm": 19.25800132751465,
|
| 63 |
+
"learning_rate": 3.96e-06,
|
| 64 |
+
"loss": 1.4243,
|
| 65 |
+
"step": 200
|
| 66 |
+
},
|
| 67 |
+
{
|
| 68 |
+
"epoch": 0.31824611032531824,
|
| 69 |
+
"grad_norm": 17.486873626708984,
|
| 70 |
+
"learning_rate": 4.4600000000000005e-06,
|
| 71 |
+
"loss": 1.3264,
|
| 72 |
+
"step": 225
|
| 73 |
+
},
|
| 74 |
+
{
|
| 75 |
+
"epoch": 0.3536067892503536,
|
| 76 |
+
"grad_norm": 17.029237747192383,
|
| 77 |
+
"learning_rate": 4.960000000000001e-06,
|
| 78 |
+
"loss": 1.2738,
|
| 79 |
+
"step": 250
|
| 80 |
+
},
|
| 81 |
+
{
|
| 82 |
+
"epoch": 0.38896746817538896,
|
| 83 |
+
"grad_norm": 18.300344467163086,
|
| 84 |
+
"learning_rate": 5.460000000000001e-06,
|
| 85 |
+
"loss": 1.188,
|
| 86 |
+
"step": 275
|
| 87 |
+
},
|
| 88 |
+
{
|
| 89 |
+
"epoch": 0.4243281471004243,
|
| 90 |
+
"grad_norm": 18.427331924438477,
|
| 91 |
+
"learning_rate": 5.9600000000000005e-06,
|
| 92 |
+
"loss": 1.1332,
|
| 93 |
+
"step": 300
|
| 94 |
+
},
|
| 95 |
+
{
|
| 96 |
+
"epoch": 0.4596888260254597,
|
| 97 |
+
"grad_norm": 17.452590942382812,
|
| 98 |
+
"learning_rate": 6.460000000000001e-06,
|
| 99 |
+
"loss": 1.1342,
|
| 100 |
+
"step": 325
|
| 101 |
+
},
|
| 102 |
+
{
|
| 103 |
+
"epoch": 0.49504950495049505,
|
| 104 |
+
"grad_norm": 17.23073387145996,
|
| 105 |
+
"learning_rate": 6.96e-06,
|
| 106 |
+
"loss": 1.0711,
|
| 107 |
+
"step": 350
|
| 108 |
+
},
|
| 109 |
+
{
|
| 110 |
+
"epoch": 0.5304101838755304,
|
| 111 |
+
"grad_norm": 17.96349334716797,
|
| 112 |
+
"learning_rate": 7.4600000000000006e-06,
|
| 113 |
+
"loss": 0.9895,
|
| 114 |
+
"step": 375
|
| 115 |
+
},
|
| 116 |
+
{
|
| 117 |
+
"epoch": 0.5657708628005658,
|
| 118 |
+
"grad_norm": 18.00710105895996,
|
| 119 |
+
"learning_rate": 7.960000000000002e-06,
|
| 120 |
+
"loss": 0.9815,
|
| 121 |
+
"step": 400
|
| 122 |
+
},
|
| 123 |
+
{
|
| 124 |
+
"epoch": 0.6011315417256011,
|
| 125 |
+
"grad_norm": 14.980706214904785,
|
| 126 |
+
"learning_rate": 8.46e-06,
|
| 127 |
+
"loss": 0.9336,
|
| 128 |
+
"step": 425
|
| 129 |
+
},
|
| 130 |
+
{
|
| 131 |
+
"epoch": 0.6364922206506365,
|
| 132 |
+
"grad_norm": 17.89024543762207,
|
| 133 |
+
"learning_rate": 8.96e-06,
|
| 134 |
+
"loss": 0.9098,
|
| 135 |
+
"step": 450
|
| 136 |
+
},
|
| 137 |
+
{
|
| 138 |
+
"epoch": 0.6718528995756718,
|
| 139 |
+
"grad_norm": 16.27644157409668,
|
| 140 |
+
"learning_rate": 9.460000000000001e-06,
|
| 141 |
+
"loss": 0.8745,
|
| 142 |
+
"step": 475
|
| 143 |
+
},
|
| 144 |
+
{
|
| 145 |
+
"epoch": 0.7072135785007072,
|
| 146 |
+
"grad_norm": 16.346195220947266,
|
| 147 |
+
"learning_rate": 9.960000000000001e-06,
|
| 148 |
+
"loss": 0.8709,
|
| 149 |
+
"step": 500
|
| 150 |
+
},
|
| 151 |
+
{
|
| 152 |
+
"epoch": 0.7425742574257426,
|
| 153 |
+
"grad_norm": 15.123006820678711,
|
| 154 |
+
"learning_rate": 9.846666666666668e-06,
|
| 155 |
+
"loss": 0.8402,
|
| 156 |
+
"step": 525
|
| 157 |
+
},
|
| 158 |
+
{
|
| 159 |
+
"epoch": 0.7779349363507779,
|
| 160 |
+
"grad_norm": 16.96986961364746,
|
| 161 |
+
"learning_rate": 9.68e-06,
|
| 162 |
+
"loss": 0.8082,
|
| 163 |
+
"step": 550
|
| 164 |
+
},
|
| 165 |
+
{
|
| 166 |
+
"epoch": 0.8132956152758133,
|
| 167 |
+
"grad_norm": 16.53630256652832,
|
| 168 |
+
"learning_rate": 9.513333333333334e-06,
|
| 169 |
+
"loss": 0.7627,
|
| 170 |
+
"step": 575
|
| 171 |
+
},
|
| 172 |
+
{
|
| 173 |
+
"epoch": 0.8486562942008486,
|
| 174 |
+
"grad_norm": 16.62044906616211,
|
| 175 |
+
"learning_rate": 9.346666666666666e-06,
|
| 176 |
+
"loss": 0.7566,
|
| 177 |
+
"step": 600
|
| 178 |
+
},
|
| 179 |
+
{
|
| 180 |
+
"epoch": 0.884016973125884,
|
| 181 |
+
"grad_norm": 14.903979301452637,
|
| 182 |
+
"learning_rate": 9.180000000000002e-06,
|
| 183 |
+
"loss": 0.7467,
|
| 184 |
+
"step": 625
|
| 185 |
+
},
|
| 186 |
+
{
|
| 187 |
+
"epoch": 0.9193776520509194,
|
| 188 |
+
"grad_norm": 13.78823184967041,
|
| 189 |
+
"learning_rate": 9.013333333333334e-06,
|
| 190 |
+
"loss": 0.7248,
|
| 191 |
+
"step": 650
|
| 192 |
+
},
|
| 193 |
+
{
|
| 194 |
+
"epoch": 0.9547383309759547,
|
| 195 |
+
"grad_norm": 15.028746604919434,
|
| 196 |
+
"learning_rate": 8.846666666666668e-06,
|
| 197 |
+
"loss": 0.7335,
|
| 198 |
+
"step": 675
|
| 199 |
+
},
|
| 200 |
+
{
|
| 201 |
+
"epoch": 0.9900990099009901,
|
| 202 |
+
"grad_norm": 15.316567420959473,
|
| 203 |
+
"learning_rate": 8.68e-06,
|
| 204 |
+
"loss": 0.6937,
|
| 205 |
+
"step": 700
|
| 206 |
+
},
|
| 207 |
+
{
|
| 208 |
+
"epoch": 1.0254596888260255,
|
| 209 |
+
"grad_norm": 14.481536865234375,
|
| 210 |
+
"learning_rate": 8.513333333333335e-06,
|
| 211 |
+
"loss": 0.6605,
|
| 212 |
+
"step": 725
|
| 213 |
+
},
|
| 214 |
+
{
|
| 215 |
+
"epoch": 1.0608203677510608,
|
| 216 |
+
"grad_norm": 13.36230754852295,
|
| 217 |
+
"learning_rate": 8.346666666666668e-06,
|
| 218 |
+
"loss": 0.6326,
|
| 219 |
+
"step": 750
|
| 220 |
+
},
|
| 221 |
+
{
|
| 222 |
+
"epoch": 1.0961810466760962,
|
| 223 |
+
"grad_norm": 15.339876174926758,
|
| 224 |
+
"learning_rate": 8.18e-06,
|
| 225 |
+
"loss": 0.5874,
|
| 226 |
+
"step": 775
|
| 227 |
+
},
|
| 228 |
+
{
|
| 229 |
+
"epoch": 1.1315417256011315,
|
| 230 |
+
"grad_norm": 13.94893741607666,
|
| 231 |
+
"learning_rate": 8.013333333333333e-06,
|
| 232 |
+
"loss": 0.6046,
|
| 233 |
+
"step": 800
|
| 234 |
+
},
|
| 235 |
+
{
|
| 236 |
+
"epoch": 1.166902404526167,
|
| 237 |
+
"grad_norm": 13.017638206481934,
|
| 238 |
+
"learning_rate": 7.846666666666667e-06,
|
| 239 |
+
"loss": 0.6161,
|
| 240 |
+
"step": 825
|
| 241 |
+
},
|
| 242 |
+
{
|
| 243 |
+
"epoch": 1.2022630834512023,
|
| 244 |
+
"grad_norm": 13.231209754943848,
|
| 245 |
+
"learning_rate": 7.680000000000001e-06,
|
| 246 |
+
"loss": 0.5698,
|
| 247 |
+
"step": 850
|
| 248 |
+
},
|
| 249 |
+
{
|
| 250 |
+
"epoch": 1.2376237623762376,
|
| 251 |
+
"grad_norm": 12.817441940307617,
|
| 252 |
+
"learning_rate": 7.513333333333334e-06,
|
| 253 |
+
"loss": 0.5713,
|
| 254 |
+
"step": 875
|
| 255 |
+
},
|
| 256 |
+
{
|
| 257 |
+
"epoch": 1.272984441301273,
|
| 258 |
+
"grad_norm": 13.46290397644043,
|
| 259 |
+
"learning_rate": 7.346666666666668e-06,
|
| 260 |
+
"loss": 0.5654,
|
| 261 |
+
"step": 900
|
| 262 |
+
},
|
| 263 |
+
{
|
| 264 |
+
"epoch": 1.3083451202263083,
|
| 265 |
+
"grad_norm": 14.10335922241211,
|
| 266 |
+
"learning_rate": 7.180000000000001e-06,
|
| 267 |
+
"loss": 0.573,
|
| 268 |
+
"step": 925
|
| 269 |
+
},
|
| 270 |
+
{
|
| 271 |
+
"epoch": 1.3437057991513437,
|
| 272 |
+
"grad_norm": 13.611617088317871,
|
| 273 |
+
"learning_rate": 7.0133333333333345e-06,
|
| 274 |
+
"loss": 0.5811,
|
| 275 |
+
"step": 950
|
| 276 |
+
},
|
| 277 |
+
{
|
| 278 |
+
"epoch": 1.379066478076379,
|
| 279 |
+
"grad_norm": 14.193683624267578,
|
| 280 |
+
"learning_rate": 6.846666666666667e-06,
|
| 281 |
+
"loss": 0.553,
|
| 282 |
+
"step": 975
|
| 283 |
+
},
|
| 284 |
+
{
|
| 285 |
+
"epoch": 1.4144271570014144,
|
| 286 |
+
"grad_norm": 11.151966094970703,
|
| 287 |
+
"learning_rate": 6.680000000000001e-06,
|
| 288 |
+
"loss": 0.5465,
|
| 289 |
+
"step": 1000
|
| 290 |
+
},
|
| 291 |
+
{
|
| 292 |
+
"epoch": 1.4144271570014144,
|
| 293 |
+
"eval_loss": 0.6222829222679138,
|
| 294 |
+
"eval_runtime": 620.3275,
|
| 295 |
+
"eval_samples_per_second": 8.674,
|
| 296 |
+
"eval_steps_per_second": 0.543,
|
| 297 |
+
"eval_wer": 0.47019762217185573,
|
| 298 |
+
"step": 1000
|
| 299 |
+
},
|
| 300 |
+
{
|
| 301 |
+
"epoch": 1.4497878359264498,
|
| 302 |
+
"grad_norm": 14.845292091369629,
|
| 303 |
+
"learning_rate": 6.513333333333333e-06,
|
| 304 |
+
"loss": 0.5215,
|
| 305 |
+
"step": 1025
|
| 306 |
+
},
|
| 307 |
+
{
|
| 308 |
+
"epoch": 1.4851485148514851,
|
| 309 |
+
"grad_norm": 14.059162139892578,
|
| 310 |
+
"learning_rate": 6.346666666666668e-06,
|
| 311 |
+
"loss": 0.5362,
|
| 312 |
+
"step": 1050
|
| 313 |
+
},
|
| 314 |
+
{
|
| 315 |
+
"epoch": 1.5205091937765205,
|
| 316 |
+
"grad_norm": 12.1607027053833,
|
| 317 |
+
"learning_rate": 6.18e-06,
|
| 318 |
+
"loss": 0.5198,
|
| 319 |
+
"step": 1075
|
| 320 |
+
},
|
| 321 |
+
{
|
| 322 |
+
"epoch": 1.5558698727015559,
|
| 323 |
+
"grad_norm": 14.104994773864746,
|
| 324 |
+
"learning_rate": 6.013333333333335e-06,
|
| 325 |
+
"loss": 0.522,
|
| 326 |
+
"step": 1100
|
| 327 |
+
},
|
| 328 |
+
{
|
| 329 |
+
"epoch": 1.5912305516265912,
|
| 330 |
+
"grad_norm": 14.609585762023926,
|
| 331 |
+
"learning_rate": 5.846666666666667e-06,
|
| 332 |
+
"loss": 0.5456,
|
| 333 |
+
"step": 1125
|
| 334 |
+
},
|
| 335 |
+
{
|
| 336 |
+
"epoch": 1.6265912305516266,
|
| 337 |
+
"grad_norm": 13.004403114318848,
|
| 338 |
+
"learning_rate": 5.68e-06,
|
| 339 |
+
"loss": 0.5216,
|
| 340 |
+
"step": 1150
|
| 341 |
+
},
|
| 342 |
+
{
|
| 343 |
+
"epoch": 1.661951909476662,
|
| 344 |
+
"grad_norm": 13.968815803527832,
|
| 345 |
+
"learning_rate": 5.513333333333334e-06,
|
| 346 |
+
"loss": 0.5101,
|
| 347 |
+
"step": 1175
|
| 348 |
+
},
|
| 349 |
+
{
|
| 350 |
+
"epoch": 1.6973125884016973,
|
| 351 |
+
"grad_norm": 12.793135643005371,
|
| 352 |
+
"learning_rate": 5.346666666666667e-06,
|
| 353 |
+
"loss": 0.5054,
|
| 354 |
+
"step": 1200
|
| 355 |
+
},
|
| 356 |
+
{
|
| 357 |
+
"epoch": 1.7326732673267327,
|
| 358 |
+
"grad_norm": 12.969812393188477,
|
| 359 |
+
"learning_rate": 5.18e-06,
|
| 360 |
+
"loss": 0.5078,
|
| 361 |
+
"step": 1225
|
| 362 |
+
},
|
| 363 |
+
{
|
| 364 |
+
"epoch": 1.768033946251768,
|
| 365 |
+
"grad_norm": 12.749104499816895,
|
| 366 |
+
"learning_rate": 5.013333333333333e-06,
|
| 367 |
+
"loss": 0.4944,
|
| 368 |
+
"step": 1250
|
| 369 |
+
},
|
| 370 |
+
{
|
| 371 |
+
"epoch": 1.8033946251768034,
|
| 372 |
+
"grad_norm": 13.452000617980957,
|
| 373 |
+
"learning_rate": 4.846666666666667e-06,
|
| 374 |
+
"loss": 0.5047,
|
| 375 |
+
"step": 1275
|
| 376 |
+
},
|
| 377 |
+
{
|
| 378 |
+
"epoch": 1.8387553041018387,
|
| 379 |
+
"grad_norm": 13.211859703063965,
|
| 380 |
+
"learning_rate": 4.680000000000001e-06,
|
| 381 |
+
"loss": 0.5024,
|
| 382 |
+
"step": 1300
|
| 383 |
+
},
|
| 384 |
+
{
|
| 385 |
+
"epoch": 1.874115983026874,
|
| 386 |
+
"grad_norm": 12.375421524047852,
|
| 387 |
+
"learning_rate": 4.513333333333333e-06,
|
| 388 |
+
"loss": 0.4817,
|
| 389 |
+
"step": 1325
|
| 390 |
+
},
|
| 391 |
+
{
|
| 392 |
+
"epoch": 1.9094766619519095,
|
| 393 |
+
"grad_norm": 13.229231834411621,
|
| 394 |
+
"learning_rate": 4.346666666666667e-06,
|
| 395 |
+
"loss": 0.47,
|
| 396 |
+
"step": 1350
|
| 397 |
+
},
|
| 398 |
+
{
|
| 399 |
+
"epoch": 1.9448373408769448,
|
| 400 |
+
"grad_norm": 11.967358589172363,
|
| 401 |
+
"learning_rate": 4.18e-06,
|
| 402 |
+
"loss": 0.4717,
|
| 403 |
+
"step": 1375
|
| 404 |
+
},
|
| 405 |
+
{
|
| 406 |
+
"epoch": 1.9801980198019802,
|
| 407 |
+
"grad_norm": 13.194028854370117,
|
| 408 |
+
"learning_rate": 4.013333333333334e-06,
|
| 409 |
+
"loss": 0.4871,
|
| 410 |
+
"step": 1400
|
| 411 |
+
},
|
| 412 |
+
{
|
| 413 |
+
"epoch": 2.0155586987270158,
|
| 414 |
+
"grad_norm": 10.578360557556152,
|
| 415 |
+
"learning_rate": 3.8466666666666665e-06,
|
| 416 |
+
"loss": 0.4601,
|
| 417 |
+
"step": 1425
|
| 418 |
+
},
|
| 419 |
+
{
|
| 420 |
+
"epoch": 2.050919377652051,
|
| 421 |
+
"grad_norm": 10.83725357055664,
|
| 422 |
+
"learning_rate": 3.6800000000000003e-06,
|
| 423 |
+
"loss": 0.4081,
|
| 424 |
+
"step": 1450
|
| 425 |
+
},
|
| 426 |
+
{
|
| 427 |
+
"epoch": 2.0862800565770865,
|
| 428 |
+
"grad_norm": 11.61172866821289,
|
| 429 |
+
"learning_rate": 3.5133333333333337e-06,
|
| 430 |
+
"loss": 0.4086,
|
| 431 |
+
"step": 1475
|
| 432 |
+
},
|
| 433 |
+
{
|
| 434 |
+
"epoch": 2.1216407355021216,
|
| 435 |
+
"grad_norm": 13.022510528564453,
|
| 436 |
+
"learning_rate": 3.346666666666667e-06,
|
| 437 |
+
"loss": 0.3962,
|
| 438 |
+
"step": 1500
|
| 439 |
+
},
|
| 440 |
+
{
|
| 441 |
+
"epoch": 2.157001414427157,
|
| 442 |
+
"grad_norm": 11.560221672058105,
|
| 443 |
+
"learning_rate": 3.1800000000000005e-06,
|
| 444 |
+
"loss": 0.4214,
|
| 445 |
+
"step": 1525
|
| 446 |
+
},
|
| 447 |
+
{
|
| 448 |
+
"epoch": 2.1923620933521923,
|
| 449 |
+
"grad_norm": 11.080381393432617,
|
| 450 |
+
"learning_rate": 3.013333333333334e-06,
|
| 451 |
+
"loss": 0.3891,
|
| 452 |
+
"step": 1550
|
| 453 |
+
},
|
| 454 |
+
{
|
| 455 |
+
"epoch": 2.227722772277228,
|
| 456 |
+
"grad_norm": 11.685514450073242,
|
| 457 |
+
"learning_rate": 2.8466666666666672e-06,
|
| 458 |
+
"loss": 0.4066,
|
| 459 |
+
"step": 1575
|
| 460 |
+
},
|
| 461 |
+
{
|
| 462 |
+
"epoch": 2.263083451202263,
|
| 463 |
+
"grad_norm": 10.105015754699707,
|
| 464 |
+
"learning_rate": 2.68e-06,
|
| 465 |
+
"loss": 0.4023,
|
| 466 |
+
"step": 1600
|
| 467 |
+
},
|
| 468 |
+
{
|
| 469 |
+
"epoch": 2.298444130127298,
|
| 470 |
+
"grad_norm": 10.330554008483887,
|
| 471 |
+
"learning_rate": 2.5133333333333336e-06,
|
| 472 |
+
"loss": 0.3896,
|
| 473 |
+
"step": 1625
|
| 474 |
+
},
|
| 475 |
+
{
|
| 476 |
+
"epoch": 2.333804809052334,
|
| 477 |
+
"grad_norm": 12.053783416748047,
|
| 478 |
+
"learning_rate": 2.346666666666667e-06,
|
| 479 |
+
"loss": 0.4,
|
| 480 |
+
"step": 1650
|
| 481 |
+
},
|
| 482 |
+
{
|
| 483 |
+
"epoch": 2.3691654879773694,
|
| 484 |
+
"grad_norm": 11.276220321655273,
|
| 485 |
+
"learning_rate": 2.1800000000000003e-06,
|
| 486 |
+
"loss": 0.3984,
|
| 487 |
+
"step": 1675
|
| 488 |
+
},
|
| 489 |
+
{
|
| 490 |
+
"epoch": 2.4045261669024045,
|
| 491 |
+
"grad_norm": 13.366302490234375,
|
| 492 |
+
"learning_rate": 2.0133333333333337e-06,
|
| 493 |
+
"loss": 0.3986,
|
| 494 |
+
"step": 1700
|
| 495 |
+
},
|
| 496 |
+
{
|
| 497 |
+
"epoch": 2.4398868458274396,
|
| 498 |
+
"grad_norm": 10.583386421203613,
|
| 499 |
+
"learning_rate": 1.8466666666666668e-06,
|
| 500 |
+
"loss": 0.4013,
|
| 501 |
+
"step": 1725
|
| 502 |
+
},
|
| 503 |
+
{
|
| 504 |
+
"epoch": 2.4752475247524752,
|
| 505 |
+
"grad_norm": 11.777633666992188,
|
| 506 |
+
"learning_rate": 1.6800000000000002e-06,
|
| 507 |
+
"loss": 0.4076,
|
| 508 |
+
"step": 1750
|
| 509 |
+
},
|
| 510 |
+
{
|
| 511 |
+
"epoch": 2.510608203677511,
|
| 512 |
+
"grad_norm": 12.717354774475098,
|
| 513 |
+
"learning_rate": 1.5133333333333334e-06,
|
| 514 |
+
"loss": 0.3901,
|
| 515 |
+
"step": 1775
|
| 516 |
+
},
|
| 517 |
+
{
|
| 518 |
+
"epoch": 2.545968882602546,
|
| 519 |
+
"grad_norm": 10.099494934082031,
|
| 520 |
+
"learning_rate": 1.3466666666666668e-06,
|
| 521 |
+
"loss": 0.3914,
|
| 522 |
+
"step": 1800
|
| 523 |
+
},
|
| 524 |
+
{
|
| 525 |
+
"epoch": 2.581329561527581,
|
| 526 |
+
"grad_norm": 12.377460479736328,
|
| 527 |
+
"learning_rate": 1.1800000000000001e-06,
|
| 528 |
+
"loss": 0.3904,
|
| 529 |
+
"step": 1825
|
| 530 |
+
},
|
| 531 |
+
{
|
| 532 |
+
"epoch": 2.6166902404526167,
|
| 533 |
+
"grad_norm": 11.621073722839355,
|
| 534 |
+
"learning_rate": 1.0133333333333333e-06,
|
| 535 |
+
"loss": 0.3865,
|
| 536 |
+
"step": 1850
|
| 537 |
+
},
|
| 538 |
+
{
|
| 539 |
+
"epoch": 2.6520509193776522,
|
| 540 |
+
"grad_norm": 12.402543067932129,
|
| 541 |
+
"learning_rate": 8.466666666666668e-07,
|
| 542 |
+
"loss": 0.3993,
|
| 543 |
+
"step": 1875
|
| 544 |
+
},
|
| 545 |
+
{
|
| 546 |
+
"epoch": 2.6874115983026874,
|
| 547 |
+
"grad_norm": 12.710662841796875,
|
| 548 |
+
"learning_rate": 6.800000000000001e-07,
|
| 549 |
+
"loss": 0.3892,
|
| 550 |
+
"step": 1900
|
| 551 |
+
},
|
| 552 |
+
{
|
| 553 |
+
"epoch": 2.7227722772277225,
|
| 554 |
+
"grad_norm": 11.998794555664062,
|
| 555 |
+
"learning_rate": 5.133333333333334e-07,
|
| 556 |
+
"loss": 0.3725,
|
| 557 |
+
"step": 1925
|
| 558 |
+
},
|
| 559 |
+
{
|
| 560 |
+
"epoch": 2.758132956152758,
|
| 561 |
+
"grad_norm": 11.460661888122559,
|
| 562 |
+
"learning_rate": 3.466666666666667e-07,
|
| 563 |
+
"loss": 0.3947,
|
| 564 |
+
"step": 1950
|
| 565 |
+
},
|
| 566 |
+
{
|
| 567 |
+
"epoch": 2.7934936350777937,
|
| 568 |
+
"grad_norm": 10.511993408203125,
|
| 569 |
+
"learning_rate": 1.8e-07,
|
| 570 |
+
"loss": 0.3839,
|
| 571 |
+
"step": 1975
|
| 572 |
+
},
|
| 573 |
+
{
|
| 574 |
+
"epoch": 2.828854314002829,
|
| 575 |
+
"grad_norm": 11.295842170715332,
|
| 576 |
+
"learning_rate": 1.3333333333333334e-08,
|
| 577 |
+
"loss": 0.3781,
|
| 578 |
+
"step": 2000
|
| 579 |
+
},
|
| 580 |
+
{
|
| 581 |
+
"epoch": 2.828854314002829,
|
| 582 |
+
"eval_loss": 0.5415101647377014,
|
| 583 |
+
"eval_runtime": 593.481,
|
| 584 |
+
"eval_samples_per_second": 9.067,
|
| 585 |
+
"eval_steps_per_second": 0.568,
|
| 586 |
+
"eval_wer": 0.4117459938764961,
|
| 587 |
+
"step": 2000
|
| 588 |
+
},
|
| 589 |
+
{
|
| 590 |
+
"epoch": 2.828854314002829,
|
| 591 |
+
"step": 2000,
|
| 592 |
+
"total_flos": 4.14935339433984e+18,
|
| 593 |
+
"train_loss": 0.729869236946106,
|
| 594 |
+
"train_runtime": 7273.2114,
|
| 595 |
+
"train_samples_per_second": 8.799,
|
| 596 |
+
"train_steps_per_second": 0.275
|
| 597 |
+
}
|
| 598 |
+
],
|
| 599 |
+
"logging_steps": 25,
|
| 600 |
+
"max_steps": 2000,
|
| 601 |
+
"num_input_tokens_seen": 0,
|
| 602 |
+
"num_train_epochs": 3,
|
| 603 |
+
"save_steps": 1000,
|
| 604 |
+
"stateful_callbacks": {
|
| 605 |
+
"TrainerControl": {
|
| 606 |
+
"args": {
|
| 607 |
+
"should_epoch_stop": false,
|
| 608 |
+
"should_evaluate": false,
|
| 609 |
+
"should_log": false,
|
| 610 |
+
"should_save": true,
|
| 611 |
+
"should_training_stop": true
|
| 612 |
+
},
|
| 613 |
+
"attributes": {}
|
| 614 |
+
}
|
| 615 |
+
},
|
| 616 |
+
"total_flos": 4.14935339433984e+18,
|
| 617 |
+
"train_batch_size": 16,
|
| 618 |
+
"trial_name": null,
|
| 619 |
+
"trial_params": null
|
| 620 |
+
}
|