Commit
•
3701e18
1
Parent(s):
7f80c42
End of training
Browse files- README.md +1 -1
- all_results.json +15 -0
- eval_results.json +9 -0
- runs/Nov07_16-24-20_4b49612c0fe9/events.out.tfevents.1731011318.4b49612c0fe9.2373.1 +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-en
|
| 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: 0.5813
|
| 22 |
- Wer: 1.2543
|
|
|
|
| 16 |
|
| 17 |
# whisper-base-ft-cv-cy-en
|
| 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_en train main dataset.
|
| 20 |
It achieves the following results on the evaluation set:
|
| 21 |
- Loss: 0.5813
|
| 22 |
- Wer: 1.2543
|
all_results.json
ADDED
|
@@ -0,0 +1,15 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"epoch": 1.4149274849663955,
|
| 3 |
+
"eval_loss": 0.5812551379203796,
|
| 4 |
+
"eval_runtime": 1650.3013,
|
| 5 |
+
"eval_samples": 10763,
|
| 6 |
+
"eval_samples_per_second": 6.522,
|
| 7 |
+
"eval_steps_per_second": 0.408,
|
| 8 |
+
"eval_wer": 1.254253696315419,
|
| 9 |
+
"total_flos": 4.15045601427456e+18,
|
| 10 |
+
"train_loss": 1.393887762069702,
|
| 11 |
+
"train_runtime": 12787.317,
|
| 12 |
+
"train_samples": 45223,
|
| 13 |
+
"train_samples_per_second": 5.005,
|
| 14 |
+
"train_steps_per_second": 0.156
|
| 15 |
+
}
|
eval_results.json
ADDED
|
@@ -0,0 +1,9 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"epoch": 1.4149274849663955,
|
| 3 |
+
"eval_loss": 0.5812551379203796,
|
| 4 |
+
"eval_runtime": 1650.3013,
|
| 5 |
+
"eval_samples": 10763,
|
| 6 |
+
"eval_samples_per_second": 6.522,
|
| 7 |
+
"eval_steps_per_second": 0.408,
|
| 8 |
+
"eval_wer": 1.254253696315419
|
| 9 |
+
}
|
runs/Nov07_16-24-20_4b49612c0fe9/events.out.tfevents.1731011318.4b49612c0fe9.2373.1
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:332dfb6f9f9d583f22386ae0dc9c1996bd6c95a817315e9e24a621381aa51f89
|
| 3 |
+
size 406
|
train_results.json
ADDED
|
@@ -0,0 +1,9 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"epoch": 1.4149274849663955,
|
| 3 |
+
"total_flos": 4.15045601427456e+18,
|
| 4 |
+
"train_loss": 1.393887762069702,
|
| 5 |
+
"train_runtime": 12787.317,
|
| 6 |
+
"train_samples": 45223,
|
| 7 |
+
"train_samples_per_second": 5.005,
|
| 8 |
+
"train_steps_per_second": 0.156
|
| 9 |
+
}
|
trainer_state.json
ADDED
|
@@ -0,0 +1,620 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"best_metric": null,
|
| 3 |
+
"best_model_checkpoint": null,
|
| 4 |
+
"epoch": 1.4149274849663955,
|
| 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": 0.017686593562079942,
|
| 13 |
+
"grad_norm": 28.968021392822266,
|
| 14 |
+
"learning_rate": 4.6000000000000004e-07,
|
| 15 |
+
"loss": 3.698,
|
| 16 |
+
"step": 25
|
| 17 |
+
},
|
| 18 |
+
{
|
| 19 |
+
"epoch": 0.035373187124159884,
|
| 20 |
+
"grad_norm": 21.002214431762695,
|
| 21 |
+
"learning_rate": 9.600000000000001e-07,
|
| 22 |
+
"loss": 3.5124,
|
| 23 |
+
"step": 50
|
| 24 |
+
},
|
| 25 |
+
{
|
| 26 |
+
"epoch": 0.05305978068623983,
|
| 27 |
+
"grad_norm": 21.64297866821289,
|
| 28 |
+
"learning_rate": 1.46e-06,
|
| 29 |
+
"loss": 3.2335,
|
| 30 |
+
"step": 75
|
| 31 |
+
},
|
| 32 |
+
{
|
| 33 |
+
"epoch": 0.07074637424831977,
|
| 34 |
+
"grad_norm": 19.800973892211914,
|
| 35 |
+
"learning_rate": 1.9600000000000003e-06,
|
| 36 |
+
"loss": 2.9933,
|
| 37 |
+
"step": 100
|
| 38 |
+
},
|
| 39 |
+
{
|
| 40 |
+
"epoch": 0.08843296781039972,
|
| 41 |
+
"grad_norm": 19.281307220458984,
|
| 42 |
+
"learning_rate": 2.46e-06,
|
| 43 |
+
"loss": 2.5097,
|
| 44 |
+
"step": 125
|
| 45 |
+
},
|
| 46 |
+
{
|
| 47 |
+
"epoch": 0.10611956137247966,
|
| 48 |
+
"grad_norm": 19.187774658203125,
|
| 49 |
+
"learning_rate": 2.96e-06,
|
| 50 |
+
"loss": 2.3572,
|
| 51 |
+
"step": 150
|
| 52 |
+
},
|
| 53 |
+
{
|
| 54 |
+
"epoch": 0.1238061549345596,
|
| 55 |
+
"grad_norm": 17.483562469482422,
|
| 56 |
+
"learning_rate": 3.46e-06,
|
| 57 |
+
"loss": 2.1479,
|
| 58 |
+
"step": 175
|
| 59 |
+
},
|
| 60 |
+
{
|
| 61 |
+
"epoch": 0.14149274849663954,
|
| 62 |
+
"grad_norm": 17.524105072021484,
|
| 63 |
+
"learning_rate": 3.96e-06,
|
| 64 |
+
"loss": 2.2237,
|
| 65 |
+
"step": 200
|
| 66 |
+
},
|
| 67 |
+
{
|
| 68 |
+
"epoch": 0.1591793420587195,
|
| 69 |
+
"grad_norm": 17.354618072509766,
|
| 70 |
+
"learning_rate": 4.4600000000000005e-06,
|
| 71 |
+
"loss": 2.0436,
|
| 72 |
+
"step": 225
|
| 73 |
+
},
|
| 74 |
+
{
|
| 75 |
+
"epoch": 0.17686593562079944,
|
| 76 |
+
"grad_norm": 18.539562225341797,
|
| 77 |
+
"learning_rate": 4.960000000000001e-06,
|
| 78 |
+
"loss": 2.0212,
|
| 79 |
+
"step": 250
|
| 80 |
+
},
|
| 81 |
+
{
|
| 82 |
+
"epoch": 0.19455252918287938,
|
| 83 |
+
"grad_norm": 18.09463882446289,
|
| 84 |
+
"learning_rate": 5.460000000000001e-06,
|
| 85 |
+
"loss": 1.9828,
|
| 86 |
+
"step": 275
|
| 87 |
+
},
|
| 88 |
+
{
|
| 89 |
+
"epoch": 0.21223912274495932,
|
| 90 |
+
"grad_norm": 16.73234748840332,
|
| 91 |
+
"learning_rate": 5.9600000000000005e-06,
|
| 92 |
+
"loss": 1.9199,
|
| 93 |
+
"step": 300
|
| 94 |
+
},
|
| 95 |
+
{
|
| 96 |
+
"epoch": 0.22992571630703926,
|
| 97 |
+
"grad_norm": 17.19611930847168,
|
| 98 |
+
"learning_rate": 6.460000000000001e-06,
|
| 99 |
+
"loss": 1.8943,
|
| 100 |
+
"step": 325
|
| 101 |
+
},
|
| 102 |
+
{
|
| 103 |
+
"epoch": 0.2476123098691192,
|
| 104 |
+
"grad_norm": 18.489797592163086,
|
| 105 |
+
"learning_rate": 6.96e-06,
|
| 106 |
+
"loss": 1.7673,
|
| 107 |
+
"step": 350
|
| 108 |
+
},
|
| 109 |
+
{
|
| 110 |
+
"epoch": 0.26529890343119916,
|
| 111 |
+
"grad_norm": 17.019506454467773,
|
| 112 |
+
"learning_rate": 7.4600000000000006e-06,
|
| 113 |
+
"loss": 1.6931,
|
| 114 |
+
"step": 375
|
| 115 |
+
},
|
| 116 |
+
{
|
| 117 |
+
"epoch": 0.2829854969932791,
|
| 118 |
+
"grad_norm": 16.345083236694336,
|
| 119 |
+
"learning_rate": 7.960000000000002e-06,
|
| 120 |
+
"loss": 1.6946,
|
| 121 |
+
"step": 400
|
| 122 |
+
},
|
| 123 |
+
{
|
| 124 |
+
"epoch": 0.30067209055535904,
|
| 125 |
+
"grad_norm": 16.138513565063477,
|
| 126 |
+
"learning_rate": 8.46e-06,
|
| 127 |
+
"loss": 1.6849,
|
| 128 |
+
"step": 425
|
| 129 |
+
},
|
| 130 |
+
{
|
| 131 |
+
"epoch": 0.318358684117439,
|
| 132 |
+
"grad_norm": 17.647178649902344,
|
| 133 |
+
"learning_rate": 8.96e-06,
|
| 134 |
+
"loss": 1.677,
|
| 135 |
+
"step": 450
|
| 136 |
+
},
|
| 137 |
+
{
|
| 138 |
+
"epoch": 0.3360452776795189,
|
| 139 |
+
"grad_norm": 17.51073455810547,
|
| 140 |
+
"learning_rate": 9.460000000000001e-06,
|
| 141 |
+
"loss": 1.608,
|
| 142 |
+
"step": 475
|
| 143 |
+
},
|
| 144 |
+
{
|
| 145 |
+
"epoch": 0.3537318712415989,
|
| 146 |
+
"grad_norm": 15.542696952819824,
|
| 147 |
+
"learning_rate": 9.960000000000001e-06,
|
| 148 |
+
"loss": 1.5617,
|
| 149 |
+
"step": 500
|
| 150 |
+
},
|
| 151 |
+
{
|
| 152 |
+
"epoch": 0.3714184648036788,
|
| 153 |
+
"grad_norm": 14.3616361618042,
|
| 154 |
+
"learning_rate": 9.846666666666668e-06,
|
| 155 |
+
"loss": 1.4832,
|
| 156 |
+
"step": 525
|
| 157 |
+
},
|
| 158 |
+
{
|
| 159 |
+
"epoch": 0.38910505836575876,
|
| 160 |
+
"grad_norm": 14.655681610107422,
|
| 161 |
+
"learning_rate": 9.68e-06,
|
| 162 |
+
"loss": 1.521,
|
| 163 |
+
"step": 550
|
| 164 |
+
},
|
| 165 |
+
{
|
| 166 |
+
"epoch": 0.40679165192783867,
|
| 167 |
+
"grad_norm": 15.943355560302734,
|
| 168 |
+
"learning_rate": 9.513333333333334e-06,
|
| 169 |
+
"loss": 1.4946,
|
| 170 |
+
"step": 575
|
| 171 |
+
},
|
| 172 |
+
{
|
| 173 |
+
"epoch": 0.42447824548991864,
|
| 174 |
+
"grad_norm": 17.380294799804688,
|
| 175 |
+
"learning_rate": 9.346666666666666e-06,
|
| 176 |
+
"loss": 1.4358,
|
| 177 |
+
"step": 600
|
| 178 |
+
},
|
| 179 |
+
{
|
| 180 |
+
"epoch": 0.4421648390519986,
|
| 181 |
+
"grad_norm": 16.160900115966797,
|
| 182 |
+
"learning_rate": 9.180000000000002e-06,
|
| 183 |
+
"loss": 1.4093,
|
| 184 |
+
"step": 625
|
| 185 |
+
},
|
| 186 |
+
{
|
| 187 |
+
"epoch": 0.4598514326140785,
|
| 188 |
+
"grad_norm": 14.763204574584961,
|
| 189 |
+
"learning_rate": 9.013333333333334e-06,
|
| 190 |
+
"loss": 1.4101,
|
| 191 |
+
"step": 650
|
| 192 |
+
},
|
| 193 |
+
{
|
| 194 |
+
"epoch": 0.4775380261761585,
|
| 195 |
+
"grad_norm": 15.164654731750488,
|
| 196 |
+
"learning_rate": 8.846666666666668e-06,
|
| 197 |
+
"loss": 1.3897,
|
| 198 |
+
"step": 675
|
| 199 |
+
},
|
| 200 |
+
{
|
| 201 |
+
"epoch": 0.4952246197382384,
|
| 202 |
+
"grad_norm": 15.769131660461426,
|
| 203 |
+
"learning_rate": 8.68e-06,
|
| 204 |
+
"loss": 1.3422,
|
| 205 |
+
"step": 700
|
| 206 |
+
},
|
| 207 |
+
{
|
| 208 |
+
"epoch": 0.5129112133003184,
|
| 209 |
+
"grad_norm": 16.859813690185547,
|
| 210 |
+
"learning_rate": 8.513333333333335e-06,
|
| 211 |
+
"loss": 1.3451,
|
| 212 |
+
"step": 725
|
| 213 |
+
},
|
| 214 |
+
{
|
| 215 |
+
"epoch": 0.5305978068623983,
|
| 216 |
+
"grad_norm": 14.078543663024902,
|
| 217 |
+
"learning_rate": 8.346666666666668e-06,
|
| 218 |
+
"loss": 1.3333,
|
| 219 |
+
"step": 750
|
| 220 |
+
},
|
| 221 |
+
{
|
| 222 |
+
"epoch": 0.5482844004244782,
|
| 223 |
+
"grad_norm": 13.418024063110352,
|
| 224 |
+
"learning_rate": 8.18e-06,
|
| 225 |
+
"loss": 1.3618,
|
| 226 |
+
"step": 775
|
| 227 |
+
},
|
| 228 |
+
{
|
| 229 |
+
"epoch": 0.5659709939865581,
|
| 230 |
+
"grad_norm": 14.887019157409668,
|
| 231 |
+
"learning_rate": 8.013333333333333e-06,
|
| 232 |
+
"loss": 1.2983,
|
| 233 |
+
"step": 800
|
| 234 |
+
},
|
| 235 |
+
{
|
| 236 |
+
"epoch": 0.5836575875486382,
|
| 237 |
+
"grad_norm": 13.075471878051758,
|
| 238 |
+
"learning_rate": 7.846666666666667e-06,
|
| 239 |
+
"loss": 1.3021,
|
| 240 |
+
"step": 825
|
| 241 |
+
},
|
| 242 |
+
{
|
| 243 |
+
"epoch": 0.6013441811107181,
|
| 244 |
+
"grad_norm": 14.660274505615234,
|
| 245 |
+
"learning_rate": 7.680000000000001e-06,
|
| 246 |
+
"loss": 1.2375,
|
| 247 |
+
"step": 850
|
| 248 |
+
},
|
| 249 |
+
{
|
| 250 |
+
"epoch": 0.619030774672798,
|
| 251 |
+
"grad_norm": 16.32628631591797,
|
| 252 |
+
"learning_rate": 7.513333333333334e-06,
|
| 253 |
+
"loss": 1.3038,
|
| 254 |
+
"step": 875
|
| 255 |
+
},
|
| 256 |
+
{
|
| 257 |
+
"epoch": 0.636717368234878,
|
| 258 |
+
"grad_norm": 17.701143264770508,
|
| 259 |
+
"learning_rate": 7.346666666666668e-06,
|
| 260 |
+
"loss": 1.2082,
|
| 261 |
+
"step": 900
|
| 262 |
+
},
|
| 263 |
+
{
|
| 264 |
+
"epoch": 0.6544039617969579,
|
| 265 |
+
"grad_norm": 14.465027809143066,
|
| 266 |
+
"learning_rate": 7.180000000000001e-06,
|
| 267 |
+
"loss": 1.2543,
|
| 268 |
+
"step": 925
|
| 269 |
+
},
|
| 270 |
+
{
|
| 271 |
+
"epoch": 0.6720905553590378,
|
| 272 |
+
"grad_norm": 12.923417091369629,
|
| 273 |
+
"learning_rate": 7.0133333333333345e-06,
|
| 274 |
+
"loss": 1.2544,
|
| 275 |
+
"step": 950
|
| 276 |
+
},
|
| 277 |
+
{
|
| 278 |
+
"epoch": 0.6897771489211177,
|
| 279 |
+
"grad_norm": 15.675958633422852,
|
| 280 |
+
"learning_rate": 6.846666666666667e-06,
|
| 281 |
+
"loss": 1.2279,
|
| 282 |
+
"step": 975
|
| 283 |
+
},
|
| 284 |
+
{
|
| 285 |
+
"epoch": 0.7074637424831978,
|
| 286 |
+
"grad_norm": 15.041681289672852,
|
| 287 |
+
"learning_rate": 6.680000000000001e-06,
|
| 288 |
+
"loss": 1.2368,
|
| 289 |
+
"step": 1000
|
| 290 |
+
},
|
| 291 |
+
{
|
| 292 |
+
"epoch": 0.7074637424831978,
|
| 293 |
+
"eval_loss": 0.6446413397789001,
|
| 294 |
+
"eval_runtime": 2060.0729,
|
| 295 |
+
"eval_samples_per_second": 5.225,
|
| 296 |
+
"eval_steps_per_second": 0.327,
|
| 297 |
+
"eval_wer": 0.965980208088868,
|
| 298 |
+
"step": 1000
|
| 299 |
+
},
|
| 300 |
+
{
|
| 301 |
+
"epoch": 0.7251503360452777,
|
| 302 |
+
"grad_norm": 15.67192554473877,
|
| 303 |
+
"learning_rate": 6.513333333333333e-06,
|
| 304 |
+
"loss": 1.1917,
|
| 305 |
+
"step": 1025
|
| 306 |
+
},
|
| 307 |
+
{
|
| 308 |
+
"epoch": 0.7428369296073576,
|
| 309 |
+
"grad_norm": 14.699666023254395,
|
| 310 |
+
"learning_rate": 6.346666666666668e-06,
|
| 311 |
+
"loss": 1.1812,
|
| 312 |
+
"step": 1050
|
| 313 |
+
},
|
| 314 |
+
{
|
| 315 |
+
"epoch": 0.7605235231694376,
|
| 316 |
+
"grad_norm": 14.321735382080078,
|
| 317 |
+
"learning_rate": 6.18e-06,
|
| 318 |
+
"loss": 1.2046,
|
| 319 |
+
"step": 1075
|
| 320 |
+
},
|
| 321 |
+
{
|
| 322 |
+
"epoch": 0.7782101167315175,
|
| 323 |
+
"grad_norm": 14.692147254943848,
|
| 324 |
+
"learning_rate": 6.013333333333335e-06,
|
| 325 |
+
"loss": 1.2071,
|
| 326 |
+
"step": 1100
|
| 327 |
+
},
|
| 328 |
+
{
|
| 329 |
+
"epoch": 0.7958967102935974,
|
| 330 |
+
"grad_norm": 13.778362274169922,
|
| 331 |
+
"learning_rate": 5.846666666666667e-06,
|
| 332 |
+
"loss": 1.1476,
|
| 333 |
+
"step": 1125
|
| 334 |
+
},
|
| 335 |
+
{
|
| 336 |
+
"epoch": 0.8135833038556773,
|
| 337 |
+
"grad_norm": 13.074742317199707,
|
| 338 |
+
"learning_rate": 5.68e-06,
|
| 339 |
+
"loss": 1.1543,
|
| 340 |
+
"step": 1150
|
| 341 |
+
},
|
| 342 |
+
{
|
| 343 |
+
"epoch": 0.8312698974177574,
|
| 344 |
+
"grad_norm": 14.026896476745605,
|
| 345 |
+
"learning_rate": 5.513333333333334e-06,
|
| 346 |
+
"loss": 1.1841,
|
| 347 |
+
"step": 1175
|
| 348 |
+
},
|
| 349 |
+
{
|
| 350 |
+
"epoch": 0.8489564909798373,
|
| 351 |
+
"grad_norm": 13.244227409362793,
|
| 352 |
+
"learning_rate": 5.346666666666667e-06,
|
| 353 |
+
"loss": 1.1698,
|
| 354 |
+
"step": 1200
|
| 355 |
+
},
|
| 356 |
+
{
|
| 357 |
+
"epoch": 0.8666430845419172,
|
| 358 |
+
"grad_norm": 15.449938774108887,
|
| 359 |
+
"learning_rate": 5.18e-06,
|
| 360 |
+
"loss": 1.1213,
|
| 361 |
+
"step": 1225
|
| 362 |
+
},
|
| 363 |
+
{
|
| 364 |
+
"epoch": 0.8843296781039972,
|
| 365 |
+
"grad_norm": 13.706199645996094,
|
| 366 |
+
"learning_rate": 5.013333333333333e-06,
|
| 367 |
+
"loss": 1.1935,
|
| 368 |
+
"step": 1250
|
| 369 |
+
},
|
| 370 |
+
{
|
| 371 |
+
"epoch": 0.9020162716660771,
|
| 372 |
+
"grad_norm": 14.426807403564453,
|
| 373 |
+
"learning_rate": 4.846666666666667e-06,
|
| 374 |
+
"loss": 1.0659,
|
| 375 |
+
"step": 1275
|
| 376 |
+
},
|
| 377 |
+
{
|
| 378 |
+
"epoch": 0.919702865228157,
|
| 379 |
+
"grad_norm": 14.945075035095215,
|
| 380 |
+
"learning_rate": 4.680000000000001e-06,
|
| 381 |
+
"loss": 1.1169,
|
| 382 |
+
"step": 1300
|
| 383 |
+
},
|
| 384 |
+
{
|
| 385 |
+
"epoch": 0.937389458790237,
|
| 386 |
+
"grad_norm": 13.335543632507324,
|
| 387 |
+
"learning_rate": 4.513333333333333e-06,
|
| 388 |
+
"loss": 1.0949,
|
| 389 |
+
"step": 1325
|
| 390 |
+
},
|
| 391 |
+
{
|
| 392 |
+
"epoch": 0.955076052352317,
|
| 393 |
+
"grad_norm": 15.83086109161377,
|
| 394 |
+
"learning_rate": 4.346666666666667e-06,
|
| 395 |
+
"loss": 1.0845,
|
| 396 |
+
"step": 1350
|
| 397 |
+
},
|
| 398 |
+
{
|
| 399 |
+
"epoch": 0.9727626459143969,
|
| 400 |
+
"grad_norm": 12.246453285217285,
|
| 401 |
+
"learning_rate": 4.18e-06,
|
| 402 |
+
"loss": 1.148,
|
| 403 |
+
"step": 1375
|
| 404 |
+
},
|
| 405 |
+
{
|
| 406 |
+
"epoch": 0.9904492394764768,
|
| 407 |
+
"grad_norm": 16.202903747558594,
|
| 408 |
+
"learning_rate": 4.013333333333334e-06,
|
| 409 |
+
"loss": 1.1118,
|
| 410 |
+
"step": 1400
|
| 411 |
+
},
|
| 412 |
+
{
|
| 413 |
+
"epoch": 1.0081358330385568,
|
| 414 |
+
"grad_norm": 12.500658988952637,
|
| 415 |
+
"learning_rate": 3.8466666666666665e-06,
|
| 416 |
+
"loss": 1.0352,
|
| 417 |
+
"step": 1425
|
| 418 |
+
},
|
| 419 |
+
{
|
| 420 |
+
"epoch": 1.0258224266006368,
|
| 421 |
+
"grad_norm": 11.863042831420898,
|
| 422 |
+
"learning_rate": 3.6800000000000003e-06,
|
| 423 |
+
"loss": 1.0224,
|
| 424 |
+
"step": 1450
|
| 425 |
+
},
|
| 426 |
+
{
|
| 427 |
+
"epoch": 1.0435090201627166,
|
| 428 |
+
"grad_norm": 13.093315124511719,
|
| 429 |
+
"learning_rate": 3.5133333333333337e-06,
|
| 430 |
+
"loss": 0.966,
|
| 431 |
+
"step": 1475
|
| 432 |
+
},
|
| 433 |
+
{
|
| 434 |
+
"epoch": 1.0611956137247966,
|
| 435 |
+
"grad_norm": 11.886918067932129,
|
| 436 |
+
"learning_rate": 3.346666666666667e-06,
|
| 437 |
+
"loss": 0.9643,
|
| 438 |
+
"step": 1500
|
| 439 |
+
},
|
| 440 |
+
{
|
| 441 |
+
"epoch": 1.0788822072868764,
|
| 442 |
+
"grad_norm": 15.925506591796875,
|
| 443 |
+
"learning_rate": 3.1800000000000005e-06,
|
| 444 |
+
"loss": 0.8792,
|
| 445 |
+
"step": 1525
|
| 446 |
+
},
|
| 447 |
+
{
|
| 448 |
+
"epoch": 1.0965688008489565,
|
| 449 |
+
"grad_norm": 12.430258750915527,
|
| 450 |
+
"learning_rate": 3.013333333333334e-06,
|
| 451 |
+
"loss": 0.9462,
|
| 452 |
+
"step": 1550
|
| 453 |
+
},
|
| 454 |
+
{
|
| 455 |
+
"epoch": 1.1142553944110365,
|
| 456 |
+
"grad_norm": 13.365354537963867,
|
| 457 |
+
"learning_rate": 2.8466666666666672e-06,
|
| 458 |
+
"loss": 0.9246,
|
| 459 |
+
"step": 1575
|
| 460 |
+
},
|
| 461 |
+
{
|
| 462 |
+
"epoch": 1.1319419879731163,
|
| 463 |
+
"grad_norm": 13.153193473815918,
|
| 464 |
+
"learning_rate": 2.68e-06,
|
| 465 |
+
"loss": 0.8996,
|
| 466 |
+
"step": 1600
|
| 467 |
+
},
|
| 468 |
+
{
|
| 469 |
+
"epoch": 1.1496285815351963,
|
| 470 |
+
"grad_norm": 13.20574951171875,
|
| 471 |
+
"learning_rate": 2.5133333333333336e-06,
|
| 472 |
+
"loss": 0.943,
|
| 473 |
+
"step": 1625
|
| 474 |
+
},
|
| 475 |
+
{
|
| 476 |
+
"epoch": 1.1673151750972763,
|
| 477 |
+
"grad_norm": 10.716314315795898,
|
| 478 |
+
"learning_rate": 2.346666666666667e-06,
|
| 479 |
+
"loss": 0.9071,
|
| 480 |
+
"step": 1650
|
| 481 |
+
},
|
| 482 |
+
{
|
| 483 |
+
"epoch": 1.1850017686593561,
|
| 484 |
+
"grad_norm": 15.43217658996582,
|
| 485 |
+
"learning_rate": 2.1800000000000003e-06,
|
| 486 |
+
"loss": 0.9232,
|
| 487 |
+
"step": 1675
|
| 488 |
+
},
|
| 489 |
+
{
|
| 490 |
+
"epoch": 1.2026883622214362,
|
| 491 |
+
"grad_norm": 14.106558799743652,
|
| 492 |
+
"learning_rate": 2.0133333333333337e-06,
|
| 493 |
+
"loss": 0.8998,
|
| 494 |
+
"step": 1700
|
| 495 |
+
},
|
| 496 |
+
{
|
| 497 |
+
"epoch": 1.2203749557835162,
|
| 498 |
+
"grad_norm": 12.580400466918945,
|
| 499 |
+
"learning_rate": 1.8466666666666668e-06,
|
| 500 |
+
"loss": 0.928,
|
| 501 |
+
"step": 1725
|
| 502 |
+
},
|
| 503 |
+
{
|
| 504 |
+
"epoch": 1.238061549345596,
|
| 505 |
+
"grad_norm": 12.888578414916992,
|
| 506 |
+
"learning_rate": 1.6800000000000002e-06,
|
| 507 |
+
"loss": 0.9354,
|
| 508 |
+
"step": 1750
|
| 509 |
+
},
|
| 510 |
+
{
|
| 511 |
+
"epoch": 1.255748142907676,
|
| 512 |
+
"grad_norm": 11.692194938659668,
|
| 513 |
+
"learning_rate": 1.5133333333333334e-06,
|
| 514 |
+
"loss": 0.8441,
|
| 515 |
+
"step": 1775
|
| 516 |
+
},
|
| 517 |
+
{
|
| 518 |
+
"epoch": 1.2734347364697558,
|
| 519 |
+
"grad_norm": 14.173469543457031,
|
| 520 |
+
"learning_rate": 1.3466666666666668e-06,
|
| 521 |
+
"loss": 0.925,
|
| 522 |
+
"step": 1800
|
| 523 |
+
},
|
| 524 |
+
{
|
| 525 |
+
"epoch": 1.2911213300318358,
|
| 526 |
+
"grad_norm": 12.65889835357666,
|
| 527 |
+
"learning_rate": 1.1800000000000001e-06,
|
| 528 |
+
"loss": 0.905,
|
| 529 |
+
"step": 1825
|
| 530 |
+
},
|
| 531 |
+
{
|
| 532 |
+
"epoch": 1.3088079235939158,
|
| 533 |
+
"grad_norm": 12.559049606323242,
|
| 534 |
+
"learning_rate": 1.0133333333333333e-06,
|
| 535 |
+
"loss": 0.8771,
|
| 536 |
+
"step": 1850
|
| 537 |
+
},
|
| 538 |
+
{
|
| 539 |
+
"epoch": 1.3264945171559956,
|
| 540 |
+
"grad_norm": 11.348992347717285,
|
| 541 |
+
"learning_rate": 8.466666666666668e-07,
|
| 542 |
+
"loss": 0.9033,
|
| 543 |
+
"step": 1875
|
| 544 |
+
},
|
| 545 |
+
{
|
| 546 |
+
"epoch": 1.3441811107180757,
|
| 547 |
+
"grad_norm": 11.88202953338623,
|
| 548 |
+
"learning_rate": 6.800000000000001e-07,
|
| 549 |
+
"loss": 0.8451,
|
| 550 |
+
"step": 1900
|
| 551 |
+
},
|
| 552 |
+
{
|
| 553 |
+
"epoch": 1.3618677042801557,
|
| 554 |
+
"grad_norm": 11.244508743286133,
|
| 555 |
+
"learning_rate": 5.133333333333334e-07,
|
| 556 |
+
"loss": 0.8839,
|
| 557 |
+
"step": 1925
|
| 558 |
+
},
|
| 559 |
+
{
|
| 560 |
+
"epoch": 1.3795542978422355,
|
| 561 |
+
"grad_norm": 12.49364185333252,
|
| 562 |
+
"learning_rate": 3.466666666666667e-07,
|
| 563 |
+
"loss": 0.8928,
|
| 564 |
+
"step": 1950
|
| 565 |
+
},
|
| 566 |
+
{
|
| 567 |
+
"epoch": 1.3972408914043155,
|
| 568 |
+
"grad_norm": 11.834848403930664,
|
| 569 |
+
"learning_rate": 1.8e-07,
|
| 570 |
+
"loss": 0.9105,
|
| 571 |
+
"step": 1975
|
| 572 |
+
},
|
| 573 |
+
{
|
| 574 |
+
"epoch": 1.4149274849663955,
|
| 575 |
+
"grad_norm": 12.054742813110352,
|
| 576 |
+
"learning_rate": 1.3333333333333334e-08,
|
| 577 |
+
"loss": 0.8994,
|
| 578 |
+
"step": 2000
|
| 579 |
+
},
|
| 580 |
+
{
|
| 581 |
+
"epoch": 1.4149274849663955,
|
| 582 |
+
"eval_loss": 0.5812551379203796,
|
| 583 |
+
"eval_runtime": 1719.3277,
|
| 584 |
+
"eval_samples_per_second": 6.26,
|
| 585 |
+
"eval_steps_per_second": 0.391,
|
| 586 |
+
"eval_wer": 1.254253696315419,
|
| 587 |
+
"step": 2000
|
| 588 |
+
},
|
| 589 |
+
{
|
| 590 |
+
"epoch": 1.4149274849663955,
|
| 591 |
+
"step": 2000,
|
| 592 |
+
"total_flos": 4.15045601427456e+18,
|
| 593 |
+
"train_loss": 1.393887762069702,
|
| 594 |
+
"train_runtime": 12787.317,
|
| 595 |
+
"train_samples_per_second": 5.005,
|
| 596 |
+
"train_steps_per_second": 0.156
|
| 597 |
+
}
|
| 598 |
+
],
|
| 599 |
+
"logging_steps": 25,
|
| 600 |
+
"max_steps": 2000,
|
| 601 |
+
"num_input_tokens_seen": 0,
|
| 602 |
+
"num_train_epochs": 2,
|
| 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": 4.15045601427456e+18,
|
| 617 |
+
"train_batch_size": 16,
|
| 618 |
+
"trial_name": null,
|
| 619 |
+
"trial_params": null
|
| 620 |
+
}
|