Commit
•
7849601
1
Parent(s):
1c3f04a
End of training
Browse files- all_results.json +14 -0
- eval_results.json +9 -0
- train_results.json +8 -0
- trainer_state.json +790 -0
all_results.json
ADDED
|
@@ -0,0 +1,14 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"epoch": 100.0,
|
| 3 |
+
"eval_loss": 1.5443140268325806,
|
| 4 |
+
"eval_runtime": 12.5724,
|
| 5 |
+
"eval_samples": 341,
|
| 6 |
+
"eval_samples_per_second": 27.123,
|
| 7 |
+
"eval_steps_per_second": 3.42,
|
| 8 |
+
"eval_wer": 0.7029520295202952,
|
| 9 |
+
"train_loss": 1.1548647121354645,
|
| 10 |
+
"train_runtime": 5602.3197,
|
| 11 |
+
"train_samples": 810,
|
| 12 |
+
"train_samples_per_second": 14.458,
|
| 13 |
+
"train_steps_per_second": 0.91
|
| 14 |
+
}
|
eval_results.json
ADDED
|
@@ -0,0 +1,9 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"epoch": 100.0,
|
| 3 |
+
"eval_loss": 1.5443140268325806,
|
| 4 |
+
"eval_runtime": 12.5724,
|
| 5 |
+
"eval_samples": 341,
|
| 6 |
+
"eval_samples_per_second": 27.123,
|
| 7 |
+
"eval_steps_per_second": 3.42,
|
| 8 |
+
"eval_wer": 0.7029520295202952
|
| 9 |
+
}
|
train_results.json
ADDED
|
@@ -0,0 +1,8 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"epoch": 100.0,
|
| 3 |
+
"train_loss": 1.1548647121354645,
|
| 4 |
+
"train_runtime": 5602.3197,
|
| 5 |
+
"train_samples": 810,
|
| 6 |
+
"train_samples_per_second": 14.458,
|
| 7 |
+
"train_steps_per_second": 0.91
|
| 8 |
+
}
|
trainer_state.json
ADDED
|
@@ -0,0 +1,790 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"best_metric": null,
|
| 3 |
+
"best_model_checkpoint": null,
|
| 4 |
+
"epoch": 100.0,
|
| 5 |
+
"global_step": 5100,
|
| 6 |
+
"is_hyper_param_search": false,
|
| 7 |
+
"is_local_process_zero": true,
|
| 8 |
+
"is_world_process_zero": true,
|
| 9 |
+
"log_history": [
|
| 10 |
+
{
|
| 11 |
+
"epoch": 1.96,
|
| 12 |
+
"learning_rate": 5.069866666666666e-05,
|
| 13 |
+
"loss": 10.7052,
|
| 14 |
+
"step": 100
|
| 15 |
+
},
|
| 16 |
+
{
|
| 17 |
+
"epoch": 1.96,
|
| 18 |
+
"eval_loss": 3.4682674407958984,
|
| 19 |
+
"eval_runtime": 13.0004,
|
| 20 |
+
"eval_samples_per_second": 26.23,
|
| 21 |
+
"eval_steps_per_second": 3.308,
|
| 22 |
+
"eval_wer": 1.0,
|
| 23 |
+
"step": 100
|
| 24 |
+
},
|
| 25 |
+
{
|
| 26 |
+
"epoch": 3.92,
|
| 27 |
+
"learning_rate": 0.000102432,
|
| 28 |
+
"loss": 3.2395,
|
| 29 |
+
"step": 200
|
| 30 |
+
},
|
| 31 |
+
{
|
| 32 |
+
"epoch": 3.92,
|
| 33 |
+
"eval_loss": 3.14892840385437,
|
| 34 |
+
"eval_runtime": 12.7764,
|
| 35 |
+
"eval_samples_per_second": 26.69,
|
| 36 |
+
"eval_steps_per_second": 3.366,
|
| 37 |
+
"eval_wer": 1.0,
|
| 38 |
+
"step": 200
|
| 39 |
+
},
|
| 40 |
+
{
|
| 41 |
+
"epoch": 5.88,
|
| 42 |
+
"learning_rate": 0.00015416533333333332,
|
| 43 |
+
"loss": 2.9951,
|
| 44 |
+
"step": 300
|
| 45 |
+
},
|
| 46 |
+
{
|
| 47 |
+
"epoch": 5.88,
|
| 48 |
+
"eval_loss": 2.982297897338867,
|
| 49 |
+
"eval_runtime": 12.7866,
|
| 50 |
+
"eval_samples_per_second": 26.668,
|
| 51 |
+
"eval_steps_per_second": 3.363,
|
| 52 |
+
"eval_wer": 1.0007380073800738,
|
| 53 |
+
"step": 300
|
| 54 |
+
},
|
| 55 |
+
{
|
| 56 |
+
"epoch": 7.84,
|
| 57 |
+
"learning_rate": 0.00020589866666666665,
|
| 58 |
+
"loss": 2.3574,
|
| 59 |
+
"step": 400
|
| 60 |
+
},
|
| 61 |
+
{
|
| 62 |
+
"epoch": 7.84,
|
| 63 |
+
"eval_loss": 1.2614495754241943,
|
| 64 |
+
"eval_runtime": 12.8042,
|
| 65 |
+
"eval_samples_per_second": 26.632,
|
| 66 |
+
"eval_steps_per_second": 3.358,
|
| 67 |
+
"eval_wer": 0.7597785977859779,
|
| 68 |
+
"step": 400
|
| 69 |
+
},
|
| 70 |
+
{
|
| 71 |
+
"epoch": 9.8,
|
| 72 |
+
"learning_rate": 0.000257632,
|
| 73 |
+
"loss": 1.7287,
|
| 74 |
+
"step": 500
|
| 75 |
+
},
|
| 76 |
+
{
|
| 77 |
+
"epoch": 9.8,
|
| 78 |
+
"eval_loss": 1.1816928386688232,
|
| 79 |
+
"eval_runtime": 12.7085,
|
| 80 |
+
"eval_samples_per_second": 26.833,
|
| 81 |
+
"eval_steps_per_second": 3.384,
|
| 82 |
+
"eval_wer": 0.7420664206642067,
|
| 83 |
+
"step": 500
|
| 84 |
+
},
|
| 85 |
+
{
|
| 86 |
+
"epoch": 11.76,
|
| 87 |
+
"learning_rate": 0.00030936533333333335,
|
| 88 |
+
"loss": 1.6144,
|
| 89 |
+
"step": 600
|
| 90 |
+
},
|
| 91 |
+
{
|
| 92 |
+
"epoch": 11.76,
|
| 93 |
+
"eval_loss": 1.131492257118225,
|
| 94 |
+
"eval_runtime": 12.8371,
|
| 95 |
+
"eval_samples_per_second": 26.564,
|
| 96 |
+
"eval_steps_per_second": 3.35,
|
| 97 |
+
"eval_wer": 0.7321033210332103,
|
| 98 |
+
"step": 600
|
| 99 |
+
},
|
| 100 |
+
{
|
| 101 |
+
"epoch": 13.73,
|
| 102 |
+
"learning_rate": 0.00036109866666666666,
|
| 103 |
+
"loss": 1.5598,
|
| 104 |
+
"step": 700
|
| 105 |
+
},
|
| 106 |
+
{
|
| 107 |
+
"epoch": 13.73,
|
| 108 |
+
"eval_loss": 1.232160210609436,
|
| 109 |
+
"eval_runtime": 12.7503,
|
| 110 |
+
"eval_samples_per_second": 26.745,
|
| 111 |
+
"eval_steps_per_second": 3.372,
|
| 112 |
+
"eval_wer": 0.7549815498154981,
|
| 113 |
+
"step": 700
|
| 114 |
+
},
|
| 115 |
+
{
|
| 116 |
+
"epoch": 15.69,
|
| 117 |
+
"learning_rate": 0.0003837186206896552,
|
| 118 |
+
"loss": 1.5418,
|
| 119 |
+
"step": 800
|
| 120 |
+
},
|
| 121 |
+
{
|
| 122 |
+
"epoch": 15.69,
|
| 123 |
+
"eval_loss": 1.272075891494751,
|
| 124 |
+
"eval_runtime": 12.8108,
|
| 125 |
+
"eval_samples_per_second": 26.618,
|
| 126 |
+
"eval_steps_per_second": 3.357,
|
| 127 |
+
"eval_wer": 0.7819188191881918,
|
| 128 |
+
"step": 800
|
| 129 |
+
},
|
| 130 |
+
{
|
| 131 |
+
"epoch": 17.65,
|
| 132 |
+
"learning_rate": 0.0003747990804597701,
|
| 133 |
+
"loss": 1.4578,
|
| 134 |
+
"step": 900
|
| 135 |
+
},
|
| 136 |
+
{
|
| 137 |
+
"epoch": 17.65,
|
| 138 |
+
"eval_loss": 1.1709508895874023,
|
| 139 |
+
"eval_runtime": 12.7848,
|
| 140 |
+
"eval_samples_per_second": 26.672,
|
| 141 |
+
"eval_steps_per_second": 3.363,
|
| 142 |
+
"eval_wer": 0.7531365313653137,
|
| 143 |
+
"step": 900
|
| 144 |
+
},
|
| 145 |
+
{
|
| 146 |
+
"epoch": 19.61,
|
| 147 |
+
"learning_rate": 0.0003658795402298851,
|
| 148 |
+
"loss": 1.4311,
|
| 149 |
+
"step": 1000
|
| 150 |
+
},
|
| 151 |
+
{
|
| 152 |
+
"epoch": 19.61,
|
| 153 |
+
"eval_loss": 1.2042367458343506,
|
| 154 |
+
"eval_runtime": 12.6761,
|
| 155 |
+
"eval_samples_per_second": 26.901,
|
| 156 |
+
"eval_steps_per_second": 3.392,
|
| 157 |
+
"eval_wer": 0.7490774907749077,
|
| 158 |
+
"step": 1000
|
| 159 |
+
},
|
| 160 |
+
{
|
| 161 |
+
"epoch": 21.57,
|
| 162 |
+
"learning_rate": 0.00035696,
|
| 163 |
+
"loss": 1.3483,
|
| 164 |
+
"step": 1100
|
| 165 |
+
},
|
| 166 |
+
{
|
| 167 |
+
"epoch": 21.57,
|
| 168 |
+
"eval_loss": 1.1702170372009277,
|
| 169 |
+
"eval_runtime": 12.5762,
|
| 170 |
+
"eval_samples_per_second": 27.115,
|
| 171 |
+
"eval_steps_per_second": 3.419,
|
| 172 |
+
"eval_wer": 0.7464944649446494,
|
| 173 |
+
"step": 1100
|
| 174 |
+
},
|
| 175 |
+
{
|
| 176 |
+
"epoch": 23.53,
|
| 177 |
+
"learning_rate": 0.0003480404597701149,
|
| 178 |
+
"loss": 1.3078,
|
| 179 |
+
"step": 1200
|
| 180 |
+
},
|
| 181 |
+
{
|
| 182 |
+
"epoch": 23.53,
|
| 183 |
+
"eval_loss": 1.1963412761688232,
|
| 184 |
+
"eval_runtime": 12.6937,
|
| 185 |
+
"eval_samples_per_second": 26.864,
|
| 186 |
+
"eval_steps_per_second": 3.387,
|
| 187 |
+
"eval_wer": 0.7420664206642067,
|
| 188 |
+
"step": 1200
|
| 189 |
+
},
|
| 190 |
+
{
|
| 191 |
+
"epoch": 25.49,
|
| 192 |
+
"learning_rate": 0.00033912091954022987,
|
| 193 |
+
"loss": 1.2576,
|
| 194 |
+
"step": 1300
|
| 195 |
+
},
|
| 196 |
+
{
|
| 197 |
+
"epoch": 25.49,
|
| 198 |
+
"eval_loss": 1.1501450538635254,
|
| 199 |
+
"eval_runtime": 12.6626,
|
| 200 |
+
"eval_samples_per_second": 26.93,
|
| 201 |
+
"eval_steps_per_second": 3.396,
|
| 202 |
+
"eval_wer": 0.7280442804428044,
|
| 203 |
+
"step": 1300
|
| 204 |
+
},
|
| 205 |
+
{
|
| 206 |
+
"epoch": 27.45,
|
| 207 |
+
"learning_rate": 0.0003302013793103448,
|
| 208 |
+
"loss": 1.2173,
|
| 209 |
+
"step": 1400
|
| 210 |
+
},
|
| 211 |
+
{
|
| 212 |
+
"epoch": 27.45,
|
| 213 |
+
"eval_loss": 1.2525919675827026,
|
| 214 |
+
"eval_runtime": 12.6262,
|
| 215 |
+
"eval_samples_per_second": 27.007,
|
| 216 |
+
"eval_steps_per_second": 3.406,
|
| 217 |
+
"eval_wer": 0.7298892988929889,
|
| 218 |
+
"step": 1400
|
| 219 |
+
},
|
| 220 |
+
{
|
| 221 |
+
"epoch": 29.41,
|
| 222 |
+
"learning_rate": 0.00032128183908045977,
|
| 223 |
+
"loss": 1.2217,
|
| 224 |
+
"step": 1500
|
| 225 |
+
},
|
| 226 |
+
{
|
| 227 |
+
"epoch": 29.41,
|
| 228 |
+
"eval_loss": 1.2478744983673096,
|
| 229 |
+
"eval_runtime": 12.8026,
|
| 230 |
+
"eval_samples_per_second": 26.635,
|
| 231 |
+
"eval_steps_per_second": 3.359,
|
| 232 |
+
"eval_wer": 0.7309963099630996,
|
| 233 |
+
"step": 1500
|
| 234 |
+
},
|
| 235 |
+
{
|
| 236 |
+
"epoch": 31.37,
|
| 237 |
+
"learning_rate": 0.0003123622988505747,
|
| 238 |
+
"loss": 1.1536,
|
| 239 |
+
"step": 1600
|
| 240 |
+
},
|
| 241 |
+
{
|
| 242 |
+
"epoch": 31.37,
|
| 243 |
+
"eval_loss": 1.2567418813705444,
|
| 244 |
+
"eval_runtime": 12.6734,
|
| 245 |
+
"eval_samples_per_second": 26.907,
|
| 246 |
+
"eval_steps_per_second": 3.393,
|
| 247 |
+
"eval_wer": 0.7431734317343174,
|
| 248 |
+
"step": 1600
|
| 249 |
+
},
|
| 250 |
+
{
|
| 251 |
+
"epoch": 33.33,
|
| 252 |
+
"learning_rate": 0.00030344275862068966,
|
| 253 |
+
"loss": 1.0939,
|
| 254 |
+
"step": 1700
|
| 255 |
+
},
|
| 256 |
+
{
|
| 257 |
+
"epoch": 33.33,
|
| 258 |
+
"eval_loss": 1.2800976037979126,
|
| 259 |
+
"eval_runtime": 12.5686,
|
| 260 |
+
"eval_samples_per_second": 27.131,
|
| 261 |
+
"eval_steps_per_second": 3.421,
|
| 262 |
+
"eval_wer": 0.7247232472324723,
|
| 263 |
+
"step": 1700
|
| 264 |
+
},
|
| 265 |
+
{
|
| 266 |
+
"epoch": 35.29,
|
| 267 |
+
"learning_rate": 0.0002945232183908046,
|
| 268 |
+
"loss": 1.0745,
|
| 269 |
+
"step": 1800
|
| 270 |
+
},
|
| 271 |
+
{
|
| 272 |
+
"epoch": 35.29,
|
| 273 |
+
"eval_loss": 1.2340304851531982,
|
| 274 |
+
"eval_runtime": 12.6385,
|
| 275 |
+
"eval_samples_per_second": 26.981,
|
| 276 |
+
"eval_steps_per_second": 3.402,
|
| 277 |
+
"eval_wer": 0.7151291512915129,
|
| 278 |
+
"step": 1800
|
| 279 |
+
},
|
| 280 |
+
{
|
| 281 |
+
"epoch": 37.25,
|
| 282 |
+
"learning_rate": 0.00028560367816091956,
|
| 283 |
+
"loss": 1.0454,
|
| 284 |
+
"step": 1900
|
| 285 |
+
},
|
| 286 |
+
{
|
| 287 |
+
"epoch": 37.25,
|
| 288 |
+
"eval_loss": 1.237194299697876,
|
| 289 |
+
"eval_runtime": 12.4912,
|
| 290 |
+
"eval_samples_per_second": 27.299,
|
| 291 |
+
"eval_steps_per_second": 3.442,
|
| 292 |
+
"eval_wer": 0.7151291512915129,
|
| 293 |
+
"step": 1900
|
| 294 |
+
},
|
| 295 |
+
{
|
| 296 |
+
"epoch": 39.22,
|
| 297 |
+
"learning_rate": 0.00027668413793103446,
|
| 298 |
+
"loss": 1.0101,
|
| 299 |
+
"step": 2000
|
| 300 |
+
},
|
| 301 |
+
{
|
| 302 |
+
"epoch": 39.22,
|
| 303 |
+
"eval_loss": 1.2461133003234863,
|
| 304 |
+
"eval_runtime": 12.6855,
|
| 305 |
+
"eval_samples_per_second": 26.881,
|
| 306 |
+
"eval_steps_per_second": 3.39,
|
| 307 |
+
"eval_wer": 0.7376383763837638,
|
| 308 |
+
"step": 2000
|
| 309 |
+
},
|
| 310 |
+
{
|
| 311 |
+
"epoch": 41.18,
|
| 312 |
+
"learning_rate": 0.0002677645977011494,
|
| 313 |
+
"loss": 0.9833,
|
| 314 |
+
"step": 2100
|
| 315 |
+
},
|
| 316 |
+
{
|
| 317 |
+
"epoch": 41.18,
|
| 318 |
+
"eval_loss": 1.2552708387374878,
|
| 319 |
+
"eval_runtime": 12.6273,
|
| 320 |
+
"eval_samples_per_second": 27.005,
|
| 321 |
+
"eval_steps_per_second": 3.405,
|
| 322 |
+
"eval_wer": 0.7269372693726938,
|
| 323 |
+
"step": 2100
|
| 324 |
+
},
|
| 325 |
+
{
|
| 326 |
+
"epoch": 43.14,
|
| 327 |
+
"learning_rate": 0.00025884505747126435,
|
| 328 |
+
"loss": 0.9314,
|
| 329 |
+
"step": 2200
|
| 330 |
+
},
|
| 331 |
+
{
|
| 332 |
+
"epoch": 43.14,
|
| 333 |
+
"eval_loss": 1.2371633052825928,
|
| 334 |
+
"eval_runtime": 12.5794,
|
| 335 |
+
"eval_samples_per_second": 27.108,
|
| 336 |
+
"eval_steps_per_second": 3.418,
|
| 337 |
+
"eval_wer": 0.7014760147601476,
|
| 338 |
+
"step": 2200
|
| 339 |
+
},
|
| 340 |
+
{
|
| 341 |
+
"epoch": 45.1,
|
| 342 |
+
"learning_rate": 0.0002499255172413793,
|
| 343 |
+
"loss": 0.9147,
|
| 344 |
+
"step": 2300
|
| 345 |
+
},
|
| 346 |
+
{
|
| 347 |
+
"epoch": 45.1,
|
| 348 |
+
"eval_loss": 1.3035242557525635,
|
| 349 |
+
"eval_runtime": 12.8232,
|
| 350 |
+
"eval_samples_per_second": 26.592,
|
| 351 |
+
"eval_steps_per_second": 3.353,
|
| 352 |
+
"eval_wer": 0.7357933579335794,
|
| 353 |
+
"step": 2300
|
| 354 |
+
},
|
| 355 |
+
{
|
| 356 |
+
"epoch": 47.06,
|
| 357 |
+
"learning_rate": 0.00024109517241379313,
|
| 358 |
+
"loss": 0.8758,
|
| 359 |
+
"step": 2400
|
| 360 |
+
},
|
| 361 |
+
{
|
| 362 |
+
"epoch": 47.06,
|
| 363 |
+
"eval_loss": 1.2598013877868652,
|
| 364 |
+
"eval_runtime": 12.7812,
|
| 365 |
+
"eval_samples_per_second": 26.68,
|
| 366 |
+
"eval_steps_per_second": 3.364,
|
| 367 |
+
"eval_wer": 0.7092250922509226,
|
| 368 |
+
"step": 2400
|
| 369 |
+
},
|
| 370 |
+
{
|
| 371 |
+
"epoch": 49.02,
|
| 372 |
+
"learning_rate": 0.00023217563218390802,
|
| 373 |
+
"loss": 0.8356,
|
| 374 |
+
"step": 2500
|
| 375 |
+
},
|
| 376 |
+
{
|
| 377 |
+
"epoch": 49.02,
|
| 378 |
+
"eval_loss": 1.255703091621399,
|
| 379 |
+
"eval_runtime": 12.549,
|
| 380 |
+
"eval_samples_per_second": 27.173,
|
| 381 |
+
"eval_steps_per_second": 3.427,
|
| 382 |
+
"eval_wer": 0.7143911439114391,
|
| 383 |
+
"step": 2500
|
| 384 |
+
},
|
| 385 |
+
{
|
| 386 |
+
"epoch": 50.98,
|
| 387 |
+
"learning_rate": 0.00022325609195402297,
|
| 388 |
+
"loss": 0.8105,
|
| 389 |
+
"step": 2600
|
| 390 |
+
},
|
| 391 |
+
{
|
| 392 |
+
"epoch": 50.98,
|
| 393 |
+
"eval_loss": 1.2618709802627563,
|
| 394 |
+
"eval_runtime": 12.4348,
|
| 395 |
+
"eval_samples_per_second": 27.423,
|
| 396 |
+
"eval_steps_per_second": 3.458,
|
| 397 |
+
"eval_wer": 0.7236162361623616,
|
| 398 |
+
"step": 2600
|
| 399 |
+
},
|
| 400 |
+
{
|
| 401 |
+
"epoch": 52.94,
|
| 402 |
+
"learning_rate": 0.00021433655172413795,
|
| 403 |
+
"loss": 0.7947,
|
| 404 |
+
"step": 2700
|
| 405 |
+
},
|
| 406 |
+
{
|
| 407 |
+
"epoch": 52.94,
|
| 408 |
+
"eval_loss": 1.399444818496704,
|
| 409 |
+
"eval_runtime": 12.9361,
|
| 410 |
+
"eval_samples_per_second": 26.36,
|
| 411 |
+
"eval_steps_per_second": 3.324,
|
| 412 |
+
"eval_wer": 0.7490774907749077,
|
| 413 |
+
"step": 2700
|
| 414 |
+
},
|
| 415 |
+
{
|
| 416 |
+
"epoch": 54.9,
|
| 417 |
+
"learning_rate": 0.0002054170114942529,
|
| 418 |
+
"loss": 0.7623,
|
| 419 |
+
"step": 2800
|
| 420 |
+
},
|
| 421 |
+
{
|
| 422 |
+
"epoch": 54.9,
|
| 423 |
+
"eval_loss": 1.2931541204452515,
|
| 424 |
+
"eval_runtime": 12.7092,
|
| 425 |
+
"eval_samples_per_second": 26.831,
|
| 426 |
+
"eval_steps_per_second": 3.383,
|
| 427 |
+
"eval_wer": 0.7132841328413284,
|
| 428 |
+
"step": 2800
|
| 429 |
+
},
|
| 430 |
+
{
|
| 431 |
+
"epoch": 56.86,
|
| 432 |
+
"learning_rate": 0.0001964974712643678,
|
| 433 |
+
"loss": 0.7282,
|
| 434 |
+
"step": 2900
|
| 435 |
+
},
|
| 436 |
+
{
|
| 437 |
+
"epoch": 56.86,
|
| 438 |
+
"eval_loss": 1.2799276113510132,
|
| 439 |
+
"eval_runtime": 12.7788,
|
| 440 |
+
"eval_samples_per_second": 26.685,
|
| 441 |
+
"eval_steps_per_second": 3.365,
|
| 442 |
+
"eval_wer": 0.7088560885608856,
|
| 443 |
+
"step": 2900
|
| 444 |
+
},
|
| 445 |
+
{
|
| 446 |
+
"epoch": 58.82,
|
| 447 |
+
"learning_rate": 0.00018757793103448274,
|
| 448 |
+
"loss": 0.7108,
|
| 449 |
+
"step": 3000
|
| 450 |
+
},
|
| 451 |
+
{
|
| 452 |
+
"epoch": 58.82,
|
| 453 |
+
"eval_loss": 1.3615078926086426,
|
| 454 |
+
"eval_runtime": 12.6509,
|
| 455 |
+
"eval_samples_per_second": 26.955,
|
| 456 |
+
"eval_steps_per_second": 3.399,
|
| 457 |
+
"eval_wer": 0.714760147601476,
|
| 458 |
+
"step": 3000
|
| 459 |
+
},
|
| 460 |
+
{
|
| 461 |
+
"epoch": 60.78,
|
| 462 |
+
"learning_rate": 0.00017865839080459772,
|
| 463 |
+
"loss": 0.6896,
|
| 464 |
+
"step": 3100
|
| 465 |
+
},
|
| 466 |
+
{
|
| 467 |
+
"epoch": 60.78,
|
| 468 |
+
"eval_loss": 1.312876582145691,
|
| 469 |
+
"eval_runtime": 12.6527,
|
| 470 |
+
"eval_samples_per_second": 26.951,
|
| 471 |
+
"eval_steps_per_second": 3.398,
|
| 472 |
+
"eval_wer": 0.7040590405904059,
|
| 473 |
+
"step": 3100
|
| 474 |
+
},
|
| 475 |
+
{
|
| 476 |
+
"epoch": 62.75,
|
| 477 |
+
"learning_rate": 0.00016973885057471264,
|
| 478 |
+
"loss": 0.6496,
|
| 479 |
+
"step": 3200
|
| 480 |
+
},
|
| 481 |
+
{
|
| 482 |
+
"epoch": 62.75,
|
| 483 |
+
"eval_loss": 1.4050244092941284,
|
| 484 |
+
"eval_runtime": 12.6982,
|
| 485 |
+
"eval_samples_per_second": 26.854,
|
| 486 |
+
"eval_steps_per_second": 3.386,
|
| 487 |
+
"eval_wer": 0.6933579335793358,
|
| 488 |
+
"step": 3200
|
| 489 |
+
},
|
| 490 |
+
{
|
| 491 |
+
"epoch": 64.71,
|
| 492 |
+
"learning_rate": 0.0001608193103448276,
|
| 493 |
+
"loss": 0.6075,
|
| 494 |
+
"step": 3300
|
| 495 |
+
},
|
| 496 |
+
{
|
| 497 |
+
"epoch": 64.71,
|
| 498 |
+
"eval_loss": 1.35708749294281,
|
| 499 |
+
"eval_runtime": 12.752,
|
| 500 |
+
"eval_samples_per_second": 26.741,
|
| 501 |
+
"eval_steps_per_second": 3.372,
|
| 502 |
+
"eval_wer": 0.7025830258302583,
|
| 503 |
+
"step": 3300
|
| 504 |
+
},
|
| 505 |
+
{
|
| 506 |
+
"epoch": 66.67,
|
| 507 |
+
"learning_rate": 0.00015189977011494254,
|
| 508 |
+
"loss": 0.6242,
|
| 509 |
+
"step": 3400
|
| 510 |
+
},
|
| 511 |
+
{
|
| 512 |
+
"epoch": 66.67,
|
| 513 |
+
"eval_loss": 1.3368754386901855,
|
| 514 |
+
"eval_runtime": 12.8446,
|
| 515 |
+
"eval_samples_per_second": 26.548,
|
| 516 |
+
"eval_steps_per_second": 3.348,
|
| 517 |
+
"eval_wer": 0.7062730627306273,
|
| 518 |
+
"step": 3400
|
| 519 |
+
},
|
| 520 |
+
{
|
| 521 |
+
"epoch": 68.63,
|
| 522 |
+
"learning_rate": 0.00014298022988505749,
|
| 523 |
+
"loss": 0.5865,
|
| 524 |
+
"step": 3500
|
| 525 |
+
},
|
| 526 |
+
{
|
| 527 |
+
"epoch": 68.63,
|
| 528 |
+
"eval_loss": 1.4367624521255493,
|
| 529 |
+
"eval_runtime": 12.5885,
|
| 530 |
+
"eval_samples_per_second": 27.088,
|
| 531 |
+
"eval_steps_per_second": 3.416,
|
| 532 |
+
"eval_wer": 0.7140221402214022,
|
| 533 |
+
"step": 3500
|
| 534 |
+
},
|
| 535 |
+
{
|
| 536 |
+
"epoch": 70.59,
|
| 537 |
+
"learning_rate": 0.0001340606896551724,
|
| 538 |
+
"loss": 0.5721,
|
| 539 |
+
"step": 3600
|
| 540 |
+
},
|
| 541 |
+
{
|
| 542 |
+
"epoch": 70.59,
|
| 543 |
+
"eval_loss": 1.4223829507827759,
|
| 544 |
+
"eval_runtime": 12.6692,
|
| 545 |
+
"eval_samples_per_second": 26.916,
|
| 546 |
+
"eval_steps_per_second": 3.394,
|
| 547 |
+
"eval_wer": 0.7066420664206642,
|
| 548 |
+
"step": 3600
|
| 549 |
+
},
|
| 550 |
+
{
|
| 551 |
+
"epoch": 72.55,
|
| 552 |
+
"learning_rate": 0.00012514114942528736,
|
| 553 |
+
"loss": 0.5475,
|
| 554 |
+
"step": 3700
|
| 555 |
+
},
|
| 556 |
+
{
|
| 557 |
+
"epoch": 72.55,
|
| 558 |
+
"eval_loss": 1.4797747135162354,
|
| 559 |
+
"eval_runtime": 12.8068,
|
| 560 |
+
"eval_samples_per_second": 26.626,
|
| 561 |
+
"eval_steps_per_second": 3.358,
|
| 562 |
+
"eval_wer": 0.7118081180811808,
|
| 563 |
+
"step": 3700
|
| 564 |
+
},
|
| 565 |
+
{
|
| 566 |
+
"epoch": 74.51,
|
| 567 |
+
"learning_rate": 0.00011622160919540229,
|
| 568 |
+
"loss": 0.5086,
|
| 569 |
+
"step": 3800
|
| 570 |
+
},
|
| 571 |
+
{
|
| 572 |
+
"epoch": 74.51,
|
| 573 |
+
"eval_loss": 1.510679841041565,
|
| 574 |
+
"eval_runtime": 12.607,
|
| 575 |
+
"eval_samples_per_second": 27.048,
|
| 576 |
+
"eval_steps_per_second": 3.411,
|
| 577 |
+
"eval_wer": 0.7232472324723247,
|
| 578 |
+
"step": 3800
|
| 579 |
+
},
|
| 580 |
+
{
|
| 581 |
+
"epoch": 76.47,
|
| 582 |
+
"learning_rate": 0.00010730206896551725,
|
| 583 |
+
"loss": 0.4958,
|
| 584 |
+
"step": 3900
|
| 585 |
+
},
|
| 586 |
+
{
|
| 587 |
+
"epoch": 76.47,
|
| 588 |
+
"eval_loss": 1.4849300384521484,
|
| 589 |
+
"eval_runtime": 12.6845,
|
| 590 |
+
"eval_samples_per_second": 26.883,
|
| 591 |
+
"eval_steps_per_second": 3.39,
|
| 592 |
+
"eval_wer": 0.7088560885608856,
|
| 593 |
+
"step": 3900
|
| 594 |
+
},
|
| 595 |
+
{
|
| 596 |
+
"epoch": 78.43,
|
| 597 |
+
"learning_rate": 9.838252873563218e-05,
|
| 598 |
+
"loss": 0.5046,
|
| 599 |
+
"step": 4000
|
| 600 |
+
},
|
| 601 |
+
{
|
| 602 |
+
"epoch": 78.43,
|
| 603 |
+
"eval_loss": 1.4450523853302002,
|
| 604 |
+
"eval_runtime": 12.6526,
|
| 605 |
+
"eval_samples_per_second": 26.951,
|
| 606 |
+
"eval_steps_per_second": 3.399,
|
| 607 |
+
"eval_wer": 0.7114391143911439,
|
| 608 |
+
"step": 4000
|
| 609 |
+
},
|
| 610 |
+
{
|
| 611 |
+
"epoch": 80.39,
|
| 612 |
+
"learning_rate": 8.946298850574712e-05,
|
| 613 |
+
"loss": 0.4694,
|
| 614 |
+
"step": 4100
|
| 615 |
+
},
|
| 616 |
+
{
|
| 617 |
+
"epoch": 80.39,
|
| 618 |
+
"eval_loss": 1.4674367904663086,
|
| 619 |
+
"eval_runtime": 12.49,
|
| 620 |
+
"eval_samples_per_second": 27.302,
|
| 621 |
+
"eval_steps_per_second": 3.443,
|
| 622 |
+
"eval_wer": 0.7088560885608856,
|
| 623 |
+
"step": 4100
|
| 624 |
+
},
|
| 625 |
+
{
|
| 626 |
+
"epoch": 82.35,
|
| 627 |
+
"learning_rate": 8.054344827586206e-05,
|
| 628 |
+
"loss": 0.4386,
|
| 629 |
+
"step": 4200
|
| 630 |
+
},
|
| 631 |
+
{
|
| 632 |
+
"epoch": 82.35,
|
| 633 |
+
"eval_loss": 1.524474859237671,
|
| 634 |
+
"eval_runtime": 12.6393,
|
| 635 |
+
"eval_samples_per_second": 26.979,
|
| 636 |
+
"eval_steps_per_second": 3.402,
|
| 637 |
+
"eval_wer": 0.7103321033210332,
|
| 638 |
+
"step": 4200
|
| 639 |
+
},
|
| 640 |
+
{
|
| 641 |
+
"epoch": 84.31,
|
| 642 |
+
"learning_rate": 7.162390804597701e-05,
|
| 643 |
+
"loss": 0.4516,
|
| 644 |
+
"step": 4300
|
| 645 |
+
},
|
| 646 |
+
{
|
| 647 |
+
"epoch": 84.31,
|
| 648 |
+
"eval_loss": 1.5031787157058716,
|
| 649 |
+
"eval_runtime": 12.6415,
|
| 650 |
+
"eval_samples_per_second": 26.975,
|
| 651 |
+
"eval_steps_per_second": 3.401,
|
| 652 |
+
"eval_wer": 0.7103321033210332,
|
| 653 |
+
"step": 4300
|
| 654 |
+
},
|
| 655 |
+
{
|
| 656 |
+
"epoch": 86.27,
|
| 657 |
+
"learning_rate": 6.27935632183908e-05,
|
| 658 |
+
"loss": 0.4113,
|
| 659 |
+
"step": 4400
|
| 660 |
+
},
|
| 661 |
+
{
|
| 662 |
+
"epoch": 86.27,
|
| 663 |
+
"eval_loss": 1.5246329307556152,
|
| 664 |
+
"eval_runtime": 12.6508,
|
| 665 |
+
"eval_samples_per_second": 26.955,
|
| 666 |
+
"eval_steps_per_second": 3.399,
|
| 667 |
+
"eval_wer": 0.7195571955719557,
|
| 668 |
+
"step": 4400
|
| 669 |
+
},
|
| 670 |
+
{
|
| 671 |
+
"epoch": 88.24,
|
| 672 |
+
"learning_rate": 5.3874022988505745e-05,
|
| 673 |
+
"loss": 0.3972,
|
| 674 |
+
"step": 4500
|
| 675 |
+
},
|
| 676 |
+
{
|
| 677 |
+
"epoch": 88.24,
|
| 678 |
+
"eval_loss": 1.5318336486816406,
|
| 679 |
+
"eval_runtime": 12.5704,
|
| 680 |
+
"eval_samples_per_second": 27.127,
|
| 681 |
+
"eval_steps_per_second": 3.421,
|
| 682 |
+
"eval_wer": 0.7114391143911439,
|
| 683 |
+
"step": 4500
|
| 684 |
+
},
|
| 685 |
+
{
|
| 686 |
+
"epoch": 90.2,
|
| 687 |
+
"learning_rate": 4.4954482758620694e-05,
|
| 688 |
+
"loss": 0.4006,
|
| 689 |
+
"step": 4600
|
| 690 |
+
},
|
| 691 |
+
{
|
| 692 |
+
"epoch": 90.2,
|
| 693 |
+
"eval_loss": 1.554287314414978,
|
| 694 |
+
"eval_runtime": 12.5754,
|
| 695 |
+
"eval_samples_per_second": 27.116,
|
| 696 |
+
"eval_steps_per_second": 3.419,
|
| 697 |
+
"eval_wer": 0.6981549815498155,
|
| 698 |
+
"step": 4600
|
| 699 |
+
},
|
| 700 |
+
{
|
| 701 |
+
"epoch": 92.16,
|
| 702 |
+
"learning_rate": 3.603494252873563e-05,
|
| 703 |
+
"loss": 0.4014,
|
| 704 |
+
"step": 4700
|
| 705 |
+
},
|
| 706 |
+
{
|
| 707 |
+
"epoch": 92.16,
|
| 708 |
+
"eval_loss": 1.5442076921463013,
|
| 709 |
+
"eval_runtime": 12.6146,
|
| 710 |
+
"eval_samples_per_second": 27.032,
|
| 711 |
+
"eval_steps_per_second": 3.409,
|
| 712 |
+
"eval_wer": 0.7047970479704797,
|
| 713 |
+
"step": 4700
|
| 714 |
+
},
|
| 715 |
+
{
|
| 716 |
+
"epoch": 94.12,
|
| 717 |
+
"learning_rate": 2.711540229885057e-05,
|
| 718 |
+
"loss": 0.3672,
|
| 719 |
+
"step": 4800
|
| 720 |
+
},
|
| 721 |
+
{
|
| 722 |
+
"epoch": 94.12,
|
| 723 |
+
"eval_loss": 1.5541517734527588,
|
| 724 |
+
"eval_runtime": 12.6543,
|
| 725 |
+
"eval_samples_per_second": 26.947,
|
| 726 |
+
"eval_steps_per_second": 3.398,
|
| 727 |
+
"eval_wer": 0.7136531365313653,
|
| 728 |
+
"step": 4800
|
| 729 |
+
},
|
| 730 |
+
{
|
| 731 |
+
"epoch": 96.08,
|
| 732 |
+
"learning_rate": 1.8195862068965517e-05,
|
| 733 |
+
"loss": 0.3666,
|
| 734 |
+
"step": 4900
|
| 735 |
+
},
|
| 736 |
+
{
|
| 737 |
+
"epoch": 96.08,
|
| 738 |
+
"eval_loss": 1.5414179563522339,
|
| 739 |
+
"eval_runtime": 12.5239,
|
| 740 |
+
"eval_samples_per_second": 27.228,
|
| 741 |
+
"eval_steps_per_second": 3.433,
|
| 742 |
+
"eval_wer": 0.7018450184501845,
|
| 743 |
+
"step": 4900
|
| 744 |
+
},
|
| 745 |
+
{
|
| 746 |
+
"epoch": 98.04,
|
| 747 |
+
"learning_rate": 9.27632183908046e-06,
|
| 748 |
+
"loss": 0.3574,
|
| 749 |
+
"step": 5000
|
| 750 |
+
},
|
| 751 |
+
{
|
| 752 |
+
"epoch": 98.04,
|
| 753 |
+
"eval_loss": 1.5465455055236816,
|
| 754 |
+
"eval_runtime": 12.7119,
|
| 755 |
+
"eval_samples_per_second": 26.825,
|
| 756 |
+
"eval_steps_per_second": 3.383,
|
| 757 |
+
"eval_wer": 0.7059040590405904,
|
| 758 |
+
"step": 5000
|
| 759 |
+
},
|
| 760 |
+
{
|
| 761 |
+
"epoch": 100.0,
|
| 762 |
+
"learning_rate": 3.567816091954023e-07,
|
| 763 |
+
"loss": 0.3428,
|
| 764 |
+
"step": 5100
|
| 765 |
+
},
|
| 766 |
+
{
|
| 767 |
+
"epoch": 100.0,
|
| 768 |
+
"eval_loss": 1.5443140268325806,
|
| 769 |
+
"eval_runtime": 12.8582,
|
| 770 |
+
"eval_samples_per_second": 26.52,
|
| 771 |
+
"eval_steps_per_second": 3.344,
|
| 772 |
+
"eval_wer": 0.7029520295202952,
|
| 773 |
+
"step": 5100
|
| 774 |
+
},
|
| 775 |
+
{
|
| 776 |
+
"epoch": 100.0,
|
| 777 |
+
"step": 5100,
|
| 778 |
+
"total_flos": 9.838577578075728e+18,
|
| 779 |
+
"train_loss": 1.1548647121354645,
|
| 780 |
+
"train_runtime": 5602.3197,
|
| 781 |
+
"train_samples_per_second": 14.458,
|
| 782 |
+
"train_steps_per_second": 0.91
|
| 783 |
+
}
|
| 784 |
+
],
|
| 785 |
+
"max_steps": 5100,
|
| 786 |
+
"num_train_epochs": 100,
|
| 787 |
+
"total_flos": 9.838577578075728e+18,
|
| 788 |
+
"trial_name": null,
|
| 789 |
+
"trial_params": null
|
| 790 |
+
}
|