End of training
Browse files- all_results.json +14 -0
- eval_results.json +9 -0
- pytorch_model.bin +1 -1
- train_results.json +8 -0
- trainer_state.json +751 -0
all_results.json
ADDED
|
@@ -0,0 +1,14 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"epoch": 50.0,
|
| 3 |
+
"eval_loss": 0.24299392104148865,
|
| 4 |
+
"eval_runtime": 142.3605,
|
| 5 |
+
"eval_samples": 2744,
|
| 6 |
+
"eval_samples_per_second": 19.275,
|
| 7 |
+
"eval_steps_per_second": 2.409,
|
| 8 |
+
"eval_wer": 0.3803796021038189,
|
| 9 |
+
"train_loss": 1.5738848333155855,
|
| 10 |
+
"train_runtime": 24806.553,
|
| 11 |
+
"train_samples": 6034,
|
| 12 |
+
"train_samples_per_second": 12.162,
|
| 13 |
+
"train_steps_per_second": 0.379
|
| 14 |
+
}
|
eval_results.json
ADDED
|
@@ -0,0 +1,9 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"epoch": 50.0,
|
| 3 |
+
"eval_loss": 0.24299392104148865,
|
| 4 |
+
"eval_runtime": 142.3605,
|
| 5 |
+
"eval_samples": 2744,
|
| 6 |
+
"eval_samples_per_second": 19.275,
|
| 7 |
+
"eval_steps_per_second": 2.409,
|
| 8 |
+
"eval_wer": 0.3803796021038189
|
| 9 |
+
}
|
pytorch_model.bin
CHANGED
|
@@ -1,3 +1,3 @@
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
-
oid sha256:
|
| 3 |
size 1262112241
|
|
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:a07b4dd396459cd0f012358a86158ae1d7b371ce3ae2cbc67f4c43b9694dbd2c
|
| 3 |
size 1262112241
|
train_results.json
ADDED
|
@@ -0,0 +1,8 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"epoch": 50.0,
|
| 3 |
+
"train_loss": 1.5738848333155855,
|
| 4 |
+
"train_runtime": 24806.553,
|
| 5 |
+
"train_samples": 6034,
|
| 6 |
+
"train_samples_per_second": 12.162,
|
| 7 |
+
"train_steps_per_second": 0.379
|
| 8 |
+
}
|
trainer_state.json
ADDED
|
@@ -0,0 +1,751 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"best_metric": null,
|
| 3 |
+
"best_model_checkpoint": null,
|
| 4 |
+
"epoch": 49.99602649006623,
|
| 5 |
+
"global_step": 9400,
|
| 6 |
+
"is_hyper_param_search": false,
|
| 7 |
+
"is_local_process_zero": true,
|
| 8 |
+
"is_world_process_zero": true,
|
| 9 |
+
"log_history": [
|
| 10 |
+
{
|
| 11 |
+
"epoch": 0.53,
|
| 12 |
+
"learning_rate": 3.7125e-06,
|
| 13 |
+
"loss": 15.287,
|
| 14 |
+
"step": 100
|
| 15 |
+
},
|
| 16 |
+
{
|
| 17 |
+
"epoch": 1.06,
|
| 18 |
+
"learning_rate": 7.4625e-06,
|
| 19 |
+
"loss": 7.6558,
|
| 20 |
+
"step": 200
|
| 21 |
+
},
|
| 22 |
+
{
|
| 23 |
+
"epoch": 1.59,
|
| 24 |
+
"learning_rate": 1.1212499999999998e-05,
|
| 25 |
+
"loss": 4.9409,
|
| 26 |
+
"step": 300
|
| 27 |
+
},
|
| 28 |
+
{
|
| 29 |
+
"epoch": 2.13,
|
| 30 |
+
"learning_rate": 1.49625e-05,
|
| 31 |
+
"loss": 4.2283,
|
| 32 |
+
"step": 400
|
| 33 |
+
},
|
| 34 |
+
{
|
| 35 |
+
"epoch": 2.66,
|
| 36 |
+
"learning_rate": 1.8712499999999997e-05,
|
| 37 |
+
"loss": 3.6871,
|
| 38 |
+
"step": 500
|
| 39 |
+
},
|
| 40 |
+
{
|
| 41 |
+
"epoch": 2.66,
|
| 42 |
+
"eval_loss": 3.5374293327331543,
|
| 43 |
+
"eval_runtime": 140.96,
|
| 44 |
+
"eval_samples_per_second": 19.467,
|
| 45 |
+
"eval_steps_per_second": 2.433,
|
| 46 |
+
"eval_wer": 1.0,
|
| 47 |
+
"step": 500
|
| 48 |
+
},
|
| 49 |
+
{
|
| 50 |
+
"epoch": 3.19,
|
| 51 |
+
"learning_rate": 2.2462499999999997e-05,
|
| 52 |
+
"loss": 3.4073,
|
| 53 |
+
"step": 600
|
| 54 |
+
},
|
| 55 |
+
{
|
| 56 |
+
"epoch": 3.72,
|
| 57 |
+
"learning_rate": 2.6212499999999997e-05,
|
| 58 |
+
"loss": 3.2613,
|
| 59 |
+
"step": 700
|
| 60 |
+
},
|
| 61 |
+
{
|
| 62 |
+
"epoch": 4.25,
|
| 63 |
+
"learning_rate": 2.99625e-05,
|
| 64 |
+
"loss": 3.2504,
|
| 65 |
+
"step": 800
|
| 66 |
+
},
|
| 67 |
+
{
|
| 68 |
+
"epoch": 4.78,
|
| 69 |
+
"learning_rate": 3.37125e-05,
|
| 70 |
+
"loss": 3.1732,
|
| 71 |
+
"step": 900
|
| 72 |
+
},
|
| 73 |
+
{
|
| 74 |
+
"epoch": 5.32,
|
| 75 |
+
"learning_rate": 3.7462499999999996e-05,
|
| 76 |
+
"loss": 3.1501,
|
| 77 |
+
"step": 1000
|
| 78 |
+
},
|
| 79 |
+
{
|
| 80 |
+
"epoch": 5.32,
|
| 81 |
+
"eval_loss": 3.127795934677124,
|
| 82 |
+
"eval_runtime": 140.7753,
|
| 83 |
+
"eval_samples_per_second": 19.492,
|
| 84 |
+
"eval_steps_per_second": 2.437,
|
| 85 |
+
"eval_wer": 1.0,
|
| 86 |
+
"step": 1000
|
| 87 |
+
},
|
| 88 |
+
{
|
| 89 |
+
"epoch": 5.85,
|
| 90 |
+
"learning_rate": 4.12125e-05,
|
| 91 |
+
"loss": 3.0666,
|
| 92 |
+
"step": 1100
|
| 93 |
+
},
|
| 94 |
+
{
|
| 95 |
+
"epoch": 6.38,
|
| 96 |
+
"learning_rate": 4.4962499999999995e-05,
|
| 97 |
+
"loss": 2.7047,
|
| 98 |
+
"step": 1200
|
| 99 |
+
},
|
| 100 |
+
{
|
| 101 |
+
"epoch": 6.91,
|
| 102 |
+
"learning_rate": 4.871249999999999e-05,
|
| 103 |
+
"loss": 2.0895,
|
| 104 |
+
"step": 1300
|
| 105 |
+
},
|
| 106 |
+
{
|
| 107 |
+
"epoch": 7.45,
|
| 108 |
+
"learning_rate": 5.2462499999999994e-05,
|
| 109 |
+
"loss": 1.7459,
|
| 110 |
+
"step": 1400
|
| 111 |
+
},
|
| 112 |
+
{
|
| 113 |
+
"epoch": 7.97,
|
| 114 |
+
"learning_rate": 5.62125e-05,
|
| 115 |
+
"loss": 1.5843,
|
| 116 |
+
"step": 1500
|
| 117 |
+
},
|
| 118 |
+
{
|
| 119 |
+
"epoch": 7.97,
|
| 120 |
+
"eval_loss": 0.6358404755592346,
|
| 121 |
+
"eval_runtime": 138.8212,
|
| 122 |
+
"eval_samples_per_second": 19.766,
|
| 123 |
+
"eval_steps_per_second": 2.471,
|
| 124 |
+
"eval_wer": 0.6914246512691516,
|
| 125 |
+
"step": 1500
|
| 126 |
+
},
|
| 127 |
+
{
|
| 128 |
+
"epoch": 8.51,
|
| 129 |
+
"learning_rate": 5.9962499999999994e-05,
|
| 130 |
+
"loss": 1.5166,
|
| 131 |
+
"step": 1600
|
| 132 |
+
},
|
| 133 |
+
{
|
| 134 |
+
"epoch": 9.04,
|
| 135 |
+
"learning_rate": 6.37125e-05,
|
| 136 |
+
"loss": 1.4555,
|
| 137 |
+
"step": 1700
|
| 138 |
+
},
|
| 139 |
+
{
|
| 140 |
+
"epoch": 9.57,
|
| 141 |
+
"learning_rate": 6.746249999999999e-05,
|
| 142 |
+
"loss": 1.4013,
|
| 143 |
+
"step": 1800
|
| 144 |
+
},
|
| 145 |
+
{
|
| 146 |
+
"epoch": 10.11,
|
| 147 |
+
"learning_rate": 7.121249999999999e-05,
|
| 148 |
+
"loss": 1.3841,
|
| 149 |
+
"step": 1900
|
| 150 |
+
},
|
| 151 |
+
{
|
| 152 |
+
"epoch": 10.64,
|
| 153 |
+
"learning_rate": 7.49625e-05,
|
| 154 |
+
"loss": 1.3378,
|
| 155 |
+
"step": 2000
|
| 156 |
+
},
|
| 157 |
+
{
|
| 158 |
+
"epoch": 10.64,
|
| 159 |
+
"eval_loss": 0.442169189453125,
|
| 160 |
+
"eval_runtime": 140.0078,
|
| 161 |
+
"eval_samples_per_second": 19.599,
|
| 162 |
+
"eval_steps_per_second": 2.45,
|
| 163 |
+
"eval_wer": 0.5924536931168534,
|
| 164 |
+
"step": 2000
|
| 165 |
+
},
|
| 166 |
+
{
|
| 167 |
+
"epoch": 11.17,
|
| 168 |
+
"learning_rate": 7.400675675675676e-05,
|
| 169 |
+
"loss": 1.3249,
|
| 170 |
+
"step": 2100
|
| 171 |
+
},
|
| 172 |
+
{
|
| 173 |
+
"epoch": 11.7,
|
| 174 |
+
"learning_rate": 7.299324324324324e-05,
|
| 175 |
+
"loss": 1.2996,
|
| 176 |
+
"step": 2200
|
| 177 |
+
},
|
| 178 |
+
{
|
| 179 |
+
"epoch": 12.23,
|
| 180 |
+
"learning_rate": 7.197972972972972e-05,
|
| 181 |
+
"loss": 1.2952,
|
| 182 |
+
"step": 2300
|
| 183 |
+
},
|
| 184 |
+
{
|
| 185 |
+
"epoch": 12.76,
|
| 186 |
+
"learning_rate": 7.096621621621621e-05,
|
| 187 |
+
"loss": 1.2626,
|
| 188 |
+
"step": 2400
|
| 189 |
+
},
|
| 190 |
+
{
|
| 191 |
+
"epoch": 13.3,
|
| 192 |
+
"learning_rate": 6.99527027027027e-05,
|
| 193 |
+
"loss": 1.2595,
|
| 194 |
+
"step": 2500
|
| 195 |
+
},
|
| 196 |
+
{
|
| 197 |
+
"epoch": 13.3,
|
| 198 |
+
"eval_loss": 0.3921487033367157,
|
| 199 |
+
"eval_runtime": 140.1536,
|
| 200 |
+
"eval_samples_per_second": 19.579,
|
| 201 |
+
"eval_steps_per_second": 2.447,
|
| 202 |
+
"eval_wer": 0.5511548136290876,
|
| 203 |
+
"step": 2500
|
| 204 |
+
},
|
| 205 |
+
{
|
| 206 |
+
"epoch": 13.83,
|
| 207 |
+
"learning_rate": 6.893918918918919e-05,
|
| 208 |
+
"loss": 1.2282,
|
| 209 |
+
"step": 2600
|
| 210 |
+
},
|
| 211 |
+
{
|
| 212 |
+
"epoch": 14.36,
|
| 213 |
+
"learning_rate": 6.792567567567567e-05,
|
| 214 |
+
"loss": 1.2352,
|
| 215 |
+
"step": 2700
|
| 216 |
+
},
|
| 217 |
+
{
|
| 218 |
+
"epoch": 14.89,
|
| 219 |
+
"learning_rate": 6.691216216216216e-05,
|
| 220 |
+
"loss": 1.2129,
|
| 221 |
+
"step": 2800
|
| 222 |
+
},
|
| 223 |
+
{
|
| 224 |
+
"epoch": 15.42,
|
| 225 |
+
"learning_rate": 6.589864864864864e-05,
|
| 226 |
+
"loss": 1.2184,
|
| 227 |
+
"step": 2900
|
| 228 |
+
},
|
| 229 |
+
{
|
| 230 |
+
"epoch": 15.95,
|
| 231 |
+
"learning_rate": 6.488513513513514e-05,
|
| 232 |
+
"loss": 1.1643,
|
| 233 |
+
"step": 3000
|
| 234 |
+
},
|
| 235 |
+
{
|
| 236 |
+
"epoch": 15.95,
|
| 237 |
+
"eval_loss": 0.35074228048324585,
|
| 238 |
+
"eval_runtime": 142.5452,
|
| 239 |
+
"eval_samples_per_second": 19.25,
|
| 240 |
+
"eval_steps_per_second": 2.406,
|
| 241 |
+
"eval_wer": 0.514932540589984,
|
| 242 |
+
"step": 3000
|
| 243 |
+
},
|
| 244 |
+
{
|
| 245 |
+
"epoch": 16.49,
|
| 246 |
+
"learning_rate": 6.388175675675675e-05,
|
| 247 |
+
"loss": 1.1704,
|
| 248 |
+
"step": 3100
|
| 249 |
+
},
|
| 250 |
+
{
|
| 251 |
+
"epoch": 17.02,
|
| 252 |
+
"learning_rate": 6.287837837837837e-05,
|
| 253 |
+
"loss": 1.1774,
|
| 254 |
+
"step": 3200
|
| 255 |
+
},
|
| 256 |
+
{
|
| 257 |
+
"epoch": 17.55,
|
| 258 |
+
"learning_rate": 6.186486486486485e-05,
|
| 259 |
+
"loss": 1.157,
|
| 260 |
+
"step": 3300
|
| 261 |
+
},
|
| 262 |
+
{
|
| 263 |
+
"epoch": 18.08,
|
| 264 |
+
"learning_rate": 6.085135135135135e-05,
|
| 265 |
+
"loss": 1.1605,
|
| 266 |
+
"step": 3400
|
| 267 |
+
},
|
| 268 |
+
{
|
| 269 |
+
"epoch": 18.61,
|
| 270 |
+
"learning_rate": 5.983783783783783e-05,
|
| 271 |
+
"loss": 1.1352,
|
| 272 |
+
"step": 3500
|
| 273 |
+
},
|
| 274 |
+
{
|
| 275 |
+
"epoch": 18.61,
|
| 276 |
+
"eval_loss": 0.3350585401058197,
|
| 277 |
+
"eval_runtime": 140.4732,
|
| 278 |
+
"eval_samples_per_second": 19.534,
|
| 279 |
+
"eval_steps_per_second": 2.442,
|
| 280 |
+
"eval_wer": 0.5018980105190944,
|
| 281 |
+
"step": 3500
|
| 282 |
+
},
|
| 283 |
+
{
|
| 284 |
+
"epoch": 19.15,
|
| 285 |
+
"learning_rate": 5.882432432432432e-05,
|
| 286 |
+
"loss": 1.1474,
|
| 287 |
+
"step": 3600
|
| 288 |
+
},
|
| 289 |
+
{
|
| 290 |
+
"epoch": 19.68,
|
| 291 |
+
"learning_rate": 5.781081081081081e-05,
|
| 292 |
+
"loss": 1.1273,
|
| 293 |
+
"step": 3700
|
| 294 |
+
},
|
| 295 |
+
{
|
| 296 |
+
"epoch": 20.21,
|
| 297 |
+
"learning_rate": 5.679729729729729e-05,
|
| 298 |
+
"loss": 1.133,
|
| 299 |
+
"step": 3800
|
| 300 |
+
},
|
| 301 |
+
{
|
| 302 |
+
"epoch": 20.74,
|
| 303 |
+
"learning_rate": 5.578378378378378e-05,
|
| 304 |
+
"loss": 1.1112,
|
| 305 |
+
"step": 3900
|
| 306 |
+
},
|
| 307 |
+
{
|
| 308 |
+
"epoch": 21.28,
|
| 309 |
+
"learning_rate": 5.477027027027026e-05,
|
| 310 |
+
"loss": 1.1113,
|
| 311 |
+
"step": 4000
|
| 312 |
+
},
|
| 313 |
+
{
|
| 314 |
+
"epoch": 21.28,
|
| 315 |
+
"eval_loss": 0.3152759373188019,
|
| 316 |
+
"eval_runtime": 139.3947,
|
| 317 |
+
"eval_samples_per_second": 19.685,
|
| 318 |
+
"eval_steps_per_second": 2.461,
|
| 319 |
+
"eval_wer": 0.4845186370912417,
|
| 320 |
+
"step": 4000
|
| 321 |
+
},
|
| 322 |
+
{
|
| 323 |
+
"epoch": 21.81,
|
| 324 |
+
"learning_rate": 5.375675675675675e-05,
|
| 325 |
+
"loss": 1.1029,
|
| 326 |
+
"step": 4100
|
| 327 |
+
},
|
| 328 |
+
{
|
| 329 |
+
"epoch": 22.34,
|
| 330 |
+
"learning_rate": 5.274324324324324e-05,
|
| 331 |
+
"loss": 1.1124,
|
| 332 |
+
"step": 4200
|
| 333 |
+
},
|
| 334 |
+
{
|
| 335 |
+
"epoch": 22.87,
|
| 336 |
+
"learning_rate": 5.172972972972972e-05,
|
| 337 |
+
"loss": 1.091,
|
| 338 |
+
"step": 4300
|
| 339 |
+
},
|
| 340 |
+
{
|
| 341 |
+
"epoch": 23.4,
|
| 342 |
+
"learning_rate": 5.071621621621621e-05,
|
| 343 |
+
"loss": 1.097,
|
| 344 |
+
"step": 4400
|
| 345 |
+
},
|
| 346 |
+
{
|
| 347 |
+
"epoch": 23.93,
|
| 348 |
+
"learning_rate": 4.97027027027027e-05,
|
| 349 |
+
"loss": 1.0914,
|
| 350 |
+
"step": 4500
|
| 351 |
+
},
|
| 352 |
+
{
|
| 353 |
+
"epoch": 23.93,
|
| 354 |
+
"eval_loss": 0.3050296902656555,
|
| 355 |
+
"eval_runtime": 137.2603,
|
| 356 |
+
"eval_samples_per_second": 19.991,
|
| 357 |
+
"eval_steps_per_second": 2.499,
|
| 358 |
+
"eval_wer": 0.4594100160073176,
|
| 359 |
+
"step": 4500
|
| 360 |
+
},
|
| 361 |
+
{
|
| 362 |
+
"epoch": 24.47,
|
| 363 |
+
"learning_rate": 4.8689189189189184e-05,
|
| 364 |
+
"loss": 1.0749,
|
| 365 |
+
"step": 4600
|
| 366 |
+
},
|
| 367 |
+
{
|
| 368 |
+
"epoch": 25.0,
|
| 369 |
+
"learning_rate": 4.767567567567567e-05,
|
| 370 |
+
"loss": 1.0807,
|
| 371 |
+
"step": 4700
|
| 372 |
+
},
|
| 373 |
+
{
|
| 374 |
+
"epoch": 25.53,
|
| 375 |
+
"learning_rate": 4.666216216216216e-05,
|
| 376 |
+
"loss": 1.0647,
|
| 377 |
+
"step": 4800
|
| 378 |
+
},
|
| 379 |
+
{
|
| 380 |
+
"epoch": 26.06,
|
| 381 |
+
"learning_rate": 4.564864864864864e-05,
|
| 382 |
+
"loss": 1.068,
|
| 383 |
+
"step": 4900
|
| 384 |
+
},
|
| 385 |
+
{
|
| 386 |
+
"epoch": 26.59,
|
| 387 |
+
"learning_rate": 4.463513513513513e-05,
|
| 388 |
+
"loss": 1.0468,
|
| 389 |
+
"step": 5000
|
| 390 |
+
},
|
| 391 |
+
{
|
| 392 |
+
"epoch": 26.59,
|
| 393 |
+
"eval_loss": 0.2889558672904968,
|
| 394 |
+
"eval_runtime": 137.6066,
|
| 395 |
+
"eval_samples_per_second": 19.941,
|
| 396 |
+
"eval_steps_per_second": 2.493,
|
| 397 |
+
"eval_wer": 0.44701577864166475,
|
| 398 |
+
"step": 5000
|
| 399 |
+
},
|
| 400 |
+
{
|
| 401 |
+
"epoch": 27.13,
|
| 402 |
+
"learning_rate": 4.3621621621621624e-05,
|
| 403 |
+
"loss": 1.0548,
|
| 404 |
+
"step": 5100
|
| 405 |
+
},
|
| 406 |
+
{
|
| 407 |
+
"epoch": 27.66,
|
| 408 |
+
"learning_rate": 4.26081081081081e-05,
|
| 409 |
+
"loss": 1.0423,
|
| 410 |
+
"step": 5200
|
| 411 |
+
},
|
| 412 |
+
{
|
| 413 |
+
"epoch": 28.19,
|
| 414 |
+
"learning_rate": 4.160472972972973e-05,
|
| 415 |
+
"loss": 1.0436,
|
| 416 |
+
"step": 5300
|
| 417 |
+
},
|
| 418 |
+
{
|
| 419 |
+
"epoch": 28.72,
|
| 420 |
+
"learning_rate": 4.059121621621621e-05,
|
| 421 |
+
"loss": 1.0338,
|
| 422 |
+
"step": 5400
|
| 423 |
+
},
|
| 424 |
+
{
|
| 425 |
+
"epoch": 29.25,
|
| 426 |
+
"learning_rate": 3.95777027027027e-05,
|
| 427 |
+
"loss": 1.0473,
|
| 428 |
+
"step": 5500
|
| 429 |
+
},
|
| 430 |
+
{
|
| 431 |
+
"epoch": 29.25,
|
| 432 |
+
"eval_loss": 0.27552109956741333,
|
| 433 |
+
"eval_runtime": 139.7371,
|
| 434 |
+
"eval_samples_per_second": 19.637,
|
| 435 |
+
"eval_steps_per_second": 2.455,
|
| 436 |
+
"eval_wer": 0.43306654470615136,
|
| 437 |
+
"step": 5500
|
| 438 |
+
},
|
| 439 |
+
{
|
| 440 |
+
"epoch": 29.78,
|
| 441 |
+
"learning_rate": 3.856418918918919e-05,
|
| 442 |
+
"loss": 1.027,
|
| 443 |
+
"step": 5600
|
| 444 |
+
},
|
| 445 |
+
{
|
| 446 |
+
"epoch": 30.32,
|
| 447 |
+
"learning_rate": 3.755067567567568e-05,
|
| 448 |
+
"loss": 1.0418,
|
| 449 |
+
"step": 5700
|
| 450 |
+
},
|
| 451 |
+
{
|
| 452 |
+
"epoch": 30.85,
|
| 453 |
+
"learning_rate": 3.653716216216216e-05,
|
| 454 |
+
"loss": 1.0147,
|
| 455 |
+
"step": 5800
|
| 456 |
+
},
|
| 457 |
+
{
|
| 458 |
+
"epoch": 31.38,
|
| 459 |
+
"learning_rate": 3.552364864864865e-05,
|
| 460 |
+
"loss": 1.0197,
|
| 461 |
+
"step": 5900
|
| 462 |
+
},
|
| 463 |
+
{
|
| 464 |
+
"epoch": 31.91,
|
| 465 |
+
"learning_rate": 3.451013513513513e-05,
|
| 466 |
+
"loss": 1.0065,
|
| 467 |
+
"step": 6000
|
| 468 |
+
},
|
| 469 |
+
{
|
| 470 |
+
"epoch": 31.91,
|
| 471 |
+
"eval_loss": 0.27181389927864075,
|
| 472 |
+
"eval_runtime": 138.4335,
|
| 473 |
+
"eval_samples_per_second": 19.822,
|
| 474 |
+
"eval_steps_per_second": 2.478,
|
| 475 |
+
"eval_wer": 0.42638920649439743,
|
| 476 |
+
"step": 6000
|
| 477 |
+
},
|
| 478 |
+
{
|
| 479 |
+
"epoch": 32.45,
|
| 480 |
+
"learning_rate": 3.3496621621621615e-05,
|
| 481 |
+
"loss": 1.0138,
|
| 482 |
+
"step": 6100
|
| 483 |
+
},
|
| 484 |
+
{
|
| 485 |
+
"epoch": 32.97,
|
| 486 |
+
"learning_rate": 3.2483108108108105e-05,
|
| 487 |
+
"loss": 1.0027,
|
| 488 |
+
"step": 6200
|
| 489 |
+
},
|
| 490 |
+
{
|
| 491 |
+
"epoch": 33.51,
|
| 492 |
+
"learning_rate": 3.146959459459459e-05,
|
| 493 |
+
"loss": 1.0157,
|
| 494 |
+
"step": 6300
|
| 495 |
+
},
|
| 496 |
+
{
|
| 497 |
+
"epoch": 34.04,
|
| 498 |
+
"learning_rate": 3.045608108108108e-05,
|
| 499 |
+
"loss": 1.0034,
|
| 500 |
+
"step": 6400
|
| 501 |
+
},
|
| 502 |
+
{
|
| 503 |
+
"epoch": 34.57,
|
| 504 |
+
"learning_rate": 2.9442567567567563e-05,
|
| 505 |
+
"loss": 0.9794,
|
| 506 |
+
"step": 6500
|
| 507 |
+
},
|
| 508 |
+
{
|
| 509 |
+
"epoch": 34.57,
|
| 510 |
+
"eval_loss": 0.2646064758300781,
|
| 511 |
+
"eval_runtime": 138.3623,
|
| 512 |
+
"eval_samples_per_second": 19.832,
|
| 513 |
+
"eval_steps_per_second": 2.479,
|
| 514 |
+
"eval_wer": 0.41934598673679396,
|
| 515 |
+
"step": 6500
|
| 516 |
+
},
|
| 517 |
+
{
|
| 518 |
+
"epoch": 35.11,
|
| 519 |
+
"learning_rate": 2.8429054054054054e-05,
|
| 520 |
+
"loss": 0.9911,
|
| 521 |
+
"step": 6600
|
| 522 |
+
},
|
| 523 |
+
{
|
| 524 |
+
"epoch": 35.64,
|
| 525 |
+
"learning_rate": 2.7415540540540538e-05,
|
| 526 |
+
"loss": 0.9807,
|
| 527 |
+
"step": 6700
|
| 528 |
+
},
|
| 529 |
+
{
|
| 530 |
+
"epoch": 36.17,
|
| 531 |
+
"learning_rate": 2.6402027027027025e-05,
|
| 532 |
+
"loss": 0.9843,
|
| 533 |
+
"step": 6800
|
| 534 |
+
},
|
| 535 |
+
{
|
| 536 |
+
"epoch": 36.7,
|
| 537 |
+
"learning_rate": 2.538851351351351e-05,
|
| 538 |
+
"loss": 0.9735,
|
| 539 |
+
"step": 6900
|
| 540 |
+
},
|
| 541 |
+
{
|
| 542 |
+
"epoch": 37.23,
|
| 543 |
+
"learning_rate": 2.4375e-05,
|
| 544 |
+
"loss": 0.9849,
|
| 545 |
+
"step": 7000
|
| 546 |
+
},
|
| 547 |
+
{
|
| 548 |
+
"epoch": 37.23,
|
| 549 |
+
"eval_loss": 0.2609545886516571,
|
| 550 |
+
"eval_runtime": 144.7382,
|
| 551 |
+
"eval_samples_per_second": 18.958,
|
| 552 |
+
"eval_steps_per_second": 2.37,
|
| 553 |
+
"eval_wer": 0.4058083695403613,
|
| 554 |
+
"step": 7000
|
| 555 |
+
},
|
| 556 |
+
{
|
| 557 |
+
"epoch": 37.76,
|
| 558 |
+
"learning_rate": 2.3361486486486483e-05,
|
| 559 |
+
"loss": 0.983,
|
| 560 |
+
"step": 7100
|
| 561 |
+
},
|
| 562 |
+
{
|
| 563 |
+
"epoch": 38.3,
|
| 564 |
+
"learning_rate": 2.234797297297297e-05,
|
| 565 |
+
"loss": 0.9775,
|
| 566 |
+
"step": 7200
|
| 567 |
+
},
|
| 568 |
+
{
|
| 569 |
+
"epoch": 38.83,
|
| 570 |
+
"learning_rate": 2.1334459459459458e-05,
|
| 571 |
+
"loss": 0.9597,
|
| 572 |
+
"step": 7300
|
| 573 |
+
},
|
| 574 |
+
{
|
| 575 |
+
"epoch": 39.36,
|
| 576 |
+
"learning_rate": 2.0320945945945945e-05,
|
| 577 |
+
"loss": 0.9606,
|
| 578 |
+
"step": 7400
|
| 579 |
+
},
|
| 580 |
+
{
|
| 581 |
+
"epoch": 39.89,
|
| 582 |
+
"learning_rate": 1.930743243243243e-05,
|
| 583 |
+
"loss": 0.9496,
|
| 584 |
+
"step": 7500
|
| 585 |
+
},
|
| 586 |
+
{
|
| 587 |
+
"epoch": 39.89,
|
| 588 |
+
"eval_loss": 0.2522386610507965,
|
| 589 |
+
"eval_runtime": 139.7036,
|
| 590 |
+
"eval_samples_per_second": 19.642,
|
| 591 |
+
"eval_steps_per_second": 2.455,
|
| 592 |
+
"eval_wer": 0.3984907386233707,
|
| 593 |
+
"step": 7500
|
| 594 |
+
},
|
| 595 |
+
{
|
| 596 |
+
"epoch": 40.42,
|
| 597 |
+
"learning_rate": 1.829391891891892e-05,
|
| 598 |
+
"loss": 0.9581,
|
| 599 |
+
"step": 7600
|
| 600 |
+
},
|
| 601 |
+
{
|
| 602 |
+
"epoch": 40.95,
|
| 603 |
+
"learning_rate": 1.7280405405405403e-05,
|
| 604 |
+
"loss": 0.9477,
|
| 605 |
+
"step": 7700
|
| 606 |
+
},
|
| 607 |
+
{
|
| 608 |
+
"epoch": 41.49,
|
| 609 |
+
"learning_rate": 1.626689189189189e-05,
|
| 610 |
+
"loss": 0.9538,
|
| 611 |
+
"step": 7800
|
| 612 |
+
},
|
| 613 |
+
{
|
| 614 |
+
"epoch": 42.02,
|
| 615 |
+
"learning_rate": 1.5253378378378378e-05,
|
| 616 |
+
"loss": 0.9631,
|
| 617 |
+
"step": 7900
|
| 618 |
+
},
|
| 619 |
+
{
|
| 620 |
+
"epoch": 42.55,
|
| 621 |
+
"learning_rate": 1.4239864864864863e-05,
|
| 622 |
+
"loss": 0.9367,
|
| 623 |
+
"step": 8000
|
| 624 |
+
},
|
| 625 |
+
{
|
| 626 |
+
"epoch": 42.55,
|
| 627 |
+
"eval_loss": 0.25142449140548706,
|
| 628 |
+
"eval_runtime": 138.6762,
|
| 629 |
+
"eval_samples_per_second": 19.787,
|
| 630 |
+
"eval_steps_per_second": 2.473,
|
| 631 |
+
"eval_wer": 0.3946947175851818,
|
| 632 |
+
"step": 8000
|
| 633 |
+
},
|
| 634 |
+
{
|
| 635 |
+
"epoch": 43.08,
|
| 636 |
+
"learning_rate": 1.322635135135135e-05,
|
| 637 |
+
"loss": 0.9498,
|
| 638 |
+
"step": 8100
|
| 639 |
+
},
|
| 640 |
+
{
|
| 641 |
+
"epoch": 43.61,
|
| 642 |
+
"learning_rate": 1.2212837837837838e-05,
|
| 643 |
+
"loss": 0.9389,
|
| 644 |
+
"step": 8200
|
| 645 |
+
},
|
| 646 |
+
{
|
| 647 |
+
"epoch": 44.15,
|
| 648 |
+
"learning_rate": 1.1199324324324323e-05,
|
| 649 |
+
"loss": 0.937,
|
| 650 |
+
"step": 8300
|
| 651 |
+
},
|
| 652 |
+
{
|
| 653 |
+
"epoch": 44.68,
|
| 654 |
+
"learning_rate": 1.0195945945945945e-05,
|
| 655 |
+
"loss": 0.9394,
|
| 656 |
+
"step": 8400
|
| 657 |
+
},
|
| 658 |
+
{
|
| 659 |
+
"epoch": 45.21,
|
| 660 |
+
"learning_rate": 9.182432432432432e-06,
|
| 661 |
+
"loss": 0.9295,
|
| 662 |
+
"step": 8500
|
| 663 |
+
},
|
| 664 |
+
{
|
| 665 |
+
"epoch": 45.21,
|
| 666 |
+
"eval_loss": 0.24582630395889282,
|
| 667 |
+
"eval_runtime": 140.7119,
|
| 668 |
+
"eval_samples_per_second": 19.501,
|
| 669 |
+
"eval_steps_per_second": 2.438,
|
| 670 |
+
"eval_wer": 0.3883375257260462,
|
| 671 |
+
"step": 8500
|
| 672 |
+
},
|
| 673 |
+
{
|
| 674 |
+
"epoch": 45.74,
|
| 675 |
+
"learning_rate": 8.168918918918917e-06,
|
| 676 |
+
"loss": 0.9244,
|
| 677 |
+
"step": 8600
|
| 678 |
+
},
|
| 679 |
+
{
|
| 680 |
+
"epoch": 46.28,
|
| 681 |
+
"learning_rate": 7.1554054054054045e-06,
|
| 682 |
+
"loss": 0.9273,
|
| 683 |
+
"step": 8700
|
| 684 |
+
},
|
| 685 |
+
{
|
| 686 |
+
"epoch": 46.81,
|
| 687 |
+
"learning_rate": 6.141891891891891e-06,
|
| 688 |
+
"loss": 0.9229,
|
| 689 |
+
"step": 8800
|
| 690 |
+
},
|
| 691 |
+
{
|
| 692 |
+
"epoch": 47.34,
|
| 693 |
+
"learning_rate": 5.128378378378377e-06,
|
| 694 |
+
"loss": 0.9392,
|
| 695 |
+
"step": 8900
|
| 696 |
+
},
|
| 697 |
+
{
|
| 698 |
+
"epoch": 47.87,
|
| 699 |
+
"learning_rate": 4.1148648648648645e-06,
|
| 700 |
+
"loss": 0.9187,
|
| 701 |
+
"step": 9000
|
| 702 |
+
},
|
| 703 |
+
{
|
| 704 |
+
"epoch": 47.87,
|
| 705 |
+
"eval_loss": 0.24390903115272522,
|
| 706 |
+
"eval_runtime": 139.6323,
|
| 707 |
+
"eval_samples_per_second": 19.652,
|
| 708 |
+
"eval_steps_per_second": 2.456,
|
| 709 |
+
"eval_wer": 0.3832609192773839,
|
| 710 |
+
"step": 9000
|
| 711 |
+
},
|
| 712 |
+
{
|
| 713 |
+
"epoch": 48.4,
|
| 714 |
+
"learning_rate": 3.1013513513513513e-06,
|
| 715 |
+
"loss": 0.9194,
|
| 716 |
+
"step": 9100
|
| 717 |
+
},
|
| 718 |
+
{
|
| 719 |
+
"epoch": 48.93,
|
| 720 |
+
"learning_rate": 2.0878378378378376e-06,
|
| 721 |
+
"loss": 0.9234,
|
| 722 |
+
"step": 9200
|
| 723 |
+
},
|
| 724 |
+
{
|
| 725 |
+
"epoch": 49.47,
|
| 726 |
+
"learning_rate": 1.0743243243243242e-06,
|
| 727 |
+
"loss": 0.9257,
|
| 728 |
+
"step": 9300
|
| 729 |
+
},
|
| 730 |
+
{
|
| 731 |
+
"epoch": 50.0,
|
| 732 |
+
"learning_rate": 6.08108108108108e-08,
|
| 733 |
+
"loss": 0.9097,
|
| 734 |
+
"step": 9400
|
| 735 |
+
},
|
| 736 |
+
{
|
| 737 |
+
"epoch": 50.0,
|
| 738 |
+
"step": 9400,
|
| 739 |
+
"total_flos": 5.3948756860365595e+19,
|
| 740 |
+
"train_loss": 1.5738848333155855,
|
| 741 |
+
"train_runtime": 24806.553,
|
| 742 |
+
"train_samples_per_second": 12.162,
|
| 743 |
+
"train_steps_per_second": 0.379
|
| 744 |
+
}
|
| 745 |
+
],
|
| 746 |
+
"max_steps": 9400,
|
| 747 |
+
"num_train_epochs": 50,
|
| 748 |
+
"total_flos": 5.3948756860365595e+19,
|
| 749 |
+
"trial_name": null,
|
| 750 |
+
"trial_params": null
|
| 751 |
+
}
|