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 +706 -0
all_results.json
ADDED
|
@@ -0,0 +1,14 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"epoch": 400.0,
|
| 3 |
+
"eval_loss": 0.5950954556465149,
|
| 4 |
+
"eval_runtime": 13.595,
|
| 5 |
+
"eval_samples": 306,
|
| 6 |
+
"eval_samples_per_second": 22.508,
|
| 7 |
+
"eval_steps_per_second": 2.869,
|
| 8 |
+
"eval_wer": 0.5435218612113919,
|
| 9 |
+
"train_loss": 1.3667902833765204,
|
| 10 |
+
"train_runtime": 19151.8999,
|
| 11 |
+
"train_samples": 698,
|
| 12 |
+
"train_samples_per_second": 14.578,
|
| 13 |
+
"train_steps_per_second": 0.459
|
| 14 |
+
}
|
eval_results.json
ADDED
|
@@ -0,0 +1,9 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"epoch": 400.0,
|
| 3 |
+
"eval_loss": 0.5950954556465149,
|
| 4 |
+
"eval_runtime": 13.595,
|
| 5 |
+
"eval_samples": 306,
|
| 6 |
+
"eval_samples_per_second": 22.508,
|
| 7 |
+
"eval_steps_per_second": 2.869,
|
| 8 |
+
"eval_wer": 0.5435218612113919
|
| 9 |
+
}
|
pytorch_model.bin
CHANGED
|
@@ -1,3 +1,3 @@
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
-
oid sha256:
|
| 3 |
size 1262218865
|
|
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:d7af4645b6f25facedab2878125586571da40b2d0b39eb271102dbc14914fdfa
|
| 3 |
size 1262218865
|
train_results.json
ADDED
|
@@ -0,0 +1,8 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"epoch": 400.0,
|
| 3 |
+
"train_loss": 1.3667902833765204,
|
| 4 |
+
"train_runtime": 19151.8999,
|
| 5 |
+
"train_samples": 698,
|
| 6 |
+
"train_samples_per_second": 14.578,
|
| 7 |
+
"train_steps_per_second": 0.459
|
| 8 |
+
}
|
trainer_state.json
ADDED
|
@@ -0,0 +1,706 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"best_metric": null,
|
| 3 |
+
"best_model_checkpoint": null,
|
| 4 |
+
"epoch": 400.0,
|
| 5 |
+
"global_step": 8800,
|
| 6 |
+
"is_hyper_param_search": false,
|
| 7 |
+
"is_local_process_zero": true,
|
| 8 |
+
"is_world_process_zero": true,
|
| 9 |
+
"log_history": [
|
| 10 |
+
{
|
| 11 |
+
"epoch": 4.55,
|
| 12 |
+
"learning_rate": 3.7125e-06,
|
| 13 |
+
"loss": 16.0016,
|
| 14 |
+
"step": 100
|
| 15 |
+
},
|
| 16 |
+
{
|
| 17 |
+
"epoch": 9.09,
|
| 18 |
+
"learning_rate": 7.4625e-06,
|
| 19 |
+
"loss": 8.8896,
|
| 20 |
+
"step": 200
|
| 21 |
+
},
|
| 22 |
+
{
|
| 23 |
+
"epoch": 13.64,
|
| 24 |
+
"learning_rate": 1.1212499999999998e-05,
|
| 25 |
+
"loss": 5.8994,
|
| 26 |
+
"step": 300
|
| 27 |
+
},
|
| 28 |
+
{
|
| 29 |
+
"epoch": 18.18,
|
| 30 |
+
"learning_rate": 1.49625e-05,
|
| 31 |
+
"loss": 4.9866,
|
| 32 |
+
"step": 400
|
| 33 |
+
},
|
| 34 |
+
{
|
| 35 |
+
"epoch": 22.73,
|
| 36 |
+
"learning_rate": 1.8712499999999997e-05,
|
| 37 |
+
"loss": 4.2706,
|
| 38 |
+
"step": 500
|
| 39 |
+
},
|
| 40 |
+
{
|
| 41 |
+
"epoch": 22.73,
|
| 42 |
+
"eval_loss": 4.017421722412109,
|
| 43 |
+
"eval_runtime": 13.7125,
|
| 44 |
+
"eval_samples_per_second": 22.315,
|
| 45 |
+
"eval_steps_per_second": 2.844,
|
| 46 |
+
"eval_wer": 1.0,
|
| 47 |
+
"step": 500
|
| 48 |
+
},
|
| 49 |
+
{
|
| 50 |
+
"epoch": 27.27,
|
| 51 |
+
"learning_rate": 2.2462499999999997e-05,
|
| 52 |
+
"loss": 3.7391,
|
| 53 |
+
"step": 600
|
| 54 |
+
},
|
| 55 |
+
{
|
| 56 |
+
"epoch": 31.82,
|
| 57 |
+
"learning_rate": 2.6212499999999997e-05,
|
| 58 |
+
"loss": 3.4669,
|
| 59 |
+
"step": 700
|
| 60 |
+
},
|
| 61 |
+
{
|
| 62 |
+
"epoch": 36.36,
|
| 63 |
+
"learning_rate": 2.99625e-05,
|
| 64 |
+
"loss": 3.3769,
|
| 65 |
+
"step": 800
|
| 66 |
+
},
|
| 67 |
+
{
|
| 68 |
+
"epoch": 40.91,
|
| 69 |
+
"learning_rate": 3.37125e-05,
|
| 70 |
+
"loss": 3.304,
|
| 71 |
+
"step": 900
|
| 72 |
+
},
|
| 73 |
+
{
|
| 74 |
+
"epoch": 45.45,
|
| 75 |
+
"learning_rate": 3.7462499999999996e-05,
|
| 76 |
+
"loss": 3.2492,
|
| 77 |
+
"step": 1000
|
| 78 |
+
},
|
| 79 |
+
{
|
| 80 |
+
"epoch": 45.45,
|
| 81 |
+
"eval_loss": 3.2308928966522217,
|
| 82 |
+
"eval_runtime": 13.6766,
|
| 83 |
+
"eval_samples_per_second": 22.374,
|
| 84 |
+
"eval_steps_per_second": 2.852,
|
| 85 |
+
"eval_wer": 0.9907741676694746,
|
| 86 |
+
"step": 1000
|
| 87 |
+
},
|
| 88 |
+
{
|
| 89 |
+
"epoch": 50.0,
|
| 90 |
+
"learning_rate": 4.12125e-05,
|
| 91 |
+
"loss": 3.1542,
|
| 92 |
+
"step": 1100
|
| 93 |
+
},
|
| 94 |
+
{
|
| 95 |
+
"epoch": 54.55,
|
| 96 |
+
"learning_rate": 4.4962499999999995e-05,
|
| 97 |
+
"loss": 2.9723,
|
| 98 |
+
"step": 1200
|
| 99 |
+
},
|
| 100 |
+
{
|
| 101 |
+
"epoch": 59.09,
|
| 102 |
+
"learning_rate": 4.871249999999999e-05,
|
| 103 |
+
"loss": 2.6187,
|
| 104 |
+
"step": 1300
|
| 105 |
+
},
|
| 106 |
+
{
|
| 107 |
+
"epoch": 63.64,
|
| 108 |
+
"learning_rate": 5.2462499999999994e-05,
|
| 109 |
+
"loss": 2.2201,
|
| 110 |
+
"step": 1400
|
| 111 |
+
},
|
| 112 |
+
{
|
| 113 |
+
"epoch": 68.18,
|
| 114 |
+
"learning_rate": 5.62125e-05,
|
| 115 |
+
"loss": 1.9709,
|
| 116 |
+
"step": 1500
|
| 117 |
+
},
|
| 118 |
+
{
|
| 119 |
+
"epoch": 68.18,
|
| 120 |
+
"eval_loss": 1.0651201009750366,
|
| 121 |
+
"eval_runtime": 13.4664,
|
| 122 |
+
"eval_samples_per_second": 22.723,
|
| 123 |
+
"eval_steps_per_second": 2.896,
|
| 124 |
+
"eval_wer": 0.8439630966706779,
|
| 125 |
+
"step": 1500
|
| 126 |
+
},
|
| 127 |
+
{
|
| 128 |
+
"epoch": 72.73,
|
| 129 |
+
"learning_rate": 5.9962499999999994e-05,
|
| 130 |
+
"loss": 1.7822,
|
| 131 |
+
"step": 1600
|
| 132 |
+
},
|
| 133 |
+
{
|
| 134 |
+
"epoch": 77.27,
|
| 135 |
+
"learning_rate": 6.37125e-05,
|
| 136 |
+
"loss": 1.6511,
|
| 137 |
+
"step": 1700
|
| 138 |
+
},
|
| 139 |
+
{
|
| 140 |
+
"epoch": 81.82,
|
| 141 |
+
"learning_rate": 6.746249999999999e-05,
|
| 142 |
+
"loss": 1.5519,
|
| 143 |
+
"step": 1800
|
| 144 |
+
},
|
| 145 |
+
{
|
| 146 |
+
"epoch": 86.36,
|
| 147 |
+
"learning_rate": 7.121249999999999e-05,
|
| 148 |
+
"loss": 1.4612,
|
| 149 |
+
"step": 1900
|
| 150 |
+
},
|
| 151 |
+
{
|
| 152 |
+
"epoch": 90.91,
|
| 153 |
+
"learning_rate": 7.49625e-05,
|
| 154 |
+
"loss": 1.4088,
|
| 155 |
+
"step": 2000
|
| 156 |
+
},
|
| 157 |
+
{
|
| 158 |
+
"epoch": 90.91,
|
| 159 |
+
"eval_loss": 0.576468288898468,
|
| 160 |
+
"eval_runtime": 13.6104,
|
| 161 |
+
"eval_samples_per_second": 22.483,
|
| 162 |
+
"eval_steps_per_second": 2.865,
|
| 163 |
+
"eval_wer": 0.6550340954673085,
|
| 164 |
+
"step": 2000
|
| 165 |
+
},
|
| 166 |
+
{
|
| 167 |
+
"epoch": 95.45,
|
| 168 |
+
"learning_rate": 7.390808823529411e-05,
|
| 169 |
+
"loss": 1.3302,
|
| 170 |
+
"step": 2100
|
| 171 |
+
},
|
| 172 |
+
{
|
| 173 |
+
"epoch": 100.0,
|
| 174 |
+
"learning_rate": 7.280514705882352e-05,
|
| 175 |
+
"loss": 1.2739,
|
| 176 |
+
"step": 2200
|
| 177 |
+
},
|
| 178 |
+
{
|
| 179 |
+
"epoch": 104.55,
|
| 180 |
+
"learning_rate": 7.171323529411764e-05,
|
| 181 |
+
"loss": 1.2181,
|
| 182 |
+
"step": 2300
|
| 183 |
+
},
|
| 184 |
+
{
|
| 185 |
+
"epoch": 109.09,
|
| 186 |
+
"learning_rate": 7.062132352941176e-05,
|
| 187 |
+
"loss": 1.1776,
|
| 188 |
+
"step": 2400
|
| 189 |
+
},
|
| 190 |
+
{
|
| 191 |
+
"epoch": 113.64,
|
| 192 |
+
"learning_rate": 6.951838235294117e-05,
|
| 193 |
+
"loss": 1.1326,
|
| 194 |
+
"step": 2500
|
| 195 |
+
},
|
| 196 |
+
{
|
| 197 |
+
"epoch": 113.64,
|
| 198 |
+
"eval_loss": 0.48421531915664673,
|
| 199 |
+
"eval_runtime": 13.6288,
|
| 200 |
+
"eval_samples_per_second": 22.452,
|
| 201 |
+
"eval_steps_per_second": 2.862,
|
| 202 |
+
"eval_wer": 0.5760128359406338,
|
| 203 |
+
"step": 2500
|
| 204 |
+
},
|
| 205 |
+
{
|
| 206 |
+
"epoch": 118.18,
|
| 207 |
+
"learning_rate": 6.841544117647059e-05,
|
| 208 |
+
"loss": 1.0897,
|
| 209 |
+
"step": 2600
|
| 210 |
+
},
|
| 211 |
+
{
|
| 212 |
+
"epoch": 122.73,
|
| 213 |
+
"learning_rate": 6.731249999999999e-05,
|
| 214 |
+
"loss": 1.0634,
|
| 215 |
+
"step": 2700
|
| 216 |
+
},
|
| 217 |
+
{
|
| 218 |
+
"epoch": 127.27,
|
| 219 |
+
"learning_rate": 6.62095588235294e-05,
|
| 220 |
+
"loss": 1.0321,
|
| 221 |
+
"step": 2800
|
| 222 |
+
},
|
| 223 |
+
{
|
| 224 |
+
"epoch": 131.82,
|
| 225 |
+
"learning_rate": 6.510661764705882e-05,
|
| 226 |
+
"loss": 0.9962,
|
| 227 |
+
"step": 2900
|
| 228 |
+
},
|
| 229 |
+
{
|
| 230 |
+
"epoch": 136.36,
|
| 231 |
+
"learning_rate": 6.400367647058824e-05,
|
| 232 |
+
"loss": 0.9709,
|
| 233 |
+
"step": 3000
|
| 234 |
+
},
|
| 235 |
+
{
|
| 236 |
+
"epoch": 136.36,
|
| 237 |
+
"eval_loss": 0.4784719944000244,
|
| 238 |
+
"eval_runtime": 13.5638,
|
| 239 |
+
"eval_samples_per_second": 22.56,
|
| 240 |
+
"eval_steps_per_second": 2.875,
|
| 241 |
+
"eval_wer": 0.6012835940633775,
|
| 242 |
+
"step": 3000
|
| 243 |
+
},
|
| 244 |
+
{
|
| 245 |
+
"epoch": 140.91,
|
| 246 |
+
"learning_rate": 6.290073529411764e-05,
|
| 247 |
+
"loss": 0.935,
|
| 248 |
+
"step": 3100
|
| 249 |
+
},
|
| 250 |
+
{
|
| 251 |
+
"epoch": 145.45,
|
| 252 |
+
"learning_rate": 6.179779411764705e-05,
|
| 253 |
+
"loss": 0.9179,
|
| 254 |
+
"step": 3200
|
| 255 |
+
},
|
| 256 |
+
{
|
| 257 |
+
"epoch": 150.0,
|
| 258 |
+
"learning_rate": 6.069485294117646e-05,
|
| 259 |
+
"loss": 0.8854,
|
| 260 |
+
"step": 3300
|
| 261 |
+
},
|
| 262 |
+
{
|
| 263 |
+
"epoch": 154.55,
|
| 264 |
+
"learning_rate": 5.9591911764705876e-05,
|
| 265 |
+
"loss": 0.87,
|
| 266 |
+
"step": 3400
|
| 267 |
+
},
|
| 268 |
+
{
|
| 269 |
+
"epoch": 159.09,
|
| 270 |
+
"learning_rate": 5.848897058823529e-05,
|
| 271 |
+
"loss": 0.8433,
|
| 272 |
+
"step": 3500
|
| 273 |
+
},
|
| 274 |
+
{
|
| 275 |
+
"epoch": 159.09,
|
| 276 |
+
"eval_loss": 0.5048245191574097,
|
| 277 |
+
"eval_runtime": 14.3488,
|
| 278 |
+
"eval_samples_per_second": 21.326,
|
| 279 |
+
"eval_steps_per_second": 2.718,
|
| 280 |
+
"eval_wer": 0.5419173686321701,
|
| 281 |
+
"step": 3500
|
| 282 |
+
},
|
| 283 |
+
{
|
| 284 |
+
"epoch": 163.64,
|
| 285 |
+
"learning_rate": 5.73860294117647e-05,
|
| 286 |
+
"loss": 0.8145,
|
| 287 |
+
"step": 3600
|
| 288 |
+
},
|
| 289 |
+
{
|
| 290 |
+
"epoch": 168.18,
|
| 291 |
+
"learning_rate": 5.6283088235294115e-05,
|
| 292 |
+
"loss": 0.7986,
|
| 293 |
+
"step": 3700
|
| 294 |
+
},
|
| 295 |
+
{
|
| 296 |
+
"epoch": 172.73,
|
| 297 |
+
"learning_rate": 5.5180147058823523e-05,
|
| 298 |
+
"loss": 0.7804,
|
| 299 |
+
"step": 3800
|
| 300 |
+
},
|
| 301 |
+
{
|
| 302 |
+
"epoch": 177.27,
|
| 303 |
+
"learning_rate": 5.407720588235294e-05,
|
| 304 |
+
"loss": 0.7639,
|
| 305 |
+
"step": 3900
|
| 306 |
+
},
|
| 307 |
+
{
|
| 308 |
+
"epoch": 181.82,
|
| 309 |
+
"learning_rate": 5.297426470588235e-05,
|
| 310 |
+
"loss": 0.7404,
|
| 311 |
+
"step": 4000
|
| 312 |
+
},
|
| 313 |
+
{
|
| 314 |
+
"epoch": 181.82,
|
| 315 |
+
"eval_loss": 0.5051876902580261,
|
| 316 |
+
"eval_runtime": 14.251,
|
| 317 |
+
"eval_samples_per_second": 21.472,
|
| 318 |
+
"eval_steps_per_second": 2.737,
|
| 319 |
+
"eval_wer": 0.533894905736061,
|
| 320 |
+
"step": 4000
|
| 321 |
+
},
|
| 322 |
+
{
|
| 323 |
+
"epoch": 186.36,
|
| 324 |
+
"learning_rate": 5.187132352941176e-05,
|
| 325 |
+
"loss": 0.7352,
|
| 326 |
+
"step": 4100
|
| 327 |
+
},
|
| 328 |
+
{
|
| 329 |
+
"epoch": 190.91,
|
| 330 |
+
"learning_rate": 5.076838235294117e-05,
|
| 331 |
+
"loss": 0.7073,
|
| 332 |
+
"step": 4200
|
| 333 |
+
},
|
| 334 |
+
{
|
| 335 |
+
"epoch": 195.45,
|
| 336 |
+
"learning_rate": 4.9665441176470586e-05,
|
| 337 |
+
"loss": 0.6933,
|
| 338 |
+
"step": 4300
|
| 339 |
+
},
|
| 340 |
+
{
|
| 341 |
+
"epoch": 200.0,
|
| 342 |
+
"learning_rate": 4.8562499999999995e-05,
|
| 343 |
+
"loss": 0.6729,
|
| 344 |
+
"step": 4400
|
| 345 |
+
},
|
| 346 |
+
{
|
| 347 |
+
"epoch": 204.55,
|
| 348 |
+
"learning_rate": 4.747058823529411e-05,
|
| 349 |
+
"loss": 0.6589,
|
| 350 |
+
"step": 4500
|
| 351 |
+
},
|
| 352 |
+
{
|
| 353 |
+
"epoch": 204.55,
|
| 354 |
+
"eval_loss": 0.5237030982971191,
|
| 355 |
+
"eval_runtime": 13.4774,
|
| 356 |
+
"eval_samples_per_second": 22.705,
|
| 357 |
+
"eval_steps_per_second": 2.894,
|
| 358 |
+
"eval_wer": 0.5896510228640193,
|
| 359 |
+
"step": 4500
|
| 360 |
+
},
|
| 361 |
+
{
|
| 362 |
+
"epoch": 209.09,
|
| 363 |
+
"learning_rate": 4.6367647058823526e-05,
|
| 364 |
+
"loss": 0.644,
|
| 365 |
+
"step": 4600
|
| 366 |
+
},
|
| 367 |
+
{
|
| 368 |
+
"epoch": 213.64,
|
| 369 |
+
"learning_rate": 4.5264705882352935e-05,
|
| 370 |
+
"loss": 0.6221,
|
| 371 |
+
"step": 4700
|
| 372 |
+
},
|
| 373 |
+
{
|
| 374 |
+
"epoch": 218.18,
|
| 375 |
+
"learning_rate": 4.416176470588235e-05,
|
| 376 |
+
"loss": 0.6119,
|
| 377 |
+
"step": 4800
|
| 378 |
+
},
|
| 379 |
+
{
|
| 380 |
+
"epoch": 222.73,
|
| 381 |
+
"learning_rate": 4.305882352941176e-05,
|
| 382 |
+
"loss": 0.601,
|
| 383 |
+
"step": 4900
|
| 384 |
+
},
|
| 385 |
+
{
|
| 386 |
+
"epoch": 227.27,
|
| 387 |
+
"learning_rate": 4.196691176470588e-05,
|
| 388 |
+
"loss": 0.5831,
|
| 389 |
+
"step": 5000
|
| 390 |
+
},
|
| 391 |
+
{
|
| 392 |
+
"epoch": 227.27,
|
| 393 |
+
"eval_loss": 0.5166385769844055,
|
| 394 |
+
"eval_runtime": 13.5084,
|
| 395 |
+
"eval_samples_per_second": 22.653,
|
| 396 |
+
"eval_steps_per_second": 2.887,
|
| 397 |
+
"eval_wer": 0.5447252306458082,
|
| 398 |
+
"step": 5000
|
| 399 |
+
},
|
| 400 |
+
{
|
| 401 |
+
"epoch": 231.82,
|
| 402 |
+
"learning_rate": 4.086397058823529e-05,
|
| 403 |
+
"loss": 0.579,
|
| 404 |
+
"step": 5100
|
| 405 |
+
},
|
| 406 |
+
{
|
| 407 |
+
"epoch": 236.36,
|
| 408 |
+
"learning_rate": 3.9761029411764705e-05,
|
| 409 |
+
"loss": 0.5577,
|
| 410 |
+
"step": 5200
|
| 411 |
+
},
|
| 412 |
+
{
|
| 413 |
+
"epoch": 240.91,
|
| 414 |
+
"learning_rate": 3.8658088235294113e-05,
|
| 415 |
+
"loss": 0.547,
|
| 416 |
+
"step": 5300
|
| 417 |
+
},
|
| 418 |
+
{
|
| 419 |
+
"epoch": 245.45,
|
| 420 |
+
"learning_rate": 3.755514705882353e-05,
|
| 421 |
+
"loss": 0.5452,
|
| 422 |
+
"step": 5400
|
| 423 |
+
},
|
| 424 |
+
{
|
| 425 |
+
"epoch": 250.0,
|
| 426 |
+
"learning_rate": 3.645220588235294e-05,
|
| 427 |
+
"loss": 0.5375,
|
| 428 |
+
"step": 5500
|
| 429 |
+
},
|
| 430 |
+
{
|
| 431 |
+
"epoch": 250.0,
|
| 432 |
+
"eval_loss": 0.5291843414306641,
|
| 433 |
+
"eval_runtime": 13.5842,
|
| 434 |
+
"eval_samples_per_second": 22.526,
|
| 435 |
+
"eval_steps_per_second": 2.871,
|
| 436 |
+
"eval_wer": 0.5487364620938628,
|
| 437 |
+
"step": 5500
|
| 438 |
+
},
|
| 439 |
+
{
|
| 440 |
+
"epoch": 254.55,
|
| 441 |
+
"learning_rate": 3.534926470588235e-05,
|
| 442 |
+
"loss": 0.5233,
|
| 443 |
+
"step": 5600
|
| 444 |
+
},
|
| 445 |
+
{
|
| 446 |
+
"epoch": 259.09,
|
| 447 |
+
"learning_rate": 3.424632352941176e-05,
|
| 448 |
+
"loss": 0.5081,
|
| 449 |
+
"step": 5700
|
| 450 |
+
},
|
| 451 |
+
{
|
| 452 |
+
"epoch": 263.64,
|
| 453 |
+
"learning_rate": 3.314338235294117e-05,
|
| 454 |
+
"loss": 0.5033,
|
| 455 |
+
"step": 5800
|
| 456 |
+
},
|
| 457 |
+
{
|
| 458 |
+
"epoch": 268.18,
|
| 459 |
+
"learning_rate": 3.2040441176470585e-05,
|
| 460 |
+
"loss": 0.4857,
|
| 461 |
+
"step": 5900
|
| 462 |
+
},
|
| 463 |
+
{
|
| 464 |
+
"epoch": 272.73,
|
| 465 |
+
"learning_rate": 3.093749999999999e-05,
|
| 466 |
+
"loss": 0.4784,
|
| 467 |
+
"step": 6000
|
| 468 |
+
},
|
| 469 |
+
{
|
| 470 |
+
"epoch": 272.73,
|
| 471 |
+
"eval_loss": 0.548018217086792,
|
| 472 |
+
"eval_runtime": 13.4861,
|
| 473 |
+
"eval_samples_per_second": 22.69,
|
| 474 |
+
"eval_steps_per_second": 2.892,
|
| 475 |
+
"eval_wer": 0.5595667870036101,
|
| 476 |
+
"step": 6000
|
| 477 |
+
},
|
| 478 |
+
{
|
| 479 |
+
"epoch": 277.27,
|
| 480 |
+
"learning_rate": 2.9834558823529408e-05,
|
| 481 |
+
"loss": 0.4693,
|
| 482 |
+
"step": 6100
|
| 483 |
+
},
|
| 484 |
+
{
|
| 485 |
+
"epoch": 281.82,
|
| 486 |
+
"learning_rate": 2.8731617647058824e-05,
|
| 487 |
+
"loss": 0.4673,
|
| 488 |
+
"step": 6200
|
| 489 |
+
},
|
| 490 |
+
{
|
| 491 |
+
"epoch": 286.36,
|
| 492 |
+
"learning_rate": 2.7628676470588235e-05,
|
| 493 |
+
"loss": 0.4582,
|
| 494 |
+
"step": 6300
|
| 495 |
+
},
|
| 496 |
+
{
|
| 497 |
+
"epoch": 290.91,
|
| 498 |
+
"learning_rate": 2.6525735294117647e-05,
|
| 499 |
+
"loss": 0.4397,
|
| 500 |
+
"step": 6400
|
| 501 |
+
},
|
| 502 |
+
{
|
| 503 |
+
"epoch": 295.45,
|
| 504 |
+
"learning_rate": 2.542279411764706e-05,
|
| 505 |
+
"loss": 0.4421,
|
| 506 |
+
"step": 6500
|
| 507 |
+
},
|
| 508 |
+
{
|
| 509 |
+
"epoch": 295.45,
|
| 510 |
+
"eval_loss": 0.5682183504104614,
|
| 511 |
+
"eval_runtime": 13.4936,
|
| 512 |
+
"eval_samples_per_second": 22.677,
|
| 513 |
+
"eval_steps_per_second": 2.89,
|
| 514 |
+
"eval_wer": 0.5467308463698355,
|
| 515 |
+
"step": 6500
|
| 516 |
+
},
|
| 517 |
+
{
|
| 518 |
+
"epoch": 300.0,
|
| 519 |
+
"learning_rate": 2.431985294117647e-05,
|
| 520 |
+
"loss": 0.4361,
|
| 521 |
+
"step": 6600
|
| 522 |
+
},
|
| 523 |
+
{
|
| 524 |
+
"epoch": 304.55,
|
| 525 |
+
"learning_rate": 2.321691176470588e-05,
|
| 526 |
+
"loss": 0.4287,
|
| 527 |
+
"step": 6700
|
| 528 |
+
},
|
| 529 |
+
{
|
| 530 |
+
"epoch": 309.09,
|
| 531 |
+
"learning_rate": 2.211397058823529e-05,
|
| 532 |
+
"loss": 0.4181,
|
| 533 |
+
"step": 6800
|
| 534 |
+
},
|
| 535 |
+
{
|
| 536 |
+
"epoch": 313.64,
|
| 537 |
+
"learning_rate": 2.1011029411764703e-05,
|
| 538 |
+
"loss": 0.4156,
|
| 539 |
+
"step": 6900
|
| 540 |
+
},
|
| 541 |
+
{
|
| 542 |
+
"epoch": 318.18,
|
| 543 |
+
"learning_rate": 1.9908088235294115e-05,
|
| 544 |
+
"loss": 0.4047,
|
| 545 |
+
"step": 7000
|
| 546 |
+
},
|
| 547 |
+
{
|
| 548 |
+
"epoch": 318.18,
|
| 549 |
+
"eval_loss": 0.5681235194206238,
|
| 550 |
+
"eval_runtime": 13.3347,
|
| 551 |
+
"eval_samples_per_second": 22.948,
|
| 552 |
+
"eval_steps_per_second": 2.925,
|
| 553 |
+
"eval_wer": 0.5447252306458082,
|
| 554 |
+
"step": 7000
|
| 555 |
+
},
|
| 556 |
+
{
|
| 557 |
+
"epoch": 322.73,
|
| 558 |
+
"learning_rate": 1.8805147058823527e-05,
|
| 559 |
+
"loss": 0.3978,
|
| 560 |
+
"step": 7100
|
| 561 |
+
},
|
| 562 |
+
{
|
| 563 |
+
"epoch": 327.27,
|
| 564 |
+
"learning_rate": 1.770220588235294e-05,
|
| 565 |
+
"loss": 0.3874,
|
| 566 |
+
"step": 7200
|
| 567 |
+
},
|
| 568 |
+
{
|
| 569 |
+
"epoch": 331.82,
|
| 570 |
+
"learning_rate": 1.659926470588235e-05,
|
| 571 |
+
"loss": 0.385,
|
| 572 |
+
"step": 7300
|
| 573 |
+
},
|
| 574 |
+
{
|
| 575 |
+
"epoch": 336.36,
|
| 576 |
+
"learning_rate": 1.5496323529411763e-05,
|
| 577 |
+
"loss": 0.3817,
|
| 578 |
+
"step": 7400
|
| 579 |
+
},
|
| 580 |
+
{
|
| 581 |
+
"epoch": 340.91,
|
| 582 |
+
"learning_rate": 1.4393382352941176e-05,
|
| 583 |
+
"loss": 0.3779,
|
| 584 |
+
"step": 7500
|
| 585 |
+
},
|
| 586 |
+
{
|
| 587 |
+
"epoch": 340.91,
|
| 588 |
+
"eval_loss": 0.578274667263031,
|
| 589 |
+
"eval_runtime": 13.5305,
|
| 590 |
+
"eval_samples_per_second": 22.616,
|
| 591 |
+
"eval_steps_per_second": 2.882,
|
| 592 |
+
"eval_wer": 0.5346971520256719,
|
| 593 |
+
"step": 7500
|
| 594 |
+
},
|
| 595 |
+
{
|
| 596 |
+
"epoch": 345.45,
|
| 597 |
+
"learning_rate": 1.3290441176470588e-05,
|
| 598 |
+
"loss": 0.3702,
|
| 599 |
+
"step": 7600
|
| 600 |
+
},
|
| 601 |
+
{
|
| 602 |
+
"epoch": 350.0,
|
| 603 |
+
"learning_rate": 1.21875e-05,
|
| 604 |
+
"loss": 0.3723,
|
| 605 |
+
"step": 7700
|
| 606 |
+
},
|
| 607 |
+
{
|
| 608 |
+
"epoch": 354.55,
|
| 609 |
+
"learning_rate": 1.1084558823529412e-05,
|
| 610 |
+
"loss": 0.3646,
|
| 611 |
+
"step": 7800
|
| 612 |
+
},
|
| 613 |
+
{
|
| 614 |
+
"epoch": 359.09,
|
| 615 |
+
"learning_rate": 9.981617647058822e-06,
|
| 616 |
+
"loss": 0.3478,
|
| 617 |
+
"step": 7900
|
| 618 |
+
},
|
| 619 |
+
{
|
| 620 |
+
"epoch": 363.64,
|
| 621 |
+
"learning_rate": 8.878676470588234e-06,
|
| 622 |
+
"loss": 0.3525,
|
| 623 |
+
"step": 8000
|
| 624 |
+
},
|
| 625 |
+
{
|
| 626 |
+
"epoch": 363.64,
|
| 627 |
+
"eval_loss": 0.5856263041496277,
|
| 628 |
+
"eval_runtime": 13.4874,
|
| 629 |
+
"eval_samples_per_second": 22.688,
|
| 630 |
+
"eval_steps_per_second": 2.892,
|
| 631 |
+
"eval_wer": 0.5367027677496992,
|
| 632 |
+
"step": 8000
|
| 633 |
+
},
|
| 634 |
+
{
|
| 635 |
+
"epoch": 368.18,
|
| 636 |
+
"learning_rate": 7.775735294117647e-06,
|
| 637 |
+
"loss": 0.3574,
|
| 638 |
+
"step": 8100
|
| 639 |
+
},
|
| 640 |
+
{
|
| 641 |
+
"epoch": 372.73,
|
| 642 |
+
"learning_rate": 6.672794117647058e-06,
|
| 643 |
+
"loss": 0.3495,
|
| 644 |
+
"step": 8200
|
| 645 |
+
},
|
| 646 |
+
{
|
| 647 |
+
"epoch": 377.27,
|
| 648 |
+
"learning_rate": 5.56985294117647e-06,
|
| 649 |
+
"loss": 0.3576,
|
| 650 |
+
"step": 8300
|
| 651 |
+
},
|
| 652 |
+
{
|
| 653 |
+
"epoch": 381.82,
|
| 654 |
+
"learning_rate": 4.466911764705882e-06,
|
| 655 |
+
"loss": 0.3441,
|
| 656 |
+
"step": 8400
|
| 657 |
+
},
|
| 658 |
+
{
|
| 659 |
+
"epoch": 386.36,
|
| 660 |
+
"learning_rate": 3.363970588235294e-06,
|
| 661 |
+
"loss": 0.3393,
|
| 662 |
+
"step": 8500
|
| 663 |
+
},
|
| 664 |
+
{
|
| 665 |
+
"epoch": 386.36,
|
| 666 |
+
"eval_loss": 0.5959608554840088,
|
| 667 |
+
"eval_runtime": 13.5943,
|
| 668 |
+
"eval_samples_per_second": 22.509,
|
| 669 |
+
"eval_steps_per_second": 2.869,
|
| 670 |
+
"eval_wer": 0.5359005214600883,
|
| 671 |
+
"step": 8500
|
| 672 |
+
},
|
| 673 |
+
{
|
| 674 |
+
"epoch": 390.91,
|
| 675 |
+
"learning_rate": 2.2610294117647057e-06,
|
| 676 |
+
"loss": 0.3366,
|
| 677 |
+
"step": 8600
|
| 678 |
+
},
|
| 679 |
+
{
|
| 680 |
+
"epoch": 395.45,
|
| 681 |
+
"learning_rate": 1.1580882352941175e-06,
|
| 682 |
+
"loss": 0.3404,
|
| 683 |
+
"step": 8700
|
| 684 |
+
},
|
| 685 |
+
{
|
| 686 |
+
"epoch": 400.0,
|
| 687 |
+
"learning_rate": 5.514705882352941e-08,
|
| 688 |
+
"loss": 0.3312,
|
| 689 |
+
"step": 8800
|
| 690 |
+
},
|
| 691 |
+
{
|
| 692 |
+
"epoch": 400.0,
|
| 693 |
+
"step": 8800,
|
| 694 |
+
"total_flos": 4.6124470831742304e+19,
|
| 695 |
+
"train_loss": 1.3667902833765204,
|
| 696 |
+
"train_runtime": 19151.8999,
|
| 697 |
+
"train_samples_per_second": 14.578,
|
| 698 |
+
"train_steps_per_second": 0.459
|
| 699 |
+
}
|
| 700 |
+
],
|
| 701 |
+
"max_steps": 8800,
|
| 702 |
+
"num_train_epochs": 400,
|
| 703 |
+
"total_flos": 4.6124470831742304e+19,
|
| 704 |
+
"trial_name": null,
|
| 705 |
+
"trial_params": null
|
| 706 |
+
}
|