Commit
•
458e515
1
Parent(s):
dd8406b
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 +634 -0
all_results.json
ADDED
|
@@ -0,0 +1,14 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"epoch": 100.0,
|
| 3 |
+
"eval_loss": 0.44874435663223267,
|
| 4 |
+
"eval_runtime": 41.047,
|
| 5 |
+
"eval_samples": 953,
|
| 6 |
+
"eval_samples_per_second": 23.217,
|
| 7 |
+
"eval_steps_per_second": 0.731,
|
| 8 |
+
"eval_wer": 0.4673555126603838,
|
| 9 |
+
"train_loss": 1.2487838832034341,
|
| 10 |
+
"train_runtime": 15247.9542,
|
| 11 |
+
"train_samples": 2498,
|
| 12 |
+
"train_samples_per_second": 16.383,
|
| 13 |
+
"train_steps_per_second": 0.518
|
| 14 |
+
}
|
eval_results.json
ADDED
|
@@ -0,0 +1,9 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"epoch": 100.0,
|
| 3 |
+
"eval_loss": 0.44874435663223267,
|
| 4 |
+
"eval_runtime": 41.047,
|
| 5 |
+
"eval_samples": 953,
|
| 6 |
+
"eval_samples_per_second": 23.217,
|
| 7 |
+
"eval_steps_per_second": 0.731,
|
| 8 |
+
"eval_wer": 0.4673555126603838
|
| 9 |
+
}
|
pytorch_model.bin
CHANGED
|
@@ -1,3 +1,3 @@
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
-
oid sha256:
|
| 3 |
size 1262091761
|
|
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:1b975e71a6323abae5507c98a37766f95f014603c3f8cb1b48aa478c6b3f2a1f
|
| 3 |
size 1262091761
|
train_results.json
ADDED
|
@@ -0,0 +1,8 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"epoch": 100.0,
|
| 3 |
+
"train_loss": 1.2487838832034341,
|
| 4 |
+
"train_runtime": 15247.9542,
|
| 5 |
+
"train_samples": 2498,
|
| 6 |
+
"train_samples_per_second": 16.383,
|
| 7 |
+
"train_steps_per_second": 0.518
|
| 8 |
+
}
|
trainer_state.json
ADDED
|
@@ -0,0 +1,634 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"best_metric": null,
|
| 3 |
+
"best_model_checkpoint": null,
|
| 4 |
+
"epoch": 100.0,
|
| 5 |
+
"global_step": 7900,
|
| 6 |
+
"is_hyper_param_search": false,
|
| 7 |
+
"is_local_process_zero": true,
|
| 8 |
+
"is_world_process_zero": true,
|
| 9 |
+
"log_history": [
|
| 10 |
+
{
|
| 11 |
+
"epoch": 1.27,
|
| 12 |
+
"learning_rate": 1.3719999999999999e-05,
|
| 13 |
+
"loss": 10.7972,
|
| 14 |
+
"step": 100
|
| 15 |
+
},
|
| 16 |
+
{
|
| 17 |
+
"epoch": 2.53,
|
| 18 |
+
"learning_rate": 2.772e-05,
|
| 19 |
+
"loss": 4.2661,
|
| 20 |
+
"step": 200
|
| 21 |
+
},
|
| 22 |
+
{
|
| 23 |
+
"epoch": 3.8,
|
| 24 |
+
"learning_rate": 4.1719999999999994e-05,
|
| 25 |
+
"loss": 3.3145,
|
| 26 |
+
"step": 300
|
| 27 |
+
},
|
| 28 |
+
{
|
| 29 |
+
"epoch": 5.06,
|
| 30 |
+
"learning_rate": 5.5719999999999995e-05,
|
| 31 |
+
"loss": 3.039,
|
| 32 |
+
"step": 400
|
| 33 |
+
},
|
| 34 |
+
{
|
| 35 |
+
"epoch": 6.33,
|
| 36 |
+
"learning_rate": 6.971999999999999e-05,
|
| 37 |
+
"loss": 2.9774,
|
| 38 |
+
"step": 500
|
| 39 |
+
},
|
| 40 |
+
{
|
| 41 |
+
"epoch": 6.33,
|
| 42 |
+
"eval_loss": 2.9768528938293457,
|
| 43 |
+
"eval_runtime": 42.649,
|
| 44 |
+
"eval_samples_per_second": 22.345,
|
| 45 |
+
"eval_steps_per_second": 0.703,
|
| 46 |
+
"eval_wer": 1.0,
|
| 47 |
+
"step": 500
|
| 48 |
+
},
|
| 49 |
+
{
|
| 50 |
+
"epoch": 7.59,
|
| 51 |
+
"learning_rate": 6.907297297297297e-05,
|
| 52 |
+
"loss": 2.9216,
|
| 53 |
+
"step": 600
|
| 54 |
+
},
|
| 55 |
+
{
|
| 56 |
+
"epoch": 8.86,
|
| 57 |
+
"learning_rate": 6.812702702702703e-05,
|
| 58 |
+
"loss": 2.453,
|
| 59 |
+
"step": 700
|
| 60 |
+
},
|
| 61 |
+
{
|
| 62 |
+
"epoch": 10.13,
|
| 63 |
+
"learning_rate": 6.718108108108108e-05,
|
| 64 |
+
"loss": 1.7073,
|
| 65 |
+
"step": 800
|
| 66 |
+
},
|
| 67 |
+
{
|
| 68 |
+
"epoch": 11.39,
|
| 69 |
+
"learning_rate": 6.623513513513513e-05,
|
| 70 |
+
"loss": 1.4457,
|
| 71 |
+
"step": 900
|
| 72 |
+
},
|
| 73 |
+
{
|
| 74 |
+
"epoch": 12.66,
|
| 75 |
+
"learning_rate": 6.528918918918918e-05,
|
| 76 |
+
"loss": 1.3453,
|
| 77 |
+
"step": 1000
|
| 78 |
+
},
|
| 79 |
+
{
|
| 80 |
+
"epoch": 12.66,
|
| 81 |
+
"eval_loss": 0.6522914171218872,
|
| 82 |
+
"eval_runtime": 42.9729,
|
| 83 |
+
"eval_samples_per_second": 22.177,
|
| 84 |
+
"eval_steps_per_second": 0.698,
|
| 85 |
+
"eval_wer": 0.6979675258317247,
|
| 86 |
+
"step": 1000
|
| 87 |
+
},
|
| 88 |
+
{
|
| 89 |
+
"epoch": 13.92,
|
| 90 |
+
"learning_rate": 6.434324324324325e-05,
|
| 91 |
+
"loss": 1.2867,
|
| 92 |
+
"step": 1100
|
| 93 |
+
},
|
| 94 |
+
{
|
| 95 |
+
"epoch": 15.19,
|
| 96 |
+
"learning_rate": 6.339729729729728e-05,
|
| 97 |
+
"loss": 1.2507,
|
| 98 |
+
"step": 1200
|
| 99 |
+
},
|
| 100 |
+
{
|
| 101 |
+
"epoch": 16.46,
|
| 102 |
+
"learning_rate": 6.245135135135135e-05,
|
| 103 |
+
"loss": 1.21,
|
| 104 |
+
"step": 1300
|
| 105 |
+
},
|
| 106 |
+
{
|
| 107 |
+
"epoch": 17.72,
|
| 108 |
+
"learning_rate": 6.15054054054054e-05,
|
| 109 |
+
"loss": 1.1863,
|
| 110 |
+
"step": 1400
|
| 111 |
+
},
|
| 112 |
+
{
|
| 113 |
+
"epoch": 18.99,
|
| 114 |
+
"learning_rate": 6.055945945945946e-05,
|
| 115 |
+
"loss": 1.1658,
|
| 116 |
+
"step": 1500
|
| 117 |
+
},
|
| 118 |
+
{
|
| 119 |
+
"epoch": 18.99,
|
| 120 |
+
"eval_loss": 0.5635793805122375,
|
| 121 |
+
"eval_runtime": 43.1337,
|
| 122 |
+
"eval_samples_per_second": 22.094,
|
| 123 |
+
"eval_steps_per_second": 0.696,
|
| 124 |
+
"eval_wer": 0.6358578403542636,
|
| 125 |
+
"step": 1500
|
| 126 |
+
},
|
| 127 |
+
{
|
| 128 |
+
"epoch": 20.25,
|
| 129 |
+
"learning_rate": 5.9613513513513504e-05,
|
| 130 |
+
"loss": 1.1476,
|
| 131 |
+
"step": 1600
|
| 132 |
+
},
|
| 133 |
+
{
|
| 134 |
+
"epoch": 21.52,
|
| 135 |
+
"learning_rate": 5.866756756756756e-05,
|
| 136 |
+
"loss": 1.1346,
|
| 137 |
+
"step": 1700
|
| 138 |
+
},
|
| 139 |
+
{
|
| 140 |
+
"epoch": 22.78,
|
| 141 |
+
"learning_rate": 5.772162162162162e-05,
|
| 142 |
+
"loss": 1.1074,
|
| 143 |
+
"step": 1800
|
| 144 |
+
},
|
| 145 |
+
{
|
| 146 |
+
"epoch": 24.05,
|
| 147 |
+
"learning_rate": 5.677567567567567e-05,
|
| 148 |
+
"loss": 1.0892,
|
| 149 |
+
"step": 1900
|
| 150 |
+
},
|
| 151 |
+
{
|
| 152 |
+
"epoch": 25.32,
|
| 153 |
+
"learning_rate": 5.5829729729729724e-05,
|
| 154 |
+
"loss": 1.0797,
|
| 155 |
+
"step": 2000
|
| 156 |
+
},
|
| 157 |
+
{
|
| 158 |
+
"epoch": 25.32,
|
| 159 |
+
"eval_loss": 0.5003558993339539,
|
| 160 |
+
"eval_runtime": 43.4309,
|
| 161 |
+
"eval_samples_per_second": 21.943,
|
| 162 |
+
"eval_steps_per_second": 0.691,
|
| 163 |
+
"eval_wer": 0.5759055296922903,
|
| 164 |
+
"step": 2000
|
| 165 |
+
},
|
| 166 |
+
{
|
| 167 |
+
"epoch": 26.58,
|
| 168 |
+
"learning_rate": 5.489324324324324e-05,
|
| 169 |
+
"loss": 1.0695,
|
| 170 |
+
"step": 2100
|
| 171 |
+
},
|
| 172 |
+
{
|
| 173 |
+
"epoch": 27.85,
|
| 174 |
+
"learning_rate": 5.394729729729729e-05,
|
| 175 |
+
"loss": 1.065,
|
| 176 |
+
"step": 2200
|
| 177 |
+
},
|
| 178 |
+
{
|
| 179 |
+
"epoch": 29.11,
|
| 180 |
+
"learning_rate": 5.300135135135134e-05,
|
| 181 |
+
"loss": 1.0534,
|
| 182 |
+
"step": 2300
|
| 183 |
+
},
|
| 184 |
+
{
|
| 185 |
+
"epoch": 30.38,
|
| 186 |
+
"learning_rate": 5.20554054054054e-05,
|
| 187 |
+
"loss": 1.0323,
|
| 188 |
+
"step": 2400
|
| 189 |
+
},
|
| 190 |
+
{
|
| 191 |
+
"epoch": 31.65,
|
| 192 |
+
"learning_rate": 5.110945945945946e-05,
|
| 193 |
+
"loss": 1.044,
|
| 194 |
+
"step": 2500
|
| 195 |
+
},
|
| 196 |
+
{
|
| 197 |
+
"epoch": 31.65,
|
| 198 |
+
"eval_loss": 0.49577832221984863,
|
| 199 |
+
"eval_runtime": 42.5512,
|
| 200 |
+
"eval_samples_per_second": 22.397,
|
| 201 |
+
"eval_steps_per_second": 0.705,
|
| 202 |
+
"eval_wer": 0.5569433405245827,
|
| 203 |
+
"step": 2500
|
| 204 |
+
},
|
| 205 |
+
{
|
| 206 |
+
"epoch": 32.91,
|
| 207 |
+
"learning_rate": 5.0163513513513505e-05,
|
| 208 |
+
"loss": 1.0187,
|
| 209 |
+
"step": 2600
|
| 210 |
+
},
|
| 211 |
+
{
|
| 212 |
+
"epoch": 34.18,
|
| 213 |
+
"learning_rate": 4.921756756756756e-05,
|
| 214 |
+
"loss": 1.0094,
|
| 215 |
+
"step": 2700
|
| 216 |
+
},
|
| 217 |
+
{
|
| 218 |
+
"epoch": 35.44,
|
| 219 |
+
"learning_rate": 4.8271621621621615e-05,
|
| 220 |
+
"loss": 1.0002,
|
| 221 |
+
"step": 2800
|
| 222 |
+
},
|
| 223 |
+
{
|
| 224 |
+
"epoch": 36.71,
|
| 225 |
+
"learning_rate": 4.732567567567567e-05,
|
| 226 |
+
"loss": 0.9978,
|
| 227 |
+
"step": 2900
|
| 228 |
+
},
|
| 229 |
+
{
|
| 230 |
+
"epoch": 37.97,
|
| 231 |
+
"learning_rate": 4.6379729729729725e-05,
|
| 232 |
+
"loss": 0.9915,
|
| 233 |
+
"step": 3000
|
| 234 |
+
},
|
| 235 |
+
{
|
| 236 |
+
"epoch": 37.97,
|
| 237 |
+
"eval_loss": 0.49712273478507996,
|
| 238 |
+
"eval_runtime": 41.1197,
|
| 239 |
+
"eval_samples_per_second": 23.176,
|
| 240 |
+
"eval_steps_per_second": 0.73,
|
| 241 |
+
"eval_wer": 0.5350289542409447,
|
| 242 |
+
"step": 3000
|
| 243 |
+
},
|
| 244 |
+
{
|
| 245 |
+
"epoch": 39.24,
|
| 246 |
+
"learning_rate": 4.5433783783783776e-05,
|
| 247 |
+
"loss": 0.9786,
|
| 248 |
+
"step": 3100
|
| 249 |
+
},
|
| 250 |
+
{
|
| 251 |
+
"epoch": 40.51,
|
| 252 |
+
"learning_rate": 4.4487837837837834e-05,
|
| 253 |
+
"loss": 0.9819,
|
| 254 |
+
"step": 3200
|
| 255 |
+
},
|
| 256 |
+
{
|
| 257 |
+
"epoch": 41.77,
|
| 258 |
+
"learning_rate": 4.354189189189189e-05,
|
| 259 |
+
"loss": 0.9735,
|
| 260 |
+
"step": 3300
|
| 261 |
+
},
|
| 262 |
+
{
|
| 263 |
+
"epoch": 43.04,
|
| 264 |
+
"learning_rate": 4.259594594594594e-05,
|
| 265 |
+
"loss": 0.9624,
|
| 266 |
+
"step": 3400
|
| 267 |
+
},
|
| 268 |
+
{
|
| 269 |
+
"epoch": 44.3,
|
| 270 |
+
"learning_rate": 4.1649999999999996e-05,
|
| 271 |
+
"loss": 0.9429,
|
| 272 |
+
"step": 3500
|
| 273 |
+
},
|
| 274 |
+
{
|
| 275 |
+
"epoch": 44.3,
|
| 276 |
+
"eval_loss": 0.48288407921791077,
|
| 277 |
+
"eval_runtime": 41.6865,
|
| 278 |
+
"eval_samples_per_second": 22.861,
|
| 279 |
+
"eval_steps_per_second": 0.72,
|
| 280 |
+
"eval_wer": 0.5228795276484615,
|
| 281 |
+
"step": 3500
|
| 282 |
+
},
|
| 283 |
+
{
|
| 284 |
+
"epoch": 45.57,
|
| 285 |
+
"learning_rate": 4.070405405405405e-05,
|
| 286 |
+
"loss": 0.9448,
|
| 287 |
+
"step": 3600
|
| 288 |
+
},
|
| 289 |
+
{
|
| 290 |
+
"epoch": 46.84,
|
| 291 |
+
"learning_rate": 3.9758108108108106e-05,
|
| 292 |
+
"loss": 0.9383,
|
| 293 |
+
"step": 3700
|
| 294 |
+
},
|
| 295 |
+
{
|
| 296 |
+
"epoch": 48.1,
|
| 297 |
+
"learning_rate": 3.8812162162162164e-05,
|
| 298 |
+
"loss": 0.939,
|
| 299 |
+
"step": 3800
|
| 300 |
+
},
|
| 301 |
+
{
|
| 302 |
+
"epoch": 49.37,
|
| 303 |
+
"learning_rate": 3.786621621621621e-05,
|
| 304 |
+
"loss": 0.9322,
|
| 305 |
+
"step": 3900
|
| 306 |
+
},
|
| 307 |
+
{
|
| 308 |
+
"epoch": 50.63,
|
| 309 |
+
"learning_rate": 3.692027027027027e-05,
|
| 310 |
+
"loss": 0.9266,
|
| 311 |
+
"step": 4000
|
| 312 |
+
},
|
| 313 |
+
{
|
| 314 |
+
"epoch": 50.63,
|
| 315 |
+
"eval_loss": 0.4515039622783661,
|
| 316 |
+
"eval_runtime": 41.6834,
|
| 317 |
+
"eval_samples_per_second": 22.863,
|
| 318 |
+
"eval_steps_per_second": 0.72,
|
| 319 |
+
"eval_wer": 0.5074372658112865,
|
| 320 |
+
"step": 4000
|
| 321 |
+
},
|
| 322 |
+
{
|
| 323 |
+
"epoch": 51.9,
|
| 324 |
+
"learning_rate": 3.5974324324324326e-05,
|
| 325 |
+
"loss": 0.9147,
|
| 326 |
+
"step": 4100
|
| 327 |
+
},
|
| 328 |
+
{
|
| 329 |
+
"epoch": 53.16,
|
| 330 |
+
"learning_rate": 3.502837837837838e-05,
|
| 331 |
+
"loss": 0.9129,
|
| 332 |
+
"step": 4200
|
| 333 |
+
},
|
| 334 |
+
{
|
| 335 |
+
"epoch": 54.43,
|
| 336 |
+
"learning_rate": 3.4091891891891893e-05,
|
| 337 |
+
"loss": 0.9071,
|
| 338 |
+
"step": 4300
|
| 339 |
+
},
|
| 340 |
+
{
|
| 341 |
+
"epoch": 55.7,
|
| 342 |
+
"learning_rate": 3.3145945945945945e-05,
|
| 343 |
+
"loss": 0.9098,
|
| 344 |
+
"step": 4400
|
| 345 |
+
},
|
| 346 |
+
{
|
| 347 |
+
"epoch": 56.96,
|
| 348 |
+
"learning_rate": 3.220945945945946e-05,
|
| 349 |
+
"loss": 0.8965,
|
| 350 |
+
"step": 4500
|
| 351 |
+
},
|
| 352 |
+
{
|
| 353 |
+
"epoch": 56.96,
|
| 354 |
+
"eval_loss": 0.45992422103881836,
|
| 355 |
+
"eval_runtime": 41.9829,
|
| 356 |
+
"eval_samples_per_second": 22.7,
|
| 357 |
+
"eval_steps_per_second": 0.715,
|
| 358 |
+
"eval_wer": 0.5039173384807539,
|
| 359 |
+
"step": 4500
|
| 360 |
+
},
|
| 361 |
+
{
|
| 362 |
+
"epoch": 58.23,
|
| 363 |
+
"learning_rate": 3.126351351351351e-05,
|
| 364 |
+
"loss": 0.8992,
|
| 365 |
+
"step": 4600
|
| 366 |
+
},
|
| 367 |
+
{
|
| 368 |
+
"epoch": 59.49,
|
| 369 |
+
"learning_rate": 3.0327027027027026e-05,
|
| 370 |
+
"loss": 0.9006,
|
| 371 |
+
"step": 4700
|
| 372 |
+
},
|
| 373 |
+
{
|
| 374 |
+
"epoch": 60.76,
|
| 375 |
+
"learning_rate": 2.9381081081081077e-05,
|
| 376 |
+
"loss": 0.8931,
|
| 377 |
+
"step": 4800
|
| 378 |
+
},
|
| 379 |
+
{
|
| 380 |
+
"epoch": 62.03,
|
| 381 |
+
"learning_rate": 2.8435135135135132e-05,
|
| 382 |
+
"loss": 0.8846,
|
| 383 |
+
"step": 4900
|
| 384 |
+
},
|
| 385 |
+
{
|
| 386 |
+
"epoch": 63.29,
|
| 387 |
+
"learning_rate": 2.7489189189189187e-05,
|
| 388 |
+
"loss": 0.878,
|
| 389 |
+
"step": 5000
|
| 390 |
+
},
|
| 391 |
+
{
|
| 392 |
+
"epoch": 63.29,
|
| 393 |
+
"eval_loss": 0.4735390543937683,
|
| 394 |
+
"eval_runtime": 41.3368,
|
| 395 |
+
"eval_samples_per_second": 23.054,
|
| 396 |
+
"eval_steps_per_second": 0.726,
|
| 397 |
+
"eval_wer": 0.4954013852617236,
|
| 398 |
+
"step": 5000
|
| 399 |
+
},
|
| 400 |
+
{
|
| 401 |
+
"epoch": 64.56,
|
| 402 |
+
"learning_rate": 2.6543243243243242e-05,
|
| 403 |
+
"loss": 0.8696,
|
| 404 |
+
"step": 5100
|
| 405 |
+
},
|
| 406 |
+
{
|
| 407 |
+
"epoch": 65.82,
|
| 408 |
+
"learning_rate": 2.5597297297297294e-05,
|
| 409 |
+
"loss": 0.8756,
|
| 410 |
+
"step": 5200
|
| 411 |
+
},
|
| 412 |
+
{
|
| 413 |
+
"epoch": 67.09,
|
| 414 |
+
"learning_rate": 2.465135135135135e-05,
|
| 415 |
+
"loss": 0.8588,
|
| 416 |
+
"step": 5300
|
| 417 |
+
},
|
| 418 |
+
{
|
| 419 |
+
"epoch": 68.35,
|
| 420 |
+
"learning_rate": 2.3705405405405404e-05,
|
| 421 |
+
"loss": 0.8616,
|
| 422 |
+
"step": 5400
|
| 423 |
+
},
|
| 424 |
+
{
|
| 425 |
+
"epoch": 69.62,
|
| 426 |
+
"learning_rate": 2.275945945945946e-05,
|
| 427 |
+
"loss": 0.8494,
|
| 428 |
+
"step": 5500
|
| 429 |
+
},
|
| 430 |
+
{
|
| 431 |
+
"epoch": 69.62,
|
| 432 |
+
"eval_loss": 0.44596442580223083,
|
| 433 |
+
"eval_runtime": 41.8864,
|
| 434 |
+
"eval_samples_per_second": 22.752,
|
| 435 |
+
"eval_steps_per_second": 0.716,
|
| 436 |
+
"eval_wer": 0.4877938003860565,
|
| 437 |
+
"step": 5500
|
| 438 |
+
},
|
| 439 |
+
{
|
| 440 |
+
"epoch": 70.89,
|
| 441 |
+
"learning_rate": 2.181351351351351e-05,
|
| 442 |
+
"loss": 0.8535,
|
| 443 |
+
"step": 5600
|
| 444 |
+
},
|
| 445 |
+
{
|
| 446 |
+
"epoch": 72.15,
|
| 447 |
+
"learning_rate": 2.0867567567567565e-05,
|
| 448 |
+
"loss": 0.8511,
|
| 449 |
+
"step": 5700
|
| 450 |
+
},
|
| 451 |
+
{
|
| 452 |
+
"epoch": 73.42,
|
| 453 |
+
"learning_rate": 1.992162162162162e-05,
|
| 454 |
+
"loss": 0.8436,
|
| 455 |
+
"step": 5800
|
| 456 |
+
},
|
| 457 |
+
{
|
| 458 |
+
"epoch": 74.68,
|
| 459 |
+
"learning_rate": 1.8975675675675675e-05,
|
| 460 |
+
"loss": 0.8527,
|
| 461 |
+
"step": 5900
|
| 462 |
+
},
|
| 463 |
+
{
|
| 464 |
+
"epoch": 75.95,
|
| 465 |
+
"learning_rate": 1.8029729729729727e-05,
|
| 466 |
+
"loss": 0.8343,
|
| 467 |
+
"step": 6000
|
| 468 |
+
},
|
| 469 |
+
{
|
| 470 |
+
"epoch": 75.95,
|
| 471 |
+
"eval_loss": 0.4510088264942169,
|
| 472 |
+
"eval_runtime": 41.5079,
|
| 473 |
+
"eval_samples_per_second": 22.959,
|
| 474 |
+
"eval_steps_per_second": 0.723,
|
| 475 |
+
"eval_wer": 0.479504939252867,
|
| 476 |
+
"step": 6000
|
| 477 |
+
},
|
| 478 |
+
{
|
| 479 |
+
"epoch": 77.22,
|
| 480 |
+
"learning_rate": 1.708378378378378e-05,
|
| 481 |
+
"loss": 0.8474,
|
| 482 |
+
"step": 6100
|
| 483 |
+
},
|
| 484 |
+
{
|
| 485 |
+
"epoch": 78.48,
|
| 486 |
+
"learning_rate": 1.6137837837837837e-05,
|
| 487 |
+
"loss": 0.83,
|
| 488 |
+
"step": 6200
|
| 489 |
+
},
|
| 490 |
+
{
|
| 491 |
+
"epoch": 79.75,
|
| 492 |
+
"learning_rate": 1.5191891891891891e-05,
|
| 493 |
+
"loss": 0.8168,
|
| 494 |
+
"step": 6300
|
| 495 |
+
},
|
| 496 |
+
{
|
| 497 |
+
"epoch": 81.01,
|
| 498 |
+
"learning_rate": 1.4245945945945945e-05,
|
| 499 |
+
"loss": 0.8363,
|
| 500 |
+
"step": 6400
|
| 501 |
+
},
|
| 502 |
+
{
|
| 503 |
+
"epoch": 82.28,
|
| 504 |
+
"learning_rate": 1.33e-05,
|
| 505 |
+
"loss": 0.8236,
|
| 506 |
+
"step": 6500
|
| 507 |
+
},
|
| 508 |
+
{
|
| 509 |
+
"epoch": 82.28,
|
| 510 |
+
"eval_loss": 0.4537811577320099,
|
| 511 |
+
"eval_runtime": 40.81,
|
| 512 |
+
"eval_samples_per_second": 23.352,
|
| 513 |
+
"eval_steps_per_second": 0.735,
|
| 514 |
+
"eval_wer": 0.47893720903826503,
|
| 515 |
+
"step": 6500
|
| 516 |
+
},
|
| 517 |
+
{
|
| 518 |
+
"epoch": 83.54,
|
| 519 |
+
"learning_rate": 1.2354054054054053e-05,
|
| 520 |
+
"loss": 0.8189,
|
| 521 |
+
"step": 6600
|
| 522 |
+
},
|
| 523 |
+
{
|
| 524 |
+
"epoch": 84.81,
|
| 525 |
+
"learning_rate": 1.1408108108108108e-05,
|
| 526 |
+
"loss": 0.8121,
|
| 527 |
+
"step": 6700
|
| 528 |
+
},
|
| 529 |
+
{
|
| 530 |
+
"epoch": 86.08,
|
| 531 |
+
"learning_rate": 1.0462162162162161e-05,
|
| 532 |
+
"loss": 0.8098,
|
| 533 |
+
"step": 6800
|
| 534 |
+
},
|
| 535 |
+
{
|
| 536 |
+
"epoch": 87.34,
|
| 537 |
+
"learning_rate": 9.516216216216216e-06,
|
| 538 |
+
"loss": 0.8091,
|
| 539 |
+
"step": 6900
|
| 540 |
+
},
|
| 541 |
+
{
|
| 542 |
+
"epoch": 88.61,
|
| 543 |
+
"learning_rate": 8.57027027027027e-06,
|
| 544 |
+
"loss": 0.8069,
|
| 545 |
+
"step": 7000
|
| 546 |
+
},
|
| 547 |
+
{
|
| 548 |
+
"epoch": 88.61,
|
| 549 |
+
"eval_loss": 0.45264118909835815,
|
| 550 |
+
"eval_runtime": 41.4122,
|
| 551 |
+
"eval_samples_per_second": 23.013,
|
| 552 |
+
"eval_steps_per_second": 0.724,
|
| 553 |
+
"eval_wer": 0.47484955149313046,
|
| 554 |
+
"step": 7000
|
| 555 |
+
},
|
| 556 |
+
{
|
| 557 |
+
"epoch": 89.87,
|
| 558 |
+
"learning_rate": 7.6243243243243236e-06,
|
| 559 |
+
"loss": 0.8042,
|
| 560 |
+
"step": 7100
|
| 561 |
+
},
|
| 562 |
+
{
|
| 563 |
+
"epoch": 91.14,
|
| 564 |
+
"learning_rate": 6.678378378378378e-06,
|
| 565 |
+
"loss": 0.8042,
|
| 566 |
+
"step": 7200
|
| 567 |
+
},
|
| 568 |
+
{
|
| 569 |
+
"epoch": 92.41,
|
| 570 |
+
"learning_rate": 5.732432432432432e-06,
|
| 571 |
+
"loss": 0.7968,
|
| 572 |
+
"step": 7300
|
| 573 |
+
},
|
| 574 |
+
{
|
| 575 |
+
"epoch": 93.67,
|
| 576 |
+
"learning_rate": 4.786486486486486e-06,
|
| 577 |
+
"loss": 0.798,
|
| 578 |
+
"step": 7400
|
| 579 |
+
},
|
| 580 |
+
{
|
| 581 |
+
"epoch": 94.94,
|
| 582 |
+
"learning_rate": 3.84054054054054e-06,
|
| 583 |
+
"loss": 0.7958,
|
| 584 |
+
"step": 7500
|
| 585 |
+
},
|
| 586 |
+
{
|
| 587 |
+
"epoch": 94.94,
|
| 588 |
+
"eval_loss": 0.449596643447876,
|
| 589 |
+
"eval_runtime": 41.0608,
|
| 590 |
+
"eval_samples_per_second": 23.209,
|
| 591 |
+
"eval_steps_per_second": 0.731,
|
| 592 |
+
"eval_wer": 0.4699670716475531,
|
| 593 |
+
"step": 7500
|
| 594 |
+
},
|
| 595 |
+
{
|
| 596 |
+
"epoch": 96.2,
|
| 597 |
+
"learning_rate": 2.8945945945945945e-06,
|
| 598 |
+
"loss": 0.7953,
|
| 599 |
+
"step": 7600
|
| 600 |
+
},
|
| 601 |
+
{
|
| 602 |
+
"epoch": 97.47,
|
| 603 |
+
"learning_rate": 1.9486486486486487e-06,
|
| 604 |
+
"loss": 0.7917,
|
| 605 |
+
"step": 7700
|
| 606 |
+
},
|
| 607 |
+
{
|
| 608 |
+
"epoch": 98.73,
|
| 609 |
+
"learning_rate": 1.0027027027027026e-06,
|
| 610 |
+
"loss": 0.7881,
|
| 611 |
+
"step": 7800
|
| 612 |
+
},
|
| 613 |
+
{
|
| 614 |
+
"epoch": 100.0,
|
| 615 |
+
"learning_rate": 5.675675675675675e-08,
|
| 616 |
+
"loss": 0.795,
|
| 617 |
+
"step": 7900
|
| 618 |
+
},
|
| 619 |
+
{
|
| 620 |
+
"epoch": 100.0,
|
| 621 |
+
"step": 7900,
|
| 622 |
+
"total_flos": 3.986125981994791e+19,
|
| 623 |
+
"train_loss": 1.2487838832034341,
|
| 624 |
+
"train_runtime": 15247.9542,
|
| 625 |
+
"train_samples_per_second": 16.383,
|
| 626 |
+
"train_steps_per_second": 0.518
|
| 627 |
+
}
|
| 628 |
+
],
|
| 629 |
+
"max_steps": 7900,
|
| 630 |
+
"num_train_epochs": 100,
|
| 631 |
+
"total_flos": 3.986125981994791e+19,
|
| 632 |
+
"trial_name": null,
|
| 633 |
+
"trial_params": null
|
| 634 |
+
}
|