Commit
•
236f82b
1
Parent(s):
423d9bc
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": 70.0,
|
| 3 |
+
"eval_loss": 0.23074762523174286,
|
| 4 |
+
"eval_runtime": 48.9134,
|
| 5 |
+
"eval_samples": 1599,
|
| 6 |
+
"eval_samples_per_second": 32.69,
|
| 7 |
+
"eval_steps_per_second": 1.022,
|
| 8 |
+
"eval_wer": 0.2983537016462984,
|
| 9 |
+
"train_loss": 1.2354308326740313,
|
| 10 |
+
"train_runtime": 11524.5807,
|
| 11 |
+
"train_samples": 3627,
|
| 12 |
+
"train_samples_per_second": 22.03,
|
| 13 |
+
"train_steps_per_second": 0.692
|
| 14 |
+
}
|
eval_results.json
ADDED
|
@@ -0,0 +1,9 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"epoch": 70.0,
|
| 3 |
+
"eval_loss": 0.23074762523174286,
|
| 4 |
+
"eval_runtime": 48.9134,
|
| 5 |
+
"eval_samples": 1599,
|
| 6 |
+
"eval_samples_per_second": 32.69,
|
| 7 |
+
"eval_steps_per_second": 1.022,
|
| 8 |
+
"eval_wer": 0.2983537016462984
|
| 9 |
+
}
|
pytorch_model.bin
CHANGED
|
@@ -1,3 +1,3 @@
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
-
oid sha256:
|
| 3 |
size 1262067185
|
|
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:1a1e4face17e6f97b42c070d4fa992971a18c5788603ecd7e1ee2b1dd92d0aa5
|
| 3 |
size 1262067185
|
train_results.json
ADDED
|
@@ -0,0 +1,8 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"epoch": 70.0,
|
| 3 |
+
"train_loss": 1.2354308326740313,
|
| 4 |
+
"train_runtime": 11524.5807,
|
| 5 |
+
"train_samples": 3627,
|
| 6 |
+
"train_samples_per_second": 22.03,
|
| 7 |
+
"train_steps_per_second": 0.692
|
| 8 |
+
}
|
trainer_state.json
ADDED
|
@@ -0,0 +1,634 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"best_metric": null,
|
| 3 |
+
"best_model_checkpoint": null,
|
| 4 |
+
"epoch": 70.0,
|
| 5 |
+
"global_step": 7980,
|
| 6 |
+
"is_hyper_param_search": false,
|
| 7 |
+
"is_local_process_zero": true,
|
| 8 |
+
"is_world_process_zero": true,
|
| 9 |
+
"log_history": [
|
| 10 |
+
{
|
| 11 |
+
"epoch": 0.88,
|
| 12 |
+
"learning_rate": 1.3719999999999999e-05,
|
| 13 |
+
"loss": 8.2074,
|
| 14 |
+
"step": 100
|
| 15 |
+
},
|
| 16 |
+
{
|
| 17 |
+
"epoch": 1.75,
|
| 18 |
+
"learning_rate": 2.772e-05,
|
| 19 |
+
"loss": 3.7205,
|
| 20 |
+
"step": 200
|
| 21 |
+
},
|
| 22 |
+
{
|
| 23 |
+
"epoch": 2.63,
|
| 24 |
+
"learning_rate": 4.1719999999999994e-05,
|
| 25 |
+
"loss": 3.1583,
|
| 26 |
+
"step": 300
|
| 27 |
+
},
|
| 28 |
+
{
|
| 29 |
+
"epoch": 3.51,
|
| 30 |
+
"learning_rate": 5.5719999999999995e-05,
|
| 31 |
+
"loss": 2.9766,
|
| 32 |
+
"step": 400
|
| 33 |
+
},
|
| 34 |
+
{
|
| 35 |
+
"epoch": 4.39,
|
| 36 |
+
"learning_rate": 6.971999999999999e-05,
|
| 37 |
+
"loss": 2.9032,
|
| 38 |
+
"step": 500
|
| 39 |
+
},
|
| 40 |
+
{
|
| 41 |
+
"epoch": 4.39,
|
| 42 |
+
"eval_loss": 2.8768463134765625,
|
| 43 |
+
"eval_runtime": 54.8752,
|
| 44 |
+
"eval_samples_per_second": 29.139,
|
| 45 |
+
"eval_steps_per_second": 0.911,
|
| 46 |
+
"eval_wer": 1.0,
|
| 47 |
+
"step": 500
|
| 48 |
+
},
|
| 49 |
+
{
|
| 50 |
+
"epoch": 5.26,
|
| 51 |
+
"learning_rate": 6.908288770053476e-05,
|
| 52 |
+
"loss": 2.8346,
|
| 53 |
+
"step": 600
|
| 54 |
+
},
|
| 55 |
+
{
|
| 56 |
+
"epoch": 6.14,
|
| 57 |
+
"learning_rate": 6.81470588235294e-05,
|
| 58 |
+
"loss": 2.7766,
|
| 59 |
+
"step": 700
|
| 60 |
+
},
|
| 61 |
+
{
|
| 62 |
+
"epoch": 7.02,
|
| 63 |
+
"learning_rate": 6.721122994652407e-05,
|
| 64 |
+
"loss": 2.3776,
|
| 65 |
+
"step": 800
|
| 66 |
+
},
|
| 67 |
+
{
|
| 68 |
+
"epoch": 7.89,
|
| 69 |
+
"learning_rate": 6.627540106951871e-05,
|
| 70 |
+
"loss": 1.8024,
|
| 71 |
+
"step": 900
|
| 72 |
+
},
|
| 73 |
+
{
|
| 74 |
+
"epoch": 8.77,
|
| 75 |
+
"learning_rate": 6.533957219251336e-05,
|
| 76 |
+
"loss": 1.5724,
|
| 77 |
+
"step": 1000
|
| 78 |
+
},
|
| 79 |
+
{
|
| 80 |
+
"epoch": 8.77,
|
| 81 |
+
"eval_loss": 0.5638473033905029,
|
| 82 |
+
"eval_runtime": 54.3288,
|
| 83 |
+
"eval_samples_per_second": 29.432,
|
| 84 |
+
"eval_steps_per_second": 0.92,
|
| 85 |
+
"eval_wer": 0.6437733562266438,
|
| 86 |
+
"step": 1000
|
| 87 |
+
},
|
| 88 |
+
{
|
| 89 |
+
"epoch": 9.65,
|
| 90 |
+
"learning_rate": 6.440374331550802e-05,
|
| 91 |
+
"loss": 1.4498,
|
| 92 |
+
"step": 1100
|
| 93 |
+
},
|
| 94 |
+
{
|
| 95 |
+
"epoch": 10.53,
|
| 96 |
+
"learning_rate": 6.346791443850267e-05,
|
| 97 |
+
"loss": 1.3426,
|
| 98 |
+
"step": 1200
|
| 99 |
+
},
|
| 100 |
+
{
|
| 101 |
+
"epoch": 11.4,
|
| 102 |
+
"learning_rate": 6.253208556149732e-05,
|
| 103 |
+
"loss": 1.2806,
|
| 104 |
+
"step": 1300
|
| 105 |
+
},
|
| 106 |
+
{
|
| 107 |
+
"epoch": 12.28,
|
| 108 |
+
"learning_rate": 6.159625668449198e-05,
|
| 109 |
+
"loss": 1.2595,
|
| 110 |
+
"step": 1400
|
| 111 |
+
},
|
| 112 |
+
{
|
| 113 |
+
"epoch": 13.16,
|
| 114 |
+
"learning_rate": 6.0660427807486626e-05,
|
| 115 |
+
"loss": 1.1818,
|
| 116 |
+
"step": 1500
|
| 117 |
+
},
|
| 118 |
+
{
|
| 119 |
+
"epoch": 13.16,
|
| 120 |
+
"eval_loss": 0.3338354229927063,
|
| 121 |
+
"eval_runtime": 54.9878,
|
| 122 |
+
"eval_samples_per_second": 29.079,
|
| 123 |
+
"eval_steps_per_second": 0.909,
|
| 124 |
+
"eval_wer": 0.4759115240884759,
|
| 125 |
+
"step": 1500
|
| 126 |
+
},
|
| 127 |
+
{
|
| 128 |
+
"epoch": 14.04,
|
| 129 |
+
"learning_rate": 5.972459893048127e-05,
|
| 130 |
+
"loss": 1.1818,
|
| 131 |
+
"step": 1600
|
| 132 |
+
},
|
| 133 |
+
{
|
| 134 |
+
"epoch": 14.91,
|
| 135 |
+
"learning_rate": 5.8788770053475934e-05,
|
| 136 |
+
"loss": 1.133,
|
| 137 |
+
"step": 1700
|
| 138 |
+
},
|
| 139 |
+
{
|
| 140 |
+
"epoch": 15.79,
|
| 141 |
+
"learning_rate": 5.785294117647058e-05,
|
| 142 |
+
"loss": 1.1239,
|
| 143 |
+
"step": 1800
|
| 144 |
+
},
|
| 145 |
+
{
|
| 146 |
+
"epoch": 16.67,
|
| 147 |
+
"learning_rate": 5.6917112299465236e-05,
|
| 148 |
+
"loss": 1.0991,
|
| 149 |
+
"step": 1900
|
| 150 |
+
},
|
| 151 |
+
{
|
| 152 |
+
"epoch": 17.54,
|
| 153 |
+
"learning_rate": 5.5981283422459884e-05,
|
| 154 |
+
"loss": 1.0798,
|
| 155 |
+
"step": 2000
|
| 156 |
+
},
|
| 157 |
+
{
|
| 158 |
+
"epoch": 17.54,
|
| 159 |
+
"eval_loss": 0.2876473665237427,
|
| 160 |
+
"eval_runtime": 49.892,
|
| 161 |
+
"eval_samples_per_second": 32.049,
|
| 162 |
+
"eval_steps_per_second": 1.002,
|
| 163 |
+
"eval_wer": 0.40864559135440864,
|
| 164 |
+
"step": 2000
|
| 165 |
+
},
|
| 166 |
+
{
|
| 167 |
+
"epoch": 18.42,
|
| 168 |
+
"learning_rate": 5.5045454545454545e-05,
|
| 169 |
+
"loss": 1.0756,
|
| 170 |
+
"step": 2100
|
| 171 |
+
},
|
| 172 |
+
{
|
| 173 |
+
"epoch": 19.3,
|
| 174 |
+
"learning_rate": 5.410962566844919e-05,
|
| 175 |
+
"loss": 1.0653,
|
| 176 |
+
"step": 2200
|
| 177 |
+
},
|
| 178 |
+
{
|
| 179 |
+
"epoch": 20.18,
|
| 180 |
+
"learning_rate": 5.317379679144385e-05,
|
| 181 |
+
"loss": 1.0472,
|
| 182 |
+
"step": 2300
|
| 183 |
+
},
|
| 184 |
+
{
|
| 185 |
+
"epoch": 21.05,
|
| 186 |
+
"learning_rate": 5.2237967914438494e-05,
|
| 187 |
+
"loss": 1.03,
|
| 188 |
+
"step": 2400
|
| 189 |
+
},
|
| 190 |
+
{
|
| 191 |
+
"epoch": 21.93,
|
| 192 |
+
"learning_rate": 5.1302139037433155e-05,
|
| 193 |
+
"loss": 1.0296,
|
| 194 |
+
"step": 2500
|
| 195 |
+
},
|
| 196 |
+
{
|
| 197 |
+
"epoch": 21.93,
|
| 198 |
+
"eval_loss": 0.2693549394607544,
|
| 199 |
+
"eval_runtime": 49.7815,
|
| 200 |
+
"eval_samples_per_second": 32.12,
|
| 201 |
+
"eval_steps_per_second": 1.004,
|
| 202 |
+
"eval_wer": 0.4248055751944248,
|
| 203 |
+
"step": 2500
|
| 204 |
+
},
|
| 205 |
+
{
|
| 206 |
+
"epoch": 22.81,
|
| 207 |
+
"learning_rate": 5.03663101604278e-05,
|
| 208 |
+
"loss": 1.0121,
|
| 209 |
+
"step": 2600
|
| 210 |
+
},
|
| 211 |
+
{
|
| 212 |
+
"epoch": 23.68,
|
| 213 |
+
"learning_rate": 4.943048128342246e-05,
|
| 214 |
+
"loss": 1.0139,
|
| 215 |
+
"step": 2700
|
| 216 |
+
},
|
| 217 |
+
{
|
| 218 |
+
"epoch": 24.56,
|
| 219 |
+
"learning_rate": 4.8494652406417105e-05,
|
| 220 |
+
"loss": 1.0081,
|
| 221 |
+
"step": 2800
|
| 222 |
+
},
|
| 223 |
+
{
|
| 224 |
+
"epoch": 25.44,
|
| 225 |
+
"learning_rate": 4.755882352941176e-05,
|
| 226 |
+
"loss": 0.994,
|
| 227 |
+
"step": 2900
|
| 228 |
+
},
|
| 229 |
+
{
|
| 230 |
+
"epoch": 26.32,
|
| 231 |
+
"learning_rate": 4.6622994652406414e-05,
|
| 232 |
+
"loss": 1.0014,
|
| 233 |
+
"step": 3000
|
| 234 |
+
},
|
| 235 |
+
{
|
| 236 |
+
"epoch": 26.32,
|
| 237 |
+
"eval_loss": 0.26258525252342224,
|
| 238 |
+
"eval_runtime": 49.6202,
|
| 239 |
+
"eval_samples_per_second": 32.225,
|
| 240 |
+
"eval_steps_per_second": 1.008,
|
| 241 |
+
"eval_wer": 0.3732956267043733,
|
| 242 |
+
"step": 3000
|
| 243 |
+
},
|
| 244 |
+
{
|
| 245 |
+
"epoch": 27.19,
|
| 246 |
+
"learning_rate": 4.569652406417112e-05,
|
| 247 |
+
"loss": 0.991,
|
| 248 |
+
"step": 3100
|
| 249 |
+
},
|
| 250 |
+
{
|
| 251 |
+
"epoch": 28.07,
|
| 252 |
+
"learning_rate": 4.4760695187165766e-05,
|
| 253 |
+
"loss": 0.9786,
|
| 254 |
+
"step": 3200
|
| 255 |
+
},
|
| 256 |
+
{
|
| 257 |
+
"epoch": 28.95,
|
| 258 |
+
"learning_rate": 4.382486631016043e-05,
|
| 259 |
+
"loss": 0.9689,
|
| 260 |
+
"step": 3300
|
| 261 |
+
},
|
| 262 |
+
{
|
| 263 |
+
"epoch": 29.82,
|
| 264 |
+
"learning_rate": 4.2889037433155075e-05,
|
| 265 |
+
"loss": 0.9631,
|
| 266 |
+
"step": 3400
|
| 267 |
+
},
|
| 268 |
+
{
|
| 269 |
+
"epoch": 30.7,
|
| 270 |
+
"learning_rate": 4.195320855614973e-05,
|
| 271 |
+
"loss": 0.9616,
|
| 272 |
+
"step": 3500
|
| 273 |
+
},
|
| 274 |
+
{
|
| 275 |
+
"epoch": 30.7,
|
| 276 |
+
"eval_loss": 0.23906062543392181,
|
| 277 |
+
"eval_runtime": 49.6357,
|
| 278 |
+
"eval_samples_per_second": 32.215,
|
| 279 |
+
"eval_steps_per_second": 1.007,
|
| 280 |
+
"eval_wer": 0.32936067063932933,
|
| 281 |
+
"step": 3500
|
| 282 |
+
},
|
| 283 |
+
{
|
| 284 |
+
"epoch": 31.58,
|
| 285 |
+
"learning_rate": 4.101737967914438e-05,
|
| 286 |
+
"loss": 0.9375,
|
| 287 |
+
"step": 3600
|
| 288 |
+
},
|
| 289 |
+
{
|
| 290 |
+
"epoch": 32.46,
|
| 291 |
+
"learning_rate": 4.008155080213904e-05,
|
| 292 |
+
"loss": 0.9629,
|
| 293 |
+
"step": 3700
|
| 294 |
+
},
|
| 295 |
+
{
|
| 296 |
+
"epoch": 33.33,
|
| 297 |
+
"learning_rate": 3.9145721925133686e-05,
|
| 298 |
+
"loss": 0.946,
|
| 299 |
+
"step": 3800
|
| 300 |
+
},
|
| 301 |
+
{
|
| 302 |
+
"epoch": 34.21,
|
| 303 |
+
"learning_rate": 3.820989304812834e-05,
|
| 304 |
+
"loss": 0.9511,
|
| 305 |
+
"step": 3900
|
| 306 |
+
},
|
| 307 |
+
{
|
| 308 |
+
"epoch": 35.09,
|
| 309 |
+
"learning_rate": 3.727406417112299e-05,
|
| 310 |
+
"loss": 0.9303,
|
| 311 |
+
"step": 4000
|
| 312 |
+
},
|
| 313 |
+
{
|
| 314 |
+
"epoch": 35.09,
|
| 315 |
+
"eval_loss": 0.23517830669879913,
|
| 316 |
+
"eval_runtime": 50.0462,
|
| 317 |
+
"eval_samples_per_second": 31.95,
|
| 318 |
+
"eval_steps_per_second": 0.999,
|
| 319 |
+
"eval_wer": 0.3217856782143218,
|
| 320 |
+
"step": 4000
|
| 321 |
+
},
|
| 322 |
+
{
|
| 323 |
+
"epoch": 35.96,
|
| 324 |
+
"learning_rate": 3.633823529411765e-05,
|
| 325 |
+
"loss": 0.9236,
|
| 326 |
+
"step": 4100
|
| 327 |
+
},
|
| 328 |
+
{
|
| 329 |
+
"epoch": 36.84,
|
| 330 |
+
"learning_rate": 3.5402406417112296e-05,
|
| 331 |
+
"loss": 0.9278,
|
| 332 |
+
"step": 4200
|
| 333 |
+
},
|
| 334 |
+
{
|
| 335 |
+
"epoch": 37.72,
|
| 336 |
+
"learning_rate": 3.446657754010695e-05,
|
| 337 |
+
"loss": 0.9218,
|
| 338 |
+
"step": 4300
|
| 339 |
+
},
|
| 340 |
+
{
|
| 341 |
+
"epoch": 38.6,
|
| 342 |
+
"learning_rate": 3.35307486631016e-05,
|
| 343 |
+
"loss": 0.922,
|
| 344 |
+
"step": 4400
|
| 345 |
+
},
|
| 346 |
+
{
|
| 347 |
+
"epoch": 39.47,
|
| 348 |
+
"learning_rate": 3.259491978609625e-05,
|
| 349 |
+
"loss": 0.9248,
|
| 350 |
+
"step": 4500
|
| 351 |
+
},
|
| 352 |
+
{
|
| 353 |
+
"epoch": 39.47,
|
| 354 |
+
"eval_loss": 0.23506982624530792,
|
| 355 |
+
"eval_runtime": 49.0987,
|
| 356 |
+
"eval_samples_per_second": 32.567,
|
| 357 |
+
"eval_steps_per_second": 1.018,
|
| 358 |
+
"eval_wer": 0.3206746793253207,
|
| 359 |
+
"step": 4500
|
| 360 |
+
},
|
| 361 |
+
{
|
| 362 |
+
"epoch": 40.35,
|
| 363 |
+
"learning_rate": 3.165909090909091e-05,
|
| 364 |
+
"loss": 0.8909,
|
| 365 |
+
"step": 4600
|
| 366 |
+
},
|
| 367 |
+
{
|
| 368 |
+
"epoch": 41.23,
|
| 369 |
+
"learning_rate": 3.072326203208556e-05,
|
| 370 |
+
"loss": 0.912,
|
| 371 |
+
"step": 4700
|
| 372 |
+
},
|
| 373 |
+
{
|
| 374 |
+
"epoch": 42.11,
|
| 375 |
+
"learning_rate": 2.9787433155080212e-05,
|
| 376 |
+
"loss": 0.8901,
|
| 377 |
+
"step": 4800
|
| 378 |
+
},
|
| 379 |
+
{
|
| 380 |
+
"epoch": 42.98,
|
| 381 |
+
"learning_rate": 2.8851604278074863e-05,
|
| 382 |
+
"loss": 0.9007,
|
| 383 |
+
"step": 4900
|
| 384 |
+
},
|
| 385 |
+
{
|
| 386 |
+
"epoch": 43.86,
|
| 387 |
+
"learning_rate": 2.7915775401069517e-05,
|
| 388 |
+
"loss": 0.8837,
|
| 389 |
+
"step": 5000
|
| 390 |
+
},
|
| 391 |
+
{
|
| 392 |
+
"epoch": 43.86,
|
| 393 |
+
"eval_loss": 0.23407745361328125,
|
| 394 |
+
"eval_runtime": 49.1749,
|
| 395 |
+
"eval_samples_per_second": 32.517,
|
| 396 |
+
"eval_steps_per_second": 1.017,
|
| 397 |
+
"eval_wer": 0.31027168972831026,
|
| 398 |
+
"step": 5000
|
| 399 |
+
},
|
| 400 |
+
{
|
| 401 |
+
"epoch": 44.74,
|
| 402 |
+
"learning_rate": 2.6979946524064168e-05,
|
| 403 |
+
"loss": 0.8891,
|
| 404 |
+
"step": 5100
|
| 405 |
+
},
|
| 406 |
+
{
|
| 407 |
+
"epoch": 45.61,
|
| 408 |
+
"learning_rate": 2.6044117647058823e-05,
|
| 409 |
+
"loss": 0.882,
|
| 410 |
+
"step": 5200
|
| 411 |
+
},
|
| 412 |
+
{
|
| 413 |
+
"epoch": 46.49,
|
| 414 |
+
"learning_rate": 2.5108288770053474e-05,
|
| 415 |
+
"loss": 0.8921,
|
| 416 |
+
"step": 5300
|
| 417 |
+
},
|
| 418 |
+
{
|
| 419 |
+
"epoch": 47.37,
|
| 420 |
+
"learning_rate": 2.4172459893048128e-05,
|
| 421 |
+
"loss": 0.8794,
|
| 422 |
+
"step": 5400
|
| 423 |
+
},
|
| 424 |
+
{
|
| 425 |
+
"epoch": 48.25,
|
| 426 |
+
"learning_rate": 2.323663101604278e-05,
|
| 427 |
+
"loss": 0.8887,
|
| 428 |
+
"step": 5500
|
| 429 |
+
},
|
| 430 |
+
{
|
| 431 |
+
"epoch": 48.25,
|
| 432 |
+
"eval_loss": 0.23113244771957397,
|
| 433 |
+
"eval_runtime": 48.9674,
|
| 434 |
+
"eval_samples_per_second": 32.654,
|
| 435 |
+
"eval_steps_per_second": 1.021,
|
| 436 |
+
"eval_wer": 0.3114836885163115,
|
| 437 |
+
"step": 5500
|
| 438 |
+
},
|
| 439 |
+
{
|
| 440 |
+
"epoch": 49.12,
|
| 441 |
+
"learning_rate": 2.2300802139037433e-05,
|
| 442 |
+
"loss": 0.8553,
|
| 443 |
+
"step": 5600
|
| 444 |
+
},
|
| 445 |
+
{
|
| 446 |
+
"epoch": 50.0,
|
| 447 |
+
"learning_rate": 2.1364973262032084e-05,
|
| 448 |
+
"loss": 0.8751,
|
| 449 |
+
"step": 5700
|
| 450 |
+
},
|
| 451 |
+
{
|
| 452 |
+
"epoch": 50.88,
|
| 453 |
+
"learning_rate": 2.042914438502674e-05,
|
| 454 |
+
"loss": 0.8557,
|
| 455 |
+
"step": 5800
|
| 456 |
+
},
|
| 457 |
+
{
|
| 458 |
+
"epoch": 51.75,
|
| 459 |
+
"learning_rate": 1.949331550802139e-05,
|
| 460 |
+
"loss": 0.8626,
|
| 461 |
+
"step": 5900
|
| 462 |
+
},
|
| 463 |
+
{
|
| 464 |
+
"epoch": 52.63,
|
| 465 |
+
"learning_rate": 1.8557486631016044e-05,
|
| 466 |
+
"loss": 0.8529,
|
| 467 |
+
"step": 6000
|
| 468 |
+
},
|
| 469 |
+
{
|
| 470 |
+
"epoch": 52.63,
|
| 471 |
+
"eval_loss": 0.22297005355358124,
|
| 472 |
+
"eval_runtime": 49.2944,
|
| 473 |
+
"eval_samples_per_second": 32.438,
|
| 474 |
+
"eval_steps_per_second": 1.014,
|
| 475 |
+
"eval_wer": 0.3000706999293001,
|
| 476 |
+
"step": 6000
|
| 477 |
+
},
|
| 478 |
+
{
|
| 479 |
+
"epoch": 53.51,
|
| 480 |
+
"learning_rate": 1.762165775401069e-05,
|
| 481 |
+
"loss": 0.8487,
|
| 482 |
+
"step": 6100
|
| 483 |
+
},
|
| 484 |
+
{
|
| 485 |
+
"epoch": 54.39,
|
| 486 |
+
"learning_rate": 1.6685828877005346e-05,
|
| 487 |
+
"loss": 0.8585,
|
| 488 |
+
"step": 6200
|
| 489 |
+
},
|
| 490 |
+
{
|
| 491 |
+
"epoch": 55.26,
|
| 492 |
+
"learning_rate": 1.575e-05,
|
| 493 |
+
"loss": 0.8412,
|
| 494 |
+
"step": 6300
|
| 495 |
+
},
|
| 496 |
+
{
|
| 497 |
+
"epoch": 56.14,
|
| 498 |
+
"learning_rate": 1.4814171122994651e-05,
|
| 499 |
+
"loss": 0.8457,
|
| 500 |
+
"step": 6400
|
| 501 |
+
},
|
| 502 |
+
{
|
| 503 |
+
"epoch": 57.02,
|
| 504 |
+
"learning_rate": 1.3878342245989304e-05,
|
| 505 |
+
"loss": 0.8404,
|
| 506 |
+
"step": 6500
|
| 507 |
+
},
|
| 508 |
+
{
|
| 509 |
+
"epoch": 57.02,
|
| 510 |
+
"eval_loss": 0.22788779437541962,
|
| 511 |
+
"eval_runtime": 48.6265,
|
| 512 |
+
"eval_samples_per_second": 32.883,
|
| 513 |
+
"eval_steps_per_second": 1.028,
|
| 514 |
+
"eval_wer": 0.3054236945763054,
|
| 515 |
+
"step": 6500
|
| 516 |
+
},
|
| 517 |
+
{
|
| 518 |
+
"epoch": 57.89,
|
| 519 |
+
"learning_rate": 1.2942513368983956e-05,
|
| 520 |
+
"loss": 0.8342,
|
| 521 |
+
"step": 6600
|
| 522 |
+
},
|
| 523 |
+
{
|
| 524 |
+
"epoch": 58.77,
|
| 525 |
+
"learning_rate": 1.2006684491978609e-05,
|
| 526 |
+
"loss": 0.844,
|
| 527 |
+
"step": 6700
|
| 528 |
+
},
|
| 529 |
+
{
|
| 530 |
+
"epoch": 59.65,
|
| 531 |
+
"learning_rate": 1.107085561497326e-05,
|
| 532 |
+
"loss": 0.8276,
|
| 533 |
+
"step": 6800
|
| 534 |
+
},
|
| 535 |
+
{
|
| 536 |
+
"epoch": 60.53,
|
| 537 |
+
"learning_rate": 1.0144385026737967e-05,
|
| 538 |
+
"loss": 0.8264,
|
| 539 |
+
"step": 6900
|
| 540 |
+
},
|
| 541 |
+
{
|
| 542 |
+
"epoch": 61.4,
|
| 543 |
+
"learning_rate": 9.20855614973262e-06,
|
| 544 |
+
"loss": 0.8242,
|
| 545 |
+
"step": 7000
|
| 546 |
+
},
|
| 547 |
+
{
|
| 548 |
+
"epoch": 61.4,
|
| 549 |
+
"eval_loss": 0.2298082411289215,
|
| 550 |
+
"eval_runtime": 48.4266,
|
| 551 |
+
"eval_samples_per_second": 33.019,
|
| 552 |
+
"eval_steps_per_second": 1.032,
|
| 553 |
+
"eval_wer": 0.30057569942430057,
|
| 554 |
+
"step": 7000
|
| 555 |
+
},
|
| 556 |
+
{
|
| 557 |
+
"epoch": 62.28,
|
| 558 |
+
"learning_rate": 8.272727272727272e-06,
|
| 559 |
+
"loss": 0.8291,
|
| 560 |
+
"step": 7100
|
| 561 |
+
},
|
| 562 |
+
{
|
| 563 |
+
"epoch": 63.16,
|
| 564 |
+
"learning_rate": 7.336898395721925e-06,
|
| 565 |
+
"loss": 0.8213,
|
| 566 |
+
"step": 7200
|
| 567 |
+
},
|
| 568 |
+
{
|
| 569 |
+
"epoch": 64.04,
|
| 570 |
+
"learning_rate": 6.401069518716577e-06,
|
| 571 |
+
"loss": 0.8139,
|
| 572 |
+
"step": 7300
|
| 573 |
+
},
|
| 574 |
+
{
|
| 575 |
+
"epoch": 64.91,
|
| 576 |
+
"learning_rate": 5.46524064171123e-06,
|
| 577 |
+
"loss": 0.8177,
|
| 578 |
+
"step": 7400
|
| 579 |
+
},
|
| 580 |
+
{
|
| 581 |
+
"epoch": 65.79,
|
| 582 |
+
"learning_rate": 4.529411764705883e-06,
|
| 583 |
+
"loss": 0.8288,
|
| 584 |
+
"step": 7500
|
| 585 |
+
},
|
| 586 |
+
{
|
| 587 |
+
"epoch": 65.79,
|
| 588 |
+
"eval_loss": 0.2333229035139084,
|
| 589 |
+
"eval_runtime": 48.8974,
|
| 590 |
+
"eval_samples_per_second": 32.701,
|
| 591 |
+
"eval_steps_per_second": 1.023,
|
| 592 |
+
"eval_wer": 0.2996667003332997,
|
| 593 |
+
"step": 7500
|
| 594 |
+
},
|
| 595 |
+
{
|
| 596 |
+
"epoch": 66.67,
|
| 597 |
+
"learning_rate": 3.5935828877005344e-06,
|
| 598 |
+
"loss": 0.8184,
|
| 599 |
+
"step": 7600
|
| 600 |
+
},
|
| 601 |
+
{
|
| 602 |
+
"epoch": 67.54,
|
| 603 |
+
"learning_rate": 2.657754010695187e-06,
|
| 604 |
+
"loss": 0.8178,
|
| 605 |
+
"step": 7700
|
| 606 |
+
},
|
| 607 |
+
{
|
| 608 |
+
"epoch": 68.42,
|
| 609 |
+
"learning_rate": 1.7219251336898393e-06,
|
| 610 |
+
"loss": 0.8214,
|
| 611 |
+
"step": 7800
|
| 612 |
+
},
|
| 613 |
+
{
|
| 614 |
+
"epoch": 69.3,
|
| 615 |
+
"learning_rate": 7.860962566844919e-07,
|
| 616 |
+
"loss": 0.8133,
|
| 617 |
+
"step": 7900
|
| 618 |
+
},
|
| 619 |
+
{
|
| 620 |
+
"epoch": 70.0,
|
| 621 |
+
"step": 7980,
|
| 622 |
+
"total_flos": 2.7520612954833367e+19,
|
| 623 |
+
"train_loss": 1.2354308326740313,
|
| 624 |
+
"train_runtime": 11524.5807,
|
| 625 |
+
"train_samples_per_second": 22.03,
|
| 626 |
+
"train_steps_per_second": 0.692
|
| 627 |
+
}
|
| 628 |
+
],
|
| 629 |
+
"max_steps": 7980,
|
| 630 |
+
"num_train_epochs": 70,
|
| 631 |
+
"total_flos": 2.7520612954833367e+19,
|
| 632 |
+
"trial_name": null,
|
| 633 |
+
"trial_params": null
|
| 634 |
+
}
|