End of training
Browse files- all_results.json +10 -10
- eval_results.json +6 -6
- pytorch_model.bin +1 -1
- train_results.json +5 -5
- trainer_state.json +396 -33
all_results.json
CHANGED
|
@@ -1,14 +1,14 @@
|
|
| 1 |
{
|
| 2 |
-
"epoch":
|
| 3 |
-
"eval_loss": 0.
|
| 4 |
-
"eval_runtime":
|
| 5 |
"eval_samples": 1112,
|
| 6 |
-
"eval_samples_per_second":
|
| 7 |
-
"eval_steps_per_second": 4.
|
| 8 |
-
"eval_wer": 0.
|
| 9 |
-
"train_loss":
|
| 10 |
-
"train_runtime":
|
| 11 |
"train_samples": 3336,
|
| 12 |
-
"train_samples_per_second":
|
| 13 |
-
"train_steps_per_second": 0.
|
| 14 |
}
|
|
|
|
| 1 |
{
|
| 2 |
+
"epoch": 100.0,
|
| 3 |
+
"eval_loss": 0.5928943157196045,
|
| 4 |
+
"eval_runtime": 66.288,
|
| 5 |
"eval_samples": 1112,
|
| 6 |
+
"eval_samples_per_second": 16.775,
|
| 7 |
+
"eval_steps_per_second": 4.194,
|
| 8 |
+
"eval_wer": 0.7180269058295964,
|
| 9 |
+
"train_loss": 1.2273276989276594,
|
| 10 |
+
"train_runtime": 32753.7935,
|
| 11 |
"train_samples": 3336,
|
| 12 |
+
"train_samples_per_second": 10.185,
|
| 13 |
+
"train_steps_per_second": 0.159
|
| 14 |
}
|
eval_results.json
CHANGED
|
@@ -1,9 +1,9 @@
|
|
| 1 |
{
|
| 2 |
-
"epoch":
|
| 3 |
-
"eval_loss": 0.
|
| 4 |
-
"eval_runtime":
|
| 5 |
"eval_samples": 1112,
|
| 6 |
-
"eval_samples_per_second":
|
| 7 |
-
"eval_steps_per_second": 4.
|
| 8 |
-
"eval_wer": 0.
|
| 9 |
}
|
|
|
|
| 1 |
{
|
| 2 |
+
"epoch": 100.0,
|
| 3 |
+
"eval_loss": 0.5928943157196045,
|
| 4 |
+
"eval_runtime": 66.288,
|
| 5 |
"eval_samples": 1112,
|
| 6 |
+
"eval_samples_per_second": 16.775,
|
| 7 |
+
"eval_steps_per_second": 4.194,
|
| 8 |
+
"eval_wer": 0.7180269058295964
|
| 9 |
}
|
pytorch_model.bin
CHANGED
|
@@ -1,3 +1,3 @@
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
-
oid sha256:
|
| 3 |
size 3850717489
|
|
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:e84e469f4b91a928d80c1677ce1094cf6c9c72a5e36fa367eb082b154b9c45c3
|
| 3 |
size 3850717489
|
train_results.json
CHANGED
|
@@ -1,8 +1,8 @@
|
|
| 1 |
{
|
| 2 |
-
"epoch":
|
| 3 |
-
"train_loss":
|
| 4 |
-
"train_runtime":
|
| 5 |
"train_samples": 3336,
|
| 6 |
-
"train_samples_per_second":
|
| 7 |
-
"train_steps_per_second": 0.
|
| 8 |
}
|
|
|
|
| 1 |
{
|
| 2 |
+
"epoch": 100.0,
|
| 3 |
+
"train_loss": 1.2273276989276594,
|
| 4 |
+
"train_runtime": 32753.7935,
|
| 5 |
"train_samples": 3336,
|
| 6 |
+
"train_samples_per_second": 10.185,
|
| 7 |
+
"train_steps_per_second": 0.159
|
| 8 |
}
|
trainer_state.json
CHANGED
|
@@ -1,64 +1,427 @@
|
|
| 1 |
{
|
| 2 |
"best_metric": null,
|
| 3 |
"best_model_checkpoint": null,
|
| 4 |
-
"epoch":
|
| 5 |
-
"global_step":
|
| 6 |
"is_hyper_param_search": false,
|
| 7 |
"is_local_process_zero": true,
|
| 8 |
"is_world_process_zero": true,
|
| 9 |
"log_history": [
|
| 10 |
{
|
| 11 |
-
"epoch":
|
| 12 |
-
"learning_rate":
|
| 13 |
-
"loss":
|
| 14 |
"step": 100
|
| 15 |
},
|
| 16 |
{
|
| 17 |
-
"epoch":
|
| 18 |
-
"learning_rate":
|
| 19 |
-
"loss": 3.
|
| 20 |
"step": 200
|
| 21 |
},
|
| 22 |
{
|
| 23 |
-
"epoch":
|
| 24 |
-
"learning_rate":
|
| 25 |
-
"loss":
|
| 26 |
"step": 300
|
| 27 |
},
|
| 28 |
{
|
| 29 |
-
"epoch":
|
| 30 |
-
"learning_rate":
|
| 31 |
-
"loss":
|
| 32 |
"step": 400
|
| 33 |
},
|
| 34 |
{
|
| 35 |
-
"epoch":
|
| 36 |
-
"learning_rate":
|
| 37 |
-
"loss": 1.
|
| 38 |
"step": 500
|
| 39 |
},
|
| 40 |
{
|
| 41 |
-
"epoch":
|
| 42 |
-
"eval_loss": 0.
|
| 43 |
-
"eval_runtime":
|
| 44 |
-
"eval_samples_per_second":
|
| 45 |
-
"eval_steps_per_second": 4.
|
| 46 |
-
"eval_wer": 0.
|
| 47 |
"step": 500
|
| 48 |
},
|
| 49 |
{
|
| 50 |
-
"epoch":
|
| 51 |
-
"
|
| 52 |
-
"
|
| 53 |
-
"
|
| 54 |
-
|
| 55 |
-
|
| 56 |
-
"
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 57 |
}
|
| 58 |
],
|
| 59 |
-
"max_steps":
|
| 60 |
-
"num_train_epochs":
|
| 61 |
-
"total_flos":
|
| 62 |
"trial_name": null,
|
| 63 |
"trial_params": null
|
| 64 |
}
|
|
|
|
| 1 |
{
|
| 2 |
"best_metric": null,
|
| 3 |
"best_model_checkpoint": null,
|
| 4 |
+
"epoch": 99.99521531100478,
|
| 5 |
+
"global_step": 5200,
|
| 6 |
"is_hyper_param_search": false,
|
| 7 |
"is_local_process_zero": true,
|
| 8 |
"is_world_process_zero": true,
|
| 9 |
"log_history": [
|
| 10 |
{
|
| 11 |
+
"epoch": 1.92,
|
| 12 |
+
"learning_rate": 3.2333333333333334e-06,
|
| 13 |
+
"loss": 13.5304,
|
| 14 |
"step": 100
|
| 15 |
},
|
| 16 |
{
|
| 17 |
+
"epoch": 3.84,
|
| 18 |
+
"learning_rate": 6.566666666666667e-06,
|
| 19 |
+
"loss": 3.5412,
|
| 20 |
"step": 200
|
| 21 |
},
|
| 22 |
{
|
| 23 |
+
"epoch": 5.77,
|
| 24 |
+
"learning_rate": 9.9e-06,
|
| 25 |
+
"loss": 3.1425,
|
| 26 |
"step": 300
|
| 27 |
},
|
| 28 |
{
|
| 29 |
+
"epoch": 7.69,
|
| 30 |
+
"learning_rate": 9.80204081632653e-06,
|
| 31 |
+
"loss": 2.0204,
|
| 32 |
"step": 400
|
| 33 |
},
|
| 34 |
{
|
| 35 |
+
"epoch": 9.61,
|
| 36 |
+
"learning_rate": 9.59795918367347e-06,
|
| 37 |
+
"loss": 1.5794,
|
| 38 |
"step": 500
|
| 39 |
},
|
| 40 |
{
|
| 41 |
+
"epoch": 9.61,
|
| 42 |
+
"eval_loss": 0.9859886169433594,
|
| 43 |
+
"eval_runtime": 65.6063,
|
| 44 |
+
"eval_samples_per_second": 16.95,
|
| 45 |
+
"eval_steps_per_second": 4.237,
|
| 46 |
+
"eval_wer": 0.9582062780269058,
|
| 47 |
"step": 500
|
| 48 |
},
|
| 49 |
{
|
| 50 |
+
"epoch": 11.54,
|
| 51 |
+
"learning_rate": 9.39387755102041e-06,
|
| 52 |
+
"loss": 1.4015,
|
| 53 |
+
"step": 600
|
| 54 |
+
},
|
| 55 |
+
{
|
| 56 |
+
"epoch": 13.46,
|
| 57 |
+
"learning_rate": 9.189795918367347e-06,
|
| 58 |
+
"loss": 1.2803,
|
| 59 |
+
"step": 700
|
| 60 |
+
},
|
| 61 |
+
{
|
| 62 |
+
"epoch": 15.38,
|
| 63 |
+
"learning_rate": 8.985714285714287e-06,
|
| 64 |
+
"loss": 1.1969,
|
| 65 |
+
"step": 800
|
| 66 |
+
},
|
| 67 |
+
{
|
| 68 |
+
"epoch": 17.31,
|
| 69 |
+
"learning_rate": 8.781632653061225e-06,
|
| 70 |
+
"loss": 1.1458,
|
| 71 |
+
"step": 900
|
| 72 |
+
},
|
| 73 |
+
{
|
| 74 |
+
"epoch": 19.23,
|
| 75 |
+
"learning_rate": 8.577551020408163e-06,
|
| 76 |
+
"loss": 1.0985,
|
| 77 |
+
"step": 1000
|
| 78 |
+
},
|
| 79 |
+
{
|
| 80 |
+
"epoch": 19.23,
|
| 81 |
+
"eval_loss": 0.5948446989059448,
|
| 82 |
+
"eval_runtime": 63.7089,
|
| 83 |
+
"eval_samples_per_second": 17.454,
|
| 84 |
+
"eval_steps_per_second": 4.364,
|
| 85 |
+
"eval_wer": 0.7533632286995515,
|
| 86 |
+
"step": 1000
|
| 87 |
+
},
|
| 88 |
+
{
|
| 89 |
+
"epoch": 21.15,
|
| 90 |
+
"learning_rate": 8.373469387755104e-06,
|
| 91 |
+
"loss": 1.0639,
|
| 92 |
+
"step": 1100
|
| 93 |
+
},
|
| 94 |
+
{
|
| 95 |
+
"epoch": 23.08,
|
| 96 |
+
"learning_rate": 8.169387755102042e-06,
|
| 97 |
+
"loss": 1.0214,
|
| 98 |
+
"step": 1200
|
| 99 |
+
},
|
| 100 |
+
{
|
| 101 |
+
"epoch": 25.0,
|
| 102 |
+
"learning_rate": 7.96530612244898e-06,
|
| 103 |
+
"loss": 1.0047,
|
| 104 |
+
"step": 1300
|
| 105 |
+
},
|
| 106 |
+
{
|
| 107 |
+
"epoch": 26.92,
|
| 108 |
+
"learning_rate": 7.76122448979592e-06,
|
| 109 |
+
"loss": 0.971,
|
| 110 |
+
"step": 1400
|
| 111 |
+
},
|
| 112 |
+
{
|
| 113 |
+
"epoch": 28.84,
|
| 114 |
+
"learning_rate": 7.557142857142857e-06,
|
| 115 |
+
"loss": 0.9551,
|
| 116 |
+
"step": 1500
|
| 117 |
+
},
|
| 118 |
+
{
|
| 119 |
+
"epoch": 28.84,
|
| 120 |
+
"eval_loss": 0.5892378091812134,
|
| 121 |
+
"eval_runtime": 64.7699,
|
| 122 |
+
"eval_samples_per_second": 17.168,
|
| 123 |
+
"eval_steps_per_second": 4.292,
|
| 124 |
+
"eval_wer": 0.7391928251121076,
|
| 125 |
+
"step": 1500
|
| 126 |
+
},
|
| 127 |
+
{
|
| 128 |
+
"epoch": 30.77,
|
| 129 |
+
"learning_rate": 7.353061224489797e-06,
|
| 130 |
+
"loss": 0.9387,
|
| 131 |
+
"step": 1600
|
| 132 |
+
},
|
| 133 |
+
{
|
| 134 |
+
"epoch": 32.69,
|
| 135 |
+
"learning_rate": 7.148979591836735e-06,
|
| 136 |
+
"loss": 0.9272,
|
| 137 |
+
"step": 1700
|
| 138 |
+
},
|
| 139 |
+
{
|
| 140 |
+
"epoch": 34.61,
|
| 141 |
+
"learning_rate": 6.944897959183674e-06,
|
| 142 |
+
"loss": 0.9192,
|
| 143 |
+
"step": 1800
|
| 144 |
+
},
|
| 145 |
+
{
|
| 146 |
+
"epoch": 36.54,
|
| 147 |
+
"learning_rate": 6.740816326530613e-06,
|
| 148 |
+
"loss": 0.892,
|
| 149 |
+
"step": 1900
|
| 150 |
+
},
|
| 151 |
+
{
|
| 152 |
+
"epoch": 38.46,
|
| 153 |
+
"learning_rate": 6.5367346938775515e-06,
|
| 154 |
+
"loss": 0.8699,
|
| 155 |
+
"step": 2000
|
| 156 |
+
},
|
| 157 |
+
{
|
| 158 |
+
"epoch": 38.46,
|
| 159 |
+
"eval_loss": 0.5926958322525024,
|
| 160 |
+
"eval_runtime": 64.9144,
|
| 161 |
+
"eval_samples_per_second": 17.13,
|
| 162 |
+
"eval_steps_per_second": 4.283,
|
| 163 |
+
"eval_wer": 0.72,
|
| 164 |
+
"step": 2000
|
| 165 |
+
},
|
| 166 |
+
{
|
| 167 |
+
"epoch": 40.38,
|
| 168 |
+
"learning_rate": 6.332653061224491e-06,
|
| 169 |
+
"loss": 0.8779,
|
| 170 |
+
"step": 2100
|
| 171 |
+
},
|
| 172 |
+
{
|
| 173 |
+
"epoch": 42.31,
|
| 174 |
+
"learning_rate": 6.128571428571429e-06,
|
| 175 |
+
"loss": 0.8511,
|
| 176 |
+
"step": 2200
|
| 177 |
+
},
|
| 178 |
+
{
|
| 179 |
+
"epoch": 44.23,
|
| 180 |
+
"learning_rate": 5.924489795918368e-06,
|
| 181 |
+
"loss": 0.8532,
|
| 182 |
+
"step": 2300
|
| 183 |
+
},
|
| 184 |
+
{
|
| 185 |
+
"epoch": 46.15,
|
| 186 |
+
"learning_rate": 5.720408163265306e-06,
|
| 187 |
+
"loss": 0.8442,
|
| 188 |
+
"step": 2400
|
| 189 |
+
},
|
| 190 |
+
{
|
| 191 |
+
"epoch": 48.08,
|
| 192 |
+
"learning_rate": 5.516326530612245e-06,
|
| 193 |
+
"loss": 0.8372,
|
| 194 |
+
"step": 2500
|
| 195 |
+
},
|
| 196 |
+
{
|
| 197 |
+
"epoch": 48.08,
|
| 198 |
+
"eval_loss": 0.5938696265220642,
|
| 199 |
+
"eval_runtime": 64.5641,
|
| 200 |
+
"eval_samples_per_second": 17.223,
|
| 201 |
+
"eval_steps_per_second": 4.306,
|
| 202 |
+
"eval_wer": 0.73847533632287,
|
| 203 |
+
"step": 2500
|
| 204 |
+
},
|
| 205 |
+
{
|
| 206 |
+
"epoch": 50.0,
|
| 207 |
+
"learning_rate": 5.312244897959185e-06,
|
| 208 |
+
"loss": 0.8112,
|
| 209 |
+
"step": 2600
|
| 210 |
+
},
|
| 211 |
+
{
|
| 212 |
+
"epoch": 51.92,
|
| 213 |
+
"learning_rate": 5.108163265306123e-06,
|
| 214 |
+
"loss": 0.8165,
|
| 215 |
+
"step": 2700
|
| 216 |
+
},
|
| 217 |
+
{
|
| 218 |
+
"epoch": 53.84,
|
| 219 |
+
"learning_rate": 4.904081632653061e-06,
|
| 220 |
+
"loss": 0.8045,
|
| 221 |
+
"step": 2800
|
| 222 |
+
},
|
| 223 |
+
{
|
| 224 |
+
"epoch": 55.77,
|
| 225 |
+
"learning_rate": 4.7e-06,
|
| 226 |
+
"loss": 0.7974,
|
| 227 |
+
"step": 2900
|
| 228 |
+
},
|
| 229 |
+
{
|
| 230 |
+
"epoch": 57.69,
|
| 231 |
+
"learning_rate": 4.4959183673469394e-06,
|
| 232 |
+
"loss": 0.7794,
|
| 233 |
+
"step": 3000
|
| 234 |
+
},
|
| 235 |
+
{
|
| 236 |
+
"epoch": 57.69,
|
| 237 |
+
"eval_loss": 0.5800846815109253,
|
| 238 |
+
"eval_runtime": 63.4747,
|
| 239 |
+
"eval_samples_per_second": 17.519,
|
| 240 |
+
"eval_steps_per_second": 4.38,
|
| 241 |
+
"eval_wer": 0.7388340807174888,
|
| 242 |
+
"step": 3000
|
| 243 |
+
},
|
| 244 |
+
{
|
| 245 |
+
"epoch": 59.61,
|
| 246 |
+
"learning_rate": 4.291836734693878e-06,
|
| 247 |
+
"loss": 0.7752,
|
| 248 |
+
"step": 3100
|
| 249 |
+
},
|
| 250 |
+
{
|
| 251 |
+
"epoch": 61.54,
|
| 252 |
+
"learning_rate": 4.087755102040817e-06,
|
| 253 |
+
"loss": 0.7765,
|
| 254 |
+
"step": 3200
|
| 255 |
+
},
|
| 256 |
+
{
|
| 257 |
+
"epoch": 63.46,
|
| 258 |
+
"learning_rate": 3.8836734693877556e-06,
|
| 259 |
+
"loss": 0.7584,
|
| 260 |
+
"step": 3300
|
| 261 |
+
},
|
| 262 |
+
{
|
| 263 |
+
"epoch": 65.38,
|
| 264 |
+
"learning_rate": 3.6795918367346943e-06,
|
| 265 |
+
"loss": 0.7589,
|
| 266 |
+
"step": 3400
|
| 267 |
+
},
|
| 268 |
+
{
|
| 269 |
+
"epoch": 67.31,
|
| 270 |
+
"learning_rate": 3.475510204081633e-06,
|
| 271 |
+
"loss": 0.7565,
|
| 272 |
+
"step": 3500
|
| 273 |
+
},
|
| 274 |
+
{
|
| 275 |
+
"epoch": 67.31,
|
| 276 |
+
"eval_loss": 0.582656979560852,
|
| 277 |
+
"eval_runtime": 71.1025,
|
| 278 |
+
"eval_samples_per_second": 15.639,
|
| 279 |
+
"eval_steps_per_second": 3.91,
|
| 280 |
+
"eval_wer": 0.7183856502242153,
|
| 281 |
+
"step": 3500
|
| 282 |
+
},
|
| 283 |
+
{
|
| 284 |
+
"epoch": 69.23,
|
| 285 |
+
"learning_rate": 3.2714285714285717e-06,
|
| 286 |
+
"loss": 0.7595,
|
| 287 |
+
"step": 3600
|
| 288 |
+
},
|
| 289 |
+
{
|
| 290 |
+
"epoch": 71.15,
|
| 291 |
+
"learning_rate": 3.0673469387755104e-06,
|
| 292 |
+
"loss": 0.747,
|
| 293 |
+
"step": 3700
|
| 294 |
+
},
|
| 295 |
+
{
|
| 296 |
+
"epoch": 73.08,
|
| 297 |
+
"learning_rate": 2.863265306122449e-06,
|
| 298 |
+
"loss": 0.7373,
|
| 299 |
+
"step": 3800
|
| 300 |
+
},
|
| 301 |
+
{
|
| 302 |
+
"epoch": 75.0,
|
| 303 |
+
"learning_rate": 2.6591836734693882e-06,
|
| 304 |
+
"loss": 0.7268,
|
| 305 |
+
"step": 3900
|
| 306 |
+
},
|
| 307 |
+
{
|
| 308 |
+
"epoch": 76.92,
|
| 309 |
+
"learning_rate": 2.455102040816327e-06,
|
| 310 |
+
"loss": 0.7303,
|
| 311 |
+
"step": 4000
|
| 312 |
+
},
|
| 313 |
+
{
|
| 314 |
+
"epoch": 76.92,
|
| 315 |
+
"eval_loss": 0.580017626285553,
|
| 316 |
+
"eval_runtime": 63.495,
|
| 317 |
+
"eval_samples_per_second": 17.513,
|
| 318 |
+
"eval_steps_per_second": 4.378,
|
| 319 |
+
"eval_wer": 0.7296860986547086,
|
| 320 |
+
"step": 4000
|
| 321 |
+
},
|
| 322 |
+
{
|
| 323 |
+
"epoch": 78.84,
|
| 324 |
+
"learning_rate": 2.2510204081632656e-06,
|
| 325 |
+
"loss": 0.7239,
|
| 326 |
+
"step": 4100
|
| 327 |
+
},
|
| 328 |
+
{
|
| 329 |
+
"epoch": 80.77,
|
| 330 |
+
"learning_rate": 2.0469387755102044e-06,
|
| 331 |
+
"loss": 0.7201,
|
| 332 |
+
"step": 4200
|
| 333 |
+
},
|
| 334 |
+
{
|
| 335 |
+
"epoch": 82.69,
|
| 336 |
+
"learning_rate": 1.8428571428571428e-06,
|
| 337 |
+
"loss": 0.7185,
|
| 338 |
+
"step": 4300
|
| 339 |
+
},
|
| 340 |
+
{
|
| 341 |
+
"epoch": 84.61,
|
| 342 |
+
"learning_rate": 1.6387755102040818e-06,
|
| 343 |
+
"loss": 0.7149,
|
| 344 |
+
"step": 4400
|
| 345 |
+
},
|
| 346 |
+
{
|
| 347 |
+
"epoch": 86.54,
|
| 348 |
+
"learning_rate": 1.4346938775510205e-06,
|
| 349 |
+
"loss": 0.7128,
|
| 350 |
+
"step": 4500
|
| 351 |
+
},
|
| 352 |
+
{
|
| 353 |
+
"epoch": 86.54,
|
| 354 |
+
"eval_loss": 0.5977014899253845,
|
| 355 |
+
"eval_runtime": 64.1516,
|
| 356 |
+
"eval_samples_per_second": 17.334,
|
| 357 |
+
"eval_steps_per_second": 4.333,
|
| 358 |
+
"eval_wer": 0.7149775784753363,
|
| 359 |
+
"step": 4500
|
| 360 |
+
},
|
| 361 |
+
{
|
| 362 |
+
"epoch": 88.46,
|
| 363 |
+
"learning_rate": 1.2306122448979594e-06,
|
| 364 |
+
"loss": 0.705,
|
| 365 |
+
"step": 4600
|
| 366 |
+
},
|
| 367 |
+
{
|
| 368 |
+
"epoch": 90.38,
|
| 369 |
+
"learning_rate": 1.026530612244898e-06,
|
| 370 |
+
"loss": 0.7129,
|
| 371 |
+
"step": 4700
|
| 372 |
+
},
|
| 373 |
+
{
|
| 374 |
+
"epoch": 92.31,
|
| 375 |
+
"learning_rate": 8.224489795918368e-07,
|
| 376 |
+
"loss": 0.7076,
|
| 377 |
+
"step": 4800
|
| 378 |
+
},
|
| 379 |
+
{
|
| 380 |
+
"epoch": 94.23,
|
| 381 |
+
"learning_rate": 6.183673469387755e-07,
|
| 382 |
+
"loss": 0.7126,
|
| 383 |
+
"step": 4900
|
| 384 |
+
},
|
| 385 |
+
{
|
| 386 |
+
"epoch": 96.15,
|
| 387 |
+
"learning_rate": 4.142857142857143e-07,
|
| 388 |
+
"loss": 0.6972,
|
| 389 |
+
"step": 5000
|
| 390 |
+
},
|
| 391 |
+
{
|
| 392 |
+
"epoch": 96.15,
|
| 393 |
+
"eval_loss": 0.5900700092315674,
|
| 394 |
+
"eval_runtime": 63.2743,
|
| 395 |
+
"eval_samples_per_second": 17.574,
|
| 396 |
+
"eval_steps_per_second": 4.394,
|
| 397 |
+
"eval_wer": 0.7176681614349776,
|
| 398 |
+
"step": 5000
|
| 399 |
+
},
|
| 400 |
+
{
|
| 401 |
+
"epoch": 98.08,
|
| 402 |
+
"learning_rate": 2.1020408163265306e-07,
|
| 403 |
+
"loss": 0.7001,
|
| 404 |
+
"step": 5100
|
| 405 |
+
},
|
| 406 |
+
{
|
| 407 |
+
"epoch": 100.0,
|
| 408 |
+
"learning_rate": 6.1224489795918365e-09,
|
| 409 |
+
"loss": 0.6965,
|
| 410 |
+
"step": 5200
|
| 411 |
+
},
|
| 412 |
+
{
|
| 413 |
+
"epoch": 100.0,
|
| 414 |
+
"step": 5200,
|
| 415 |
+
"total_flos": 1.4653197993726655e+20,
|
| 416 |
+
"train_loss": 1.2273276989276594,
|
| 417 |
+
"train_runtime": 32753.7935,
|
| 418 |
+
"train_samples_per_second": 10.185,
|
| 419 |
+
"train_steps_per_second": 0.159
|
| 420 |
}
|
| 421 |
],
|
| 422 |
+
"max_steps": 5200,
|
| 423 |
+
"num_train_epochs": 100,
|
| 424 |
+
"total_flos": 1.4653197993726655e+20,
|
| 425 |
"trial_name": null,
|
| 426 |
"trial_params": null
|
| 427 |
}
|