Commit
·
54d63f4
1
Parent(s):
291279f
End of training
Browse files- all_results.json +14 -0
- eval_results.json +9 -0
- train_results.json +8 -0
- trainer_state.json +493 -0
all_results.json
ADDED
|
@@ -0,0 +1,14 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"epoch": 50.0,
|
| 3 |
+
"eval_loss": 0.24730530381202698,
|
| 4 |
+
"eval_runtime": 73.7734,
|
| 5 |
+
"eval_samples": 1700,
|
| 6 |
+
"eval_samples_per_second": 23.044,
|
| 7 |
+
"eval_steps_per_second": 1.45,
|
| 8 |
+
"eval_wer": 0.3001967879134133,
|
| 9 |
+
"train_loss": 1.4450044761326002,
|
| 10 |
+
"train_runtime": 12353.9156,
|
| 11 |
+
"train_samples": 3676,
|
| 12 |
+
"train_samples_per_second": 14.878,
|
| 13 |
+
"train_steps_per_second": 0.465
|
| 14 |
+
}
|
eval_results.json
ADDED
|
@@ -0,0 +1,9 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"epoch": 50.0,
|
| 3 |
+
"eval_loss": 0.24730530381202698,
|
| 4 |
+
"eval_runtime": 73.7734,
|
| 5 |
+
"eval_samples": 1700,
|
| 6 |
+
"eval_samples_per_second": 23.044,
|
| 7 |
+
"eval_steps_per_second": 1.45,
|
| 8 |
+
"eval_wer": 0.3001967879134133
|
| 9 |
+
}
|
train_results.json
ADDED
|
@@ -0,0 +1,8 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"epoch": 50.0,
|
| 3 |
+
"train_loss": 1.4450044761326002,
|
| 4 |
+
"train_runtime": 12353.9156,
|
| 5 |
+
"train_samples": 3676,
|
| 6 |
+
"train_samples_per_second": 14.878,
|
| 7 |
+
"train_steps_per_second": 0.465
|
| 8 |
+
}
|
trainer_state.json
ADDED
|
@@ -0,0 +1,493 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"best_metric": null,
|
| 3 |
+
"best_model_checkpoint": null,
|
| 4 |
+
"epoch": 50.0,
|
| 5 |
+
"global_step": 5750,
|
| 6 |
+
"is_hyper_param_search": false,
|
| 7 |
+
"is_local_process_zero": true,
|
| 8 |
+
"is_world_process_zero": true,
|
| 9 |
+
"log_history": [
|
| 10 |
+
{
|
| 11 |
+
"epoch": 0.87,
|
| 12 |
+
"learning_rate": 7.35e-06,
|
| 13 |
+
"loss": 11.7893,
|
| 14 |
+
"step": 100
|
| 15 |
+
},
|
| 16 |
+
{
|
| 17 |
+
"epoch": 1.74,
|
| 18 |
+
"learning_rate": 1.485e-05,
|
| 19 |
+
"loss": 4.6989,
|
| 20 |
+
"step": 200
|
| 21 |
+
},
|
| 22 |
+
{
|
| 23 |
+
"epoch": 2.61,
|
| 24 |
+
"learning_rate": 2.2349999999999998e-05,
|
| 25 |
+
"loss": 3.5677,
|
| 26 |
+
"step": 300
|
| 27 |
+
},
|
| 28 |
+
{
|
| 29 |
+
"epoch": 3.48,
|
| 30 |
+
"learning_rate": 2.985e-05,
|
| 31 |
+
"loss": 3.1589,
|
| 32 |
+
"step": 400
|
| 33 |
+
},
|
| 34 |
+
{
|
| 35 |
+
"epoch": 3.48,
|
| 36 |
+
"eval_loss": 3.0830252170562744,
|
| 37 |
+
"eval_runtime": 76.0167,
|
| 38 |
+
"eval_samples_per_second": 22.363,
|
| 39 |
+
"eval_steps_per_second": 1.408,
|
| 40 |
+
"eval_wer": 1.0,
|
| 41 |
+
"step": 400
|
| 42 |
+
},
|
| 43 |
+
{
|
| 44 |
+
"epoch": 4.35,
|
| 45 |
+
"learning_rate": 3.735e-05,
|
| 46 |
+
"loss": 3.0279,
|
| 47 |
+
"step": 500
|
| 48 |
+
},
|
| 49 |
+
{
|
| 50 |
+
"epoch": 5.22,
|
| 51 |
+
"learning_rate": 4.484999999999999e-05,
|
| 52 |
+
"loss": 2.9807,
|
| 53 |
+
"step": 600
|
| 54 |
+
},
|
| 55 |
+
{
|
| 56 |
+
"epoch": 6.09,
|
| 57 |
+
"learning_rate": 5.234999999999999e-05,
|
| 58 |
+
"loss": 2.9589,
|
| 59 |
+
"step": 700
|
| 60 |
+
},
|
| 61 |
+
{
|
| 62 |
+
"epoch": 6.96,
|
| 63 |
+
"learning_rate": 5.985e-05,
|
| 64 |
+
"loss": 2.8921,
|
| 65 |
+
"step": 800
|
| 66 |
+
},
|
| 67 |
+
{
|
| 68 |
+
"epoch": 6.96,
|
| 69 |
+
"eval_loss": 2.6604607105255127,
|
| 70 |
+
"eval_runtime": 76.2841,
|
| 71 |
+
"eval_samples_per_second": 22.285,
|
| 72 |
+
"eval_steps_per_second": 1.403,
|
| 73 |
+
"eval_wer": 0.9981590808100045,
|
| 74 |
+
"step": 800
|
| 75 |
+
},
|
| 76 |
+
{
|
| 77 |
+
"epoch": 7.83,
|
| 78 |
+
"learning_rate": 6.735e-05,
|
| 79 |
+
"loss": 2.346,
|
| 80 |
+
"step": 900
|
| 81 |
+
},
|
| 82 |
+
{
|
| 83 |
+
"epoch": 8.7,
|
| 84 |
+
"learning_rate": 7.484999999999999e-05,
|
| 85 |
+
"loss": 1.6326,
|
| 86 |
+
"step": 1000
|
| 87 |
+
},
|
| 88 |
+
{
|
| 89 |
+
"epoch": 9.57,
|
| 90 |
+
"learning_rate": 7.345263157894736e-05,
|
| 91 |
+
"loss": 1.4075,
|
| 92 |
+
"step": 1100
|
| 93 |
+
},
|
| 94 |
+
{
|
| 95 |
+
"epoch": 10.43,
|
| 96 |
+
"learning_rate": 7.187368421052631e-05,
|
| 97 |
+
"loss": 1.3049,
|
| 98 |
+
"step": 1200
|
| 99 |
+
},
|
| 100 |
+
{
|
| 101 |
+
"epoch": 10.43,
|
| 102 |
+
"eval_loss": 0.5069074630737305,
|
| 103 |
+
"eval_runtime": 75.7661,
|
| 104 |
+
"eval_samples_per_second": 22.437,
|
| 105 |
+
"eval_steps_per_second": 1.412,
|
| 106 |
+
"eval_wer": 0.5706849488986225,
|
| 107 |
+
"step": 1200
|
| 108 |
+
},
|
| 109 |
+
{
|
| 110 |
+
"epoch": 11.3,
|
| 111 |
+
"learning_rate": 7.029473684210525e-05,
|
| 112 |
+
"loss": 1.2407,
|
| 113 |
+
"step": 1300
|
| 114 |
+
},
|
| 115 |
+
{
|
| 116 |
+
"epoch": 12.17,
|
| 117 |
+
"learning_rate": 6.87157894736842e-05,
|
| 118 |
+
"loss": 1.1999,
|
| 119 |
+
"step": 1400
|
| 120 |
+
},
|
| 121 |
+
{
|
| 122 |
+
"epoch": 13.04,
|
| 123 |
+
"learning_rate": 6.713684210526315e-05,
|
| 124 |
+
"loss": 1.167,
|
| 125 |
+
"step": 1500
|
| 126 |
+
},
|
| 127 |
+
{
|
| 128 |
+
"epoch": 13.91,
|
| 129 |
+
"learning_rate": 6.55578947368421e-05,
|
| 130 |
+
"loss": 1.1349,
|
| 131 |
+
"step": 1600
|
| 132 |
+
},
|
| 133 |
+
{
|
| 134 |
+
"epoch": 13.91,
|
| 135 |
+
"eval_loss": 0.4158603847026825,
|
| 136 |
+
"eval_runtime": 75.9244,
|
| 137 |
+
"eval_samples_per_second": 22.391,
|
| 138 |
+
"eval_steps_per_second": 1.409,
|
| 139 |
+
"eval_wer": 0.5040944581984383,
|
| 140 |
+
"step": 1600
|
| 141 |
+
},
|
| 142 |
+
{
|
| 143 |
+
"epoch": 14.78,
|
| 144 |
+
"learning_rate": 6.397894736842105e-05,
|
| 145 |
+
"loss": 1.1161,
|
| 146 |
+
"step": 1700
|
| 147 |
+
},
|
| 148 |
+
{
|
| 149 |
+
"epoch": 15.65,
|
| 150 |
+
"learning_rate": 6.239999999999999e-05,
|
| 151 |
+
"loss": 1.0982,
|
| 152 |
+
"step": 1800
|
| 153 |
+
},
|
| 154 |
+
{
|
| 155 |
+
"epoch": 16.52,
|
| 156 |
+
"learning_rate": 6.0821052631578945e-05,
|
| 157 |
+
"loss": 1.0841,
|
| 158 |
+
"step": 1900
|
| 159 |
+
},
|
| 160 |
+
{
|
| 161 |
+
"epoch": 17.39,
|
| 162 |
+
"learning_rate": 5.924210526315789e-05,
|
| 163 |
+
"loss": 1.0686,
|
| 164 |
+
"step": 2000
|
| 165 |
+
},
|
| 166 |
+
{
|
| 167 |
+
"epoch": 17.39,
|
| 168 |
+
"eval_loss": 0.38148975372314453,
|
| 169 |
+
"eval_runtime": 77.84,
|
| 170 |
+
"eval_samples_per_second": 21.84,
|
| 171 |
+
"eval_steps_per_second": 1.375,
|
| 172 |
+
"eval_wer": 0.4746397511585095,
|
| 173 |
+
"step": 2000
|
| 174 |
+
},
|
| 175 |
+
{
|
| 176 |
+
"epoch": 18.26,
|
| 177 |
+
"learning_rate": 5.7663157894736836e-05,
|
| 178 |
+
"loss": 1.047,
|
| 179 |
+
"step": 2100
|
| 180 |
+
},
|
| 181 |
+
{
|
| 182 |
+
"epoch": 19.13,
|
| 183 |
+
"learning_rate": 5.608421052631579e-05,
|
| 184 |
+
"loss": 1.03,
|
| 185 |
+
"step": 2200
|
| 186 |
+
},
|
| 187 |
+
{
|
| 188 |
+
"epoch": 20.0,
|
| 189 |
+
"learning_rate": 5.450526315789473e-05,
|
| 190 |
+
"loss": 1.0144,
|
| 191 |
+
"step": 2300
|
| 192 |
+
},
|
| 193 |
+
{
|
| 194 |
+
"epoch": 20.87,
|
| 195 |
+
"learning_rate": 5.292631578947368e-05,
|
| 196 |
+
"loss": 0.999,
|
| 197 |
+
"step": 2400
|
| 198 |
+
},
|
| 199 |
+
{
|
| 200 |
+
"epoch": 20.87,
|
| 201 |
+
"eval_loss": 0.3541375696659088,
|
| 202 |
+
"eval_runtime": 75.1099,
|
| 203 |
+
"eval_samples_per_second": 22.633,
|
| 204 |
+
"eval_steps_per_second": 1.425,
|
| 205 |
+
"eval_wer": 0.4343299688948137,
|
| 206 |
+
"step": 2400
|
| 207 |
+
},
|
| 208 |
+
{
|
| 209 |
+
"epoch": 21.74,
|
| 210 |
+
"learning_rate": 5.1347368421052626e-05,
|
| 211 |
+
"loss": 0.993,
|
| 212 |
+
"step": 2500
|
| 213 |
+
},
|
| 214 |
+
{
|
| 215 |
+
"epoch": 22.61,
|
| 216 |
+
"learning_rate": 4.976842105263157e-05,
|
| 217 |
+
"loss": 0.9766,
|
| 218 |
+
"step": 2600
|
| 219 |
+
},
|
| 220 |
+
{
|
| 221 |
+
"epoch": 23.48,
|
| 222 |
+
"learning_rate": 4.8189473684210524e-05,
|
| 223 |
+
"loss": 0.9573,
|
| 224 |
+
"step": 2700
|
| 225 |
+
},
|
| 226 |
+
{
|
| 227 |
+
"epoch": 24.35,
|
| 228 |
+
"learning_rate": 4.661052631578947e-05,
|
| 229 |
+
"loss": 0.945,
|
| 230 |
+
"step": 2800
|
| 231 |
+
},
|
| 232 |
+
{
|
| 233 |
+
"epoch": 24.35,
|
| 234 |
+
"eval_loss": 0.3265579342842102,
|
| 235 |
+
"eval_runtime": 75.4543,
|
| 236 |
+
"eval_samples_per_second": 22.53,
|
| 237 |
+
"eval_steps_per_second": 1.418,
|
| 238 |
+
"eval_wer": 0.4131911381958992,
|
| 239 |
+
"step": 2800
|
| 240 |
+
},
|
| 241 |
+
{
|
| 242 |
+
"epoch": 25.22,
|
| 243 |
+
"learning_rate": 4.5031578947368415e-05,
|
| 244 |
+
"loss": 0.9309,
|
| 245 |
+
"step": 2900
|
| 246 |
+
},
|
| 247 |
+
{
|
| 248 |
+
"epoch": 26.09,
|
| 249 |
+
"learning_rate": 4.345263157894736e-05,
|
| 250 |
+
"loss": 0.9151,
|
| 251 |
+
"step": 3000
|
| 252 |
+
},
|
| 253 |
+
{
|
| 254 |
+
"epoch": 26.96,
|
| 255 |
+
"learning_rate": 4.1873684210526314e-05,
|
| 256 |
+
"loss": 0.9138,
|
| 257 |
+
"step": 3100
|
| 258 |
+
},
|
| 259 |
+
{
|
| 260 |
+
"epoch": 27.83,
|
| 261 |
+
"learning_rate": 4.031052631578947e-05,
|
| 262 |
+
"loss": 0.9058,
|
| 263 |
+
"step": 3200
|
| 264 |
+
},
|
| 265 |
+
{
|
| 266 |
+
"epoch": 27.83,
|
| 267 |
+
"eval_loss": 0.2969067096710205,
|
| 268 |
+
"eval_runtime": 75.0182,
|
| 269 |
+
"eval_samples_per_second": 22.661,
|
| 270 |
+
"eval_steps_per_second": 1.426,
|
| 271 |
+
"eval_wer": 0.377071034088745,
|
| 272 |
+
"step": 3200
|
| 273 |
+
},
|
| 274 |
+
{
|
| 275 |
+
"epoch": 28.7,
|
| 276 |
+
"learning_rate": 3.8731578947368425e-05,
|
| 277 |
+
"loss": 0.9001,
|
| 278 |
+
"step": 3300
|
| 279 |
+
},
|
| 280 |
+
{
|
| 281 |
+
"epoch": 29.57,
|
| 282 |
+
"learning_rate": 3.7152631578947364e-05,
|
| 283 |
+
"loss": 0.8882,
|
| 284 |
+
"step": 3400
|
| 285 |
+
},
|
| 286 |
+
{
|
| 287 |
+
"epoch": 30.43,
|
| 288 |
+
"learning_rate": 3.557368421052631e-05,
|
| 289 |
+
"loss": 0.867,
|
| 290 |
+
"step": 3500
|
| 291 |
+
},
|
| 292 |
+
{
|
| 293 |
+
"epoch": 31.3,
|
| 294 |
+
"learning_rate": 3.399473684210526e-05,
|
| 295 |
+
"loss": 0.8672,
|
| 296 |
+
"step": 3600
|
| 297 |
+
},
|
| 298 |
+
{
|
| 299 |
+
"epoch": 31.3,
|
| 300 |
+
"eval_loss": 0.2802095413208008,
|
| 301 |
+
"eval_runtime": 75.6097,
|
| 302 |
+
"eval_samples_per_second": 22.484,
|
| 303 |
+
"eval_steps_per_second": 1.415,
|
| 304 |
+
"eval_wer": 0.35529740366914236,
|
| 305 |
+
"step": 3600
|
| 306 |
+
},
|
| 307 |
+
{
|
| 308 |
+
"epoch": 32.17,
|
| 309 |
+
"learning_rate": 3.241578947368421e-05,
|
| 310 |
+
"loss": 0.8522,
|
| 311 |
+
"step": 3700
|
| 312 |
+
},
|
| 313 |
+
{
|
| 314 |
+
"epoch": 33.04,
|
| 315 |
+
"learning_rate": 3.0836842105263153e-05,
|
| 316 |
+
"loss": 0.8391,
|
| 317 |
+
"step": 3800
|
| 318 |
+
},
|
| 319 |
+
{
|
| 320 |
+
"epoch": 33.91,
|
| 321 |
+
"learning_rate": 2.9257894736842103e-05,
|
| 322 |
+
"loss": 0.8368,
|
| 323 |
+
"step": 3900
|
| 324 |
+
},
|
| 325 |
+
{
|
| 326 |
+
"epoch": 34.78,
|
| 327 |
+
"learning_rate": 2.767894736842105e-05,
|
| 328 |
+
"loss": 0.8313,
|
| 329 |
+
"step": 4000
|
| 330 |
+
},
|
| 331 |
+
{
|
| 332 |
+
"epoch": 34.78,
|
| 333 |
+
"eval_loss": 0.26624372601509094,
|
| 334 |
+
"eval_runtime": 74.7639,
|
| 335 |
+
"eval_samples_per_second": 22.738,
|
| 336 |
+
"eval_steps_per_second": 1.431,
|
| 337 |
+
"eval_wer": 0.3379673712943566,
|
| 338 |
+
"step": 4000
|
| 339 |
+
},
|
| 340 |
+
{
|
| 341 |
+
"epoch": 35.65,
|
| 342 |
+
"learning_rate": 2.6099999999999997e-05,
|
| 343 |
+
"loss": 0.8188,
|
| 344 |
+
"step": 4100
|
| 345 |
+
},
|
| 346 |
+
{
|
| 347 |
+
"epoch": 36.52,
|
| 348 |
+
"learning_rate": 2.4521052631578947e-05,
|
| 349 |
+
"loss": 0.8168,
|
| 350 |
+
"step": 4200
|
| 351 |
+
},
|
| 352 |
+
{
|
| 353 |
+
"epoch": 37.39,
|
| 354 |
+
"learning_rate": 2.2942105263157892e-05,
|
| 355 |
+
"loss": 0.8103,
|
| 356 |
+
"step": 4300
|
| 357 |
+
},
|
| 358 |
+
{
|
| 359 |
+
"epoch": 38.26,
|
| 360 |
+
"learning_rate": 2.136315789473684e-05,
|
| 361 |
+
"loss": 0.8068,
|
| 362 |
+
"step": 4400
|
| 363 |
+
},
|
| 364 |
+
{
|
| 365 |
+
"epoch": 38.26,
|
| 366 |
+
"eval_loss": 0.25275924801826477,
|
| 367 |
+
"eval_runtime": 75.264,
|
| 368 |
+
"eval_samples_per_second": 22.587,
|
| 369 |
+
"eval_steps_per_second": 1.422,
|
| 370 |
+
"eval_wer": 0.3180981400368184,
|
| 371 |
+
"step": 4400
|
| 372 |
+
},
|
| 373 |
+
{
|
| 374 |
+
"epoch": 39.13,
|
| 375 |
+
"learning_rate": 1.9784210526315787e-05,
|
| 376 |
+
"loss": 0.8046,
|
| 377 |
+
"step": 4500
|
| 378 |
+
},
|
| 379 |
+
{
|
| 380 |
+
"epoch": 40.0,
|
| 381 |
+
"learning_rate": 1.8205263157894736e-05,
|
| 382 |
+
"loss": 0.7929,
|
| 383 |
+
"step": 4600
|
| 384 |
+
},
|
| 385 |
+
{
|
| 386 |
+
"epoch": 40.87,
|
| 387 |
+
"learning_rate": 1.6626315789473682e-05,
|
| 388 |
+
"loss": 0.7817,
|
| 389 |
+
"step": 4700
|
| 390 |
+
},
|
| 391 |
+
{
|
| 392 |
+
"epoch": 41.74,
|
| 393 |
+
"learning_rate": 1.504736842105263e-05,
|
| 394 |
+
"loss": 0.7796,
|
| 395 |
+
"step": 4800
|
| 396 |
+
},
|
| 397 |
+
{
|
| 398 |
+
"epoch": 41.74,
|
| 399 |
+
"eval_loss": 0.25369659066200256,
|
| 400 |
+
"eval_runtime": 74.2196,
|
| 401 |
+
"eval_samples_per_second": 22.905,
|
| 402 |
+
"eval_steps_per_second": 1.442,
|
| 403 |
+
"eval_wer": 0.3073065447851203,
|
| 404 |
+
"step": 4800
|
| 405 |
+
},
|
| 406 |
+
{
|
| 407 |
+
"epoch": 42.61,
|
| 408 |
+
"learning_rate": 1.3468421052631578e-05,
|
| 409 |
+
"loss": 0.7789,
|
| 410 |
+
"step": 4900
|
| 411 |
+
},
|
| 412 |
+
{
|
| 413 |
+
"epoch": 43.48,
|
| 414 |
+
"learning_rate": 1.1889473684210526e-05,
|
| 415 |
+
"loss": 0.7736,
|
| 416 |
+
"step": 5000
|
| 417 |
+
},
|
| 418 |
+
{
|
| 419 |
+
"epoch": 44.35,
|
| 420 |
+
"learning_rate": 1.0310526315789472e-05,
|
| 421 |
+
"loss": 0.783,
|
| 422 |
+
"step": 5100
|
| 423 |
+
},
|
| 424 |
+
{
|
| 425 |
+
"epoch": 45.22,
|
| 426 |
+
"learning_rate": 8.73157894736842e-06,
|
| 427 |
+
"loss": 0.7621,
|
| 428 |
+
"step": 5200
|
| 429 |
+
},
|
| 430 |
+
{
|
| 431 |
+
"epoch": 45.22,
|
| 432 |
+
"eval_loss": 0.2502739131450653,
|
| 433 |
+
"eval_runtime": 74.1963,
|
| 434 |
+
"eval_samples_per_second": 22.912,
|
| 435 |
+
"eval_steps_per_second": 1.442,
|
| 436 |
+
"eval_wer": 0.3035612264330604,
|
| 437 |
+
"step": 5200
|
| 438 |
+
},
|
| 439 |
+
{
|
| 440 |
+
"epoch": 46.09,
|
| 441 |
+
"learning_rate": 7.168421052631579e-06,
|
| 442 |
+
"loss": 0.767,
|
| 443 |
+
"step": 5300
|
| 444 |
+
},
|
| 445 |
+
{
|
| 446 |
+
"epoch": 46.96,
|
| 447 |
+
"learning_rate": 5.589473684210526e-06,
|
| 448 |
+
"loss": 0.7659,
|
| 449 |
+
"step": 5400
|
| 450 |
+
},
|
| 451 |
+
{
|
| 452 |
+
"epoch": 47.83,
|
| 453 |
+
"learning_rate": 4.0105263157894735e-06,
|
| 454 |
+
"loss": 0.7622,
|
| 455 |
+
"step": 5500
|
| 456 |
+
},
|
| 457 |
+
{
|
| 458 |
+
"epoch": 48.7,
|
| 459 |
+
"learning_rate": 2.431578947368421e-06,
|
| 460 |
+
"loss": 0.7611,
|
| 461 |
+
"step": 5600
|
| 462 |
+
},
|
| 463 |
+
{
|
| 464 |
+
"epoch": 48.7,
|
| 465 |
+
"eval_loss": 0.24768942594528198,
|
| 466 |
+
"eval_runtime": 74.3172,
|
| 467 |
+
"eval_samples_per_second": 22.875,
|
| 468 |
+
"eval_steps_per_second": 1.44,
|
| 469 |
+
"eval_wer": 0.29905414841617467,
|
| 470 |
+
"step": 5600
|
| 471 |
+
},
|
| 472 |
+
{
|
| 473 |
+
"epoch": 49.57,
|
| 474 |
+
"learning_rate": 8.526315789473684e-07,
|
| 475 |
+
"loss": 0.7494,
|
| 476 |
+
"step": 5700
|
| 477 |
+
},
|
| 478 |
+
{
|
| 479 |
+
"epoch": 50.0,
|
| 480 |
+
"step": 5750,
|
| 481 |
+
"total_flos": 2.9609555446330925e+19,
|
| 482 |
+
"train_loss": 1.4450044761326002,
|
| 483 |
+
"train_runtime": 12353.9156,
|
| 484 |
+
"train_samples_per_second": 14.878,
|
| 485 |
+
"train_steps_per_second": 0.465
|
| 486 |
+
}
|
| 487 |
+
],
|
| 488 |
+
"max_steps": 5750,
|
| 489 |
+
"num_train_epochs": 50,
|
| 490 |
+
"total_flos": 2.9609555446330925e+19,
|
| 491 |
+
"trial_name": null,
|
| 492 |
+
"trial_params": null
|
| 493 |
+
}
|