End of training
Browse files
all_results.json
ADDED
|
@@ -0,0 +1,12 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"epoch": 116.74,
|
| 3 |
+
"eval_loss": 0.8623208999633789,
|
| 4 |
+
"eval_runtime": 2294.4381,
|
| 5 |
+
"eval_samples_per_second": 0.223,
|
| 6 |
+
"eval_steps_per_second": 0.056,
|
| 7 |
+
"eval_wer": 54.06850588964662,
|
| 8 |
+
"train_loss": 0.0002270207000356938,
|
| 9 |
+
"train_runtime": 19.3242,
|
| 10 |
+
"train_samples_per_second": 2318.342,
|
| 11 |
+
"train_steps_per_second": 36.224
|
| 12 |
+
}
|
eval_results.json
ADDED
|
@@ -0,0 +1,8 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"epoch": 116.74,
|
| 3 |
+
"eval_loss": 0.8623208999633789,
|
| 4 |
+
"eval_runtime": 2294.4381,
|
| 5 |
+
"eval_samples_per_second": 0.223,
|
| 6 |
+
"eval_steps_per_second": 0.056,
|
| 7 |
+
"eval_wer": 54.06850588964662
|
| 8 |
+
}
|
runs/Dec14_08-20-14_129-213-22-31/events.out.tfevents.1671008657.129-213-22-31.75139.2
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:f9298768b7fb3d3db7480fdc749c9e805bff7ded6b578189a03ed79b419ca463
|
| 3 |
+
size 358
|
train_results.json
ADDED
|
@@ -0,0 +1,7 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"epoch": 116.74,
|
| 3 |
+
"train_loss": 0.0002270207000356938,
|
| 4 |
+
"train_runtime": 19.3242,
|
| 5 |
+
"train_samples_per_second": 2318.342,
|
| 6 |
+
"train_steps_per_second": 36.224
|
| 7 |
+
}
|
trainer_state.json
ADDED
|
@@ -0,0 +1,508 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"best_metric": 54.06850588964662,
|
| 3 |
+
"best_model_checkpoint": "./checkpoint-400",
|
| 4 |
+
"epoch": 116.74074074074075,
|
| 5 |
+
"global_step": 701,
|
| 6 |
+
"is_hyper_param_search": false,
|
| 7 |
+
"is_local_process_zero": true,
|
| 8 |
+
"is_world_process_zero": true,
|
| 9 |
+
"log_history": [
|
| 10 |
+
{
|
| 11 |
+
"epoch": 1.59,
|
| 12 |
+
"learning_rate": 4.2e-08,
|
| 13 |
+
"loss": 2.5631,
|
| 14 |
+
"step": 10
|
| 15 |
+
},
|
| 16 |
+
{
|
| 17 |
+
"epoch": 3.3,
|
| 18 |
+
"learning_rate": 1.02e-07,
|
| 19 |
+
"loss": 2.7338,
|
| 20 |
+
"step": 20
|
| 21 |
+
},
|
| 22 |
+
{
|
| 23 |
+
"epoch": 4.89,
|
| 24 |
+
"learning_rate": 1.56e-07,
|
| 25 |
+
"loss": 2.5033,
|
| 26 |
+
"step": 30
|
| 27 |
+
},
|
| 28 |
+
{
|
| 29 |
+
"epoch": 6.59,
|
| 30 |
+
"learning_rate": 2.1599999999999998e-07,
|
| 31 |
+
"loss": 2.5045,
|
| 32 |
+
"step": 40
|
| 33 |
+
},
|
| 34 |
+
{
|
| 35 |
+
"epoch": 8.3,
|
| 36 |
+
"learning_rate": 2.76e-07,
|
| 37 |
+
"loss": 2.1675,
|
| 38 |
+
"step": 50
|
| 39 |
+
},
|
| 40 |
+
{
|
| 41 |
+
"epoch": 9.89,
|
| 42 |
+
"learning_rate": 2.9810526315789476e-07,
|
| 43 |
+
"loss": 1.7642,
|
| 44 |
+
"step": 60
|
| 45 |
+
},
|
| 46 |
+
{
|
| 47 |
+
"epoch": 11.59,
|
| 48 |
+
"learning_rate": 2.949473684210526e-07,
|
| 49 |
+
"loss": 1.6382,
|
| 50 |
+
"step": 70
|
| 51 |
+
},
|
| 52 |
+
{
|
| 53 |
+
"epoch": 13.3,
|
| 54 |
+
"learning_rate": 2.917894736842105e-07,
|
| 55 |
+
"loss": 1.469,
|
| 56 |
+
"step": 80
|
| 57 |
+
},
|
| 58 |
+
{
|
| 59 |
+
"epoch": 14.89,
|
| 60 |
+
"learning_rate": 2.886315789473684e-07,
|
| 61 |
+
"loss": 1.2432,
|
| 62 |
+
"step": 90
|
| 63 |
+
},
|
| 64 |
+
{
|
| 65 |
+
"epoch": 16.59,
|
| 66 |
+
"learning_rate": 2.8547368421052633e-07,
|
| 67 |
+
"loss": 1.2281,
|
| 68 |
+
"step": 100
|
| 69 |
+
},
|
| 70 |
+
{
|
| 71 |
+
"epoch": 16.59,
|
| 72 |
+
"eval_loss": 1.0950794219970703,
|
| 73 |
+
"eval_runtime": 2208.5997,
|
| 74 |
+
"eval_samples_per_second": 0.232,
|
| 75 |
+
"eval_steps_per_second": 0.058,
|
| 76 |
+
"eval_wer": 69.31184128952262,
|
| 77 |
+
"step": 100
|
| 78 |
+
},
|
| 79 |
+
{
|
| 80 |
+
"epoch": 18.3,
|
| 81 |
+
"learning_rate": 2.823157894736842e-07,
|
| 82 |
+
"loss": 1.1316,
|
| 83 |
+
"step": 110
|
| 84 |
+
},
|
| 85 |
+
{
|
| 86 |
+
"epoch": 19.89,
|
| 87 |
+
"learning_rate": 2.7915789473684206e-07,
|
| 88 |
+
"loss": 0.9889,
|
| 89 |
+
"step": 120
|
| 90 |
+
},
|
| 91 |
+
{
|
| 92 |
+
"epoch": 21.59,
|
| 93 |
+
"learning_rate": 2.76e-07,
|
| 94 |
+
"loss": 0.9964,
|
| 95 |
+
"step": 130
|
| 96 |
+
},
|
| 97 |
+
{
|
| 98 |
+
"epoch": 23.3,
|
| 99 |
+
"learning_rate": 2.728421052631579e-07,
|
| 100 |
+
"loss": 0.9559,
|
| 101 |
+
"step": 140
|
| 102 |
+
},
|
| 103 |
+
{
|
| 104 |
+
"epoch": 24.89,
|
| 105 |
+
"learning_rate": 2.6968421052631577e-07,
|
| 106 |
+
"loss": 0.8493,
|
| 107 |
+
"step": 150
|
| 108 |
+
},
|
| 109 |
+
{
|
| 110 |
+
"epoch": 26.59,
|
| 111 |
+
"learning_rate": 2.665263157894737e-07,
|
| 112 |
+
"loss": 0.8809,
|
| 113 |
+
"step": 160
|
| 114 |
+
},
|
| 115 |
+
{
|
| 116 |
+
"epoch": 28.3,
|
| 117 |
+
"learning_rate": 2.6336842105263155e-07,
|
| 118 |
+
"loss": 0.8447,
|
| 119 |
+
"step": 170
|
| 120 |
+
},
|
| 121 |
+
{
|
| 122 |
+
"epoch": 29.89,
|
| 123 |
+
"learning_rate": 2.602105263157895e-07,
|
| 124 |
+
"loss": 0.7582,
|
| 125 |
+
"step": 180
|
| 126 |
+
},
|
| 127 |
+
{
|
| 128 |
+
"epoch": 31.59,
|
| 129 |
+
"learning_rate": 2.5705263157894734e-07,
|
| 130 |
+
"loss": 0.789,
|
| 131 |
+
"step": 190
|
| 132 |
+
},
|
| 133 |
+
{
|
| 134 |
+
"epoch": 33.3,
|
| 135 |
+
"learning_rate": 2.5389473684210526e-07,
|
| 136 |
+
"loss": 0.7529,
|
| 137 |
+
"step": 200
|
| 138 |
+
},
|
| 139 |
+
{
|
| 140 |
+
"epoch": 33.3,
|
| 141 |
+
"eval_loss": 0.8693189024925232,
|
| 142 |
+
"eval_runtime": 2268.4256,
|
| 143 |
+
"eval_samples_per_second": 0.226,
|
| 144 |
+
"eval_steps_per_second": 0.056,
|
| 145 |
+
"eval_wer": 57.563546187228766,
|
| 146 |
+
"step": 200
|
| 147 |
+
},
|
| 148 |
+
{
|
| 149 |
+
"epoch": 34.89,
|
| 150 |
+
"learning_rate": 2.5073684210526313e-07,
|
| 151 |
+
"loss": 0.6902,
|
| 152 |
+
"step": 210
|
| 153 |
+
},
|
| 154 |
+
{
|
| 155 |
+
"epoch": 36.59,
|
| 156 |
+
"learning_rate": 2.4757894736842105e-07,
|
| 157 |
+
"loss": 0.7184,
|
| 158 |
+
"step": 220
|
| 159 |
+
},
|
| 160 |
+
{
|
| 161 |
+
"epoch": 38.3,
|
| 162 |
+
"learning_rate": 2.4442105263157897e-07,
|
| 163 |
+
"loss": 0.6878,
|
| 164 |
+
"step": 230
|
| 165 |
+
},
|
| 166 |
+
{
|
| 167 |
+
"epoch": 39.89,
|
| 168 |
+
"learning_rate": 2.4126315789473683e-07,
|
| 169 |
+
"loss": 0.6384,
|
| 170 |
+
"step": 240
|
| 171 |
+
},
|
| 172 |
+
{
|
| 173 |
+
"epoch": 41.59,
|
| 174 |
+
"learning_rate": 2.3810526315789472e-07,
|
| 175 |
+
"loss": 0.6526,
|
| 176 |
+
"step": 250
|
| 177 |
+
},
|
| 178 |
+
{
|
| 179 |
+
"epoch": 43.3,
|
| 180 |
+
"learning_rate": 2.3494736842105262e-07,
|
| 181 |
+
"loss": 0.6423,
|
| 182 |
+
"step": 260
|
| 183 |
+
},
|
| 184 |
+
{
|
| 185 |
+
"epoch": 44.89,
|
| 186 |
+
"learning_rate": 2.317894736842105e-07,
|
| 187 |
+
"loss": 0.5752,
|
| 188 |
+
"step": 270
|
| 189 |
+
},
|
| 190 |
+
{
|
| 191 |
+
"epoch": 46.59,
|
| 192 |
+
"learning_rate": 2.286315789473684e-07,
|
| 193 |
+
"loss": 0.6023,
|
| 194 |
+
"step": 280
|
| 195 |
+
},
|
| 196 |
+
{
|
| 197 |
+
"epoch": 48.3,
|
| 198 |
+
"learning_rate": 2.254736842105263e-07,
|
| 199 |
+
"loss": 0.5882,
|
| 200 |
+
"step": 290
|
| 201 |
+
},
|
| 202 |
+
{
|
| 203 |
+
"epoch": 49.89,
|
| 204 |
+
"learning_rate": 2.2231578947368422e-07,
|
| 205 |
+
"loss": 0.5372,
|
| 206 |
+
"step": 300
|
| 207 |
+
},
|
| 208 |
+
{
|
| 209 |
+
"epoch": 49.89,
|
| 210 |
+
"eval_loss": 0.8399219512939453,
|
| 211 |
+
"eval_runtime": 2313.3276,
|
| 212 |
+
"eval_samples_per_second": 0.221,
|
| 213 |
+
"eval_steps_per_second": 0.055,
|
| 214 |
+
"eval_wer": 54.73496590204587,
|
| 215 |
+
"step": 300
|
| 216 |
+
},
|
| 217 |
+
{
|
| 218 |
+
"epoch": 51.59,
|
| 219 |
+
"learning_rate": 2.191578947368421e-07,
|
| 220 |
+
"loss": 0.5521,
|
| 221 |
+
"step": 310
|
| 222 |
+
},
|
| 223 |
+
{
|
| 224 |
+
"epoch": 53.3,
|
| 225 |
+
"learning_rate": 2.1599999999999998e-07,
|
| 226 |
+
"loss": 0.5448,
|
| 227 |
+
"step": 320
|
| 228 |
+
},
|
| 229 |
+
{
|
| 230 |
+
"epoch": 54.89,
|
| 231 |
+
"learning_rate": 2.128421052631579e-07,
|
| 232 |
+
"loss": 0.4926,
|
| 233 |
+
"step": 330
|
| 234 |
+
},
|
| 235 |
+
{
|
| 236 |
+
"epoch": 56.59,
|
| 237 |
+
"learning_rate": 2.096842105263158e-07,
|
| 238 |
+
"loss": 0.5193,
|
| 239 |
+
"step": 340
|
| 240 |
+
},
|
| 241 |
+
{
|
| 242 |
+
"epoch": 58.3,
|
| 243 |
+
"learning_rate": 2.0652631578947368e-07,
|
| 244 |
+
"loss": 0.4955,
|
| 245 |
+
"step": 350
|
| 246 |
+
},
|
| 247 |
+
{
|
| 248 |
+
"epoch": 59.89,
|
| 249 |
+
"learning_rate": 2.0336842105263155e-07,
|
| 250 |
+
"loss": 0.4537,
|
| 251 |
+
"step": 360
|
| 252 |
+
},
|
| 253 |
+
{
|
| 254 |
+
"epoch": 61.59,
|
| 255 |
+
"learning_rate": 2.0021052631578947e-07,
|
| 256 |
+
"loss": 0.4724,
|
| 257 |
+
"step": 370
|
| 258 |
+
},
|
| 259 |
+
{
|
| 260 |
+
"epoch": 63.3,
|
| 261 |
+
"learning_rate": 1.9705263157894736e-07,
|
| 262 |
+
"loss": 0.461,
|
| 263 |
+
"step": 380
|
| 264 |
+
},
|
| 265 |
+
{
|
| 266 |
+
"epoch": 64.89,
|
| 267 |
+
"learning_rate": 1.9389473684210525e-07,
|
| 268 |
+
"loss": 0.4134,
|
| 269 |
+
"step": 390
|
| 270 |
+
},
|
| 271 |
+
{
|
| 272 |
+
"epoch": 66.59,
|
| 273 |
+
"learning_rate": 1.9073684210526317e-07,
|
| 274 |
+
"loss": 0.4398,
|
| 275 |
+
"step": 400
|
| 276 |
+
},
|
| 277 |
+
{
|
| 278 |
+
"epoch": 66.59,
|
| 279 |
+
"eval_loss": 0.8623208999633789,
|
| 280 |
+
"eval_runtime": 2300.4889,
|
| 281 |
+
"eval_samples_per_second": 0.223,
|
| 282 |
+
"eval_steps_per_second": 0.056,
|
| 283 |
+
"eval_wer": 54.06850588964662,
|
| 284 |
+
"step": 400
|
| 285 |
+
},
|
| 286 |
+
{
|
| 287 |
+
"epoch": 68.3,
|
| 288 |
+
"learning_rate": 1.8757894736842104e-07,
|
| 289 |
+
"loss": 0.4255,
|
| 290 |
+
"step": 410
|
| 291 |
+
},
|
| 292 |
+
{
|
| 293 |
+
"epoch": 69.89,
|
| 294 |
+
"learning_rate": 1.8442105263157893e-07,
|
| 295 |
+
"loss": 0.3835,
|
| 296 |
+
"step": 420
|
| 297 |
+
},
|
| 298 |
+
{
|
| 299 |
+
"epoch": 71.59,
|
| 300 |
+
"learning_rate": 1.8126315789473682e-07,
|
| 301 |
+
"loss": 0.4041,
|
| 302 |
+
"step": 430
|
| 303 |
+
},
|
| 304 |
+
{
|
| 305 |
+
"epoch": 73.3,
|
| 306 |
+
"learning_rate": 1.7810526315789474e-07,
|
| 307 |
+
"loss": 0.3895,
|
| 308 |
+
"step": 440
|
| 309 |
+
},
|
| 310 |
+
{
|
| 311 |
+
"epoch": 74.89,
|
| 312 |
+
"learning_rate": 1.749473684210526e-07,
|
| 313 |
+
"loss": 0.35,
|
| 314 |
+
"step": 450
|
| 315 |
+
},
|
| 316 |
+
{
|
| 317 |
+
"epoch": 76.59,
|
| 318 |
+
"learning_rate": 1.717894736842105e-07,
|
| 319 |
+
"loss": 0.3722,
|
| 320 |
+
"step": 460
|
| 321 |
+
},
|
| 322 |
+
{
|
| 323 |
+
"epoch": 78.3,
|
| 324 |
+
"learning_rate": 1.6863157894736842e-07,
|
| 325 |
+
"loss": 0.3569,
|
| 326 |
+
"step": 470
|
| 327 |
+
},
|
| 328 |
+
{
|
| 329 |
+
"epoch": 79.89,
|
| 330 |
+
"learning_rate": 1.6547368421052632e-07,
|
| 331 |
+
"loss": 0.3314,
|
| 332 |
+
"step": 480
|
| 333 |
+
},
|
| 334 |
+
{
|
| 335 |
+
"epoch": 81.59,
|
| 336 |
+
"learning_rate": 1.6231578947368418e-07,
|
| 337 |
+
"loss": 0.341,
|
| 338 |
+
"step": 490
|
| 339 |
+
},
|
| 340 |
+
{
|
| 341 |
+
"epoch": 83.3,
|
| 342 |
+
"learning_rate": 1.591578947368421e-07,
|
| 343 |
+
"loss": 0.3244,
|
| 344 |
+
"step": 500
|
| 345 |
+
},
|
| 346 |
+
{
|
| 347 |
+
"epoch": 83.3,
|
| 348 |
+
"eval_loss": 0.9098332524299622,
|
| 349 |
+
"eval_runtime": 2322.6222,
|
| 350 |
+
"eval_samples_per_second": 0.22,
|
| 351 |
+
"eval_steps_per_second": 0.055,
|
| 352 |
+
"eval_wer": 54.750464972101675,
|
| 353 |
+
"step": 500
|
| 354 |
+
},
|
| 355 |
+
{
|
| 356 |
+
"epoch": 84.89,
|
| 357 |
+
"learning_rate": 1.56e-07,
|
| 358 |
+
"loss": 0.3066,
|
| 359 |
+
"step": 510
|
| 360 |
+
},
|
| 361 |
+
{
|
| 362 |
+
"epoch": 86.59,
|
| 363 |
+
"learning_rate": 1.528421052631579e-07,
|
| 364 |
+
"loss": 0.3158,
|
| 365 |
+
"step": 520
|
| 366 |
+
},
|
| 367 |
+
{
|
| 368 |
+
"epoch": 88.3,
|
| 369 |
+
"learning_rate": 1.4968421052631578e-07,
|
| 370 |
+
"loss": 0.304,
|
| 371 |
+
"step": 530
|
| 372 |
+
},
|
| 373 |
+
{
|
| 374 |
+
"epoch": 89.89,
|
| 375 |
+
"learning_rate": 1.4652631578947367e-07,
|
| 376 |
+
"loss": 0.2782,
|
| 377 |
+
"step": 540
|
| 378 |
+
},
|
| 379 |
+
{
|
| 380 |
+
"epoch": 91.59,
|
| 381 |
+
"learning_rate": 1.4336842105263157e-07,
|
| 382 |
+
"loss": 0.293,
|
| 383 |
+
"step": 550
|
| 384 |
+
},
|
| 385 |
+
{
|
| 386 |
+
"epoch": 93.3,
|
| 387 |
+
"learning_rate": 1.4021052631578949e-07,
|
| 388 |
+
"loss": 0.2797,
|
| 389 |
+
"step": 560
|
| 390 |
+
},
|
| 391 |
+
{
|
| 392 |
+
"epoch": 94.89,
|
| 393 |
+
"learning_rate": 1.3705263157894735e-07,
|
| 394 |
+
"loss": 0.2595,
|
| 395 |
+
"step": 570
|
| 396 |
+
},
|
| 397 |
+
{
|
| 398 |
+
"epoch": 96.59,
|
| 399 |
+
"learning_rate": 1.3389473684210527e-07,
|
| 400 |
+
"loss": 0.2648,
|
| 401 |
+
"step": 580
|
| 402 |
+
},
|
| 403 |
+
{
|
| 404 |
+
"epoch": 98.3,
|
| 405 |
+
"learning_rate": 1.3073684210526314e-07,
|
| 406 |
+
"loss": 0.2648,
|
| 407 |
+
"step": 590
|
| 408 |
+
},
|
| 409 |
+
{
|
| 410 |
+
"epoch": 99.89,
|
| 411 |
+
"learning_rate": 1.2757894736842106e-07,
|
| 412 |
+
"loss": 0.238,
|
| 413 |
+
"step": 600
|
| 414 |
+
},
|
| 415 |
+
{
|
| 416 |
+
"epoch": 99.89,
|
| 417 |
+
"eval_loss": 0.9607409238815308,
|
| 418 |
+
"eval_runtime": 2334.4802,
|
| 419 |
+
"eval_samples_per_second": 0.219,
|
| 420 |
+
"eval_steps_per_second": 0.055,
|
| 421 |
+
"eval_wer": 55.37817730936144,
|
| 422 |
+
"step": 600
|
| 423 |
+
},
|
| 424 |
+
{
|
| 425 |
+
"epoch": 101.59,
|
| 426 |
+
"learning_rate": 1.2442105263157895e-07,
|
| 427 |
+
"loss": 0.2493,
|
| 428 |
+
"step": 610
|
| 429 |
+
},
|
| 430 |
+
{
|
| 431 |
+
"epoch": 103.3,
|
| 432 |
+
"learning_rate": 1.2126315789473684e-07,
|
| 433 |
+
"loss": 0.2368,
|
| 434 |
+
"step": 620
|
| 435 |
+
},
|
| 436 |
+
{
|
| 437 |
+
"epoch": 104.89,
|
| 438 |
+
"learning_rate": 1.1810526315789474e-07,
|
| 439 |
+
"loss": 0.2222,
|
| 440 |
+
"step": 630
|
| 441 |
+
},
|
| 442 |
+
{
|
| 443 |
+
"epoch": 106.59,
|
| 444 |
+
"learning_rate": 1.1494736842105262e-07,
|
| 445 |
+
"loss": 0.2325,
|
| 446 |
+
"step": 640
|
| 447 |
+
},
|
| 448 |
+
{
|
| 449 |
+
"epoch": 108.3,
|
| 450 |
+
"learning_rate": 1.1178947368421052e-07,
|
| 451 |
+
"loss": 0.2216,
|
| 452 |
+
"step": 650
|
| 453 |
+
},
|
| 454 |
+
{
|
| 455 |
+
"epoch": 109.89,
|
| 456 |
+
"learning_rate": 1.0863157894736843e-07,
|
| 457 |
+
"loss": 0.2045,
|
| 458 |
+
"step": 660
|
| 459 |
+
},
|
| 460 |
+
{
|
| 461 |
+
"epoch": 111.59,
|
| 462 |
+
"learning_rate": 1.0547368421052631e-07,
|
| 463 |
+
"loss": 0.2168,
|
| 464 |
+
"step": 670
|
| 465 |
+
},
|
| 466 |
+
{
|
| 467 |
+
"epoch": 113.3,
|
| 468 |
+
"learning_rate": 1.0231578947368421e-07,
|
| 469 |
+
"loss": 0.2104,
|
| 470 |
+
"step": 680
|
| 471 |
+
},
|
| 472 |
+
{
|
| 473 |
+
"epoch": 114.89,
|
| 474 |
+
"learning_rate": 9.915789473684209e-08,
|
| 475 |
+
"loss": 0.1888,
|
| 476 |
+
"step": 690
|
| 477 |
+
},
|
| 478 |
+
{
|
| 479 |
+
"epoch": 116.59,
|
| 480 |
+
"learning_rate": 9.6e-08,
|
| 481 |
+
"loss": 0.2014,
|
| 482 |
+
"step": 700
|
| 483 |
+
},
|
| 484 |
+
{
|
| 485 |
+
"epoch": 116.59,
|
| 486 |
+
"eval_loss": 1.007683515548706,
|
| 487 |
+
"eval_runtime": 2341.2281,
|
| 488 |
+
"eval_samples_per_second": 0.219,
|
| 489 |
+
"eval_steps_per_second": 0.055,
|
| 490 |
+
"eval_wer": 55.92064476131432,
|
| 491 |
+
"step": 700
|
| 492 |
+
},
|
| 493 |
+
{
|
| 494 |
+
"epoch": 116.74,
|
| 495 |
+
"step": 701,
|
| 496 |
+
"total_flos": 1.068294730825728e+20,
|
| 497 |
+
"train_loss": 0.0002270207000356938,
|
| 498 |
+
"train_runtime": 19.3242,
|
| 499 |
+
"train_samples_per_second": 2318.342,
|
| 500 |
+
"train_steps_per_second": 36.224
|
| 501 |
+
}
|
| 502 |
+
],
|
| 503 |
+
"max_steps": 700,
|
| 504 |
+
"num_train_epochs": 117,
|
| 505 |
+
"total_flos": 1.068294730825728e+20,
|
| 506 |
+
"trial_name": null,
|
| 507 |
+
"trial_params": null
|
| 508 |
+
}
|