End of training
Browse files
all_results.json
CHANGED
|
@@ -1,12 +1,12 @@
|
|
| 1 |
{
|
| 2 |
-
"epoch":
|
| 3 |
-
"eval_loss": 0.
|
| 4 |
-
"eval_runtime":
|
| 5 |
-
"eval_samples_per_second": 4.
|
| 6 |
"eval_steps_per_second": 0.133,
|
| 7 |
-
"eval_wer": 32.
|
| 8 |
-
"train_loss": 0.
|
| 9 |
-
"train_runtime":
|
| 10 |
-
"train_samples_per_second":
|
| 11 |
-
"train_steps_per_second": 0.
|
| 12 |
}
|
|
|
|
| 1 |
{
|
| 2 |
+
"epoch": 13.04,
|
| 3 |
+
"eval_loss": 0.4333903193473816,
|
| 4 |
+
"eval_runtime": 1109.1543,
|
| 5 |
+
"eval_samples_per_second": 4.246,
|
| 6 |
"eval_steps_per_second": 0.133,
|
| 7 |
+
"eval_wer": 32.04202832343535,
|
| 8 |
+
"train_loss": 0.006580068808048963,
|
| 9 |
+
"train_runtime": 18902.2125,
|
| 10 |
+
"train_samples_per_second": 6.772,
|
| 11 |
+
"train_steps_per_second": 0.106
|
| 12 |
}
|
eval_results.json
CHANGED
|
@@ -1,8 +1,8 @@
|
|
| 1 |
{
|
| 2 |
-
"epoch":
|
| 3 |
-
"eval_loss": 0.
|
| 4 |
-
"eval_runtime":
|
| 5 |
-
"eval_samples_per_second": 4.
|
| 6 |
"eval_steps_per_second": 0.133,
|
| 7 |
-
"eval_wer": 32.
|
| 8 |
}
|
|
|
|
| 1 |
{
|
| 2 |
+
"epoch": 13.04,
|
| 3 |
+
"eval_loss": 0.4333903193473816,
|
| 4 |
+
"eval_runtime": 1109.1543,
|
| 5 |
+
"eval_samples_per_second": 4.246,
|
| 6 |
"eval_steps_per_second": 0.133,
|
| 7 |
+
"eval_wer": 32.04202832343535
|
| 8 |
}
|
runs/Dec14_01-26-33_129-213-23-45/events.out.tfevents.1671001275.129-213-23-45.2845517.2
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:d5080d66f70737b41adf816e2f5508247c99a789dd3fb2b70d2b24bf439b6281
|
| 3 |
+
size 358
|
train_results.json
CHANGED
|
@@ -1,7 +1,7 @@
|
|
| 1 |
{
|
| 2 |
-
"epoch":
|
| 3 |
-
"train_loss": 0.
|
| 4 |
-
"train_runtime":
|
| 5 |
-
"train_samples_per_second":
|
| 6 |
-
"train_steps_per_second": 0.
|
| 7 |
}
|
|
|
|
| 1 |
{
|
| 2 |
+
"epoch": 13.04,
|
| 3 |
+
"train_loss": 0.006580068808048963,
|
| 4 |
+
"train_runtime": 18902.2125,
|
| 5 |
+
"train_samples_per_second": 6.772,
|
| 6 |
+
"train_steps_per_second": 0.106
|
| 7 |
}
|
trainer_state.json
CHANGED
|
@@ -1,310 +1,550 @@
|
|
| 1 |
{
|
| 2 |
-
"best_metric": 32.
|
| 3 |
-
"best_model_checkpoint": "./checkpoint-
|
| 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": 0.
|
| 12 |
-
"learning_rate": 2.
|
| 13 |
-
"loss":
|
| 14 |
"step": 25
|
| 15 |
},
|
| 16 |
{
|
| 17 |
-
"epoch": 0.
|
| 18 |
-
"learning_rate":
|
| 19 |
-
"loss": 0.
|
| 20 |
"step": 50
|
| 21 |
},
|
| 22 |
{
|
| 23 |
-
"epoch": 0.
|
| 24 |
-
"learning_rate":
|
| 25 |
-
"loss": 0.
|
| 26 |
"step": 75
|
| 27 |
},
|
| 28 |
{
|
| 29 |
-
"epoch": 0.
|
| 30 |
-
"learning_rate":
|
| 31 |
-
"loss": 0.
|
| 32 |
"step": 100
|
| 33 |
},
|
| 34 |
{
|
| 35 |
-
"epoch": 0.
|
| 36 |
-
"learning_rate": 1.
|
| 37 |
-
"loss": 0.
|
| 38 |
"step": 125
|
| 39 |
},
|
| 40 |
{
|
| 41 |
"epoch": 1.0,
|
| 42 |
-
"learning_rate": 1.
|
| 43 |
-
"loss": 0.
|
| 44 |
"step": 150
|
| 45 |
},
|
| 46 |
{
|
| 47 |
-
"epoch": 1.
|
| 48 |
-
"learning_rate":
|
| 49 |
-
"loss": 0.
|
| 50 |
"step": 175
|
| 51 |
},
|
| 52 |
{
|
| 53 |
-
"epoch": 1.
|
| 54 |
-
"learning_rate":
|
| 55 |
-
"loss": 0.
|
| 56 |
-
"step": 200
|
| 57 |
-
},
|
| 58 |
-
{
|
| 59 |
-
"epoch": 1.05,
|
| 60 |
-
"eval_loss": 0.3499109745025635,
|
| 61 |
-
"eval_runtime": 1155.2135,
|
| 62 |
-
"eval_samples_per_second": 4.076,
|
| 63 |
-
"eval_steps_per_second": 0.128,
|
| 64 |
-
"eval_wer": 41.93238921882138,
|
| 65 |
"step": 200
|
| 66 |
},
|
| 67 |
{
|
| 68 |
-
"epoch": 1.
|
| 69 |
-
"learning_rate":
|
| 70 |
-
"loss": 0.
|
| 71 |
"step": 225
|
| 72 |
},
|
| 73 |
{
|
| 74 |
-
"epoch": 1.
|
| 75 |
-
"learning_rate":
|
| 76 |
-
"loss": 0.
|
| 77 |
"step": 250
|
| 78 |
},
|
| 79 |
{
|
| 80 |
-
"epoch": 1.
|
| 81 |
-
"learning_rate":
|
| 82 |
-
"loss": 0.
|
| 83 |
"step": 275
|
| 84 |
},
|
| 85 |
{
|
| 86 |
"epoch": 2.0,
|
| 87 |
-
"learning_rate":
|
| 88 |
-
"loss": 0.
|
| 89 |
"step": 300
|
| 90 |
},
|
| 91 |
{
|
| 92 |
-
"epoch": 2.
|
| 93 |
-
"learning_rate":
|
| 94 |
-
"loss": 0.
|
| 95 |
"step": 325
|
| 96 |
},
|
| 97 |
{
|
| 98 |
-
"epoch": 2.
|
| 99 |
-
"learning_rate":
|
| 100 |
-
"loss": 0.
|
| 101 |
"step": 350
|
| 102 |
},
|
| 103 |
{
|
| 104 |
-
"epoch": 2.
|
| 105 |
-
"learning_rate":
|
| 106 |
-
"loss": 0.
|
| 107 |
"step": 375
|
| 108 |
},
|
| 109 |
{
|
| 110 |
-
"epoch": 2.
|
| 111 |
-
"learning_rate":
|
| 112 |
-
"loss": 0.
|
| 113 |
"step": 400
|
| 114 |
},
|
| 115 |
{
|
| 116 |
-
"epoch": 2.
|
| 117 |
-
"eval_loss": 0.
|
| 118 |
-
"eval_runtime":
|
| 119 |
-
"eval_samples_per_second": 4.
|
| 120 |
-
"eval_steps_per_second": 0.
|
| 121 |
-
"eval_wer":
|
| 122 |
"step": 400
|
| 123 |
},
|
| 124 |
{
|
| 125 |
-
"epoch": 2.
|
| 126 |
-
"learning_rate":
|
| 127 |
-
"loss": 0.
|
| 128 |
"step": 425
|
| 129 |
},
|
| 130 |
{
|
| 131 |
-
"epoch": 3.
|
| 132 |
-
"learning_rate":
|
| 133 |
-
"loss": 0.
|
| 134 |
"step": 450
|
| 135 |
},
|
| 136 |
{
|
| 137 |
-
"epoch": 3.
|
| 138 |
-
"learning_rate":
|
| 139 |
-
"loss": 0.
|
| 140 |
"step": 475
|
| 141 |
},
|
| 142 |
{
|
| 143 |
-
"epoch": 3.
|
| 144 |
-
"learning_rate":
|
| 145 |
-
"loss": 0.
|
| 146 |
"step": 500
|
| 147 |
},
|
| 148 |
{
|
| 149 |
-
"epoch": 3.
|
| 150 |
-
"learning_rate":
|
| 151 |
-
"loss": 0.
|
| 152 |
"step": 525
|
| 153 |
},
|
| 154 |
{
|
| 155 |
-
"epoch": 3.
|
| 156 |
-
"learning_rate":
|
| 157 |
-
"loss": 0.
|
| 158 |
"step": 550
|
| 159 |
},
|
| 160 |
{
|
| 161 |
-
"epoch": 3.
|
| 162 |
-
"learning_rate":
|
| 163 |
-
"loss": 0.
|
| 164 |
"step": 575
|
| 165 |
},
|
| 166 |
{
|
| 167 |
-
"epoch": 4.
|
| 168 |
-
"learning_rate":
|
| 169 |
-
"loss": 0.
|
| 170 |
-
"step": 600
|
| 171 |
-
},
|
| 172 |
-
{
|
| 173 |
-
"epoch": 4.01,
|
| 174 |
-
"eval_loss": 0.29274308681488037,
|
| 175 |
-
"eval_runtime": 1130.2455,
|
| 176 |
-
"eval_samples_per_second": 4.166,
|
| 177 |
-
"eval_steps_per_second": 0.131,
|
| 178 |
-
"eval_wer": 33.72316126084971,
|
| 179 |
"step": 600
|
| 180 |
},
|
| 181 |
{
|
| 182 |
-
"epoch": 4.
|
| 183 |
-
"learning_rate":
|
| 184 |
-
"loss": 0.
|
| 185 |
"step": 625
|
| 186 |
},
|
| 187 |
{
|
| 188 |
-
"epoch": 4.
|
| 189 |
-
"learning_rate":
|
| 190 |
-
"loss": 0.
|
| 191 |
"step": 650
|
| 192 |
},
|
| 193 |
{
|
| 194 |
-
"epoch": 4.
|
| 195 |
-
"learning_rate":
|
| 196 |
-
"loss": 0.
|
| 197 |
"step": 675
|
| 198 |
},
|
| 199 |
{
|
| 200 |
-
"epoch": 4.
|
| 201 |
-
"learning_rate":
|
| 202 |
-
"loss": 0.
|
| 203 |
"step": 700
|
| 204 |
},
|
| 205 |
{
|
| 206 |
-
"epoch": 4.
|
| 207 |
-
"learning_rate":
|
| 208 |
-
"loss": 0.
|
| 209 |
"step": 725
|
| 210 |
},
|
| 211 |
{
|
| 212 |
-
"epoch": 5.
|
| 213 |
-
"learning_rate":
|
| 214 |
-
"loss": 0.
|
| 215 |
"step": 750
|
| 216 |
},
|
| 217 |
{
|
| 218 |
-
"epoch": 5.
|
| 219 |
-
"learning_rate":
|
| 220 |
-
"loss": 0.
|
| 221 |
"step": 775
|
| 222 |
},
|
| 223 |
{
|
| 224 |
-
"epoch": 5.
|
| 225 |
-
"learning_rate":
|
| 226 |
-
"loss": 0.
|
| 227 |
"step": 800
|
| 228 |
},
|
| 229 |
{
|
| 230 |
-
"epoch": 5.
|
| 231 |
-
"eval_loss": 0.
|
| 232 |
-
"eval_runtime":
|
| 233 |
-
"eval_samples_per_second": 4.
|
| 234 |
-
"eval_steps_per_second": 0.
|
| 235 |
-
"eval_wer":
|
| 236 |
"step": 800
|
| 237 |
},
|
| 238 |
{
|
| 239 |
-
"epoch": 5.
|
| 240 |
-
"learning_rate":
|
| 241 |
-
"loss": 0.
|
| 242 |
"step": 825
|
| 243 |
},
|
| 244 |
{
|
| 245 |
-
"epoch": 5.
|
| 246 |
-
"learning_rate": 7.
|
| 247 |
-
"loss": 0.
|
| 248 |
"step": 850
|
| 249 |
},
|
| 250 |
{
|
| 251 |
-
"epoch": 5.
|
| 252 |
-
"learning_rate":
|
| 253 |
-
"loss": 0.
|
| 254 |
"step": 875
|
| 255 |
},
|
| 256 |
{
|
| 257 |
"epoch": 6.01,
|
| 258 |
-
"learning_rate":
|
| 259 |
-
"loss": 0.
|
| 260 |
"step": 900
|
| 261 |
},
|
| 262 |
{
|
| 263 |
-
"epoch": 6.
|
| 264 |
-
"learning_rate":
|
| 265 |
-
"loss": 0.
|
| 266 |
"step": 925
|
| 267 |
},
|
| 268 |
{
|
| 269 |
-
"epoch": 6.
|
| 270 |
-
"learning_rate":
|
| 271 |
-
"loss": 0.
|
| 272 |
"step": 950
|
| 273 |
},
|
| 274 |
{
|
| 275 |
-
"epoch": 6.
|
| 276 |
-
"learning_rate":
|
| 277 |
-
"loss": 0.
|
| 278 |
"step": 975
|
| 279 |
},
|
| 280 |
{
|
| 281 |
-
"epoch": 6.
|
| 282 |
-
"learning_rate":
|
| 283 |
-
"loss": 0.
|
| 284 |
"step": 1000
|
| 285 |
},
|
| 286 |
{
|
| 287 |
-
"epoch": 6.
|
| 288 |
-
"
|
| 289 |
-
"
|
| 290 |
-
"
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 291 |
"eval_steps_per_second": 0.131,
|
| 292 |
-
"eval_wer":
|
| 293 |
-
"step":
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 294 |
},
|
| 295 |
{
|
| 296 |
-
"epoch":
|
| 297 |
-
"step":
|
| 298 |
-
"total_flos":
|
| 299 |
-
"train_loss": 0.
|
| 300 |
-
"train_runtime":
|
| 301 |
-
"train_samples_per_second":
|
| 302 |
-
"train_steps_per_second": 0.
|
| 303 |
}
|
| 304 |
],
|
| 305 |
-
"max_steps":
|
| 306 |
"num_train_epochs": 9223372036854775807,
|
| 307 |
-
"total_flos":
|
| 308 |
"trial_name": null,
|
| 309 |
"trial_params": null
|
| 310 |
}
|
|
|
|
| 1 |
{
|
| 2 |
+
"best_metric": 32.04202832343535,
|
| 3 |
+
"best_model_checkpoint": "./checkpoint-2000",
|
| 4 |
+
"epoch": 13.038,
|
| 5 |
+
"global_step": 2000,
|
| 6 |
"is_hyper_param_search": false,
|
| 7 |
"is_local_process_zero": true,
|
| 8 |
"is_world_process_zero": true,
|
| 9 |
"log_history": [
|
| 10 |
{
|
| 11 |
+
"epoch": 0.01,
|
| 12 |
+
"learning_rate": 2.0833333333333333e-07,
|
| 13 |
+
"loss": 0.011,
|
| 14 |
"step": 25
|
| 15 |
},
|
| 16 |
{
|
| 17 |
+
"epoch": 0.03,
|
| 18 |
+
"learning_rate": 4.1666666666666667e-07,
|
| 19 |
+
"loss": 0.0111,
|
| 20 |
"step": 50
|
| 21 |
},
|
| 22 |
{
|
| 23 |
+
"epoch": 0.04,
|
| 24 |
+
"learning_rate": 6.25e-07,
|
| 25 |
+
"loss": 0.0111,
|
| 26 |
"step": 75
|
| 27 |
},
|
| 28 |
{
|
| 29 |
+
"epoch": 0.05,
|
| 30 |
+
"learning_rate": 8.333333333333333e-07,
|
| 31 |
+
"loss": 0.0132,
|
| 32 |
"step": 100
|
| 33 |
},
|
| 34 |
{
|
| 35 |
+
"epoch": 0.06,
|
| 36 |
+
"learning_rate": 1.0416666666666667e-06,
|
| 37 |
+
"loss": 0.0139,
|
| 38 |
"step": 125
|
| 39 |
},
|
| 40 |
{
|
| 41 |
"epoch": 1.0,
|
| 42 |
+
"learning_rate": 1.25e-06,
|
| 43 |
+
"loss": 0.013,
|
| 44 |
"step": 150
|
| 45 |
},
|
| 46 |
{
|
| 47 |
+
"epoch": 1.01,
|
| 48 |
+
"learning_rate": 1.4583333333333335e-06,
|
| 49 |
+
"loss": 0.0092,
|
| 50 |
"step": 175
|
| 51 |
},
|
| 52 |
{
|
| 53 |
+
"epoch": 1.03,
|
| 54 |
+
"learning_rate": 1.6666666666666667e-06,
|
| 55 |
+
"loss": 0.0098,
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 56 |
"step": 200
|
| 57 |
},
|
| 58 |
{
|
| 59 |
+
"epoch": 1.04,
|
| 60 |
+
"learning_rate": 1.8750000000000003e-06,
|
| 61 |
+
"loss": 0.0091,
|
| 62 |
"step": 225
|
| 63 |
},
|
| 64 |
{
|
| 65 |
+
"epoch": 1.05,
|
| 66 |
+
"learning_rate": 2.0833333333333334e-06,
|
| 67 |
+
"loss": 0.0101,
|
| 68 |
"step": 250
|
| 69 |
},
|
| 70 |
{
|
| 71 |
+
"epoch": 1.06,
|
| 72 |
+
"learning_rate": 2.2916666666666666e-06,
|
| 73 |
+
"loss": 0.0092,
|
| 74 |
"step": 275
|
| 75 |
},
|
| 76 |
{
|
| 77 |
"epoch": 2.0,
|
| 78 |
+
"learning_rate": 2.5e-06,
|
| 79 |
+
"loss": 0.0084,
|
| 80 |
"step": 300
|
| 81 |
},
|
| 82 |
{
|
| 83 |
+
"epoch": 2.01,
|
| 84 |
+
"learning_rate": 2.7083333333333334e-06,
|
| 85 |
+
"loss": 0.0067,
|
| 86 |
"step": 325
|
| 87 |
},
|
| 88 |
{
|
| 89 |
+
"epoch": 2.03,
|
| 90 |
+
"learning_rate": 2.916666666666667e-06,
|
| 91 |
+
"loss": 0.0066,
|
| 92 |
"step": 350
|
| 93 |
},
|
| 94 |
{
|
| 95 |
+
"epoch": 2.04,
|
| 96 |
+
"learning_rate": 3.125e-06,
|
| 97 |
+
"loss": 0.0066,
|
| 98 |
"step": 375
|
| 99 |
},
|
| 100 |
{
|
| 101 |
+
"epoch": 2.05,
|
| 102 |
+
"learning_rate": 3.3333333333333333e-06,
|
| 103 |
+
"loss": 0.0066,
|
| 104 |
"step": 400
|
| 105 |
},
|
| 106 |
{
|
| 107 |
+
"epoch": 2.05,
|
| 108 |
+
"eval_loss": 0.374257892370224,
|
| 109 |
+
"eval_runtime": 1097.1851,
|
| 110 |
+
"eval_samples_per_second": 4.292,
|
| 111 |
+
"eval_steps_per_second": 0.135,
|
| 112 |
+
"eval_wer": 32.910004568296024,
|
| 113 |
"step": 400
|
| 114 |
},
|
| 115 |
{
|
| 116 |
+
"epoch": 2.06,
|
| 117 |
+
"learning_rate": 3.5416666666666673e-06,
|
| 118 |
+
"loss": 0.0059,
|
| 119 |
"step": 425
|
| 120 |
},
|
| 121 |
{
|
| 122 |
+
"epoch": 3.0,
|
| 123 |
+
"learning_rate": 3.7500000000000005e-06,
|
| 124 |
+
"loss": 0.0053,
|
| 125 |
"step": 450
|
| 126 |
},
|
| 127 |
{
|
| 128 |
+
"epoch": 3.02,
|
| 129 |
+
"learning_rate": 3.958333333333333e-06,
|
| 130 |
+
"loss": 0.0046,
|
| 131 |
"step": 475
|
| 132 |
},
|
| 133 |
{
|
| 134 |
+
"epoch": 3.03,
|
| 135 |
+
"learning_rate": 4.166666666666667e-06,
|
| 136 |
+
"loss": 0.0052,
|
| 137 |
"step": 500
|
| 138 |
},
|
| 139 |
{
|
| 140 |
+
"epoch": 3.04,
|
| 141 |
+
"learning_rate": 4.3750000000000005e-06,
|
| 142 |
+
"loss": 0.0049,
|
| 143 |
"step": 525
|
| 144 |
},
|
| 145 |
{
|
| 146 |
+
"epoch": 3.05,
|
| 147 |
+
"learning_rate": 4.583333333333333e-06,
|
| 148 |
+
"loss": 0.0051,
|
| 149 |
"step": 550
|
| 150 |
},
|
| 151 |
{
|
| 152 |
+
"epoch": 3.07,
|
| 153 |
+
"learning_rate": 4.791666666666668e-06,
|
| 154 |
+
"loss": 0.0047,
|
| 155 |
"step": 575
|
| 156 |
},
|
| 157 |
{
|
| 158 |
+
"epoch": 4.0,
|
| 159 |
+
"learning_rate": 5e-06,
|
| 160 |
+
"loss": 0.0052,
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 161 |
"step": 600
|
| 162 |
},
|
| 163 |
{
|
| 164 |
+
"epoch": 4.02,
|
| 165 |
+
"learning_rate": 5.208333333333334e-06,
|
| 166 |
+
"loss": 0.0049,
|
| 167 |
"step": 625
|
| 168 |
},
|
| 169 |
{
|
| 170 |
+
"epoch": 4.03,
|
| 171 |
+
"learning_rate": 5.416666666666667e-06,
|
| 172 |
+
"loss": 0.0052,
|
| 173 |
"step": 650
|
| 174 |
},
|
| 175 |
{
|
| 176 |
+
"epoch": 4.04,
|
| 177 |
+
"learning_rate": 5.625e-06,
|
| 178 |
+
"loss": 0.0059,
|
| 179 |
"step": 675
|
| 180 |
},
|
| 181 |
{
|
| 182 |
+
"epoch": 4.05,
|
| 183 |
+
"learning_rate": 5.833333333333334e-06,
|
| 184 |
+
"loss": 0.0084,
|
| 185 |
"step": 700
|
| 186 |
},
|
| 187 |
{
|
| 188 |
+
"epoch": 4.07,
|
| 189 |
+
"learning_rate": 6.041666666666667e-06,
|
| 190 |
+
"loss": 0.0101,
|
| 191 |
"step": 725
|
| 192 |
},
|
| 193 |
{
|
| 194 |
+
"epoch": 5.0,
|
| 195 |
+
"learning_rate": 6.25e-06,
|
| 196 |
+
"loss": 0.0078,
|
| 197 |
"step": 750
|
| 198 |
},
|
| 199 |
{
|
| 200 |
+
"epoch": 5.02,
|
| 201 |
+
"learning_rate": 6.458333333333334e-06,
|
| 202 |
+
"loss": 0.0077,
|
| 203 |
"step": 775
|
| 204 |
},
|
| 205 |
{
|
| 206 |
+
"epoch": 5.03,
|
| 207 |
+
"learning_rate": 6.666666666666667e-06,
|
| 208 |
+
"loss": 0.0084,
|
| 209 |
"step": 800
|
| 210 |
},
|
| 211 |
{
|
| 212 |
+
"epoch": 5.03,
|
| 213 |
+
"eval_loss": 0.3786599338054657,
|
| 214 |
+
"eval_runtime": 1048.12,
|
| 215 |
+
"eval_samples_per_second": 4.493,
|
| 216 |
+
"eval_steps_per_second": 0.141,
|
| 217 |
+
"eval_wer": 33.41708542713568,
|
| 218 |
"step": 800
|
| 219 |
},
|
| 220 |
{
|
| 221 |
+
"epoch": 5.04,
|
| 222 |
+
"learning_rate": 6.875e-06,
|
| 223 |
+
"loss": 0.0098,
|
| 224 |
"step": 825
|
| 225 |
},
|
| 226 |
{
|
| 227 |
+
"epoch": 5.05,
|
| 228 |
+
"learning_rate": 7.083333333333335e-06,
|
| 229 |
+
"loss": 0.0097,
|
| 230 |
"step": 850
|
| 231 |
},
|
| 232 |
{
|
| 233 |
+
"epoch": 5.07,
|
| 234 |
+
"learning_rate": 7.291666666666667e-06,
|
| 235 |
+
"loss": 0.0093,
|
| 236 |
"step": 875
|
| 237 |
},
|
| 238 |
{
|
| 239 |
"epoch": 6.01,
|
| 240 |
+
"learning_rate": 7.500000000000001e-06,
|
| 241 |
+
"loss": 0.0092,
|
| 242 |
"step": 900
|
| 243 |
},
|
| 244 |
{
|
| 245 |
+
"epoch": 6.02,
|
| 246 |
+
"learning_rate": 7.708333333333334e-06,
|
| 247 |
+
"loss": 0.0086,
|
| 248 |
"step": 925
|
| 249 |
},
|
| 250 |
{
|
| 251 |
+
"epoch": 6.03,
|
| 252 |
+
"learning_rate": 7.916666666666667e-06,
|
| 253 |
+
"loss": 0.0087,
|
| 254 |
"step": 950
|
| 255 |
},
|
| 256 |
{
|
| 257 |
+
"epoch": 6.04,
|
| 258 |
+
"learning_rate": 8.125000000000001e-06,
|
| 259 |
+
"loss": 0.0091,
|
| 260 |
"step": 975
|
| 261 |
},
|
| 262 |
{
|
| 263 |
+
"epoch": 6.06,
|
| 264 |
+
"learning_rate": 8.333333333333334e-06,
|
| 265 |
+
"loss": 0.0119,
|
| 266 |
"step": 1000
|
| 267 |
},
|
| 268 |
{
|
| 269 |
+
"epoch": 6.07,
|
| 270 |
+
"learning_rate": 8.541666666666666e-06,
|
| 271 |
+
"loss": 0.0115,
|
| 272 |
+
"step": 1025
|
| 273 |
+
},
|
| 274 |
+
{
|
| 275 |
+
"epoch": 7.01,
|
| 276 |
+
"learning_rate": 8.750000000000001e-06,
|
| 277 |
+
"loss": 0.0093,
|
| 278 |
+
"step": 1050
|
| 279 |
+
},
|
| 280 |
+
{
|
| 281 |
+
"epoch": 7.02,
|
| 282 |
+
"learning_rate": 8.958333333333334e-06,
|
| 283 |
+
"loss": 0.01,
|
| 284 |
+
"step": 1075
|
| 285 |
+
},
|
| 286 |
+
{
|
| 287 |
+
"epoch": 7.03,
|
| 288 |
+
"learning_rate": 9.166666666666666e-06,
|
| 289 |
+
"loss": 0.0101,
|
| 290 |
+
"step": 1100
|
| 291 |
+
},
|
| 292 |
+
{
|
| 293 |
+
"epoch": 7.04,
|
| 294 |
+
"learning_rate": 9.375000000000001e-06,
|
| 295 |
+
"loss": 0.0095,
|
| 296 |
+
"step": 1125
|
| 297 |
+
},
|
| 298 |
+
{
|
| 299 |
+
"epoch": 7.06,
|
| 300 |
+
"learning_rate": 9.583333333333335e-06,
|
| 301 |
+
"loss": 0.0103,
|
| 302 |
+
"step": 1150
|
| 303 |
+
},
|
| 304 |
+
{
|
| 305 |
+
"epoch": 7.07,
|
| 306 |
+
"learning_rate": 9.791666666666666e-06,
|
| 307 |
+
"loss": 0.0111,
|
| 308 |
+
"step": 1175
|
| 309 |
+
},
|
| 310 |
+
{
|
| 311 |
+
"epoch": 8.01,
|
| 312 |
+
"learning_rate": 1e-05,
|
| 313 |
+
"loss": 0.0098,
|
| 314 |
+
"step": 1200
|
| 315 |
+
},
|
| 316 |
+
{
|
| 317 |
+
"epoch": 8.01,
|
| 318 |
+
"eval_loss": 0.397890567779541,
|
| 319 |
+
"eval_runtime": 1133.829,
|
| 320 |
+
"eval_samples_per_second": 4.153,
|
| 321 |
"eval_steps_per_second": 0.131,
|
| 322 |
+
"eval_wer": 33.24805847418913,
|
| 323 |
+
"step": 1200
|
| 324 |
+
},
|
| 325 |
+
{
|
| 326 |
+
"epoch": 8.02,
|
| 327 |
+
"learning_rate": 9.6875e-06,
|
| 328 |
+
"loss": 0.0106,
|
| 329 |
+
"step": 1225
|
| 330 |
+
},
|
| 331 |
+
{
|
| 332 |
+
"epoch": 8.03,
|
| 333 |
+
"learning_rate": 9.375000000000001e-06,
|
| 334 |
+
"loss": 0.009,
|
| 335 |
+
"step": 1250
|
| 336 |
+
},
|
| 337 |
+
{
|
| 338 |
+
"epoch": 8.05,
|
| 339 |
+
"learning_rate": 9.0625e-06,
|
| 340 |
+
"loss": 0.0093,
|
| 341 |
+
"step": 1275
|
| 342 |
+
},
|
| 343 |
+
{
|
| 344 |
+
"epoch": 8.06,
|
| 345 |
+
"learning_rate": 8.750000000000001e-06,
|
| 346 |
+
"loss": 0.0087,
|
| 347 |
+
"step": 1300
|
| 348 |
+
},
|
| 349 |
+
{
|
| 350 |
+
"epoch": 8.07,
|
| 351 |
+
"learning_rate": 8.4375e-06,
|
| 352 |
+
"loss": 0.0098,
|
| 353 |
+
"step": 1325
|
| 354 |
+
},
|
| 355 |
+
{
|
| 356 |
+
"epoch": 9.01,
|
| 357 |
+
"learning_rate": 8.125000000000001e-06,
|
| 358 |
+
"loss": 0.0057,
|
| 359 |
+
"step": 1350
|
| 360 |
+
},
|
| 361 |
+
{
|
| 362 |
+
"epoch": 9.02,
|
| 363 |
+
"learning_rate": 7.8125e-06,
|
| 364 |
+
"loss": 0.0053,
|
| 365 |
+
"step": 1375
|
| 366 |
+
},
|
| 367 |
+
{
|
| 368 |
+
"epoch": 9.03,
|
| 369 |
+
"learning_rate": 7.500000000000001e-06,
|
| 370 |
+
"loss": 0.0061,
|
| 371 |
+
"step": 1400
|
| 372 |
+
},
|
| 373 |
+
{
|
| 374 |
+
"epoch": 9.05,
|
| 375 |
+
"learning_rate": 7.1875e-06,
|
| 376 |
+
"loss": 0.007,
|
| 377 |
+
"step": 1425
|
| 378 |
+
},
|
| 379 |
+
{
|
| 380 |
+
"epoch": 9.06,
|
| 381 |
+
"learning_rate": 6.875e-06,
|
| 382 |
+
"loss": 0.0049,
|
| 383 |
+
"step": 1450
|
| 384 |
+
},
|
| 385 |
+
{
|
| 386 |
+
"epoch": 9.07,
|
| 387 |
+
"learning_rate": 6.5625e-06,
|
| 388 |
+
"loss": 0.0042,
|
| 389 |
+
"step": 1475
|
| 390 |
+
},
|
| 391 |
+
{
|
| 392 |
+
"epoch": 10.01,
|
| 393 |
+
"learning_rate": 6.25e-06,
|
| 394 |
+
"loss": 0.003,
|
| 395 |
+
"step": 1500
|
| 396 |
+
},
|
| 397 |
+
{
|
| 398 |
+
"epoch": 10.02,
|
| 399 |
+
"learning_rate": 5.9375e-06,
|
| 400 |
+
"loss": 0.0048,
|
| 401 |
+
"step": 1525
|
| 402 |
+
},
|
| 403 |
+
{
|
| 404 |
+
"epoch": 10.04,
|
| 405 |
+
"learning_rate": 5.625e-06,
|
| 406 |
+
"loss": 0.0025,
|
| 407 |
+
"step": 1550
|
| 408 |
+
},
|
| 409 |
+
{
|
| 410 |
+
"epoch": 10.05,
|
| 411 |
+
"learning_rate": 5.3125e-06,
|
| 412 |
+
"loss": 0.0028,
|
| 413 |
+
"step": 1575
|
| 414 |
+
},
|
| 415 |
+
{
|
| 416 |
+
"epoch": 10.06,
|
| 417 |
+
"learning_rate": 5e-06,
|
| 418 |
+
"loss": 0.0019,
|
| 419 |
+
"step": 1600
|
| 420 |
+
},
|
| 421 |
+
{
|
| 422 |
+
"epoch": 10.06,
|
| 423 |
+
"eval_loss": 0.40835869312286377,
|
| 424 |
+
"eval_runtime": 1079.5315,
|
| 425 |
+
"eval_samples_per_second": 4.362,
|
| 426 |
+
"eval_steps_per_second": 0.137,
|
| 427 |
+
"eval_wer": 32.311557788944725,
|
| 428 |
+
"step": 1600
|
| 429 |
+
},
|
| 430 |
+
{
|
| 431 |
+
"epoch": 10.07,
|
| 432 |
+
"learning_rate": 4.6875000000000004e-06,
|
| 433 |
+
"loss": 0.0024,
|
| 434 |
+
"step": 1625
|
| 435 |
+
},
|
| 436 |
+
{
|
| 437 |
+
"epoch": 11.01,
|
| 438 |
+
"learning_rate": 4.3750000000000005e-06,
|
| 439 |
+
"loss": 0.0013,
|
| 440 |
+
"step": 1650
|
| 441 |
+
},
|
| 442 |
+
{
|
| 443 |
+
"epoch": 11.02,
|
| 444 |
+
"learning_rate": 4.0625000000000005e-06,
|
| 445 |
+
"loss": 0.0017,
|
| 446 |
+
"step": 1675
|
| 447 |
+
},
|
| 448 |
+
{
|
| 449 |
+
"epoch": 11.04,
|
| 450 |
+
"learning_rate": 3.7500000000000005e-06,
|
| 451 |
+
"loss": 0.0012,
|
| 452 |
+
"step": 1700
|
| 453 |
+
},
|
| 454 |
+
{
|
| 455 |
+
"epoch": 11.05,
|
| 456 |
+
"learning_rate": 3.4375e-06,
|
| 457 |
+
"loss": 0.0012,
|
| 458 |
+
"step": 1725
|
| 459 |
+
},
|
| 460 |
+
{
|
| 461 |
+
"epoch": 11.06,
|
| 462 |
+
"learning_rate": 3.125e-06,
|
| 463 |
+
"loss": 0.001,
|
| 464 |
+
"step": 1750
|
| 465 |
+
},
|
| 466 |
+
{
|
| 467 |
+
"epoch": 11.07,
|
| 468 |
+
"learning_rate": 2.8125e-06,
|
| 469 |
+
"loss": 0.0011,
|
| 470 |
+
"step": 1775
|
| 471 |
+
},
|
| 472 |
+
{
|
| 473 |
+
"epoch": 12.01,
|
| 474 |
+
"learning_rate": 2.5e-06,
|
| 475 |
+
"loss": 0.0009,
|
| 476 |
+
"step": 1800
|
| 477 |
+
},
|
| 478 |
+
{
|
| 479 |
+
"epoch": 12.02,
|
| 480 |
+
"learning_rate": 2.1875000000000002e-06,
|
| 481 |
+
"loss": 0.001,
|
| 482 |
+
"step": 1825
|
| 483 |
+
},
|
| 484 |
+
{
|
| 485 |
+
"epoch": 12.04,
|
| 486 |
+
"learning_rate": 1.8750000000000003e-06,
|
| 487 |
+
"loss": 0.0008,
|
| 488 |
+
"step": 1850
|
| 489 |
+
},
|
| 490 |
+
{
|
| 491 |
+
"epoch": 12.05,
|
| 492 |
+
"learning_rate": 1.5625e-06,
|
| 493 |
+
"loss": 0.0008,
|
| 494 |
+
"step": 1875
|
| 495 |
+
},
|
| 496 |
+
{
|
| 497 |
+
"epoch": 12.06,
|
| 498 |
+
"learning_rate": 1.25e-06,
|
| 499 |
+
"loss": 0.0009,
|
| 500 |
+
"step": 1900
|
| 501 |
+
},
|
| 502 |
+
{
|
| 503 |
+
"epoch": 13.0,
|
| 504 |
+
"learning_rate": 9.375000000000001e-07,
|
| 505 |
+
"loss": 0.0009,
|
| 506 |
+
"step": 1925
|
| 507 |
+
},
|
| 508 |
+
{
|
| 509 |
+
"epoch": 13.01,
|
| 510 |
+
"learning_rate": 6.25e-07,
|
| 511 |
+
"loss": 0.0008,
|
| 512 |
+
"step": 1950
|
| 513 |
+
},
|
| 514 |
+
{
|
| 515 |
+
"epoch": 13.03,
|
| 516 |
+
"learning_rate": 3.125e-07,
|
| 517 |
+
"loss": 0.0009,
|
| 518 |
+
"step": 1975
|
| 519 |
+
},
|
| 520 |
+
{
|
| 521 |
+
"epoch": 13.04,
|
| 522 |
+
"learning_rate": 0.0,
|
| 523 |
+
"loss": 0.0008,
|
| 524 |
+
"step": 2000
|
| 525 |
+
},
|
| 526 |
+
{
|
| 527 |
+
"epoch": 13.04,
|
| 528 |
+
"eval_loss": 0.4333903193473816,
|
| 529 |
+
"eval_runtime": 1067.3464,
|
| 530 |
+
"eval_samples_per_second": 4.412,
|
| 531 |
+
"eval_steps_per_second": 0.139,
|
| 532 |
+
"eval_wer": 32.04202832343535,
|
| 533 |
+
"step": 2000
|
| 534 |
},
|
| 535 |
{
|
| 536 |
+
"epoch": 13.04,
|
| 537 |
+
"step": 2000,
|
| 538 |
+
"total_flos": 3.673634428993536e+19,
|
| 539 |
+
"train_loss": 0.006580068808048963,
|
| 540 |
+
"train_runtime": 18902.2125,
|
| 541 |
+
"train_samples_per_second": 6.772,
|
| 542 |
+
"train_steps_per_second": 0.106
|
| 543 |
}
|
| 544 |
],
|
| 545 |
+
"max_steps": 2000,
|
| 546 |
"num_train_epochs": 9223372036854775807,
|
| 547 |
+
"total_flos": 3.673634428993536e+19,
|
| 548 |
"trial_name": null,
|
| 549 |
"trial_params": null
|
| 550 |
}
|