🍻 cheers
Browse files- all_results.json +6 -6
- train_results.json +6 -6
- trainer_state.json +667 -166
all_results.json
CHANGED
|
@@ -1,8 +1,8 @@
|
|
| 1 |
{
|
| 2 |
-
"epoch":
|
| 3 |
-
"total_flos":
|
| 4 |
-
"train_loss": 0.
|
| 5 |
-
"train_runtime":
|
| 6 |
-
"train_samples_per_second":
|
| 7 |
-
"train_steps_per_second":
|
| 8 |
}
|
|
|
|
| 1 |
{
|
| 2 |
+
"epoch": 12.0,
|
| 3 |
+
"total_flos": 1.4460005262082867e+18,
|
| 4 |
+
"train_loss": 0.013737765663116536,
|
| 5 |
+
"train_runtime": 2175.34,
|
| 6 |
+
"train_samples_per_second": 8.578,
|
| 7 |
+
"train_steps_per_second": 0.541
|
| 8 |
}
|
train_results.json
CHANGED
|
@@ -1,8 +1,8 @@
|
|
| 1 |
{
|
| 2 |
-
"epoch":
|
| 3 |
-
"total_flos":
|
| 4 |
-
"train_loss": 0.
|
| 5 |
-
"train_runtime":
|
| 6 |
-
"train_samples_per_second":
|
| 7 |
-
"train_steps_per_second":
|
| 8 |
}
|
|
|
|
| 1 |
{
|
| 2 |
+
"epoch": 12.0,
|
| 3 |
+
"total_flos": 1.4460005262082867e+18,
|
| 4 |
+
"train_loss": 0.013737765663116536,
|
| 5 |
+
"train_runtime": 2175.34,
|
| 6 |
+
"train_samples_per_second": 8.578,
|
| 7 |
+
"train_steps_per_second": 0.541
|
| 8 |
}
|
trainer_state.json
CHANGED
|
@@ -1,325 +1,826 @@
|
|
| 1 |
{
|
| 2 |
-
"best_metric": 0.
|
| 3 |
-
"best_model_checkpoint": "./vit-base-blur/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":
|
| 13 |
-
"loss": 0.
|
| 14 |
"step": 10
|
| 15 |
},
|
| 16 |
{
|
| 17 |
-
"epoch": 0.
|
| 18 |
-
"learning_rate":
|
| 19 |
-
"loss": 0.
|
| 20 |
"step": 20
|
| 21 |
},
|
| 22 |
{
|
| 23 |
-
"epoch": 0.
|
| 24 |
-
"learning_rate":
|
| 25 |
-
"loss": 0.
|
| 26 |
"step": 30
|
| 27 |
},
|
| 28 |
{
|
| 29 |
-
"epoch": 0.
|
| 30 |
-
"learning_rate":
|
| 31 |
-
"loss": 0.
|
| 32 |
"step": 40
|
| 33 |
},
|
| 34 |
{
|
| 35 |
-
"epoch":
|
| 36 |
-
"learning_rate":
|
| 37 |
-
"loss": 0.
|
| 38 |
"step": 50
|
| 39 |
},
|
| 40 |
{
|
| 41 |
-
"epoch":
|
| 42 |
-
"learning_rate":
|
| 43 |
-
"loss": 0.
|
| 44 |
"step": 60
|
| 45 |
},
|
| 46 |
{
|
| 47 |
-
"epoch":
|
| 48 |
-
"learning_rate":
|
| 49 |
-
"loss": 0.
|
| 50 |
"step": 70
|
| 51 |
},
|
| 52 |
{
|
| 53 |
-
"epoch":
|
| 54 |
-
"learning_rate":
|
| 55 |
-
"loss": 0.
|
| 56 |
"step": 80
|
| 57 |
},
|
| 58 |
{
|
| 59 |
-
"epoch":
|
| 60 |
-
"learning_rate":
|
| 61 |
-
"loss": 0.
|
| 62 |
"step": 90
|
| 63 |
},
|
| 64 |
{
|
| 65 |
-
"epoch":
|
| 66 |
-
"learning_rate":
|
| 67 |
-
"loss": 0.
|
| 68 |
"step": 100
|
| 69 |
},
|
| 70 |
{
|
| 71 |
-
"epoch":
|
| 72 |
-
"eval_accuracy": 0
|
| 73 |
-
"eval_loss": 0.
|
| 74 |
-
"eval_runtime":
|
| 75 |
-
"eval_samples_per_second":
|
| 76 |
-
"eval_steps_per_second":
|
| 77 |
"step": 100
|
| 78 |
},
|
| 79 |
{
|
| 80 |
-
"epoch":
|
| 81 |
-
"learning_rate":
|
| 82 |
-
"loss": 0.
|
| 83 |
"step": 110
|
| 84 |
},
|
| 85 |
{
|
| 86 |
-
"epoch":
|
| 87 |
-
"learning_rate":
|
| 88 |
-
"loss": 0.
|
| 89 |
"step": 120
|
| 90 |
},
|
| 91 |
{
|
| 92 |
-
"epoch":
|
| 93 |
-
"learning_rate":
|
| 94 |
-
"loss": 0.
|
| 95 |
"step": 130
|
| 96 |
},
|
| 97 |
{
|
| 98 |
-
"epoch":
|
| 99 |
-
"learning_rate":
|
| 100 |
-
"loss": 0.
|
| 101 |
"step": 140
|
| 102 |
},
|
| 103 |
{
|
| 104 |
-
"epoch":
|
| 105 |
-
"learning_rate":
|
| 106 |
-
"loss": 0.
|
| 107 |
"step": 150
|
| 108 |
},
|
| 109 |
{
|
| 110 |
-
"epoch":
|
| 111 |
-
"learning_rate":
|
| 112 |
-
"loss": 0.
|
| 113 |
"step": 160
|
| 114 |
},
|
| 115 |
{
|
| 116 |
-
"epoch":
|
| 117 |
-
"learning_rate":
|
| 118 |
-
"loss": 0.
|
| 119 |
"step": 170
|
| 120 |
},
|
| 121 |
{
|
| 122 |
-
"epoch":
|
| 123 |
-
"learning_rate":
|
| 124 |
-
"loss": 0.
|
| 125 |
"step": 180
|
| 126 |
},
|
| 127 |
{
|
| 128 |
-
"epoch":
|
| 129 |
-
"learning_rate":
|
| 130 |
-
"loss": 0.
|
| 131 |
"step": 190
|
| 132 |
},
|
| 133 |
{
|
| 134 |
-
"epoch":
|
| 135 |
-
"learning_rate":
|
| 136 |
-
"loss": 0.
|
| 137 |
"step": 200
|
| 138 |
},
|
| 139 |
{
|
| 140 |
-
"epoch":
|
| 141 |
"eval_accuracy": 1.0,
|
| 142 |
-
"eval_loss": 0.
|
| 143 |
-
"eval_runtime":
|
| 144 |
-
"eval_samples_per_second":
|
| 145 |
-
"eval_steps_per_second":
|
| 146 |
"step": 200
|
| 147 |
},
|
| 148 |
{
|
| 149 |
-
"epoch":
|
| 150 |
-
"learning_rate":
|
| 151 |
-
"loss": 0.
|
| 152 |
"step": 210
|
| 153 |
},
|
| 154 |
{
|
| 155 |
-
"epoch":
|
| 156 |
-
"learning_rate":
|
| 157 |
-
"loss": 0.
|
| 158 |
"step": 220
|
| 159 |
},
|
| 160 |
{
|
| 161 |
-
"epoch":
|
| 162 |
-
"learning_rate":
|
| 163 |
-
"loss": 0.
|
| 164 |
"step": 230
|
| 165 |
},
|
| 166 |
{
|
| 167 |
-
"epoch":
|
| 168 |
-
"learning_rate":
|
| 169 |
-
"loss": 0.
|
| 170 |
"step": 240
|
| 171 |
},
|
| 172 |
{
|
| 173 |
-
"epoch":
|
| 174 |
-
"learning_rate":
|
| 175 |
-
"loss": 0.
|
| 176 |
"step": 250
|
| 177 |
},
|
| 178 |
{
|
| 179 |
-
"epoch":
|
| 180 |
-
"learning_rate":
|
| 181 |
-
"loss": 0.
|
| 182 |
"step": 260
|
| 183 |
},
|
| 184 |
{
|
| 185 |
-
"epoch":
|
| 186 |
-
"learning_rate": 7.
|
| 187 |
-
"loss": 0.
|
| 188 |
"step": 270
|
| 189 |
},
|
| 190 |
{
|
| 191 |
-
"epoch":
|
| 192 |
-
"learning_rate": 7.
|
| 193 |
-
"loss": 0.
|
| 194 |
"step": 280
|
| 195 |
},
|
| 196 |
{
|
| 197 |
-
"epoch":
|
| 198 |
-
"learning_rate":
|
| 199 |
-
"loss": 0.
|
| 200 |
"step": 290
|
| 201 |
},
|
| 202 |
{
|
| 203 |
-
"epoch":
|
| 204 |
-
"learning_rate":
|
| 205 |
-
"loss": 0.
|
| 206 |
"step": 300
|
| 207 |
},
|
| 208 |
{
|
| 209 |
-
"epoch":
|
| 210 |
"eval_accuracy": 1.0,
|
| 211 |
-
"eval_loss": 0.
|
| 212 |
-
"eval_runtime":
|
| 213 |
-
"eval_samples_per_second":
|
| 214 |
-
"eval_steps_per_second":
|
| 215 |
"step": 300
|
| 216 |
},
|
| 217 |
{
|
| 218 |
-
"epoch":
|
| 219 |
-
"learning_rate":
|
| 220 |
-
"loss": 0.
|
| 221 |
"step": 310
|
| 222 |
},
|
| 223 |
{
|
| 224 |
-
"epoch":
|
| 225 |
-
"learning_rate":
|
| 226 |
-
"loss": 0.
|
| 227 |
"step": 320
|
| 228 |
},
|
| 229 |
{
|
| 230 |
-
"epoch":
|
| 231 |
-
"learning_rate":
|
| 232 |
-
"loss": 0.
|
| 233 |
"step": 330
|
| 234 |
},
|
| 235 |
{
|
| 236 |
-
"epoch":
|
| 237 |
-
"learning_rate":
|
| 238 |
-
"loss": 0.
|
| 239 |
"step": 340
|
| 240 |
},
|
| 241 |
{
|
| 242 |
-
"epoch":
|
| 243 |
-
"learning_rate":
|
| 244 |
-
"loss": 0.
|
| 245 |
"step": 350
|
| 246 |
},
|
| 247 |
{
|
| 248 |
-
"epoch":
|
| 249 |
-
"learning_rate":
|
| 250 |
-
"loss": 0.
|
| 251 |
"step": 360
|
| 252 |
},
|
| 253 |
{
|
| 254 |
-
"epoch":
|
| 255 |
-
"learning_rate":
|
| 256 |
-
"loss": 0.
|
| 257 |
"step": 370
|
| 258 |
},
|
| 259 |
{
|
| 260 |
-
"epoch":
|
| 261 |
-
"learning_rate":
|
| 262 |
-
"loss": 0.
|
| 263 |
"step": 380
|
| 264 |
},
|
| 265 |
{
|
| 266 |
-
"epoch":
|
| 267 |
-
"learning_rate":
|
| 268 |
-
"loss": 0.
|
| 269 |
"step": 390
|
| 270 |
},
|
| 271 |
{
|
| 272 |
-
"epoch":
|
| 273 |
-
"learning_rate":
|
| 274 |
-
"loss": 0.
|
| 275 |
"step": 400
|
| 276 |
},
|
| 277 |
{
|
| 278 |
-
"epoch":
|
| 279 |
"eval_accuracy": 1.0,
|
| 280 |
-
"eval_loss": 0.
|
| 281 |
-
"eval_runtime":
|
| 282 |
-
"eval_samples_per_second":
|
| 283 |
-
"eval_steps_per_second":
|
| 284 |
"step": 400
|
| 285 |
},
|
| 286 |
{
|
| 287 |
-
"epoch":
|
| 288 |
-
"learning_rate":
|
| 289 |
-
"loss": 0.
|
| 290 |
"step": 410
|
| 291 |
},
|
| 292 |
{
|
| 293 |
-
"epoch":
|
| 294 |
-
"learning_rate":
|
| 295 |
-
"loss": 0.
|
| 296 |
"step": 420
|
| 297 |
},
|
| 298 |
{
|
| 299 |
-
"epoch":
|
| 300 |
-
"learning_rate":
|
| 301 |
-
"loss": 0.
|
| 302 |
"step": 430
|
| 303 |
},
|
| 304 |
{
|
| 305 |
-
"epoch":
|
| 306 |
-
"learning_rate":
|
| 307 |
-
"loss": 0.
|
| 308 |
"step": 440
|
| 309 |
},
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 310 |
{
|
| 311 |
"epoch": 10.0,
|
| 312 |
-
"
|
| 313 |
-
"
|
| 314 |
-
"
|
| 315 |
-
|
| 316 |
-
|
| 317 |
-
"
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 318 |
}
|
| 319 |
],
|
| 320 |
-
"max_steps":
|
| 321 |
-
"num_train_epochs":
|
| 322 |
-
"total_flos":
|
| 323 |
"trial_name": null,
|
| 324 |
"trial_params": null
|
| 325 |
}
|
|
|
|
| 1 |
{
|
| 2 |
+
"best_metric": 0.0007819964084774256,
|
| 3 |
+
"best_model_checkpoint": "./vit-base-blur/checkpoint-1100",
|
| 4 |
+
"epoch": 12.0,
|
| 5 |
+
"global_step": 1176,
|
| 6 |
"is_hyper_param_search": false,
|
| 7 |
"is_local_process_zero": true,
|
| 8 |
"is_world_process_zero": true,
|
| 9 |
"log_history": [
|
| 10 |
{
|
| 11 |
+
"epoch": 0.1,
|
| 12 |
+
"learning_rate": 9.914965986394558e-05,
|
| 13 |
+
"loss": 0.5188,
|
| 14 |
"step": 10
|
| 15 |
},
|
| 16 |
{
|
| 17 |
+
"epoch": 0.2,
|
| 18 |
+
"learning_rate": 9.829931972789116e-05,
|
| 19 |
+
"loss": 0.1389,
|
| 20 |
"step": 20
|
| 21 |
},
|
| 22 |
{
|
| 23 |
+
"epoch": 0.31,
|
| 24 |
+
"learning_rate": 9.744897959183674e-05,
|
| 25 |
+
"loss": 0.0755,
|
| 26 |
"step": 30
|
| 27 |
},
|
| 28 |
{
|
| 29 |
+
"epoch": 0.41,
|
| 30 |
+
"learning_rate": 9.659863945578231e-05,
|
| 31 |
+
"loss": 0.1305,
|
| 32 |
"step": 40
|
| 33 |
},
|
| 34 |
{
|
| 35 |
+
"epoch": 0.51,
|
| 36 |
+
"learning_rate": 9.574829931972789e-05,
|
| 37 |
+
"loss": 0.0554,
|
| 38 |
"step": 50
|
| 39 |
},
|
| 40 |
{
|
| 41 |
+
"epoch": 0.61,
|
| 42 |
+
"learning_rate": 9.489795918367348e-05,
|
| 43 |
+
"loss": 0.0147,
|
| 44 |
"step": 60
|
| 45 |
},
|
| 46 |
{
|
| 47 |
+
"epoch": 0.71,
|
| 48 |
+
"learning_rate": 9.404761904761905e-05,
|
| 49 |
+
"loss": 0.0139,
|
| 50 |
"step": 70
|
| 51 |
},
|
| 52 |
{
|
| 53 |
+
"epoch": 0.82,
|
| 54 |
+
"learning_rate": 9.319727891156463e-05,
|
| 55 |
+
"loss": 0.0534,
|
| 56 |
"step": 80
|
| 57 |
},
|
| 58 |
{
|
| 59 |
+
"epoch": 0.92,
|
| 60 |
+
"learning_rate": 9.234693877551021e-05,
|
| 61 |
+
"loss": 0.0101,
|
| 62 |
"step": 90
|
| 63 |
},
|
| 64 |
{
|
| 65 |
+
"epoch": 1.02,
|
| 66 |
+
"learning_rate": 9.149659863945579e-05,
|
| 67 |
+
"loss": 0.0082,
|
| 68 |
"step": 100
|
| 69 |
},
|
| 70 |
{
|
| 71 |
+
"epoch": 1.02,
|
| 72 |
+
"eval_accuracy": 1.0,
|
| 73 |
+
"eval_loss": 0.010711735114455223,
|
| 74 |
+
"eval_runtime": 105.4365,
|
| 75 |
+
"eval_samples_per_second": 1.641,
|
| 76 |
+
"eval_steps_per_second": 0.209,
|
| 77 |
"step": 100
|
| 78 |
},
|
| 79 |
{
|
| 80 |
+
"epoch": 1.12,
|
| 81 |
+
"learning_rate": 9.07312925170068e-05,
|
| 82 |
+
"loss": 0.0375,
|
| 83 |
"step": 110
|
| 84 |
},
|
| 85 |
{
|
| 86 |
+
"epoch": 1.22,
|
| 87 |
+
"learning_rate": 8.988095238095238e-05,
|
| 88 |
+
"loss": 0.0296,
|
| 89 |
"step": 120
|
| 90 |
},
|
| 91 |
{
|
| 92 |
+
"epoch": 1.33,
|
| 93 |
+
"learning_rate": 8.903061224489796e-05,
|
| 94 |
+
"loss": 0.0466,
|
| 95 |
"step": 130
|
| 96 |
},
|
| 97 |
{
|
| 98 |
+
"epoch": 1.43,
|
| 99 |
+
"learning_rate": 8.818027210884354e-05,
|
| 100 |
+
"loss": 0.0289,
|
| 101 |
"step": 140
|
| 102 |
},
|
| 103 |
{
|
| 104 |
+
"epoch": 1.53,
|
| 105 |
+
"learning_rate": 8.732993197278912e-05,
|
| 106 |
+
"loss": 0.0084,
|
| 107 |
"step": 150
|
| 108 |
},
|
| 109 |
{
|
| 110 |
+
"epoch": 1.63,
|
| 111 |
+
"learning_rate": 8.64795918367347e-05,
|
| 112 |
+
"loss": 0.0185,
|
| 113 |
"step": 160
|
| 114 |
},
|
| 115 |
{
|
| 116 |
+
"epoch": 1.73,
|
| 117 |
+
"learning_rate": 8.562925170068028e-05,
|
| 118 |
+
"loss": 0.0879,
|
| 119 |
"step": 170
|
| 120 |
},
|
| 121 |
{
|
| 122 |
+
"epoch": 1.84,
|
| 123 |
+
"learning_rate": 8.477891156462585e-05,
|
| 124 |
+
"loss": 0.0612,
|
| 125 |
"step": 180
|
| 126 |
},
|
| 127 |
{
|
| 128 |
+
"epoch": 1.94,
|
| 129 |
+
"learning_rate": 8.392857142857144e-05,
|
| 130 |
+
"loss": 0.0176,
|
| 131 |
"step": 190
|
| 132 |
},
|
| 133 |
{
|
| 134 |
+
"epoch": 2.04,
|
| 135 |
+
"learning_rate": 8.3078231292517e-05,
|
| 136 |
+
"loss": 0.0079,
|
| 137 |
"step": 200
|
| 138 |
},
|
| 139 |
{
|
| 140 |
+
"epoch": 2.04,
|
| 141 |
"eval_accuracy": 1.0,
|
| 142 |
+
"eval_loss": 0.005197448190301657,
|
| 143 |
+
"eval_runtime": 5.9181,
|
| 144 |
+
"eval_samples_per_second": 29.232,
|
| 145 |
+
"eval_steps_per_second": 3.717,
|
| 146 |
"step": 200
|
| 147 |
},
|
| 148 |
{
|
| 149 |
+
"epoch": 2.14,
|
| 150 |
+
"learning_rate": 8.222789115646259e-05,
|
| 151 |
+
"loss": 0.005,
|
| 152 |
"step": 210
|
| 153 |
},
|
| 154 |
{
|
| 155 |
+
"epoch": 2.24,
|
| 156 |
+
"learning_rate": 8.137755102040817e-05,
|
| 157 |
+
"loss": 0.0226,
|
| 158 |
"step": 220
|
| 159 |
},
|
| 160 |
{
|
| 161 |
+
"epoch": 2.35,
|
| 162 |
+
"learning_rate": 8.052721088435375e-05,
|
| 163 |
+
"loss": 0.0894,
|
| 164 |
"step": 230
|
| 165 |
},
|
| 166 |
{
|
| 167 |
+
"epoch": 2.45,
|
| 168 |
+
"learning_rate": 7.967687074829932e-05,
|
| 169 |
+
"loss": 0.0067,
|
| 170 |
"step": 240
|
| 171 |
},
|
| 172 |
{
|
| 173 |
+
"epoch": 2.55,
|
| 174 |
+
"learning_rate": 7.882653061224489e-05,
|
| 175 |
+
"loss": 0.0039,
|
| 176 |
"step": 250
|
| 177 |
},
|
| 178 |
{
|
| 179 |
+
"epoch": 2.65,
|
| 180 |
+
"learning_rate": 7.797619047619048e-05,
|
| 181 |
+
"loss": 0.0035,
|
| 182 |
"step": 260
|
| 183 |
},
|
| 184 |
{
|
| 185 |
+
"epoch": 2.76,
|
| 186 |
+
"learning_rate": 7.712585034013606e-05,
|
| 187 |
+
"loss": 0.0034,
|
| 188 |
"step": 270
|
| 189 |
},
|
| 190 |
{
|
| 191 |
+
"epoch": 2.86,
|
| 192 |
+
"learning_rate": 7.627551020408163e-05,
|
| 193 |
+
"loss": 0.0032,
|
| 194 |
"step": 280
|
| 195 |
},
|
| 196 |
{
|
| 197 |
+
"epoch": 2.96,
|
| 198 |
+
"learning_rate": 7.542517006802722e-05,
|
| 199 |
+
"loss": 0.0032,
|
| 200 |
"step": 290
|
| 201 |
},
|
| 202 |
{
|
| 203 |
+
"epoch": 3.06,
|
| 204 |
+
"learning_rate": 7.457482993197279e-05,
|
| 205 |
+
"loss": 0.0029,
|
| 206 |
"step": 300
|
| 207 |
},
|
| 208 |
{
|
| 209 |
+
"epoch": 3.06,
|
| 210 |
"eval_accuracy": 1.0,
|
| 211 |
+
"eval_loss": 0.0028241325635463,
|
| 212 |
+
"eval_runtime": 4.5591,
|
| 213 |
+
"eval_samples_per_second": 37.946,
|
| 214 |
+
"eval_steps_per_second": 4.826,
|
| 215 |
"step": 300
|
| 216 |
},
|
| 217 |
{
|
| 218 |
+
"epoch": 3.16,
|
| 219 |
+
"learning_rate": 7.372448979591837e-05,
|
| 220 |
+
"loss": 0.0028,
|
| 221 |
"step": 310
|
| 222 |
},
|
| 223 |
{
|
| 224 |
+
"epoch": 3.27,
|
| 225 |
+
"learning_rate": 7.287414965986395e-05,
|
| 226 |
+
"loss": 0.0027,
|
| 227 |
"step": 320
|
| 228 |
},
|
| 229 |
{
|
| 230 |
+
"epoch": 3.37,
|
| 231 |
+
"learning_rate": 7.202380952380953e-05,
|
| 232 |
+
"loss": 0.0026,
|
| 233 |
"step": 330
|
| 234 |
},
|
| 235 |
{
|
| 236 |
+
"epoch": 3.47,
|
| 237 |
+
"learning_rate": 7.117346938775511e-05,
|
| 238 |
+
"loss": 0.0025,
|
| 239 |
"step": 340
|
| 240 |
},
|
| 241 |
{
|
| 242 |
+
"epoch": 3.57,
|
| 243 |
+
"learning_rate": 7.032312925170068e-05,
|
| 244 |
+
"loss": 0.0024,
|
| 245 |
"step": 350
|
| 246 |
},
|
| 247 |
{
|
| 248 |
+
"epoch": 3.67,
|
| 249 |
+
"learning_rate": 6.947278911564627e-05,
|
| 250 |
+
"loss": 0.0023,
|
| 251 |
"step": 360
|
| 252 |
},
|
| 253 |
{
|
| 254 |
+
"epoch": 3.78,
|
| 255 |
+
"learning_rate": 6.862244897959184e-05,
|
| 256 |
+
"loss": 0.0022,
|
| 257 |
"step": 370
|
| 258 |
},
|
| 259 |
{
|
| 260 |
+
"epoch": 3.88,
|
| 261 |
+
"learning_rate": 6.777210884353741e-05,
|
| 262 |
+
"loss": 0.0021,
|
| 263 |
"step": 380
|
| 264 |
},
|
| 265 |
{
|
| 266 |
+
"epoch": 3.98,
|
| 267 |
+
"learning_rate": 6.6921768707483e-05,
|
| 268 |
+
"loss": 0.0021,
|
| 269 |
"step": 390
|
| 270 |
},
|
| 271 |
{
|
| 272 |
+
"epoch": 4.08,
|
| 273 |
+
"learning_rate": 6.607142857142857e-05,
|
| 274 |
+
"loss": 0.002,
|
| 275 |
"step": 400
|
| 276 |
},
|
| 277 |
{
|
| 278 |
+
"epoch": 4.08,
|
| 279 |
"eval_accuracy": 1.0,
|
| 280 |
+
"eval_loss": 0.001991536468267441,
|
| 281 |
+
"eval_runtime": 5.3318,
|
| 282 |
+
"eval_samples_per_second": 32.447,
|
| 283 |
+
"eval_steps_per_second": 4.126,
|
| 284 |
"step": 400
|
| 285 |
},
|
| 286 |
{
|
| 287 |
+
"epoch": 4.18,
|
| 288 |
+
"learning_rate": 6.522108843537415e-05,
|
| 289 |
+
"loss": 0.002,
|
| 290 |
"step": 410
|
| 291 |
},
|
| 292 |
{
|
| 293 |
+
"epoch": 4.29,
|
| 294 |
+
"learning_rate": 6.437074829931973e-05,
|
| 295 |
+
"loss": 0.0019,
|
| 296 |
"step": 420
|
| 297 |
},
|
| 298 |
{
|
| 299 |
+
"epoch": 4.39,
|
| 300 |
+
"learning_rate": 6.352040816326531e-05,
|
| 301 |
+
"loss": 0.0019,
|
| 302 |
"step": 430
|
| 303 |
},
|
| 304 |
{
|
| 305 |
+
"epoch": 4.49,
|
| 306 |
+
"learning_rate": 6.267006802721088e-05,
|
| 307 |
+
"loss": 0.0018,
|
| 308 |
"step": 440
|
| 309 |
},
|
| 310 |
+
{
|
| 311 |
+
"epoch": 4.59,
|
| 312 |
+
"learning_rate": 6.181972789115646e-05,
|
| 313 |
+
"loss": 0.0018,
|
| 314 |
+
"step": 450
|
| 315 |
+
},
|
| 316 |
+
{
|
| 317 |
+
"epoch": 4.69,
|
| 318 |
+
"learning_rate": 6.0969387755102046e-05,
|
| 319 |
+
"loss": 0.0017,
|
| 320 |
+
"step": 460
|
| 321 |
+
},
|
| 322 |
+
{
|
| 323 |
+
"epoch": 4.8,
|
| 324 |
+
"learning_rate": 6.011904761904762e-05,
|
| 325 |
+
"loss": 0.0017,
|
| 326 |
+
"step": 470
|
| 327 |
+
},
|
| 328 |
+
{
|
| 329 |
+
"epoch": 4.9,
|
| 330 |
+
"learning_rate": 5.92687074829932e-05,
|
| 331 |
+
"loss": 0.0016,
|
| 332 |
+
"step": 480
|
| 333 |
+
},
|
| 334 |
+
{
|
| 335 |
+
"epoch": 5.0,
|
| 336 |
+
"learning_rate": 5.841836734693877e-05,
|
| 337 |
+
"loss": 0.0016,
|
| 338 |
+
"step": 490
|
| 339 |
+
},
|
| 340 |
+
{
|
| 341 |
+
"epoch": 5.1,
|
| 342 |
+
"learning_rate": 5.756802721088436e-05,
|
| 343 |
+
"loss": 0.0016,
|
| 344 |
+
"step": 500
|
| 345 |
+
},
|
| 346 |
+
{
|
| 347 |
+
"epoch": 5.1,
|
| 348 |
+
"eval_accuracy": 1.0,
|
| 349 |
+
"eval_loss": 0.0015397485112771392,
|
| 350 |
+
"eval_runtime": 5.4131,
|
| 351 |
+
"eval_samples_per_second": 31.959,
|
| 352 |
+
"eval_steps_per_second": 4.064,
|
| 353 |
+
"step": 500
|
| 354 |
+
},
|
| 355 |
+
{
|
| 356 |
+
"epoch": 5.2,
|
| 357 |
+
"learning_rate": 5.671768707482994e-05,
|
| 358 |
+
"loss": 0.0015,
|
| 359 |
+
"step": 510
|
| 360 |
+
},
|
| 361 |
+
{
|
| 362 |
+
"epoch": 5.31,
|
| 363 |
+
"learning_rate": 5.586734693877551e-05,
|
| 364 |
+
"loss": 0.0015,
|
| 365 |
+
"step": 520
|
| 366 |
+
},
|
| 367 |
+
{
|
| 368 |
+
"epoch": 5.41,
|
| 369 |
+
"learning_rate": 5.501700680272109e-05,
|
| 370 |
+
"loss": 0.0015,
|
| 371 |
+
"step": 530
|
| 372 |
+
},
|
| 373 |
+
{
|
| 374 |
+
"epoch": 5.51,
|
| 375 |
+
"learning_rate": 5.4166666666666664e-05,
|
| 376 |
+
"loss": 0.0014,
|
| 377 |
+
"step": 540
|
| 378 |
+
},
|
| 379 |
+
{
|
| 380 |
+
"epoch": 5.61,
|
| 381 |
+
"learning_rate": 5.331632653061225e-05,
|
| 382 |
+
"loss": 0.0014,
|
| 383 |
+
"step": 550
|
| 384 |
+
},
|
| 385 |
+
{
|
| 386 |
+
"epoch": 5.71,
|
| 387 |
+
"learning_rate": 5.246598639455783e-05,
|
| 388 |
+
"loss": 0.0014,
|
| 389 |
+
"step": 560
|
| 390 |
+
},
|
| 391 |
+
{
|
| 392 |
+
"epoch": 5.82,
|
| 393 |
+
"learning_rate": 5.1615646258503404e-05,
|
| 394 |
+
"loss": 0.0014,
|
| 395 |
+
"step": 570
|
| 396 |
+
},
|
| 397 |
+
{
|
| 398 |
+
"epoch": 5.92,
|
| 399 |
+
"learning_rate": 5.076530612244898e-05,
|
| 400 |
+
"loss": 0.0013,
|
| 401 |
+
"step": 580
|
| 402 |
+
},
|
| 403 |
+
{
|
| 404 |
+
"epoch": 6.02,
|
| 405 |
+
"learning_rate": 4.991496598639456e-05,
|
| 406 |
+
"loss": 0.0013,
|
| 407 |
+
"step": 590
|
| 408 |
+
},
|
| 409 |
+
{
|
| 410 |
+
"epoch": 6.12,
|
| 411 |
+
"learning_rate": 4.906462585034014e-05,
|
| 412 |
+
"loss": 0.0013,
|
| 413 |
+
"step": 600
|
| 414 |
+
},
|
| 415 |
+
{
|
| 416 |
+
"epoch": 6.12,
|
| 417 |
+
"eval_accuracy": 1.0,
|
| 418 |
+
"eval_loss": 0.0012603328796103597,
|
| 419 |
+
"eval_runtime": 6.365,
|
| 420 |
+
"eval_samples_per_second": 27.18,
|
| 421 |
+
"eval_steps_per_second": 3.456,
|
| 422 |
+
"step": 600
|
| 423 |
+
},
|
| 424 |
+
{
|
| 425 |
+
"epoch": 6.22,
|
| 426 |
+
"learning_rate": 4.8214285714285716e-05,
|
| 427 |
+
"loss": 0.0013,
|
| 428 |
+
"step": 610
|
| 429 |
+
},
|
| 430 |
+
{
|
| 431 |
+
"epoch": 6.33,
|
| 432 |
+
"learning_rate": 4.7363945578231296e-05,
|
| 433 |
+
"loss": 0.0012,
|
| 434 |
+
"step": 620
|
| 435 |
+
},
|
| 436 |
+
{
|
| 437 |
+
"epoch": 6.43,
|
| 438 |
+
"learning_rate": 4.651360544217687e-05,
|
| 439 |
+
"loss": 0.0012,
|
| 440 |
+
"step": 630
|
| 441 |
+
},
|
| 442 |
+
{
|
| 443 |
+
"epoch": 6.53,
|
| 444 |
+
"learning_rate": 4.566326530612245e-05,
|
| 445 |
+
"loss": 0.0012,
|
| 446 |
+
"step": 640
|
| 447 |
+
},
|
| 448 |
+
{
|
| 449 |
+
"epoch": 6.63,
|
| 450 |
+
"learning_rate": 4.4812925170068035e-05,
|
| 451 |
+
"loss": 0.0012,
|
| 452 |
+
"step": 650
|
| 453 |
+
},
|
| 454 |
+
{
|
| 455 |
+
"epoch": 6.73,
|
| 456 |
+
"learning_rate": 4.396258503401361e-05,
|
| 457 |
+
"loss": 0.0012,
|
| 458 |
+
"step": 660
|
| 459 |
+
},
|
| 460 |
+
{
|
| 461 |
+
"epoch": 6.84,
|
| 462 |
+
"learning_rate": 4.311224489795919e-05,
|
| 463 |
+
"loss": 0.0011,
|
| 464 |
+
"step": 670
|
| 465 |
+
},
|
| 466 |
+
{
|
| 467 |
+
"epoch": 6.94,
|
| 468 |
+
"learning_rate": 4.226190476190476e-05,
|
| 469 |
+
"loss": 0.0011,
|
| 470 |
+
"step": 680
|
| 471 |
+
},
|
| 472 |
+
{
|
| 473 |
+
"epoch": 7.04,
|
| 474 |
+
"learning_rate": 4.141156462585034e-05,
|
| 475 |
+
"loss": 0.0011,
|
| 476 |
+
"step": 690
|
| 477 |
+
},
|
| 478 |
+
{
|
| 479 |
+
"epoch": 7.14,
|
| 480 |
+
"learning_rate": 4.056122448979592e-05,
|
| 481 |
+
"loss": 0.0011,
|
| 482 |
+
"step": 700
|
| 483 |
+
},
|
| 484 |
+
{
|
| 485 |
+
"epoch": 7.14,
|
| 486 |
+
"eval_accuracy": 1.0,
|
| 487 |
+
"eval_loss": 0.0010760421864688396,
|
| 488 |
+
"eval_runtime": 5.8839,
|
| 489 |
+
"eval_samples_per_second": 29.402,
|
| 490 |
+
"eval_steps_per_second": 3.739,
|
| 491 |
+
"step": 700
|
| 492 |
+
},
|
| 493 |
+
{
|
| 494 |
+
"epoch": 7.24,
|
| 495 |
+
"learning_rate": 3.97108843537415e-05,
|
| 496 |
+
"loss": 0.0011,
|
| 497 |
+
"step": 710
|
| 498 |
+
},
|
| 499 |
+
{
|
| 500 |
+
"epoch": 7.35,
|
| 501 |
+
"learning_rate": 3.886054421768708e-05,
|
| 502 |
+
"loss": 0.0011,
|
| 503 |
+
"step": 720
|
| 504 |
+
},
|
| 505 |
+
{
|
| 506 |
+
"epoch": 7.45,
|
| 507 |
+
"learning_rate": 3.801020408163265e-05,
|
| 508 |
+
"loss": 0.001,
|
| 509 |
+
"step": 730
|
| 510 |
+
},
|
| 511 |
+
{
|
| 512 |
+
"epoch": 7.55,
|
| 513 |
+
"learning_rate": 3.715986394557823e-05,
|
| 514 |
+
"loss": 0.001,
|
| 515 |
+
"step": 740
|
| 516 |
+
},
|
| 517 |
+
{
|
| 518 |
+
"epoch": 7.65,
|
| 519 |
+
"learning_rate": 3.630952380952381e-05,
|
| 520 |
+
"loss": 0.001,
|
| 521 |
+
"step": 750
|
| 522 |
+
},
|
| 523 |
+
{
|
| 524 |
+
"epoch": 7.76,
|
| 525 |
+
"learning_rate": 3.5459183673469385e-05,
|
| 526 |
+
"loss": 0.001,
|
| 527 |
+
"step": 760
|
| 528 |
+
},
|
| 529 |
+
{
|
| 530 |
+
"epoch": 7.86,
|
| 531 |
+
"learning_rate": 3.460884353741497e-05,
|
| 532 |
+
"loss": 0.001,
|
| 533 |
+
"step": 770
|
| 534 |
+
},
|
| 535 |
+
{
|
| 536 |
+
"epoch": 7.96,
|
| 537 |
+
"learning_rate": 3.3758503401360545e-05,
|
| 538 |
+
"loss": 0.001,
|
| 539 |
+
"step": 780
|
| 540 |
+
},
|
| 541 |
+
{
|
| 542 |
+
"epoch": 8.06,
|
| 543 |
+
"learning_rate": 3.2908163265306125e-05,
|
| 544 |
+
"loss": 0.001,
|
| 545 |
+
"step": 790
|
| 546 |
+
},
|
| 547 |
+
{
|
| 548 |
+
"epoch": 8.16,
|
| 549 |
+
"learning_rate": 3.2057823129251705e-05,
|
| 550 |
+
"loss": 0.001,
|
| 551 |
+
"step": 800
|
| 552 |
+
},
|
| 553 |
+
{
|
| 554 |
+
"epoch": 8.16,
|
| 555 |
+
"eval_accuracy": 1.0,
|
| 556 |
+
"eval_loss": 0.0009508325601927936,
|
| 557 |
+
"eval_runtime": 4.4253,
|
| 558 |
+
"eval_samples_per_second": 39.093,
|
| 559 |
+
"eval_steps_per_second": 4.971,
|
| 560 |
+
"step": 800
|
| 561 |
+
},
|
| 562 |
+
{
|
| 563 |
+
"epoch": 8.27,
|
| 564 |
+
"learning_rate": 3.120748299319728e-05,
|
| 565 |
+
"loss": 0.0009,
|
| 566 |
+
"step": 810
|
| 567 |
+
},
|
| 568 |
+
{
|
| 569 |
+
"epoch": 8.37,
|
| 570 |
+
"learning_rate": 3.0357142857142857e-05,
|
| 571 |
+
"loss": 0.0009,
|
| 572 |
+
"step": 820
|
| 573 |
+
},
|
| 574 |
+
{
|
| 575 |
+
"epoch": 8.47,
|
| 576 |
+
"learning_rate": 2.950680272108844e-05,
|
| 577 |
+
"loss": 0.0009,
|
| 578 |
+
"step": 830
|
| 579 |
+
},
|
| 580 |
+
{
|
| 581 |
+
"epoch": 8.57,
|
| 582 |
+
"learning_rate": 2.8656462585034017e-05,
|
| 583 |
+
"loss": 0.0009,
|
| 584 |
+
"step": 840
|
| 585 |
+
},
|
| 586 |
+
{
|
| 587 |
+
"epoch": 8.67,
|
| 588 |
+
"learning_rate": 2.7806122448979593e-05,
|
| 589 |
+
"loss": 0.0009,
|
| 590 |
+
"step": 850
|
| 591 |
+
},
|
| 592 |
+
{
|
| 593 |
+
"epoch": 8.78,
|
| 594 |
+
"learning_rate": 2.695578231292517e-05,
|
| 595 |
+
"loss": 0.0009,
|
| 596 |
+
"step": 860
|
| 597 |
+
},
|
| 598 |
+
{
|
| 599 |
+
"epoch": 8.88,
|
| 600 |
+
"learning_rate": 2.610544217687075e-05,
|
| 601 |
+
"loss": 0.0009,
|
| 602 |
+
"step": 870
|
| 603 |
+
},
|
| 604 |
+
{
|
| 605 |
+
"epoch": 8.98,
|
| 606 |
+
"learning_rate": 2.5255102040816326e-05,
|
| 607 |
+
"loss": 0.0009,
|
| 608 |
+
"step": 880
|
| 609 |
+
},
|
| 610 |
+
{
|
| 611 |
+
"epoch": 9.08,
|
| 612 |
+
"learning_rate": 2.4404761904761906e-05,
|
| 613 |
+
"loss": 0.0009,
|
| 614 |
+
"step": 890
|
| 615 |
+
},
|
| 616 |
+
{
|
| 617 |
+
"epoch": 9.18,
|
| 618 |
+
"learning_rate": 2.3554421768707485e-05,
|
| 619 |
+
"loss": 0.0009,
|
| 620 |
+
"step": 900
|
| 621 |
+
},
|
| 622 |
+
{
|
| 623 |
+
"epoch": 9.18,
|
| 624 |
+
"eval_accuracy": 1.0,
|
| 625 |
+
"eval_loss": 0.0008663248154334724,
|
| 626 |
+
"eval_runtime": 5.3859,
|
| 627 |
+
"eval_samples_per_second": 32.121,
|
| 628 |
+
"eval_steps_per_second": 4.085,
|
| 629 |
+
"step": 900
|
| 630 |
+
},
|
| 631 |
+
{
|
| 632 |
+
"epoch": 9.29,
|
| 633 |
+
"learning_rate": 2.2704081632653062e-05,
|
| 634 |
+
"loss": 0.0009,
|
| 635 |
+
"step": 910
|
| 636 |
+
},
|
| 637 |
+
{
|
| 638 |
+
"epoch": 9.39,
|
| 639 |
+
"learning_rate": 2.1853741496598638e-05,
|
| 640 |
+
"loss": 0.0009,
|
| 641 |
+
"step": 920
|
| 642 |
+
},
|
| 643 |
+
{
|
| 644 |
+
"epoch": 9.49,
|
| 645 |
+
"learning_rate": 2.100340136054422e-05,
|
| 646 |
+
"loss": 0.0009,
|
| 647 |
+
"step": 930
|
| 648 |
+
},
|
| 649 |
+
{
|
| 650 |
+
"epoch": 9.59,
|
| 651 |
+
"learning_rate": 2.0153061224489798e-05,
|
| 652 |
+
"loss": 0.0008,
|
| 653 |
+
"step": 940
|
| 654 |
+
},
|
| 655 |
+
{
|
| 656 |
+
"epoch": 9.69,
|
| 657 |
+
"learning_rate": 1.9302721088435374e-05,
|
| 658 |
+
"loss": 0.0008,
|
| 659 |
+
"step": 950
|
| 660 |
+
},
|
| 661 |
+
{
|
| 662 |
+
"epoch": 9.8,
|
| 663 |
+
"learning_rate": 1.8452380952380954e-05,
|
| 664 |
+
"loss": 0.0008,
|
| 665 |
+
"step": 960
|
| 666 |
+
},
|
| 667 |
+
{
|
| 668 |
+
"epoch": 9.9,
|
| 669 |
+
"learning_rate": 1.760204081632653e-05,
|
| 670 |
+
"loss": 0.0008,
|
| 671 |
+
"step": 970
|
| 672 |
+
},
|
| 673 |
{
|
| 674 |
"epoch": 10.0,
|
| 675 |
+
"learning_rate": 1.675170068027211e-05,
|
| 676 |
+
"loss": 0.0008,
|
| 677 |
+
"step": 980
|
| 678 |
+
},
|
| 679 |
+
{
|
| 680 |
+
"epoch": 10.1,
|
| 681 |
+
"learning_rate": 1.590136054421769e-05,
|
| 682 |
+
"loss": 0.0008,
|
| 683 |
+
"step": 990
|
| 684 |
+
},
|
| 685 |
+
{
|
| 686 |
+
"epoch": 10.2,
|
| 687 |
+
"learning_rate": 1.5051020408163266e-05,
|
| 688 |
+
"loss": 0.0008,
|
| 689 |
+
"step": 1000
|
| 690 |
+
},
|
| 691 |
+
{
|
| 692 |
+
"epoch": 10.2,
|
| 693 |
+
"eval_accuracy": 1.0,
|
| 694 |
+
"eval_loss": 0.0008117146790027618,
|
| 695 |
+
"eval_runtime": 4.9611,
|
| 696 |
+
"eval_samples_per_second": 34.871,
|
| 697 |
+
"eval_steps_per_second": 4.435,
|
| 698 |
+
"step": 1000
|
| 699 |
+
},
|
| 700 |
+
{
|
| 701 |
+
"epoch": 10.31,
|
| 702 |
+
"learning_rate": 1.4200680272108843e-05,
|
| 703 |
+
"loss": 0.0008,
|
| 704 |
+
"step": 1010
|
| 705 |
+
},
|
| 706 |
+
{
|
| 707 |
+
"epoch": 10.41,
|
| 708 |
+
"learning_rate": 1.3350340136054423e-05,
|
| 709 |
+
"loss": 0.0008,
|
| 710 |
+
"step": 1020
|
| 711 |
+
},
|
| 712 |
+
{
|
| 713 |
+
"epoch": 10.51,
|
| 714 |
+
"learning_rate": 1.25e-05,
|
| 715 |
+
"loss": 0.0008,
|
| 716 |
+
"step": 1030
|
| 717 |
+
},
|
| 718 |
+
{
|
| 719 |
+
"epoch": 10.61,
|
| 720 |
+
"learning_rate": 1.1649659863945579e-05,
|
| 721 |
+
"loss": 0.0008,
|
| 722 |
+
"step": 1040
|
| 723 |
+
},
|
| 724 |
+
{
|
| 725 |
+
"epoch": 10.71,
|
| 726 |
+
"learning_rate": 1.0799319727891157e-05,
|
| 727 |
+
"loss": 0.0008,
|
| 728 |
+
"step": 1050
|
| 729 |
+
},
|
| 730 |
+
{
|
| 731 |
+
"epoch": 10.82,
|
| 732 |
+
"learning_rate": 9.948979591836735e-06,
|
| 733 |
+
"loss": 0.0008,
|
| 734 |
+
"step": 1060
|
| 735 |
+
},
|
| 736 |
+
{
|
| 737 |
+
"epoch": 10.92,
|
| 738 |
+
"learning_rate": 9.098639455782313e-06,
|
| 739 |
+
"loss": 0.0008,
|
| 740 |
+
"step": 1070
|
| 741 |
+
},
|
| 742 |
+
{
|
| 743 |
+
"epoch": 11.02,
|
| 744 |
+
"learning_rate": 8.248299319727891e-06,
|
| 745 |
+
"loss": 0.0008,
|
| 746 |
+
"step": 1080
|
| 747 |
+
},
|
| 748 |
+
{
|
| 749 |
+
"epoch": 11.12,
|
| 750 |
+
"learning_rate": 7.39795918367347e-06,
|
| 751 |
+
"loss": 0.0008,
|
| 752 |
+
"step": 1090
|
| 753 |
+
},
|
| 754 |
+
{
|
| 755 |
+
"epoch": 11.22,
|
| 756 |
+
"learning_rate": 6.547619047619048e-06,
|
| 757 |
+
"loss": 0.0008,
|
| 758 |
+
"step": 1100
|
| 759 |
+
},
|
| 760 |
+
{
|
| 761 |
+
"epoch": 11.22,
|
| 762 |
+
"eval_accuracy": 1.0,
|
| 763 |
+
"eval_loss": 0.0007819964084774256,
|
| 764 |
+
"eval_runtime": 4.6463,
|
| 765 |
+
"eval_samples_per_second": 37.234,
|
| 766 |
+
"eval_steps_per_second": 4.735,
|
| 767 |
+
"step": 1100
|
| 768 |
+
},
|
| 769 |
+
{
|
| 770 |
+
"epoch": 11.33,
|
| 771 |
+
"learning_rate": 5.697278911564626e-06,
|
| 772 |
+
"loss": 0.0008,
|
| 773 |
+
"step": 1110
|
| 774 |
+
},
|
| 775 |
+
{
|
| 776 |
+
"epoch": 11.43,
|
| 777 |
+
"learning_rate": 4.846938775510204e-06,
|
| 778 |
+
"loss": 0.0008,
|
| 779 |
+
"step": 1120
|
| 780 |
+
},
|
| 781 |
+
{
|
| 782 |
+
"epoch": 11.53,
|
| 783 |
+
"learning_rate": 3.996598639455783e-06,
|
| 784 |
+
"loss": 0.0008,
|
| 785 |
+
"step": 1130
|
| 786 |
+
},
|
| 787 |
+
{
|
| 788 |
+
"epoch": 11.63,
|
| 789 |
+
"learning_rate": 3.146258503401361e-06,
|
| 790 |
+
"loss": 0.0008,
|
| 791 |
+
"step": 1140
|
| 792 |
+
},
|
| 793 |
+
{
|
| 794 |
+
"epoch": 11.73,
|
| 795 |
+
"learning_rate": 2.295918367346939e-06,
|
| 796 |
+
"loss": 0.0008,
|
| 797 |
+
"step": 1150
|
| 798 |
+
},
|
| 799 |
+
{
|
| 800 |
+
"epoch": 11.84,
|
| 801 |
+
"learning_rate": 1.4455782312925172e-06,
|
| 802 |
+
"loss": 0.0008,
|
| 803 |
+
"step": 1160
|
| 804 |
+
},
|
| 805 |
+
{
|
| 806 |
+
"epoch": 11.94,
|
| 807 |
+
"learning_rate": 5.952380952380952e-07,
|
| 808 |
+
"loss": 0.0008,
|
| 809 |
+
"step": 1170
|
| 810 |
+
},
|
| 811 |
+
{
|
| 812 |
+
"epoch": 12.0,
|
| 813 |
+
"step": 1176,
|
| 814 |
+
"total_flos": 1.4460005262082867e+18,
|
| 815 |
+
"train_loss": 0.013737765663116536,
|
| 816 |
+
"train_runtime": 2175.34,
|
| 817 |
+
"train_samples_per_second": 8.578,
|
| 818 |
+
"train_steps_per_second": 0.541
|
| 819 |
}
|
| 820 |
],
|
| 821 |
+
"max_steps": 1176,
|
| 822 |
+
"num_train_epochs": 12,
|
| 823 |
+
"total_flos": 1.4460005262082867e+18,
|
| 824 |
"trial_name": null,
|
| 825 |
"trial_params": null
|
| 826 |
}
|