🍻 cheers
Browse files- all_results.json +11 -11
- eval_results.json +6 -6
- runs/Aug06_21-54-27_a3e0ec26426d/events.out.tfevents.1659824092.a3e0ec26426d.71.5 +3 -0
- test_results.json +8 -0
- train_results.json +6 -6
- trainer_state.json +1338 -147
all_results.json
CHANGED
|
@@ -1,13 +1,13 @@
|
|
| 1 |
{
|
| 2 |
-
"epoch":
|
| 3 |
-
"eval_accuracy": 0.
|
| 4 |
-
"eval_loss":
|
| 5 |
-
"eval_runtime": 1.
|
| 6 |
-
"eval_samples_per_second":
|
| 7 |
-
"eval_steps_per_second": 13.
|
| 8 |
-
"total_flos":
|
| 9 |
-
"train_loss": 0.
|
| 10 |
-
"train_runtime":
|
| 11 |
-
"train_samples_per_second":
|
| 12 |
-
"train_steps_per_second": 2.
|
| 13 |
}
|
|
|
|
| 1 |
{
|
| 2 |
+
"epoch": 25.0,
|
| 3 |
+
"eval_accuracy": 0.6839080459770115,
|
| 4 |
+
"eval_loss": 1.1035996675491333,
|
| 5 |
+
"eval_runtime": 1.6448,
|
| 6 |
+
"eval_samples_per_second": 105.785,
|
| 7 |
+
"eval_steps_per_second": 13.375,
|
| 8 |
+
"total_flos": 2.8110723118700544e+18,
|
| 9 |
+
"train_loss": 0.38098634693648786,
|
| 10 |
+
"train_runtime": 883.423,
|
| 11 |
+
"train_samples_per_second": 41.062,
|
| 12 |
+
"train_steps_per_second": 2.575
|
| 13 |
}
|
eval_results.json
CHANGED
|
@@ -1,8 +1,8 @@
|
|
| 1 |
{
|
| 2 |
-
"epoch":
|
| 3 |
-
"eval_accuracy": 0.
|
| 4 |
-
"eval_loss": 0.
|
| 5 |
-
"eval_runtime": 1.
|
| 6 |
-
"eval_samples_per_second":
|
| 7 |
-
"eval_steps_per_second": 13.
|
| 8 |
}
|
|
|
|
| 1 |
{
|
| 2 |
+
"epoch": 25.0,
|
| 3 |
+
"eval_accuracy": 0.711764705882353,
|
| 4 |
+
"eval_loss": 0.90943843126297,
|
| 5 |
+
"eval_runtime": 1.6048,
|
| 6 |
+
"eval_samples_per_second": 105.932,
|
| 7 |
+
"eval_steps_per_second": 13.709
|
| 8 |
}
|
runs/Aug06_21-54-27_a3e0ec26426d/events.out.tfevents.1659824092.a3e0ec26426d.71.5
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:9aee1918dd1c7b84a9ae2d3e88e049493c6628780b9db383efb7ac7d37bdcc19
|
| 3 |
+
size 1009
|
test_results.json
ADDED
|
@@ -0,0 +1,8 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"epoch": 25.0,
|
| 3 |
+
"eval_accuracy": 0.6839080459770115,
|
| 4 |
+
"eval_loss": 1.1035996675491333,
|
| 5 |
+
"eval_runtime": 1.6448,
|
| 6 |
+
"eval_samples_per_second": 105.785,
|
| 7 |
+
"eval_steps_per_second": 13.375
|
| 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": 2.
|
| 8 |
}
|
|
|
|
| 1 |
{
|
| 2 |
+
"epoch": 25.0,
|
| 3 |
+
"total_flos": 2.8110723118700544e+18,
|
| 4 |
+
"train_loss": 0.38098634693648786,
|
| 5 |
+
"train_runtime": 883.423,
|
| 6 |
+
"train_samples_per_second": 41.062,
|
| 7 |
+
"train_steps_per_second": 2.575
|
| 8 |
}
|
trainer_state.json
CHANGED
|
@@ -1,394 +1,1585 @@
|
|
| 1 |
{
|
| 2 |
-
"best_metric": 0.
|
| 3 |
-
"best_model_checkpoint": "./croupier-creature-classifier/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": 0.11,
|
| 12 |
-
"learning_rate": 0.
|
| 13 |
-
"loss": 1.
|
| 14 |
"step": 10
|
| 15 |
},
|
| 16 |
{
|
| 17 |
"epoch": 0.22,
|
| 18 |
-
"learning_rate": 0.
|
| 19 |
-
"loss": 1.
|
| 20 |
"step": 20
|
| 21 |
},
|
| 22 |
{
|
| 23 |
"epoch": 0.33,
|
| 24 |
-
"learning_rate": 0.
|
| 25 |
-
"loss": 1.
|
| 26 |
"step": 30
|
| 27 |
},
|
| 28 |
{
|
| 29 |
"epoch": 0.44,
|
| 30 |
-
"learning_rate": 0.
|
| 31 |
-
"loss": 1.
|
| 32 |
"step": 40
|
| 33 |
},
|
| 34 |
{
|
| 35 |
"epoch": 0.55,
|
| 36 |
-
"learning_rate": 0.
|
| 37 |
-
"loss": 1.
|
| 38 |
"step": 50
|
| 39 |
},
|
| 40 |
{
|
| 41 |
"epoch": 0.66,
|
| 42 |
-
"learning_rate": 0.
|
| 43 |
-
"loss": 1.
|
| 44 |
"step": 60
|
| 45 |
},
|
| 46 |
{
|
| 47 |
"epoch": 0.77,
|
| 48 |
-
"learning_rate": 0.
|
| 49 |
-
"loss": 1.
|
| 50 |
"step": 70
|
| 51 |
},
|
| 52 |
{
|
| 53 |
"epoch": 0.88,
|
| 54 |
-
"learning_rate": 0.
|
| 55 |
-
"loss": 1.
|
| 56 |
"step": 80
|
| 57 |
},
|
| 58 |
{
|
| 59 |
"epoch": 0.99,
|
| 60 |
-
"learning_rate": 0.
|
| 61 |
-
"loss": 1.
|
| 62 |
"step": 90
|
| 63 |
},
|
| 64 |
{
|
| 65 |
"epoch": 1.1,
|
| 66 |
-
"learning_rate": 0.
|
| 67 |
-
"loss":
|
| 68 |
"step": 100
|
| 69 |
},
|
| 70 |
{
|
| 71 |
"epoch": 1.1,
|
| 72 |
-
"eval_accuracy": 0.
|
| 73 |
-
"eval_loss":
|
| 74 |
-
"eval_runtime": 1.
|
| 75 |
-
"eval_samples_per_second":
|
| 76 |
-
"eval_steps_per_second": 14.
|
| 77 |
"step": 100
|
| 78 |
},
|
| 79 |
{
|
| 80 |
"epoch": 1.21,
|
| 81 |
-
"learning_rate": 0.
|
| 82 |
-
"loss":
|
| 83 |
"step": 110
|
| 84 |
},
|
| 85 |
{
|
| 86 |
"epoch": 1.32,
|
| 87 |
-
"learning_rate": 0.
|
| 88 |
-
"loss":
|
| 89 |
"step": 120
|
| 90 |
},
|
| 91 |
{
|
| 92 |
"epoch": 1.43,
|
| 93 |
-
"learning_rate": 0.
|
| 94 |
-
"loss":
|
| 95 |
"step": 130
|
| 96 |
},
|
| 97 |
{
|
| 98 |
"epoch": 1.54,
|
| 99 |
-
"learning_rate": 0.
|
| 100 |
-
"loss":
|
| 101 |
"step": 140
|
| 102 |
},
|
| 103 |
{
|
| 104 |
"epoch": 1.65,
|
| 105 |
-
"learning_rate": 0.
|
| 106 |
-
"loss":
|
| 107 |
"step": 150
|
| 108 |
},
|
| 109 |
{
|
| 110 |
"epoch": 1.76,
|
| 111 |
-
"learning_rate": 0.
|
| 112 |
-
"loss": 0.
|
| 113 |
"step": 160
|
| 114 |
},
|
| 115 |
{
|
| 116 |
"epoch": 1.87,
|
| 117 |
-
"learning_rate": 0.
|
| 118 |
-
"loss":
|
| 119 |
"step": 170
|
| 120 |
},
|
| 121 |
{
|
| 122 |
"epoch": 1.98,
|
| 123 |
-
"learning_rate": 0.
|
| 124 |
-
"loss":
|
| 125 |
"step": 180
|
| 126 |
},
|
| 127 |
{
|
| 128 |
"epoch": 2.09,
|
| 129 |
-
"learning_rate": 0.
|
| 130 |
-
"loss": 0.
|
| 131 |
"step": 190
|
| 132 |
},
|
| 133 |
{
|
| 134 |
"epoch": 2.2,
|
| 135 |
-
"learning_rate": 0.
|
| 136 |
-
"loss": 0.
|
| 137 |
"step": 200
|
| 138 |
},
|
| 139 |
{
|
| 140 |
"epoch": 2.2,
|
| 141 |
-
"eval_accuracy": 0.
|
| 142 |
-
"eval_loss": 0.
|
| 143 |
-
"eval_runtime": 1.
|
| 144 |
-
"eval_samples_per_second":
|
| 145 |
-
"eval_steps_per_second": 13.
|
| 146 |
"step": 200
|
| 147 |
},
|
| 148 |
{
|
| 149 |
"epoch": 2.31,
|
| 150 |
-
"learning_rate": 0.
|
| 151 |
-
"loss": 0.
|
| 152 |
"step": 210
|
| 153 |
},
|
| 154 |
{
|
| 155 |
"epoch": 2.42,
|
| 156 |
-
"learning_rate": 0.
|
| 157 |
-
"loss": 0.
|
| 158 |
"step": 220
|
| 159 |
},
|
| 160 |
{
|
| 161 |
"epoch": 2.53,
|
| 162 |
-
"learning_rate": 0.
|
| 163 |
-
"loss":
|
| 164 |
"step": 230
|
| 165 |
},
|
| 166 |
{
|
| 167 |
"epoch": 2.64,
|
| 168 |
-
"learning_rate": 0.
|
| 169 |
-
"loss": 0.
|
| 170 |
"step": 240
|
| 171 |
},
|
| 172 |
{
|
| 173 |
"epoch": 2.75,
|
| 174 |
-
"learning_rate": 0.
|
| 175 |
-
"loss": 0.
|
| 176 |
"step": 250
|
| 177 |
},
|
| 178 |
{
|
| 179 |
"epoch": 2.86,
|
| 180 |
-
"learning_rate": 0.
|
| 181 |
-
"loss": 0.
|
| 182 |
"step": 260
|
| 183 |
},
|
| 184 |
{
|
| 185 |
"epoch": 2.97,
|
| 186 |
-
"learning_rate": 0.
|
| 187 |
-
"loss": 0.
|
| 188 |
"step": 270
|
| 189 |
},
|
| 190 |
{
|
| 191 |
"epoch": 3.08,
|
| 192 |
-
"learning_rate":
|
| 193 |
-
"loss": 0.
|
| 194 |
"step": 280
|
| 195 |
},
|
| 196 |
{
|
| 197 |
"epoch": 3.19,
|
| 198 |
-
"learning_rate":
|
| 199 |
-
"loss": 0.
|
| 200 |
"step": 290
|
| 201 |
},
|
| 202 |
{
|
| 203 |
"epoch": 3.3,
|
| 204 |
-
"learning_rate":
|
| 205 |
-
"loss": 0.
|
| 206 |
"step": 300
|
| 207 |
},
|
| 208 |
{
|
| 209 |
"epoch": 3.3,
|
| 210 |
-
"eval_accuracy": 0.
|
| 211 |
-
"eval_loss": 0.
|
| 212 |
-
"eval_runtime": 2.
|
| 213 |
-
"eval_samples_per_second":
|
| 214 |
-
"eval_steps_per_second": 10.
|
| 215 |
"step": 300
|
| 216 |
},
|
| 217 |
{
|
| 218 |
"epoch": 3.41,
|
| 219 |
-
"learning_rate":
|
| 220 |
-
"loss": 0.
|
| 221 |
"step": 310
|
| 222 |
},
|
| 223 |
{
|
| 224 |
"epoch": 3.52,
|
| 225 |
-
"learning_rate":
|
| 226 |
-
"loss": 0.
|
| 227 |
"step": 320
|
| 228 |
},
|
| 229 |
{
|
| 230 |
"epoch": 3.63,
|
| 231 |
-
"learning_rate":
|
| 232 |
-
"loss": 0.
|
| 233 |
"step": 330
|
| 234 |
},
|
| 235 |
{
|
| 236 |
"epoch": 3.74,
|
| 237 |
-
"learning_rate":
|
| 238 |
-
"loss": 0.
|
| 239 |
"step": 340
|
| 240 |
},
|
| 241 |
{
|
| 242 |
"epoch": 3.85,
|
| 243 |
-
"learning_rate":
|
| 244 |
-
"loss": 0.
|
| 245 |
"step": 350
|
| 246 |
},
|
| 247 |
{
|
| 248 |
"epoch": 3.96,
|
| 249 |
-
"learning_rate":
|
| 250 |
-
"loss": 0.
|
| 251 |
"step": 360
|
| 252 |
},
|
| 253 |
{
|
| 254 |
"epoch": 4.07,
|
| 255 |
-
"learning_rate":
|
| 256 |
-
"loss": 0.
|
| 257 |
"step": 370
|
| 258 |
},
|
| 259 |
{
|
| 260 |
"epoch": 4.18,
|
| 261 |
-
"learning_rate":
|
| 262 |
-
"loss": 0.
|
| 263 |
"step": 380
|
| 264 |
},
|
| 265 |
{
|
| 266 |
"epoch": 4.29,
|
| 267 |
-
"learning_rate":
|
| 268 |
-
"loss": 0.
|
| 269 |
"step": 390
|
| 270 |
},
|
| 271 |
{
|
| 272 |
"epoch": 4.4,
|
| 273 |
-
"learning_rate":
|
| 274 |
-
"loss": 0.
|
| 275 |
"step": 400
|
| 276 |
},
|
| 277 |
{
|
| 278 |
"epoch": 4.4,
|
| 279 |
-
"eval_accuracy": 0.
|
| 280 |
-
"eval_loss": 0.
|
| 281 |
-
"eval_runtime": 1.
|
| 282 |
-
"eval_samples_per_second":
|
| 283 |
-
"eval_steps_per_second": 13.
|
| 284 |
"step": 400
|
| 285 |
},
|
| 286 |
{
|
| 287 |
"epoch": 4.51,
|
| 288 |
-
"learning_rate":
|
| 289 |
-
"loss": 0.
|
| 290 |
"step": 410
|
| 291 |
},
|
| 292 |
{
|
| 293 |
"epoch": 4.62,
|
| 294 |
-
"learning_rate":
|
| 295 |
-
"loss": 0.
|
| 296 |
"step": 420
|
| 297 |
},
|
| 298 |
{
|
| 299 |
"epoch": 4.73,
|
| 300 |
-
"learning_rate":
|
| 301 |
-
"loss": 0.
|
| 302 |
"step": 430
|
| 303 |
},
|
| 304 |
{
|
| 305 |
"epoch": 4.84,
|
| 306 |
-
"learning_rate":
|
| 307 |
-
"loss": 0.
|
| 308 |
"step": 440
|
| 309 |
},
|
| 310 |
{
|
| 311 |
"epoch": 4.95,
|
| 312 |
-
"learning_rate":
|
| 313 |
-
"loss": 0.
|
| 314 |
"step": 450
|
| 315 |
},
|
| 316 |
{
|
| 317 |
"epoch": 5.05,
|
| 318 |
-
"learning_rate":
|
| 319 |
-
"loss": 0.
|
| 320 |
"step": 460
|
| 321 |
},
|
| 322 |
{
|
| 323 |
"epoch": 5.16,
|
| 324 |
-
"learning_rate":
|
| 325 |
-
"loss": 0.
|
| 326 |
"step": 470
|
| 327 |
},
|
| 328 |
{
|
| 329 |
"epoch": 5.27,
|
| 330 |
-
"learning_rate":
|
| 331 |
-
"loss": 0.
|
| 332 |
"step": 480
|
| 333 |
},
|
| 334 |
{
|
| 335 |
"epoch": 5.38,
|
| 336 |
-
"learning_rate":
|
| 337 |
-
"loss": 0.
|
| 338 |
"step": 490
|
| 339 |
},
|
| 340 |
{
|
| 341 |
"epoch": 5.49,
|
| 342 |
-
"learning_rate":
|
| 343 |
-
"loss": 0.
|
| 344 |
"step": 500
|
| 345 |
},
|
| 346 |
{
|
| 347 |
"epoch": 5.49,
|
| 348 |
-
"eval_accuracy": 0.
|
| 349 |
-
"eval_loss": 0.
|
| 350 |
-
"eval_runtime": 1.
|
| 351 |
-
"eval_samples_per_second":
|
| 352 |
-
"eval_steps_per_second": 13.
|
| 353 |
"step": 500
|
| 354 |
},
|
| 355 |
{
|
| 356 |
"epoch": 5.6,
|
| 357 |
-
"learning_rate":
|
| 358 |
-
"loss": 0.
|
| 359 |
"step": 510
|
| 360 |
},
|
| 361 |
{
|
| 362 |
"epoch": 5.71,
|
| 363 |
-
"learning_rate":
|
| 364 |
-
"loss": 0.
|
| 365 |
"step": 520
|
| 366 |
},
|
| 367 |
{
|
| 368 |
"epoch": 5.82,
|
| 369 |
-
"learning_rate":
|
| 370 |
-
"loss": 0.
|
| 371 |
"step": 530
|
| 372 |
},
|
| 373 |
{
|
| 374 |
"epoch": 5.93,
|
| 375 |
-
"learning_rate":
|
| 376 |
-
"loss": 0.
|
| 377 |
"step": 540
|
| 378 |
},
|
| 379 |
{
|
| 380 |
-
"epoch": 6.
|
| 381 |
-
"
|
| 382 |
-
"
|
| 383 |
-
"
|
| 384 |
-
|
| 385 |
-
|
| 386 |
-
"
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 387 |
}
|
| 388 |
],
|
| 389 |
-
"max_steps":
|
| 390 |
-
"num_train_epochs":
|
| 391 |
-
"total_flos":
|
| 392 |
"trial_name": null,
|
| 393 |
"trial_params": null
|
| 394 |
}
|
|
|
|
| 1 |
{
|
| 2 |
+
"best_metric": 0.7445889711380005,
|
| 3 |
+
"best_model_checkpoint": "./croupier-creature-classifier/checkpoint-2200",
|
| 4 |
+
"epoch": 25.0,
|
| 5 |
+
"global_step": 2275,
|
| 6 |
"is_hyper_param_search": false,
|
| 7 |
"is_local_process_zero": true,
|
| 8 |
"is_world_process_zero": true,
|
| 9 |
"log_history": [
|
| 10 |
{
|
| 11 |
"epoch": 0.11,
|
| 12 |
+
"learning_rate": 0.00019912087912087913,
|
| 13 |
+
"loss": 1.3637,
|
| 14 |
"step": 10
|
| 15 |
},
|
| 16 |
{
|
| 17 |
"epoch": 0.22,
|
| 18 |
+
"learning_rate": 0.00019824175824175827,
|
| 19 |
+
"loss": 1.3785,
|
| 20 |
"step": 20
|
| 21 |
},
|
| 22 |
{
|
| 23 |
"epoch": 0.33,
|
| 24 |
+
"learning_rate": 0.00019736263736263738,
|
| 25 |
+
"loss": 1.3807,
|
| 26 |
"step": 30
|
| 27 |
},
|
| 28 |
{
|
| 29 |
"epoch": 0.44,
|
| 30 |
+
"learning_rate": 0.00019648351648351647,
|
| 31 |
+
"loss": 1.3582,
|
| 32 |
"step": 40
|
| 33 |
},
|
| 34 |
{
|
| 35 |
"epoch": 0.55,
|
| 36 |
+
"learning_rate": 0.00019560439560439562,
|
| 37 |
+
"loss": 1.2273,
|
| 38 |
"step": 50
|
| 39 |
},
|
| 40 |
{
|
| 41 |
"epoch": 0.66,
|
| 42 |
+
"learning_rate": 0.00019472527472527473,
|
| 43 |
+
"loss": 1.3811,
|
| 44 |
"step": 60
|
| 45 |
},
|
| 46 |
{
|
| 47 |
"epoch": 0.77,
|
| 48 |
+
"learning_rate": 0.00019384615384615385,
|
| 49 |
+
"loss": 1.3425,
|
| 50 |
"step": 70
|
| 51 |
},
|
| 52 |
{
|
| 53 |
"epoch": 0.88,
|
| 54 |
+
"learning_rate": 0.000192967032967033,
|
| 55 |
+
"loss": 1.3062,
|
| 56 |
"step": 80
|
| 57 |
},
|
| 58 |
{
|
| 59 |
"epoch": 0.99,
|
| 60 |
+
"learning_rate": 0.0001920879120879121,
|
| 61 |
+
"loss": 1.2959,
|
| 62 |
"step": 90
|
| 63 |
},
|
| 64 |
{
|
| 65 |
"epoch": 1.1,
|
| 66 |
+
"learning_rate": 0.00019120879120879122,
|
| 67 |
+
"loss": 1.1638,
|
| 68 |
"step": 100
|
| 69 |
},
|
| 70 |
{
|
| 71 |
"epoch": 1.1,
|
| 72 |
+
"eval_accuracy": 0.5470588235294118,
|
| 73 |
+
"eval_loss": 1.0563907623291016,
|
| 74 |
+
"eval_runtime": 1.5249,
|
| 75 |
+
"eval_samples_per_second": 111.483,
|
| 76 |
+
"eval_steps_per_second": 14.427,
|
| 77 |
"step": 100
|
| 78 |
},
|
| 79 |
{
|
| 80 |
"epoch": 1.21,
|
| 81 |
+
"learning_rate": 0.00019032967032967034,
|
| 82 |
+
"loss": 1.0851,
|
| 83 |
"step": 110
|
| 84 |
},
|
| 85 |
{
|
| 86 |
"epoch": 1.32,
|
| 87 |
+
"learning_rate": 0.00018945054945054945,
|
| 88 |
+
"loss": 1.1777,
|
| 89 |
"step": 120
|
| 90 |
},
|
| 91 |
{
|
| 92 |
"epoch": 1.43,
|
| 93 |
+
"learning_rate": 0.00018857142857142857,
|
| 94 |
+
"loss": 1.1352,
|
| 95 |
"step": 130
|
| 96 |
},
|
| 97 |
{
|
| 98 |
"epoch": 1.54,
|
| 99 |
+
"learning_rate": 0.0001876923076923077,
|
| 100 |
+
"loss": 1.0112,
|
| 101 |
"step": 140
|
| 102 |
},
|
| 103 |
{
|
| 104 |
"epoch": 1.65,
|
| 105 |
+
"learning_rate": 0.00018681318681318683,
|
| 106 |
+
"loss": 1.0718,
|
| 107 |
"step": 150
|
| 108 |
},
|
| 109 |
{
|
| 110 |
"epoch": 1.76,
|
| 111 |
+
"learning_rate": 0.00018593406593406595,
|
| 112 |
+
"loss": 0.9871,
|
| 113 |
"step": 160
|
| 114 |
},
|
| 115 |
{
|
| 116 |
"epoch": 1.87,
|
| 117 |
+
"learning_rate": 0.00018505494505494506,
|
| 118 |
+
"loss": 1.2279,
|
| 119 |
"step": 170
|
| 120 |
},
|
| 121 |
{
|
| 122 |
"epoch": 1.98,
|
| 123 |
+
"learning_rate": 0.00018417582417582418,
|
| 124 |
+
"loss": 1.148,
|
| 125 |
"step": 180
|
| 126 |
},
|
| 127 |
{
|
| 128 |
"epoch": 2.09,
|
| 129 |
+
"learning_rate": 0.0001832967032967033,
|
| 130 |
+
"loss": 0.8905,
|
| 131 |
"step": 190
|
| 132 |
},
|
| 133 |
{
|
| 134 |
"epoch": 2.2,
|
| 135 |
+
"learning_rate": 0.0001824175824175824,
|
| 136 |
+
"loss": 0.8524,
|
| 137 |
"step": 200
|
| 138 |
},
|
| 139 |
{
|
| 140 |
"epoch": 2.2,
|
| 141 |
+
"eval_accuracy": 0.611764705882353,
|
| 142 |
+
"eval_loss": 0.9402753114700317,
|
| 143 |
+
"eval_runtime": 1.611,
|
| 144 |
+
"eval_samples_per_second": 105.524,
|
| 145 |
+
"eval_steps_per_second": 13.656,
|
| 146 |
"step": 200
|
| 147 |
},
|
| 148 |
{
|
| 149 |
"epoch": 2.31,
|
| 150 |
+
"learning_rate": 0.00018153846153846155,
|
| 151 |
+
"loss": 0.8388,
|
| 152 |
"step": 210
|
| 153 |
},
|
| 154 |
{
|
| 155 |
"epoch": 2.42,
|
| 156 |
+
"learning_rate": 0.00018065934065934067,
|
| 157 |
+
"loss": 0.7519,
|
| 158 |
"step": 220
|
| 159 |
},
|
| 160 |
{
|
| 161 |
"epoch": 2.53,
|
| 162 |
+
"learning_rate": 0.00017978021978021978,
|
| 163 |
+
"loss": 1.0743,
|
| 164 |
"step": 230
|
| 165 |
},
|
| 166 |
{
|
| 167 |
"epoch": 2.64,
|
| 168 |
+
"learning_rate": 0.00017890109890109893,
|
| 169 |
+
"loss": 0.8687,
|
| 170 |
"step": 240
|
| 171 |
},
|
| 172 |
{
|
| 173 |
"epoch": 2.75,
|
| 174 |
+
"learning_rate": 0.00017802197802197802,
|
| 175 |
+
"loss": 0.886,
|
| 176 |
"step": 250
|
| 177 |
},
|
| 178 |
{
|
| 179 |
"epoch": 2.86,
|
| 180 |
+
"learning_rate": 0.00017714285714285713,
|
| 181 |
+
"loss": 0.8484,
|
| 182 |
"step": 260
|
| 183 |
},
|
| 184 |
{
|
| 185 |
"epoch": 2.97,
|
| 186 |
+
"learning_rate": 0.00017626373626373627,
|
| 187 |
+
"loss": 0.8394,
|
| 188 |
"step": 270
|
| 189 |
},
|
| 190 |
{
|
| 191 |
"epoch": 3.08,
|
| 192 |
+
"learning_rate": 0.0001753846153846154,
|
| 193 |
+
"loss": 0.8304,
|
| 194 |
"step": 280
|
| 195 |
},
|
| 196 |
{
|
| 197 |
"epoch": 3.19,
|
| 198 |
+
"learning_rate": 0.0001745054945054945,
|
| 199 |
+
"loss": 0.6359,
|
| 200 |
"step": 290
|
| 201 |
},
|
| 202 |
{
|
| 203 |
"epoch": 3.3,
|
| 204 |
+
"learning_rate": 0.00017362637362637365,
|
| 205 |
+
"loss": 0.8231,
|
| 206 |
"step": 300
|
| 207 |
},
|
| 208 |
{
|
| 209 |
"epoch": 3.3,
|
| 210 |
+
"eval_accuracy": 0.7176470588235294,
|
| 211 |
+
"eval_loss": 0.8282244801521301,
|
| 212 |
+
"eval_runtime": 2.0367,
|
| 213 |
+
"eval_samples_per_second": 83.467,
|
| 214 |
+
"eval_steps_per_second": 10.802,
|
| 215 |
"step": 300
|
| 216 |
},
|
| 217 |
{
|
| 218 |
"epoch": 3.41,
|
| 219 |
+
"learning_rate": 0.00017274725274725276,
|
| 220 |
+
"loss": 0.6735,
|
| 221 |
"step": 310
|
| 222 |
},
|
| 223 |
{
|
| 224 |
"epoch": 3.52,
|
| 225 |
+
"learning_rate": 0.00017186813186813185,
|
| 226 |
+
"loss": 0.6721,
|
| 227 |
"step": 320
|
| 228 |
},
|
| 229 |
{
|
| 230 |
"epoch": 3.63,
|
| 231 |
+
"learning_rate": 0.000170989010989011,
|
| 232 |
+
"loss": 0.8183,
|
| 233 |
"step": 330
|
| 234 |
},
|
| 235 |
{
|
| 236 |
"epoch": 3.74,
|
| 237 |
+
"learning_rate": 0.0001701098901098901,
|
| 238 |
+
"loss": 0.7158,
|
| 239 |
"step": 340
|
| 240 |
},
|
| 241 |
{
|
| 242 |
"epoch": 3.85,
|
| 243 |
+
"learning_rate": 0.00016923076923076923,
|
| 244 |
+
"loss": 0.796,
|
| 245 |
"step": 350
|
| 246 |
},
|
| 247 |
{
|
| 248 |
"epoch": 3.96,
|
| 249 |
+
"learning_rate": 0.00016835164835164837,
|
| 250 |
+
"loss": 0.7814,
|
| 251 |
"step": 360
|
| 252 |
},
|
| 253 |
{
|
| 254 |
"epoch": 4.07,
|
| 255 |
+
"learning_rate": 0.0001674725274725275,
|
| 256 |
+
"loss": 0.6363,
|
| 257 |
"step": 370
|
| 258 |
},
|
| 259 |
{
|
| 260 |
"epoch": 4.18,
|
| 261 |
+
"learning_rate": 0.0001665934065934066,
|
| 262 |
+
"loss": 0.57,
|
| 263 |
"step": 380
|
| 264 |
},
|
| 265 |
{
|
| 266 |
"epoch": 4.29,
|
| 267 |
+
"learning_rate": 0.00016571428571428575,
|
| 268 |
+
"loss": 0.5695,
|
| 269 |
"step": 390
|
| 270 |
},
|
| 271 |
{
|
| 272 |
"epoch": 4.4,
|
| 273 |
+
"learning_rate": 0.00016483516483516484,
|
| 274 |
+
"loss": 0.7398,
|
| 275 |
"step": 400
|
| 276 |
},
|
| 277 |
{
|
| 278 |
"epoch": 4.4,
|
| 279 |
+
"eval_accuracy": 0.6294117647058823,
|
| 280 |
+
"eval_loss": 0.9056094884872437,
|
| 281 |
+
"eval_runtime": 1.6835,
|
| 282 |
+
"eval_samples_per_second": 100.978,
|
| 283 |
+
"eval_steps_per_second": 13.068,
|
| 284 |
"step": 400
|
| 285 |
},
|
| 286 |
{
|
| 287 |
"epoch": 4.51,
|
| 288 |
+
"learning_rate": 0.00016395604395604395,
|
| 289 |
+
"loss": 0.5678,
|
| 290 |
"step": 410
|
| 291 |
},
|
| 292 |
{
|
| 293 |
"epoch": 4.62,
|
| 294 |
+
"learning_rate": 0.0001630769230769231,
|
| 295 |
+
"loss": 0.5507,
|
| 296 |
"step": 420
|
| 297 |
},
|
| 298 |
{
|
| 299 |
"epoch": 4.73,
|
| 300 |
+
"learning_rate": 0.0001621978021978022,
|
| 301 |
+
"loss": 0.4983,
|
| 302 |
"step": 430
|
| 303 |
},
|
| 304 |
{
|
| 305 |
"epoch": 4.84,
|
| 306 |
+
"learning_rate": 0.00016131868131868133,
|
| 307 |
+
"loss": 0.6078,
|
| 308 |
"step": 440
|
| 309 |
},
|
| 310 |
{
|
| 311 |
"epoch": 4.95,
|
| 312 |
+
"learning_rate": 0.00016043956043956044,
|
| 313 |
+
"loss": 0.584,
|
| 314 |
"step": 450
|
| 315 |
},
|
| 316 |
{
|
| 317 |
"epoch": 5.05,
|
| 318 |
+
"learning_rate": 0.00015956043956043958,
|
| 319 |
+
"loss": 0.4714,
|
| 320 |
"step": 460
|
| 321 |
},
|
| 322 |
{
|
| 323 |
"epoch": 5.16,
|
| 324 |
+
"learning_rate": 0.00015868131868131867,
|
| 325 |
+
"loss": 0.6575,
|
| 326 |
"step": 470
|
| 327 |
},
|
| 328 |
{
|
| 329 |
"epoch": 5.27,
|
| 330 |
+
"learning_rate": 0.00015780219780219782,
|
| 331 |
+
"loss": 0.522,
|
| 332 |
"step": 480
|
| 333 |
},
|
| 334 |
{
|
| 335 |
"epoch": 5.38,
|
| 336 |
+
"learning_rate": 0.00015692307692307693,
|
| 337 |
+
"loss": 0.5475,
|
| 338 |
"step": 490
|
| 339 |
},
|
| 340 |
{
|
| 341 |
"epoch": 5.49,
|
| 342 |
+
"learning_rate": 0.00015604395604395605,
|
| 343 |
+
"loss": 0.41,
|
| 344 |
"step": 500
|
| 345 |
},
|
| 346 |
{
|
| 347 |
"epoch": 5.49,
|
| 348 |
+
"eval_accuracy": 0.6235294117647059,
|
| 349 |
+
"eval_loss": 0.8814617991447449,
|
| 350 |
+
"eval_runtime": 1.5991,
|
| 351 |
+
"eval_samples_per_second": 106.312,
|
| 352 |
+
"eval_steps_per_second": 13.758,
|
| 353 |
"step": 500
|
| 354 |
},
|
| 355 |
{
|
| 356 |
"epoch": 5.6,
|
| 357 |
+
"learning_rate": 0.00015516483516483516,
|
| 358 |
+
"loss": 0.4955,
|
| 359 |
"step": 510
|
| 360 |
},
|
| 361 |
{
|
| 362 |
"epoch": 5.71,
|
| 363 |
+
"learning_rate": 0.0001542857142857143,
|
| 364 |
+
"loss": 0.5234,
|
| 365 |
"step": 520
|
| 366 |
},
|
| 367 |
{
|
| 368 |
"epoch": 5.82,
|
| 369 |
+
"learning_rate": 0.00015340659340659342,
|
| 370 |
+
"loss": 0.4696,
|
| 371 |
"step": 530
|
| 372 |
},
|
| 373 |
{
|
| 374 |
"epoch": 5.93,
|
| 375 |
+
"learning_rate": 0.00015252747252747254,
|
| 376 |
+
"loss": 0.4195,
|
| 377 |
"step": 540
|
| 378 |
},
|
| 379 |
{
|
| 380 |
+
"epoch": 6.04,
|
| 381 |
+
"learning_rate": 0.00015164835164835165,
|
| 382 |
+
"loss": 0.5375,
|
| 383 |
+
"step": 550
|
| 384 |
+
},
|
| 385 |
+
{
|
| 386 |
+
"epoch": 6.15,
|
| 387 |
+
"learning_rate": 0.00015076923076923077,
|
| 388 |
+
"loss": 0.3726,
|
| 389 |
+
"step": 560
|
| 390 |
+
},
|
| 391 |
+
{
|
| 392 |
+
"epoch": 6.26,
|
| 393 |
+
"learning_rate": 0.0001498901098901099,
|
| 394 |
+
"loss": 0.3705,
|
| 395 |
+
"step": 570
|
| 396 |
+
},
|
| 397 |
+
{
|
| 398 |
+
"epoch": 6.37,
|
| 399 |
+
"learning_rate": 0.00014901098901098903,
|
| 400 |
+
"loss": 0.3445,
|
| 401 |
+
"step": 580
|
| 402 |
+
},
|
| 403 |
+
{
|
| 404 |
+
"epoch": 6.48,
|
| 405 |
+
"learning_rate": 0.00014813186813186815,
|
| 406 |
+
"loss": 0.4785,
|
| 407 |
+
"step": 590
|
| 408 |
+
},
|
| 409 |
+
{
|
| 410 |
+
"epoch": 6.59,
|
| 411 |
+
"learning_rate": 0.00014725274725274726,
|
| 412 |
+
"loss": 0.4849,
|
| 413 |
+
"step": 600
|
| 414 |
+
},
|
| 415 |
+
{
|
| 416 |
+
"epoch": 6.59,
|
| 417 |
+
"eval_accuracy": 0.6294117647058823,
|
| 418 |
+
"eval_loss": 0.9504889249801636,
|
| 419 |
+
"eval_runtime": 1.7834,
|
| 420 |
+
"eval_samples_per_second": 95.325,
|
| 421 |
+
"eval_steps_per_second": 12.336,
|
| 422 |
+
"step": 600
|
| 423 |
+
},
|
| 424 |
+
{
|
| 425 |
+
"epoch": 6.7,
|
| 426 |
+
"learning_rate": 0.00014637362637362638,
|
| 427 |
+
"loss": 0.5062,
|
| 428 |
+
"step": 610
|
| 429 |
+
},
|
| 430 |
+
{
|
| 431 |
+
"epoch": 6.81,
|
| 432 |
+
"learning_rate": 0.0001454945054945055,
|
| 433 |
+
"loss": 0.4269,
|
| 434 |
+
"step": 620
|
| 435 |
+
},
|
| 436 |
+
{
|
| 437 |
+
"epoch": 6.92,
|
| 438 |
+
"learning_rate": 0.0001446153846153846,
|
| 439 |
+
"loss": 0.3566,
|
| 440 |
+
"step": 630
|
| 441 |
+
},
|
| 442 |
+
{
|
| 443 |
+
"epoch": 7.03,
|
| 444 |
+
"learning_rate": 0.00014373626373626375,
|
| 445 |
+
"loss": 0.3931,
|
| 446 |
+
"step": 640
|
| 447 |
+
},
|
| 448 |
+
{
|
| 449 |
+
"epoch": 7.14,
|
| 450 |
+
"learning_rate": 0.00014285714285714287,
|
| 451 |
+
"loss": 0.3896,
|
| 452 |
+
"step": 650
|
| 453 |
+
},
|
| 454 |
+
{
|
| 455 |
+
"epoch": 7.25,
|
| 456 |
+
"learning_rate": 0.00014197802197802198,
|
| 457 |
+
"loss": 0.3306,
|
| 458 |
+
"step": 660
|
| 459 |
+
},
|
| 460 |
+
{
|
| 461 |
+
"epoch": 7.36,
|
| 462 |
+
"learning_rate": 0.00014109890109890113,
|
| 463 |
+
"loss": 0.4898,
|
| 464 |
+
"step": 670
|
| 465 |
+
},
|
| 466 |
+
{
|
| 467 |
+
"epoch": 7.47,
|
| 468 |
+
"learning_rate": 0.00014021978021978022,
|
| 469 |
+
"loss": 0.4136,
|
| 470 |
+
"step": 680
|
| 471 |
+
},
|
| 472 |
+
{
|
| 473 |
+
"epoch": 7.58,
|
| 474 |
+
"learning_rate": 0.00013934065934065933,
|
| 475 |
+
"loss": 0.4874,
|
| 476 |
+
"step": 690
|
| 477 |
+
},
|
| 478 |
+
{
|
| 479 |
+
"epoch": 7.69,
|
| 480 |
+
"learning_rate": 0.00013846153846153847,
|
| 481 |
+
"loss": 0.3894,
|
| 482 |
+
"step": 700
|
| 483 |
+
},
|
| 484 |
+
{
|
| 485 |
+
"epoch": 7.69,
|
| 486 |
+
"eval_accuracy": 0.6882352941176471,
|
| 487 |
+
"eval_loss": 0.8051896095275879,
|
| 488 |
+
"eval_runtime": 2.0897,
|
| 489 |
+
"eval_samples_per_second": 81.352,
|
| 490 |
+
"eval_steps_per_second": 10.528,
|
| 491 |
+
"step": 700
|
| 492 |
+
},
|
| 493 |
+
{
|
| 494 |
+
"epoch": 7.8,
|
| 495 |
+
"learning_rate": 0.0001375824175824176,
|
| 496 |
+
"loss": 0.355,
|
| 497 |
+
"step": 710
|
| 498 |
+
},
|
| 499 |
+
{
|
| 500 |
+
"epoch": 7.91,
|
| 501 |
+
"learning_rate": 0.0001367032967032967,
|
| 502 |
+
"loss": 0.4024,
|
| 503 |
+
"step": 720
|
| 504 |
+
},
|
| 505 |
+
{
|
| 506 |
+
"epoch": 8.02,
|
| 507 |
+
"learning_rate": 0.00013582417582417585,
|
| 508 |
+
"loss": 0.458,
|
| 509 |
+
"step": 730
|
| 510 |
+
},
|
| 511 |
+
{
|
| 512 |
+
"epoch": 8.13,
|
| 513 |
+
"learning_rate": 0.00013494505494505497,
|
| 514 |
+
"loss": 0.2912,
|
| 515 |
+
"step": 740
|
| 516 |
+
},
|
| 517 |
+
{
|
| 518 |
+
"epoch": 8.24,
|
| 519 |
+
"learning_rate": 0.00013406593406593405,
|
| 520 |
+
"loss": 0.3608,
|
| 521 |
+
"step": 750
|
| 522 |
+
},
|
| 523 |
+
{
|
| 524 |
+
"epoch": 8.35,
|
| 525 |
+
"learning_rate": 0.0001331868131868132,
|
| 526 |
+
"loss": 0.3897,
|
| 527 |
+
"step": 760
|
| 528 |
+
},
|
| 529 |
+
{
|
| 530 |
+
"epoch": 8.46,
|
| 531 |
+
"learning_rate": 0.0001323076923076923,
|
| 532 |
+
"loss": 0.3451,
|
| 533 |
+
"step": 770
|
| 534 |
+
},
|
| 535 |
+
{
|
| 536 |
+
"epoch": 8.57,
|
| 537 |
+
"learning_rate": 0.00013142857142857143,
|
| 538 |
+
"loss": 0.3122,
|
| 539 |
+
"step": 780
|
| 540 |
+
},
|
| 541 |
+
{
|
| 542 |
+
"epoch": 8.68,
|
| 543 |
+
"learning_rate": 0.00013054945054945057,
|
| 544 |
+
"loss": 0.2771,
|
| 545 |
+
"step": 790
|
| 546 |
+
},
|
| 547 |
+
{
|
| 548 |
+
"epoch": 8.79,
|
| 549 |
+
"learning_rate": 0.0001296703296703297,
|
| 550 |
+
"loss": 0.4678,
|
| 551 |
+
"step": 800
|
| 552 |
+
},
|
| 553 |
+
{
|
| 554 |
+
"epoch": 8.79,
|
| 555 |
+
"eval_accuracy": 0.7058823529411765,
|
| 556 |
+
"eval_loss": 0.8423882722854614,
|
| 557 |
+
"eval_runtime": 1.8864,
|
| 558 |
+
"eval_samples_per_second": 90.118,
|
| 559 |
+
"eval_steps_per_second": 11.662,
|
| 560 |
+
"step": 800
|
| 561 |
+
},
|
| 562 |
+
{
|
| 563 |
+
"epoch": 8.9,
|
| 564 |
+
"learning_rate": 0.0001287912087912088,
|
| 565 |
+
"loss": 0.4709,
|
| 566 |
+
"step": 810
|
| 567 |
+
},
|
| 568 |
+
{
|
| 569 |
+
"epoch": 9.01,
|
| 570 |
+
"learning_rate": 0.00012791208791208792,
|
| 571 |
+
"loss": 0.3594,
|
| 572 |
+
"step": 820
|
| 573 |
+
},
|
| 574 |
+
{
|
| 575 |
+
"epoch": 9.12,
|
| 576 |
+
"learning_rate": 0.00012703296703296704,
|
| 577 |
+
"loss": 0.3349,
|
| 578 |
+
"step": 830
|
| 579 |
+
},
|
| 580 |
+
{
|
| 581 |
+
"epoch": 9.23,
|
| 582 |
+
"learning_rate": 0.00012615384615384615,
|
| 583 |
+
"loss": 0.4093,
|
| 584 |
+
"step": 840
|
| 585 |
+
},
|
| 586 |
+
{
|
| 587 |
+
"epoch": 9.34,
|
| 588 |
+
"learning_rate": 0.00012527472527472527,
|
| 589 |
+
"loss": 0.2627,
|
| 590 |
+
"step": 850
|
| 591 |
+
},
|
| 592 |
+
{
|
| 593 |
+
"epoch": 9.45,
|
| 594 |
+
"learning_rate": 0.0001243956043956044,
|
| 595 |
+
"loss": 0.2817,
|
| 596 |
+
"step": 860
|
| 597 |
+
},
|
| 598 |
+
{
|
| 599 |
+
"epoch": 9.56,
|
| 600 |
+
"learning_rate": 0.00012351648351648353,
|
| 601 |
+
"loss": 0.3046,
|
| 602 |
+
"step": 870
|
| 603 |
+
},
|
| 604 |
+
{
|
| 605 |
+
"epoch": 9.67,
|
| 606 |
+
"learning_rate": 0.00012263736263736264,
|
| 607 |
+
"loss": 0.5005,
|
| 608 |
+
"step": 880
|
| 609 |
+
},
|
| 610 |
+
{
|
| 611 |
+
"epoch": 9.78,
|
| 612 |
+
"learning_rate": 0.00012175824175824176,
|
| 613 |
+
"loss": 0.1866,
|
| 614 |
+
"step": 890
|
| 615 |
+
},
|
| 616 |
+
{
|
| 617 |
+
"epoch": 9.89,
|
| 618 |
+
"learning_rate": 0.00012087912087912087,
|
| 619 |
+
"loss": 0.4279,
|
| 620 |
+
"step": 900
|
| 621 |
+
},
|
| 622 |
+
{
|
| 623 |
+
"epoch": 9.89,
|
| 624 |
+
"eval_accuracy": 0.6705882352941176,
|
| 625 |
+
"eval_loss": 0.963887095451355,
|
| 626 |
+
"eval_runtime": 1.496,
|
| 627 |
+
"eval_samples_per_second": 113.639,
|
| 628 |
+
"eval_steps_per_second": 14.706,
|
| 629 |
+
"step": 900
|
| 630 |
+
},
|
| 631 |
+
{
|
| 632 |
+
"epoch": 10.0,
|
| 633 |
+
"learning_rate": 0.00012,
|
| 634 |
+
"loss": 0.4008,
|
| 635 |
+
"step": 910
|
| 636 |
+
},
|
| 637 |
+
{
|
| 638 |
+
"epoch": 10.11,
|
| 639 |
+
"learning_rate": 0.00011912087912087912,
|
| 640 |
+
"loss": 0.3001,
|
| 641 |
+
"step": 920
|
| 642 |
+
},
|
| 643 |
+
{
|
| 644 |
+
"epoch": 10.22,
|
| 645 |
+
"learning_rate": 0.00011824175824175825,
|
| 646 |
+
"loss": 0.2588,
|
| 647 |
+
"step": 930
|
| 648 |
+
},
|
| 649 |
+
{
|
| 650 |
+
"epoch": 10.33,
|
| 651 |
+
"learning_rate": 0.00011736263736263738,
|
| 652 |
+
"loss": 0.4082,
|
| 653 |
+
"step": 940
|
| 654 |
+
},
|
| 655 |
+
{
|
| 656 |
+
"epoch": 10.44,
|
| 657 |
+
"learning_rate": 0.0001164835164835165,
|
| 658 |
+
"loss": 0.3957,
|
| 659 |
+
"step": 950
|
| 660 |
+
},
|
| 661 |
+
{
|
| 662 |
+
"epoch": 10.55,
|
| 663 |
+
"learning_rate": 0.0001156043956043956,
|
| 664 |
+
"loss": 0.3988,
|
| 665 |
+
"step": 960
|
| 666 |
+
},
|
| 667 |
+
{
|
| 668 |
+
"epoch": 10.66,
|
| 669 |
+
"learning_rate": 0.00011472527472527473,
|
| 670 |
+
"loss": 0.2264,
|
| 671 |
+
"step": 970
|
| 672 |
+
},
|
| 673 |
+
{
|
| 674 |
+
"epoch": 10.77,
|
| 675 |
+
"learning_rate": 0.00011384615384615384,
|
| 676 |
+
"loss": 0.3202,
|
| 677 |
+
"step": 980
|
| 678 |
+
},
|
| 679 |
+
{
|
| 680 |
+
"epoch": 10.88,
|
| 681 |
+
"learning_rate": 0.00011296703296703297,
|
| 682 |
+
"loss": 0.2622,
|
| 683 |
+
"step": 990
|
| 684 |
+
},
|
| 685 |
+
{
|
| 686 |
+
"epoch": 10.99,
|
| 687 |
+
"learning_rate": 0.0001120879120879121,
|
| 688 |
+
"loss": 0.3461,
|
| 689 |
+
"step": 1000
|
| 690 |
+
},
|
| 691 |
+
{
|
| 692 |
+
"epoch": 10.99,
|
| 693 |
+
"eval_accuracy": 0.7058823529411765,
|
| 694 |
+
"eval_loss": 0.8496671915054321,
|
| 695 |
+
"eval_runtime": 1.4967,
|
| 696 |
+
"eval_samples_per_second": 113.586,
|
| 697 |
+
"eval_steps_per_second": 14.699,
|
| 698 |
+
"step": 1000
|
| 699 |
+
},
|
| 700 |
+
{
|
| 701 |
+
"epoch": 11.1,
|
| 702 |
+
"learning_rate": 0.00011120879120879122,
|
| 703 |
+
"loss": 0.1967,
|
| 704 |
+
"step": 1010
|
| 705 |
+
},
|
| 706 |
+
{
|
| 707 |
+
"epoch": 11.21,
|
| 708 |
+
"learning_rate": 0.00011032967032967035,
|
| 709 |
+
"loss": 0.2573,
|
| 710 |
+
"step": 1020
|
| 711 |
+
},
|
| 712 |
+
{
|
| 713 |
+
"epoch": 11.32,
|
| 714 |
+
"learning_rate": 0.00010945054945054946,
|
| 715 |
+
"loss": 0.3313,
|
| 716 |
+
"step": 1030
|
| 717 |
+
},
|
| 718 |
+
{
|
| 719 |
+
"epoch": 11.43,
|
| 720 |
+
"learning_rate": 0.00010857142857142856,
|
| 721 |
+
"loss": 0.2774,
|
| 722 |
+
"step": 1040
|
| 723 |
+
},
|
| 724 |
+
{
|
| 725 |
+
"epoch": 11.54,
|
| 726 |
+
"learning_rate": 0.0001076923076923077,
|
| 727 |
+
"loss": 0.1745,
|
| 728 |
+
"step": 1050
|
| 729 |
+
},
|
| 730 |
+
{
|
| 731 |
+
"epoch": 11.65,
|
| 732 |
+
"learning_rate": 0.00010681318681318682,
|
| 733 |
+
"loss": 0.2563,
|
| 734 |
+
"step": 1060
|
| 735 |
+
},
|
| 736 |
+
{
|
| 737 |
+
"epoch": 11.76,
|
| 738 |
+
"learning_rate": 0.00010593406593406594,
|
| 739 |
+
"loss": 0.15,
|
| 740 |
+
"step": 1070
|
| 741 |
+
},
|
| 742 |
+
{
|
| 743 |
+
"epoch": 11.87,
|
| 744 |
+
"learning_rate": 0.00010505494505494507,
|
| 745 |
+
"loss": 0.346,
|
| 746 |
+
"step": 1080
|
| 747 |
+
},
|
| 748 |
+
{
|
| 749 |
+
"epoch": 11.98,
|
| 750 |
+
"learning_rate": 0.00010417582417582418,
|
| 751 |
+
"loss": 0.3187,
|
| 752 |
+
"step": 1090
|
| 753 |
+
},
|
| 754 |
+
{
|
| 755 |
+
"epoch": 12.09,
|
| 756 |
+
"learning_rate": 0.00010329670329670331,
|
| 757 |
+
"loss": 0.2741,
|
| 758 |
+
"step": 1100
|
| 759 |
+
},
|
| 760 |
+
{
|
| 761 |
+
"epoch": 12.09,
|
| 762 |
+
"eval_accuracy": 0.7,
|
| 763 |
+
"eval_loss": 0.9090209007263184,
|
| 764 |
+
"eval_runtime": 1.5007,
|
| 765 |
+
"eval_samples_per_second": 113.284,
|
| 766 |
+
"eval_steps_per_second": 14.66,
|
| 767 |
+
"step": 1100
|
| 768 |
+
},
|
| 769 |
+
{
|
| 770 |
+
"epoch": 12.2,
|
| 771 |
+
"learning_rate": 0.00010241758241758242,
|
| 772 |
+
"loss": 0.2168,
|
| 773 |
+
"step": 1110
|
| 774 |
+
},
|
| 775 |
+
{
|
| 776 |
+
"epoch": 12.31,
|
| 777 |
+
"learning_rate": 0.00010153846153846153,
|
| 778 |
+
"loss": 0.1902,
|
| 779 |
+
"step": 1120
|
| 780 |
+
},
|
| 781 |
+
{
|
| 782 |
+
"epoch": 12.42,
|
| 783 |
+
"learning_rate": 0.00010065934065934066,
|
| 784 |
+
"loss": 0.2154,
|
| 785 |
+
"step": 1130
|
| 786 |
+
},
|
| 787 |
+
{
|
| 788 |
+
"epoch": 12.53,
|
| 789 |
+
"learning_rate": 9.978021978021979e-05,
|
| 790 |
+
"loss": 0.3392,
|
| 791 |
+
"step": 1140
|
| 792 |
+
},
|
| 793 |
+
{
|
| 794 |
+
"epoch": 12.64,
|
| 795 |
+
"learning_rate": 9.89010989010989e-05,
|
| 796 |
+
"loss": 0.2567,
|
| 797 |
+
"step": 1150
|
| 798 |
+
},
|
| 799 |
+
{
|
| 800 |
+
"epoch": 12.75,
|
| 801 |
+
"learning_rate": 9.802197802197802e-05,
|
| 802 |
+
"loss": 0.3863,
|
| 803 |
+
"step": 1160
|
| 804 |
+
},
|
| 805 |
+
{
|
| 806 |
+
"epoch": 12.86,
|
| 807 |
+
"learning_rate": 9.714285714285715e-05,
|
| 808 |
+
"loss": 0.3635,
|
| 809 |
+
"step": 1170
|
| 810 |
+
},
|
| 811 |
+
{
|
| 812 |
+
"epoch": 12.97,
|
| 813 |
+
"learning_rate": 9.626373626373627e-05,
|
| 814 |
+
"loss": 0.2915,
|
| 815 |
+
"step": 1180
|
| 816 |
+
},
|
| 817 |
+
{
|
| 818 |
+
"epoch": 13.08,
|
| 819 |
+
"learning_rate": 9.53846153846154e-05,
|
| 820 |
+
"loss": 0.3021,
|
| 821 |
+
"step": 1190
|
| 822 |
+
},
|
| 823 |
+
{
|
| 824 |
+
"epoch": 13.19,
|
| 825 |
+
"learning_rate": 9.450549450549451e-05,
|
| 826 |
+
"loss": 0.1771,
|
| 827 |
+
"step": 1200
|
| 828 |
+
},
|
| 829 |
+
{
|
| 830 |
+
"epoch": 13.19,
|
| 831 |
+
"eval_accuracy": 0.711764705882353,
|
| 832 |
+
"eval_loss": 0.8292282819747925,
|
| 833 |
+
"eval_runtime": 1.5182,
|
| 834 |
+
"eval_samples_per_second": 111.974,
|
| 835 |
+
"eval_steps_per_second": 14.491,
|
| 836 |
+
"step": 1200
|
| 837 |
+
},
|
| 838 |
+
{
|
| 839 |
+
"epoch": 13.3,
|
| 840 |
+
"learning_rate": 9.362637362637363e-05,
|
| 841 |
+
"loss": 0.1475,
|
| 842 |
+
"step": 1210
|
| 843 |
+
},
|
| 844 |
+
{
|
| 845 |
+
"epoch": 13.41,
|
| 846 |
+
"learning_rate": 9.274725274725276e-05,
|
| 847 |
+
"loss": 0.1051,
|
| 848 |
+
"step": 1220
|
| 849 |
+
},
|
| 850 |
+
{
|
| 851 |
+
"epoch": 13.52,
|
| 852 |
+
"learning_rate": 9.186813186813187e-05,
|
| 853 |
+
"loss": 0.1564,
|
| 854 |
+
"step": 1230
|
| 855 |
+
},
|
| 856 |
+
{
|
| 857 |
+
"epoch": 13.63,
|
| 858 |
+
"learning_rate": 9.098901098901099e-05,
|
| 859 |
+
"loss": 0.4018,
|
| 860 |
+
"step": 1240
|
| 861 |
+
},
|
| 862 |
+
{
|
| 863 |
+
"epoch": 13.74,
|
| 864 |
+
"learning_rate": 9.010989010989012e-05,
|
| 865 |
+
"loss": 0.3191,
|
| 866 |
+
"step": 1250
|
| 867 |
+
},
|
| 868 |
+
{
|
| 869 |
+
"epoch": 13.85,
|
| 870 |
+
"learning_rate": 8.923076923076924e-05,
|
| 871 |
+
"loss": 0.2845,
|
| 872 |
+
"step": 1260
|
| 873 |
+
},
|
| 874 |
+
{
|
| 875 |
+
"epoch": 13.96,
|
| 876 |
+
"learning_rate": 8.835164835164835e-05,
|
| 877 |
+
"loss": 0.3667,
|
| 878 |
+
"step": 1270
|
| 879 |
+
},
|
| 880 |
+
{
|
| 881 |
+
"epoch": 14.07,
|
| 882 |
+
"learning_rate": 8.747252747252748e-05,
|
| 883 |
+
"loss": 0.2433,
|
| 884 |
+
"step": 1280
|
| 885 |
+
},
|
| 886 |
+
{
|
| 887 |
+
"epoch": 14.18,
|
| 888 |
+
"learning_rate": 8.65934065934066e-05,
|
| 889 |
+
"loss": 0.1794,
|
| 890 |
+
"step": 1290
|
| 891 |
+
},
|
| 892 |
+
{
|
| 893 |
+
"epoch": 14.29,
|
| 894 |
+
"learning_rate": 8.571428571428571e-05,
|
| 895 |
+
"loss": 0.1779,
|
| 896 |
+
"step": 1300
|
| 897 |
+
},
|
| 898 |
+
{
|
| 899 |
+
"epoch": 14.29,
|
| 900 |
+
"eval_accuracy": 0.6294117647058823,
|
| 901 |
+
"eval_loss": 1.1313611268997192,
|
| 902 |
+
"eval_runtime": 1.4881,
|
| 903 |
+
"eval_samples_per_second": 114.237,
|
| 904 |
+
"eval_steps_per_second": 14.784,
|
| 905 |
+
"step": 1300
|
| 906 |
+
},
|
| 907 |
+
{
|
| 908 |
+
"epoch": 14.4,
|
| 909 |
+
"learning_rate": 8.483516483516484e-05,
|
| 910 |
+
"loss": 0.2517,
|
| 911 |
+
"step": 1310
|
| 912 |
+
},
|
| 913 |
+
{
|
| 914 |
+
"epoch": 14.51,
|
| 915 |
+
"learning_rate": 8.395604395604396e-05,
|
| 916 |
+
"loss": 0.2737,
|
| 917 |
+
"step": 1320
|
| 918 |
+
},
|
| 919 |
+
{
|
| 920 |
+
"epoch": 14.62,
|
| 921 |
+
"learning_rate": 8.307692307692309e-05,
|
| 922 |
+
"loss": 0.2092,
|
| 923 |
+
"step": 1330
|
| 924 |
+
},
|
| 925 |
+
{
|
| 926 |
+
"epoch": 14.73,
|
| 927 |
+
"learning_rate": 8.21978021978022e-05,
|
| 928 |
+
"loss": 0.2281,
|
| 929 |
+
"step": 1340
|
| 930 |
+
},
|
| 931 |
+
{
|
| 932 |
+
"epoch": 14.84,
|
| 933 |
+
"learning_rate": 8.131868131868132e-05,
|
| 934 |
+
"loss": 0.2375,
|
| 935 |
+
"step": 1350
|
| 936 |
+
},
|
| 937 |
+
{
|
| 938 |
+
"epoch": 14.95,
|
| 939 |
+
"learning_rate": 8.043956043956045e-05,
|
| 940 |
+
"loss": 0.2062,
|
| 941 |
+
"step": 1360
|
| 942 |
+
},
|
| 943 |
+
{
|
| 944 |
+
"epoch": 15.05,
|
| 945 |
+
"learning_rate": 7.956043956043956e-05,
|
| 946 |
+
"loss": 0.1698,
|
| 947 |
+
"step": 1370
|
| 948 |
+
},
|
| 949 |
+
{
|
| 950 |
+
"epoch": 15.16,
|
| 951 |
+
"learning_rate": 7.868131868131868e-05,
|
| 952 |
+
"loss": 0.1904,
|
| 953 |
+
"step": 1380
|
| 954 |
+
},
|
| 955 |
+
{
|
| 956 |
+
"epoch": 15.27,
|
| 957 |
+
"learning_rate": 7.780219780219781e-05,
|
| 958 |
+
"loss": 0.2469,
|
| 959 |
+
"step": 1390
|
| 960 |
+
},
|
| 961 |
+
{
|
| 962 |
+
"epoch": 15.38,
|
| 963 |
+
"learning_rate": 7.692307692307693e-05,
|
| 964 |
+
"loss": 0.2044,
|
| 965 |
+
"step": 1400
|
| 966 |
+
},
|
| 967 |
+
{
|
| 968 |
+
"epoch": 15.38,
|
| 969 |
+
"eval_accuracy": 0.7294117647058823,
|
| 970 |
+
"eval_loss": 0.8349283337593079,
|
| 971 |
+
"eval_runtime": 1.502,
|
| 972 |
+
"eval_samples_per_second": 113.184,
|
| 973 |
+
"eval_steps_per_second": 14.647,
|
| 974 |
+
"step": 1400
|
| 975 |
+
},
|
| 976 |
+
{
|
| 977 |
+
"epoch": 15.49,
|
| 978 |
+
"learning_rate": 7.604395604395604e-05,
|
| 979 |
+
"loss": 0.2405,
|
| 980 |
+
"step": 1410
|
| 981 |
+
},
|
| 982 |
+
{
|
| 983 |
+
"epoch": 15.6,
|
| 984 |
+
"learning_rate": 7.516483516483517e-05,
|
| 985 |
+
"loss": 0.1326,
|
| 986 |
+
"step": 1420
|
| 987 |
+
},
|
| 988 |
+
{
|
| 989 |
+
"epoch": 15.71,
|
| 990 |
+
"learning_rate": 7.428571428571429e-05,
|
| 991 |
+
"loss": 0.2253,
|
| 992 |
+
"step": 1430
|
| 993 |
+
},
|
| 994 |
+
{
|
| 995 |
+
"epoch": 15.82,
|
| 996 |
+
"learning_rate": 7.34065934065934e-05,
|
| 997 |
+
"loss": 0.2029,
|
| 998 |
+
"step": 1440
|
| 999 |
+
},
|
| 1000 |
+
{
|
| 1001 |
+
"epoch": 15.93,
|
| 1002 |
+
"learning_rate": 7.252747252747253e-05,
|
| 1003 |
+
"loss": 0.1161,
|
| 1004 |
+
"step": 1450
|
| 1005 |
+
},
|
| 1006 |
+
{
|
| 1007 |
+
"epoch": 16.04,
|
| 1008 |
+
"learning_rate": 7.164835164835165e-05,
|
| 1009 |
+
"loss": 0.3051,
|
| 1010 |
+
"step": 1460
|
| 1011 |
+
},
|
| 1012 |
+
{
|
| 1013 |
+
"epoch": 16.15,
|
| 1014 |
+
"learning_rate": 7.076923076923078e-05,
|
| 1015 |
+
"loss": 0.3765,
|
| 1016 |
+
"step": 1470
|
| 1017 |
+
},
|
| 1018 |
+
{
|
| 1019 |
+
"epoch": 16.26,
|
| 1020 |
+
"learning_rate": 6.98901098901099e-05,
|
| 1021 |
+
"loss": 0.1976,
|
| 1022 |
+
"step": 1480
|
| 1023 |
+
},
|
| 1024 |
+
{
|
| 1025 |
+
"epoch": 16.37,
|
| 1026 |
+
"learning_rate": 6.901098901098901e-05,
|
| 1027 |
+
"loss": 0.1887,
|
| 1028 |
+
"step": 1490
|
| 1029 |
+
},
|
| 1030 |
+
{
|
| 1031 |
+
"epoch": 16.48,
|
| 1032 |
+
"learning_rate": 6.813186813186814e-05,
|
| 1033 |
+
"loss": 0.1543,
|
| 1034 |
+
"step": 1500
|
| 1035 |
+
},
|
| 1036 |
+
{
|
| 1037 |
+
"epoch": 16.48,
|
| 1038 |
+
"eval_accuracy": 0.6941176470588235,
|
| 1039 |
+
"eval_loss": 0.8952152132987976,
|
| 1040 |
+
"eval_runtime": 1.5318,
|
| 1041 |
+
"eval_samples_per_second": 110.978,
|
| 1042 |
+
"eval_steps_per_second": 14.362,
|
| 1043 |
+
"step": 1500
|
| 1044 |
+
},
|
| 1045 |
+
{
|
| 1046 |
+
"epoch": 16.59,
|
| 1047 |
+
"learning_rate": 6.725274725274725e-05,
|
| 1048 |
+
"loss": 0.2324,
|
| 1049 |
+
"step": 1510
|
| 1050 |
+
},
|
| 1051 |
+
{
|
| 1052 |
+
"epoch": 16.7,
|
| 1053 |
+
"learning_rate": 6.637362637362637e-05,
|
| 1054 |
+
"loss": 0.1539,
|
| 1055 |
+
"step": 1520
|
| 1056 |
+
},
|
| 1057 |
+
{
|
| 1058 |
+
"epoch": 16.81,
|
| 1059 |
+
"learning_rate": 6.54945054945055e-05,
|
| 1060 |
+
"loss": 0.2305,
|
| 1061 |
+
"step": 1530
|
| 1062 |
+
},
|
| 1063 |
+
{
|
| 1064 |
+
"epoch": 16.92,
|
| 1065 |
+
"learning_rate": 6.461538461538462e-05,
|
| 1066 |
+
"loss": 0.1771,
|
| 1067 |
+
"step": 1540
|
| 1068 |
+
},
|
| 1069 |
+
{
|
| 1070 |
+
"epoch": 17.03,
|
| 1071 |
+
"learning_rate": 6.373626373626373e-05,
|
| 1072 |
+
"loss": 0.1604,
|
| 1073 |
+
"step": 1550
|
| 1074 |
+
},
|
| 1075 |
+
{
|
| 1076 |
+
"epoch": 17.14,
|
| 1077 |
+
"learning_rate": 6.285714285714286e-05,
|
| 1078 |
+
"loss": 0.1297,
|
| 1079 |
+
"step": 1560
|
| 1080 |
+
},
|
| 1081 |
+
{
|
| 1082 |
+
"epoch": 17.25,
|
| 1083 |
+
"learning_rate": 6.197802197802198e-05,
|
| 1084 |
+
"loss": 0.1646,
|
| 1085 |
+
"step": 1570
|
| 1086 |
+
},
|
| 1087 |
+
{
|
| 1088 |
+
"epoch": 17.36,
|
| 1089 |
+
"learning_rate": 6.10989010989011e-05,
|
| 1090 |
+
"loss": 0.1942,
|
| 1091 |
+
"step": 1580
|
| 1092 |
+
},
|
| 1093 |
+
{
|
| 1094 |
+
"epoch": 17.47,
|
| 1095 |
+
"learning_rate": 6.021978021978022e-05,
|
| 1096 |
+
"loss": 0.105,
|
| 1097 |
+
"step": 1590
|
| 1098 |
+
},
|
| 1099 |
+
{
|
| 1100 |
+
"epoch": 17.58,
|
| 1101 |
+
"learning_rate": 5.9340659340659345e-05,
|
| 1102 |
+
"loss": 0.1283,
|
| 1103 |
+
"step": 1600
|
| 1104 |
+
},
|
| 1105 |
+
{
|
| 1106 |
+
"epoch": 17.58,
|
| 1107 |
+
"eval_accuracy": 0.7352941176470589,
|
| 1108 |
+
"eval_loss": 0.8053585290908813,
|
| 1109 |
+
"eval_runtime": 1.4888,
|
| 1110 |
+
"eval_samples_per_second": 114.184,
|
| 1111 |
+
"eval_steps_per_second": 14.777,
|
| 1112 |
+
"step": 1600
|
| 1113 |
+
},
|
| 1114 |
+
{
|
| 1115 |
+
"epoch": 17.69,
|
| 1116 |
+
"learning_rate": 5.846153846153847e-05,
|
| 1117 |
+
"loss": 0.1779,
|
| 1118 |
+
"step": 1610
|
| 1119 |
+
},
|
| 1120 |
+
{
|
| 1121 |
+
"epoch": 17.8,
|
| 1122 |
+
"learning_rate": 5.7582417582417584e-05,
|
| 1123 |
+
"loss": 0.1588,
|
| 1124 |
+
"step": 1620
|
| 1125 |
+
},
|
| 1126 |
+
{
|
| 1127 |
+
"epoch": 17.91,
|
| 1128 |
+
"learning_rate": 5.6703296703296706e-05,
|
| 1129 |
+
"loss": 0.1868,
|
| 1130 |
+
"step": 1630
|
| 1131 |
+
},
|
| 1132 |
+
{
|
| 1133 |
+
"epoch": 18.02,
|
| 1134 |
+
"learning_rate": 5.582417582417583e-05,
|
| 1135 |
+
"loss": 0.1537,
|
| 1136 |
+
"step": 1640
|
| 1137 |
+
},
|
| 1138 |
+
{
|
| 1139 |
+
"epoch": 18.13,
|
| 1140 |
+
"learning_rate": 5.494505494505495e-05,
|
| 1141 |
+
"loss": 0.1426,
|
| 1142 |
+
"step": 1650
|
| 1143 |
+
},
|
| 1144 |
+
{
|
| 1145 |
+
"epoch": 18.24,
|
| 1146 |
+
"learning_rate": 5.406593406593407e-05,
|
| 1147 |
+
"loss": 0.1759,
|
| 1148 |
+
"step": 1660
|
| 1149 |
+
},
|
| 1150 |
+
{
|
| 1151 |
+
"epoch": 18.35,
|
| 1152 |
+
"learning_rate": 5.318681318681319e-05,
|
| 1153 |
+
"loss": 0.1908,
|
| 1154 |
+
"step": 1670
|
| 1155 |
+
},
|
| 1156 |
+
{
|
| 1157 |
+
"epoch": 18.46,
|
| 1158 |
+
"learning_rate": 5.230769230769231e-05,
|
| 1159 |
+
"loss": 0.1562,
|
| 1160 |
+
"step": 1680
|
| 1161 |
+
},
|
| 1162 |
+
{
|
| 1163 |
+
"epoch": 18.57,
|
| 1164 |
+
"learning_rate": 5.142857142857143e-05,
|
| 1165 |
+
"loss": 0.1137,
|
| 1166 |
+
"step": 1690
|
| 1167 |
+
},
|
| 1168 |
+
{
|
| 1169 |
+
"epoch": 18.68,
|
| 1170 |
+
"learning_rate": 5.054945054945055e-05,
|
| 1171 |
+
"loss": 0.1721,
|
| 1172 |
+
"step": 1700
|
| 1173 |
+
},
|
| 1174 |
+
{
|
| 1175 |
+
"epoch": 18.68,
|
| 1176 |
+
"eval_accuracy": 0.7235294117647059,
|
| 1177 |
+
"eval_loss": 0.9094342589378357,
|
| 1178 |
+
"eval_runtime": 1.4871,
|
| 1179 |
+
"eval_samples_per_second": 114.319,
|
| 1180 |
+
"eval_steps_per_second": 14.794,
|
| 1181 |
+
"step": 1700
|
| 1182 |
+
},
|
| 1183 |
+
{
|
| 1184 |
+
"epoch": 18.79,
|
| 1185 |
+
"learning_rate": 4.9670329670329674e-05,
|
| 1186 |
+
"loss": 0.1623,
|
| 1187 |
+
"step": 1710
|
| 1188 |
+
},
|
| 1189 |
+
{
|
| 1190 |
+
"epoch": 18.9,
|
| 1191 |
+
"learning_rate": 4.879120879120879e-05,
|
| 1192 |
+
"loss": 0.1731,
|
| 1193 |
+
"step": 1720
|
| 1194 |
+
},
|
| 1195 |
+
{
|
| 1196 |
+
"epoch": 19.01,
|
| 1197 |
+
"learning_rate": 4.791208791208792e-05,
|
| 1198 |
+
"loss": 0.1458,
|
| 1199 |
+
"step": 1730
|
| 1200 |
+
},
|
| 1201 |
+
{
|
| 1202 |
+
"epoch": 19.12,
|
| 1203 |
+
"learning_rate": 4.7032967032967035e-05,
|
| 1204 |
+
"loss": 0.1717,
|
| 1205 |
+
"step": 1740
|
| 1206 |
+
},
|
| 1207 |
+
{
|
| 1208 |
+
"epoch": 19.23,
|
| 1209 |
+
"learning_rate": 4.615384615384616e-05,
|
| 1210 |
+
"loss": 0.1125,
|
| 1211 |
+
"step": 1750
|
| 1212 |
+
},
|
| 1213 |
+
{
|
| 1214 |
+
"epoch": 19.34,
|
| 1215 |
+
"learning_rate": 4.5274725274725274e-05,
|
| 1216 |
+
"loss": 0.1776,
|
| 1217 |
+
"step": 1760
|
| 1218 |
+
},
|
| 1219 |
+
{
|
| 1220 |
+
"epoch": 19.45,
|
| 1221 |
+
"learning_rate": 4.43956043956044e-05,
|
| 1222 |
+
"loss": 0.1259,
|
| 1223 |
+
"step": 1770
|
| 1224 |
+
},
|
| 1225 |
+
{
|
| 1226 |
+
"epoch": 19.56,
|
| 1227 |
+
"learning_rate": 4.351648351648352e-05,
|
| 1228 |
+
"loss": 0.1294,
|
| 1229 |
+
"step": 1780
|
| 1230 |
+
},
|
| 1231 |
+
{
|
| 1232 |
+
"epoch": 19.67,
|
| 1233 |
+
"learning_rate": 4.2637362637362635e-05,
|
| 1234 |
+
"loss": 0.2037,
|
| 1235 |
+
"step": 1790
|
| 1236 |
+
},
|
| 1237 |
+
{
|
| 1238 |
+
"epoch": 19.78,
|
| 1239 |
+
"learning_rate": 4.1758241758241765e-05,
|
| 1240 |
+
"loss": 0.1509,
|
| 1241 |
+
"step": 1800
|
| 1242 |
+
},
|
| 1243 |
+
{
|
| 1244 |
+
"epoch": 19.78,
|
| 1245 |
+
"eval_accuracy": 0.7411764705882353,
|
| 1246 |
+
"eval_loss": 0.9168078303337097,
|
| 1247 |
+
"eval_runtime": 1.5237,
|
| 1248 |
+
"eval_samples_per_second": 111.573,
|
| 1249 |
+
"eval_steps_per_second": 14.439,
|
| 1250 |
+
"step": 1800
|
| 1251 |
+
},
|
| 1252 |
+
{
|
| 1253 |
+
"epoch": 19.89,
|
| 1254 |
+
"learning_rate": 4.087912087912088e-05,
|
| 1255 |
+
"loss": 0.0973,
|
| 1256 |
+
"step": 1810
|
| 1257 |
+
},
|
| 1258 |
+
{
|
| 1259 |
+
"epoch": 20.0,
|
| 1260 |
+
"learning_rate": 4e-05,
|
| 1261 |
+
"loss": 0.2074,
|
| 1262 |
+
"step": 1820
|
| 1263 |
+
},
|
| 1264 |
+
{
|
| 1265 |
+
"epoch": 20.11,
|
| 1266 |
+
"learning_rate": 3.912087912087912e-05,
|
| 1267 |
+
"loss": 0.185,
|
| 1268 |
+
"step": 1830
|
| 1269 |
+
},
|
| 1270 |
+
{
|
| 1271 |
+
"epoch": 20.22,
|
| 1272 |
+
"learning_rate": 3.824175824175824e-05,
|
| 1273 |
+
"loss": 0.0895,
|
| 1274 |
+
"step": 1840
|
| 1275 |
+
},
|
| 1276 |
+
{
|
| 1277 |
+
"epoch": 20.33,
|
| 1278 |
+
"learning_rate": 3.7362637362637365e-05,
|
| 1279 |
+
"loss": 0.1081,
|
| 1280 |
+
"step": 1850
|
| 1281 |
+
},
|
| 1282 |
+
{
|
| 1283 |
+
"epoch": 20.44,
|
| 1284 |
+
"learning_rate": 3.648351648351648e-05,
|
| 1285 |
+
"loss": 0.169,
|
| 1286 |
+
"step": 1860
|
| 1287 |
+
},
|
| 1288 |
+
{
|
| 1289 |
+
"epoch": 20.55,
|
| 1290 |
+
"learning_rate": 3.560439560439561e-05,
|
| 1291 |
+
"loss": 0.1601,
|
| 1292 |
+
"step": 1870
|
| 1293 |
+
},
|
| 1294 |
+
{
|
| 1295 |
+
"epoch": 20.66,
|
| 1296 |
+
"learning_rate": 3.4725274725274726e-05,
|
| 1297 |
+
"loss": 0.0853,
|
| 1298 |
+
"step": 1880
|
| 1299 |
+
},
|
| 1300 |
+
{
|
| 1301 |
+
"epoch": 20.77,
|
| 1302 |
+
"learning_rate": 3.384615384615385e-05,
|
| 1303 |
+
"loss": 0.1067,
|
| 1304 |
+
"step": 1890
|
| 1305 |
+
},
|
| 1306 |
+
{
|
| 1307 |
+
"epoch": 20.88,
|
| 1308 |
+
"learning_rate": 3.296703296703297e-05,
|
| 1309 |
+
"loss": 0.1257,
|
| 1310 |
+
"step": 1900
|
| 1311 |
+
},
|
| 1312 |
+
{
|
| 1313 |
+
"epoch": 20.88,
|
| 1314 |
+
"eval_accuracy": 0.7411764705882353,
|
| 1315 |
+
"eval_loss": 0.939464807510376,
|
| 1316 |
+
"eval_runtime": 1.4981,
|
| 1317 |
+
"eval_samples_per_second": 113.476,
|
| 1318 |
+
"eval_steps_per_second": 14.685,
|
| 1319 |
+
"step": 1900
|
| 1320 |
+
},
|
| 1321 |
+
{
|
| 1322 |
+
"epoch": 20.99,
|
| 1323 |
+
"learning_rate": 3.2087912087912094e-05,
|
| 1324 |
+
"loss": 0.2041,
|
| 1325 |
+
"step": 1910
|
| 1326 |
+
},
|
| 1327 |
+
{
|
| 1328 |
+
"epoch": 21.1,
|
| 1329 |
+
"learning_rate": 3.120879120879121e-05,
|
| 1330 |
+
"loss": 0.144,
|
| 1331 |
+
"step": 1920
|
| 1332 |
+
},
|
| 1333 |
+
{
|
| 1334 |
+
"epoch": 21.21,
|
| 1335 |
+
"learning_rate": 3.032967032967033e-05,
|
| 1336 |
+
"loss": 0.0677,
|
| 1337 |
+
"step": 1930
|
| 1338 |
+
},
|
| 1339 |
+
{
|
| 1340 |
+
"epoch": 21.32,
|
| 1341 |
+
"learning_rate": 2.945054945054945e-05,
|
| 1342 |
+
"loss": 0.1195,
|
| 1343 |
+
"step": 1940
|
| 1344 |
+
},
|
| 1345 |
+
{
|
| 1346 |
+
"epoch": 21.43,
|
| 1347 |
+
"learning_rate": 2.857142857142857e-05,
|
| 1348 |
+
"loss": 0.1599,
|
| 1349 |
+
"step": 1950
|
| 1350 |
+
},
|
| 1351 |
+
{
|
| 1352 |
+
"epoch": 21.54,
|
| 1353 |
+
"learning_rate": 2.7692307692307694e-05,
|
| 1354 |
+
"loss": 0.0681,
|
| 1355 |
+
"step": 1960
|
| 1356 |
+
},
|
| 1357 |
+
{
|
| 1358 |
+
"epoch": 21.65,
|
| 1359 |
+
"learning_rate": 2.6813186813186813e-05,
|
| 1360 |
+
"loss": 0.0854,
|
| 1361 |
+
"step": 1970
|
| 1362 |
+
},
|
| 1363 |
+
{
|
| 1364 |
+
"epoch": 21.76,
|
| 1365 |
+
"learning_rate": 2.593406593406594e-05,
|
| 1366 |
+
"loss": 0.1481,
|
| 1367 |
+
"step": 1980
|
| 1368 |
+
},
|
| 1369 |
+
{
|
| 1370 |
+
"epoch": 21.87,
|
| 1371 |
+
"learning_rate": 2.5054945054945055e-05,
|
| 1372 |
+
"loss": 0.0927,
|
| 1373 |
+
"step": 1990
|
| 1374 |
+
},
|
| 1375 |
+
{
|
| 1376 |
+
"epoch": 21.98,
|
| 1377 |
+
"learning_rate": 2.4175824175824177e-05,
|
| 1378 |
+
"loss": 0.1747,
|
| 1379 |
+
"step": 2000
|
| 1380 |
+
},
|
| 1381 |
+
{
|
| 1382 |
+
"epoch": 21.98,
|
| 1383 |
+
"eval_accuracy": 0.7470588235294118,
|
| 1384 |
+
"eval_loss": 0.8745805025100708,
|
| 1385 |
+
"eval_runtime": 1.5068,
|
| 1386 |
+
"eval_samples_per_second": 112.823,
|
| 1387 |
+
"eval_steps_per_second": 14.601,
|
| 1388 |
+
"step": 2000
|
| 1389 |
+
},
|
| 1390 |
+
{
|
| 1391 |
+
"epoch": 22.09,
|
| 1392 |
+
"learning_rate": 2.3296703296703297e-05,
|
| 1393 |
+
"loss": 0.0698,
|
| 1394 |
+
"step": 2010
|
| 1395 |
+
},
|
| 1396 |
+
{
|
| 1397 |
+
"epoch": 22.2,
|
| 1398 |
+
"learning_rate": 2.241758241758242e-05,
|
| 1399 |
+
"loss": 0.0488,
|
| 1400 |
+
"step": 2020
|
| 1401 |
+
},
|
| 1402 |
+
{
|
| 1403 |
+
"epoch": 22.31,
|
| 1404 |
+
"learning_rate": 2.1538461538461542e-05,
|
| 1405 |
+
"loss": 0.0583,
|
| 1406 |
+
"step": 2030
|
| 1407 |
+
},
|
| 1408 |
+
{
|
| 1409 |
+
"epoch": 22.42,
|
| 1410 |
+
"learning_rate": 2.0659340659340658e-05,
|
| 1411 |
+
"loss": 0.1999,
|
| 1412 |
+
"step": 2040
|
| 1413 |
+
},
|
| 1414 |
+
{
|
| 1415 |
+
"epoch": 22.53,
|
| 1416 |
+
"learning_rate": 1.978021978021978e-05,
|
| 1417 |
+
"loss": 0.0723,
|
| 1418 |
+
"step": 2050
|
| 1419 |
+
},
|
| 1420 |
+
{
|
| 1421 |
+
"epoch": 22.64,
|
| 1422 |
+
"learning_rate": 1.89010989010989e-05,
|
| 1423 |
+
"loss": 0.1596,
|
| 1424 |
+
"step": 2060
|
| 1425 |
+
},
|
| 1426 |
+
{
|
| 1427 |
+
"epoch": 22.75,
|
| 1428 |
+
"learning_rate": 1.8021978021978023e-05,
|
| 1429 |
+
"loss": 0.1527,
|
| 1430 |
+
"step": 2070
|
| 1431 |
+
},
|
| 1432 |
+
{
|
| 1433 |
+
"epoch": 22.86,
|
| 1434 |
+
"learning_rate": 1.7142857142857145e-05,
|
| 1435 |
+
"loss": 0.1554,
|
| 1436 |
+
"step": 2080
|
| 1437 |
+
},
|
| 1438 |
+
{
|
| 1439 |
+
"epoch": 22.97,
|
| 1440 |
+
"learning_rate": 1.6263736263736265e-05,
|
| 1441 |
+
"loss": 0.1464,
|
| 1442 |
+
"step": 2090
|
| 1443 |
+
},
|
| 1444 |
+
{
|
| 1445 |
+
"epoch": 23.08,
|
| 1446 |
+
"learning_rate": 1.5384615384615387e-05,
|
| 1447 |
+
"loss": 0.1506,
|
| 1448 |
+
"step": 2100
|
| 1449 |
+
},
|
| 1450 |
+
{
|
| 1451 |
+
"epoch": 23.08,
|
| 1452 |
+
"eval_accuracy": 0.7352941176470589,
|
| 1453 |
+
"eval_loss": 0.7992474436759949,
|
| 1454 |
+
"eval_runtime": 1.5262,
|
| 1455 |
+
"eval_samples_per_second": 111.385,
|
| 1456 |
+
"eval_steps_per_second": 14.415,
|
| 1457 |
+
"step": 2100
|
| 1458 |
+
},
|
| 1459 |
+
{
|
| 1460 |
+
"epoch": 23.19,
|
| 1461 |
+
"learning_rate": 1.4505494505494508e-05,
|
| 1462 |
+
"loss": 0.1008,
|
| 1463 |
+
"step": 2110
|
| 1464 |
+
},
|
| 1465 |
+
{
|
| 1466 |
+
"epoch": 23.3,
|
| 1467 |
+
"learning_rate": 1.3626373626373626e-05,
|
| 1468 |
+
"loss": 0.0949,
|
| 1469 |
+
"step": 2120
|
| 1470 |
+
},
|
| 1471 |
+
{
|
| 1472 |
+
"epoch": 23.41,
|
| 1473 |
+
"learning_rate": 1.2747252747252747e-05,
|
| 1474 |
+
"loss": 0.09,
|
| 1475 |
+
"step": 2130
|
| 1476 |
+
},
|
| 1477 |
+
{
|
| 1478 |
+
"epoch": 23.52,
|
| 1479 |
+
"learning_rate": 1.1868131868131868e-05,
|
| 1480 |
+
"loss": 0.1631,
|
| 1481 |
+
"step": 2140
|
| 1482 |
+
},
|
| 1483 |
+
{
|
| 1484 |
+
"epoch": 23.63,
|
| 1485 |
+
"learning_rate": 1.0989010989010989e-05,
|
| 1486 |
+
"loss": 0.0929,
|
| 1487 |
+
"step": 2150
|
| 1488 |
+
},
|
| 1489 |
+
{
|
| 1490 |
+
"epoch": 23.74,
|
| 1491 |
+
"learning_rate": 1.0109890109890111e-05,
|
| 1492 |
+
"loss": 0.1425,
|
| 1493 |
+
"step": 2160
|
| 1494 |
+
},
|
| 1495 |
+
{
|
| 1496 |
+
"epoch": 23.85,
|
| 1497 |
+
"learning_rate": 9.230769230769232e-06,
|
| 1498 |
+
"loss": 0.1187,
|
| 1499 |
+
"step": 2170
|
| 1500 |
+
},
|
| 1501 |
+
{
|
| 1502 |
+
"epoch": 23.96,
|
| 1503 |
+
"learning_rate": 8.351648351648352e-06,
|
| 1504 |
+
"loss": 0.1834,
|
| 1505 |
+
"step": 2180
|
| 1506 |
+
},
|
| 1507 |
+
{
|
| 1508 |
+
"epoch": 24.07,
|
| 1509 |
+
"learning_rate": 7.4725274725274726e-06,
|
| 1510 |
+
"loss": 0.1157,
|
| 1511 |
+
"step": 2190
|
| 1512 |
+
},
|
| 1513 |
+
{
|
| 1514 |
+
"epoch": 24.18,
|
| 1515 |
+
"learning_rate": 6.5934065934065935e-06,
|
| 1516 |
+
"loss": 0.1021,
|
| 1517 |
+
"step": 2200
|
| 1518 |
+
},
|
| 1519 |
+
{
|
| 1520 |
+
"epoch": 24.18,
|
| 1521 |
+
"eval_accuracy": 0.7705882352941177,
|
| 1522 |
+
"eval_loss": 0.7445889711380005,
|
| 1523 |
+
"eval_runtime": 1.5168,
|
| 1524 |
+
"eval_samples_per_second": 112.082,
|
| 1525 |
+
"eval_steps_per_second": 14.505,
|
| 1526 |
+
"step": 2200
|
| 1527 |
+
},
|
| 1528 |
+
{
|
| 1529 |
+
"epoch": 24.29,
|
| 1530 |
+
"learning_rate": 5.7142857142857145e-06,
|
| 1531 |
+
"loss": 0.1826,
|
| 1532 |
+
"step": 2210
|
| 1533 |
+
},
|
| 1534 |
+
{
|
| 1535 |
+
"epoch": 24.4,
|
| 1536 |
+
"learning_rate": 4.8351648351648355e-06,
|
| 1537 |
+
"loss": 0.1807,
|
| 1538 |
+
"step": 2220
|
| 1539 |
+
},
|
| 1540 |
+
{
|
| 1541 |
+
"epoch": 24.51,
|
| 1542 |
+
"learning_rate": 3.9560439560439565e-06,
|
| 1543 |
+
"loss": 0.0946,
|
| 1544 |
+
"step": 2230
|
| 1545 |
+
},
|
| 1546 |
+
{
|
| 1547 |
+
"epoch": 24.62,
|
| 1548 |
+
"learning_rate": 3.0769230769230774e-06,
|
| 1549 |
+
"loss": 0.1039,
|
| 1550 |
+
"step": 2240
|
| 1551 |
+
},
|
| 1552 |
+
{
|
| 1553 |
+
"epoch": 24.73,
|
| 1554 |
+
"learning_rate": 2.197802197802198e-06,
|
| 1555 |
+
"loss": 0.1729,
|
| 1556 |
+
"step": 2250
|
| 1557 |
+
},
|
| 1558 |
+
{
|
| 1559 |
+
"epoch": 24.84,
|
| 1560 |
+
"learning_rate": 1.3186813186813187e-06,
|
| 1561 |
+
"loss": 0.1052,
|
| 1562 |
+
"step": 2260
|
| 1563 |
+
},
|
| 1564 |
+
{
|
| 1565 |
+
"epoch": 24.95,
|
| 1566 |
+
"learning_rate": 4.3956043956043957e-07,
|
| 1567 |
+
"loss": 0.1003,
|
| 1568 |
+
"step": 2270
|
| 1569 |
+
},
|
| 1570 |
+
{
|
| 1571 |
+
"epoch": 25.0,
|
| 1572 |
+
"step": 2275,
|
| 1573 |
+
"total_flos": 2.8110723118700544e+18,
|
| 1574 |
+
"train_loss": 0.38098634693648786,
|
| 1575 |
+
"train_runtime": 883.423,
|
| 1576 |
+
"train_samples_per_second": 41.062,
|
| 1577 |
+
"train_steps_per_second": 2.575
|
| 1578 |
}
|
| 1579 |
],
|
| 1580 |
+
"max_steps": 2275,
|
| 1581 |
+
"num_train_epochs": 25,
|
| 1582 |
+
"total_flos": 2.8110723118700544e+18,
|
| 1583 |
"trial_name": null,
|
| 1584 |
"trial_params": null
|
| 1585 |
}
|