Training in progress, step 1000
Browse files- all_results.json +11 -11
- eval_results.json +6 -6
- model.safetensors +1 -1
- runs/Dec28_03-38-28_08b39d765768/events.out.tfevents.1703735726.08b39d765768.16933.1 +3 -0
- runs/Dec28_04-03-00_08b39d765768/events.out.tfevents.1703736181.08b39d765768.16933.2 +3 -0
- runs/Dec28_04-03-25_08b39d765768/events.out.tfevents.1703736208.08b39d765768.16933.3 +3 -0
- train_results.json +6 -6
- trainer_state.json +341 -335
- training_args.bin +1 -1
all_results.json
CHANGED
|
@@ -1,13 +1,13 @@
|
|
| 1 |
{
|
| 2 |
-
"epoch":
|
| 3 |
-
"eval_accuracy": 0.
|
| 4 |
-
"eval_loss": 0.
|
| 5 |
-
"eval_runtime":
|
| 6 |
-
"eval_samples_per_second":
|
| 7 |
-
"eval_steps_per_second":
|
| 8 |
-
"total_flos": 3.
|
| 9 |
-
"train_loss": 0.
|
| 10 |
-
"train_runtime":
|
| 11 |
-
"train_samples_per_second":
|
| 12 |
-
"train_steps_per_second":
|
| 13 |
}
|
|
|
|
| 1 |
{
|
| 2 |
+
"epoch": 5.0,
|
| 3 |
+
"eval_accuracy": 0.7858333333333334,
|
| 4 |
+
"eval_loss": 0.9599547386169434,
|
| 5 |
+
"eval_runtime": 17.8384,
|
| 6 |
+
"eval_samples_per_second": 67.27,
|
| 7 |
+
"eval_steps_per_second": 8.409,
|
| 8 |
+
"total_flos": 3.4029172406502605e+18,
|
| 9 |
+
"train_loss": 0.2756463099698551,
|
| 10 |
+
"train_runtime": 974.1891,
|
| 11 |
+
"train_samples_per_second": 45.073,
|
| 12 |
+
"train_steps_per_second": 0.939
|
| 13 |
}
|
eval_results.json
CHANGED
|
@@ -1,8 +1,8 @@
|
|
| 1 |
{
|
| 2 |
-
"epoch":
|
| 3 |
-
"eval_accuracy": 0.
|
| 4 |
-
"eval_loss": 0.
|
| 5 |
-
"eval_runtime":
|
| 6 |
-
"eval_samples_per_second":
|
| 7 |
-
"eval_steps_per_second":
|
| 8 |
}
|
|
|
|
| 1 |
{
|
| 2 |
+
"epoch": 5.0,
|
| 3 |
+
"eval_accuracy": 0.7858333333333334,
|
| 4 |
+
"eval_loss": 0.9599547386169434,
|
| 5 |
+
"eval_runtime": 17.8384,
|
| 6 |
+
"eval_samples_per_second": 67.27,
|
| 7 |
+
"eval_steps_per_second": 8.409
|
| 8 |
}
|
model.safetensors
CHANGED
|
@@ -1,3 +1,3 @@
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
-
oid sha256:
|
| 3 |
size 343248584
|
|
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:942f16f7548086ac80764f2ae86424766836d340199c377cf5807b6983160884
|
| 3 |
size 343248584
|
runs/Dec28_03-38-28_08b39d765768/events.out.tfevents.1703735726.08b39d765768.16933.1
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:c8fe4cc7e324dd14729f8952fa5e2bf6fe4d64566f18a4e4b212bda1aad155b6
|
| 3 |
+
size 411
|
runs/Dec28_04-03-00_08b39d765768/events.out.tfevents.1703736181.08b39d765768.16933.2
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:3e545de89c834b9df62c1a2666bc35a0feb7d4ed268e58b96a687845b31f83bf
|
| 3 |
+
size 5016
|
runs/Dec28_04-03-25_08b39d765768/events.out.tfevents.1703736208.08b39d765768.16933.3
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:5292a4dab296d28f1012bbbc420551b9857670c9ee76016cb111eb9d35e8f5fd
|
| 3 |
+
size 20864
|
train_results.json
CHANGED
|
@@ -1,8 +1,8 @@
|
|
| 1 |
{
|
| 2 |
-
"epoch":
|
| 3 |
-
"total_flos": 3.
|
| 4 |
-
"train_loss": 0.
|
| 5 |
-
"train_runtime":
|
| 6 |
-
"train_samples_per_second":
|
| 7 |
-
"train_steps_per_second":
|
| 8 |
}
|
|
|
|
| 1 |
{
|
| 2 |
+
"epoch": 5.0,
|
| 3 |
+
"total_flos": 3.4029172406502605e+18,
|
| 4 |
+
"train_loss": 0.2756463099698551,
|
| 5 |
+
"train_runtime": 974.1891,
|
| 6 |
+
"train_samples_per_second": 45.073,
|
| 7 |
+
"train_steps_per_second": 0.939
|
| 8 |
}
|
trainer_state.json
CHANGED
|
@@ -1,649 +1,655 @@
|
|
| 1 |
{
|
| 2 |
"best_metric": null,
|
| 3 |
"best_model_checkpoint": null,
|
| 4 |
-
"epoch":
|
| 5 |
"eval_steps": 100,
|
| 6 |
-
"global_step":
|
| 7 |
"is_hyper_param_search": false,
|
| 8 |
"is_local_process_zero": true,
|
| 9 |
"is_world_process_zero": true,
|
| 10 |
"log_history": [
|
| 11 |
{
|
| 12 |
-
"epoch": 0.
|
| 13 |
-
"learning_rate": 0.
|
| 14 |
-
"loss":
|
| 15 |
"step": 10
|
| 16 |
},
|
| 17 |
{
|
| 18 |
-
"epoch": 0.
|
| 19 |
-
"learning_rate": 0.
|
| 20 |
-
"loss": 1.
|
| 21 |
"step": 20
|
| 22 |
},
|
| 23 |
{
|
| 24 |
-
"epoch": 0.
|
| 25 |
-
"learning_rate": 0.
|
| 26 |
-
"loss": 1.
|
| 27 |
"step": 30
|
| 28 |
},
|
| 29 |
{
|
| 30 |
-
"epoch": 0.
|
| 31 |
-
"learning_rate": 0.
|
| 32 |
-
"loss": 1.
|
| 33 |
"step": 40
|
| 34 |
},
|
| 35 |
{
|
| 36 |
-
"epoch": 0.
|
| 37 |
-
"learning_rate": 0.
|
| 38 |
-
"loss": 1.
|
| 39 |
"step": 50
|
| 40 |
},
|
| 41 |
{
|
| 42 |
-
"epoch": 0.
|
| 43 |
-
"learning_rate": 0.
|
| 44 |
-
"loss":
|
| 45 |
"step": 60
|
| 46 |
},
|
| 47 |
{
|
| 48 |
-
"epoch": 0.
|
| 49 |
-
"learning_rate": 0.
|
| 50 |
-
"loss":
|
| 51 |
"step": 70
|
| 52 |
},
|
| 53 |
{
|
| 54 |
-
"epoch": 0.
|
| 55 |
-
"learning_rate": 0.
|
| 56 |
-
"loss": 0.
|
| 57 |
"step": 80
|
| 58 |
},
|
| 59 |
{
|
| 60 |
-
"epoch":
|
| 61 |
-
"learning_rate": 0.
|
| 62 |
-
"loss": 0.
|
| 63 |
"step": 90
|
| 64 |
},
|
| 65 |
{
|
| 66 |
-
"epoch":
|
| 67 |
-
"learning_rate": 0.
|
| 68 |
-
"loss": 0.
|
| 69 |
"step": 100
|
| 70 |
},
|
| 71 |
{
|
| 72 |
-
"epoch":
|
| 73 |
-
"eval_accuracy": 0.
|
| 74 |
-
"eval_loss": 0.
|
| 75 |
-
"eval_runtime":
|
| 76 |
-
"eval_samples_per_second":
|
| 77 |
-
"eval_steps_per_second":
|
| 78 |
"step": 100
|
| 79 |
},
|
| 80 |
{
|
| 81 |
-
"epoch":
|
| 82 |
-
"learning_rate": 0.
|
| 83 |
-
"loss": 0.
|
| 84 |
"step": 110
|
| 85 |
},
|
| 86 |
{
|
| 87 |
-
"epoch":
|
| 88 |
-
"learning_rate": 0.
|
| 89 |
-
"loss": 0.
|
| 90 |
"step": 120
|
| 91 |
},
|
| 92 |
{
|
| 93 |
-
"epoch":
|
| 94 |
-
"learning_rate": 0.
|
| 95 |
-
"loss": 0.
|
| 96 |
"step": 130
|
| 97 |
},
|
| 98 |
{
|
| 99 |
-
"epoch":
|
| 100 |
-
"learning_rate": 0.
|
| 101 |
-
"loss": 0.
|
| 102 |
"step": 140
|
| 103 |
},
|
| 104 |
{
|
| 105 |
-
"epoch":
|
| 106 |
-
"learning_rate": 0.
|
| 107 |
-
"loss": 0.
|
| 108 |
"step": 150
|
| 109 |
},
|
| 110 |
{
|
| 111 |
-
"epoch":
|
| 112 |
-
"learning_rate": 0.
|
| 113 |
-
"loss": 0.
|
| 114 |
"step": 160
|
| 115 |
},
|
| 116 |
{
|
| 117 |
-
"epoch":
|
| 118 |
-
"learning_rate": 0.
|
| 119 |
-
"loss": 0.
|
| 120 |
"step": 170
|
| 121 |
},
|
| 122 |
{
|
| 123 |
-
"epoch":
|
| 124 |
-
"learning_rate": 0.
|
| 125 |
-
"loss": 0.
|
| 126 |
"step": 180
|
| 127 |
},
|
| 128 |
{
|
| 129 |
-
"epoch":
|
| 130 |
-
"learning_rate": 0.
|
| 131 |
-
"loss": 0.
|
| 132 |
"step": 190
|
| 133 |
},
|
| 134 |
{
|
| 135 |
-
"epoch":
|
| 136 |
-
"learning_rate": 0.
|
| 137 |
-
"loss": 0.
|
| 138 |
"step": 200
|
| 139 |
},
|
| 140 |
{
|
| 141 |
-
"epoch":
|
| 142 |
-
"eval_accuracy": 0.
|
| 143 |
-
"eval_loss": 0.
|
| 144 |
-
"eval_runtime":
|
| 145 |
-
"eval_samples_per_second":
|
| 146 |
-
"eval_steps_per_second":
|
| 147 |
"step": 200
|
| 148 |
},
|
| 149 |
{
|
| 150 |
-
"epoch":
|
| 151 |
-
"learning_rate": 0.
|
| 152 |
-
"loss": 0.
|
| 153 |
"step": 210
|
| 154 |
},
|
| 155 |
{
|
| 156 |
-
"epoch": 2
|
| 157 |
-
"learning_rate": 0.
|
| 158 |
-
"loss": 0.
|
| 159 |
"step": 220
|
| 160 |
},
|
| 161 |
{
|
| 162 |
-
"epoch":
|
| 163 |
-
"learning_rate": 0.
|
| 164 |
-
"loss": 0.
|
| 165 |
"step": 230
|
| 166 |
},
|
| 167 |
{
|
| 168 |
-
"epoch":
|
| 169 |
-
"learning_rate": 0.
|
| 170 |
-
"loss": 0.
|
| 171 |
"step": 240
|
| 172 |
},
|
| 173 |
{
|
| 174 |
-
"epoch":
|
| 175 |
-
"learning_rate": 0.
|
| 176 |
-
"loss": 0.
|
| 177 |
"step": 250
|
| 178 |
},
|
| 179 |
{
|
| 180 |
-
"epoch":
|
| 181 |
-
"learning_rate": 0.
|
| 182 |
-
"loss": 0.
|
| 183 |
"step": 260
|
| 184 |
},
|
| 185 |
{
|
| 186 |
-
"epoch":
|
| 187 |
-
"learning_rate": 0.
|
| 188 |
-
"loss": 0.
|
| 189 |
"step": 270
|
| 190 |
},
|
| 191 |
{
|
| 192 |
-
"epoch":
|
| 193 |
-
"learning_rate": 0.
|
| 194 |
-
"loss": 0.
|
| 195 |
"step": 280
|
| 196 |
},
|
| 197 |
{
|
| 198 |
-
"epoch":
|
| 199 |
-
"learning_rate": 0.
|
| 200 |
-
"loss": 0.
|
| 201 |
"step": 290
|
| 202 |
},
|
| 203 |
{
|
| 204 |
-
"epoch":
|
| 205 |
-
"learning_rate": 0.
|
| 206 |
-
"loss": 0.
|
| 207 |
"step": 300
|
| 208 |
},
|
| 209 |
{
|
| 210 |
-
"epoch":
|
| 211 |
-
"eval_accuracy": 0.
|
| 212 |
-
"eval_loss": 0.
|
| 213 |
-
"eval_runtime":
|
| 214 |
-
"eval_samples_per_second":
|
| 215 |
-
"eval_steps_per_second":
|
| 216 |
"step": 300
|
| 217 |
},
|
| 218 |
{
|
| 219 |
-
"epoch":
|
| 220 |
-
"learning_rate": 0.
|
| 221 |
-
"loss": 0.
|
| 222 |
"step": 310
|
| 223 |
},
|
| 224 |
{
|
| 225 |
-
"epoch":
|
| 226 |
-
"learning_rate": 0.
|
| 227 |
-
"loss": 0.
|
| 228 |
"step": 320
|
| 229 |
},
|
| 230 |
{
|
| 231 |
-
"epoch":
|
| 232 |
-
"learning_rate": 0.
|
| 233 |
-
"loss": 0.
|
| 234 |
"step": 330
|
| 235 |
},
|
| 236 |
{
|
| 237 |
-
"epoch":
|
| 238 |
-
"learning_rate": 0.
|
| 239 |
-
"loss": 0.
|
| 240 |
"step": 340
|
| 241 |
},
|
| 242 |
{
|
| 243 |
-
"epoch":
|
| 244 |
-
"learning_rate": 0.
|
| 245 |
-
"loss": 0.
|
| 246 |
"step": 350
|
| 247 |
},
|
| 248 |
{
|
| 249 |
-
"epoch":
|
| 250 |
-
"learning_rate": 0.
|
| 251 |
-
"loss": 0.
|
| 252 |
"step": 360
|
| 253 |
},
|
| 254 |
{
|
| 255 |
-
"epoch":
|
| 256 |
-
"learning_rate": 0.
|
| 257 |
-
"loss": 0.
|
| 258 |
"step": 370
|
| 259 |
},
|
| 260 |
{
|
| 261 |
-
"epoch":
|
| 262 |
-
"learning_rate": 0.
|
| 263 |
-
"loss": 0.
|
| 264 |
"step": 380
|
| 265 |
},
|
| 266 |
{
|
| 267 |
-
"epoch":
|
| 268 |
-
"learning_rate": 0.
|
| 269 |
-
"loss": 0.
|
| 270 |
"step": 390
|
| 271 |
},
|
| 272 |
{
|
| 273 |
-
"epoch":
|
| 274 |
-
"learning_rate": 0.
|
| 275 |
-
"loss": 0.
|
| 276 |
"step": 400
|
| 277 |
},
|
| 278 |
{
|
| 279 |
-
"epoch":
|
| 280 |
-
"eval_accuracy": 0.
|
| 281 |
-
"eval_loss": 0.
|
| 282 |
-
"eval_runtime":
|
| 283 |
-
"eval_samples_per_second":
|
| 284 |
-
"eval_steps_per_second":
|
| 285 |
"step": 400
|
| 286 |
},
|
| 287 |
{
|
| 288 |
-
"epoch":
|
| 289 |
-
"learning_rate": 0.
|
| 290 |
-
"loss": 0.
|
| 291 |
"step": 410
|
| 292 |
},
|
| 293 |
{
|
| 294 |
-
"epoch":
|
| 295 |
-
"learning_rate": 0.
|
| 296 |
-
"loss": 0.
|
| 297 |
"step": 420
|
| 298 |
},
|
| 299 |
{
|
| 300 |
-
"epoch":
|
| 301 |
-
"learning_rate": 0.
|
| 302 |
-
"loss": 0.
|
| 303 |
"step": 430
|
| 304 |
},
|
| 305 |
{
|
| 306 |
-
"epoch": 4
|
| 307 |
-
"learning_rate": 0.
|
| 308 |
-
"loss": 0.
|
| 309 |
"step": 440
|
| 310 |
},
|
| 311 |
{
|
| 312 |
-
"epoch":
|
| 313 |
-
"learning_rate": 0.
|
| 314 |
-
"loss": 0.
|
| 315 |
"step": 450
|
| 316 |
},
|
| 317 |
{
|
| 318 |
-
"epoch":
|
| 319 |
-
"learning_rate": 9.
|
| 320 |
-
"loss": 0.
|
| 321 |
"step": 460
|
| 322 |
},
|
| 323 |
{
|
| 324 |
-
"epoch":
|
| 325 |
-
"learning_rate": 9.
|
| 326 |
-
"loss": 0.
|
| 327 |
"step": 470
|
| 328 |
},
|
| 329 |
{
|
| 330 |
-
"epoch":
|
| 331 |
-
"learning_rate": 9.
|
| 332 |
-
"loss": 0.
|
| 333 |
"step": 480
|
| 334 |
},
|
| 335 |
{
|
| 336 |
-
"epoch":
|
| 337 |
-
"learning_rate": 9.
|
| 338 |
-
"loss": 0.
|
| 339 |
"step": 490
|
| 340 |
},
|
| 341 |
{
|
| 342 |
-
"epoch":
|
| 343 |
-
"learning_rate":
|
| 344 |
-
"loss": 0.
|
| 345 |
"step": 500
|
| 346 |
},
|
| 347 |
{
|
| 348 |
-
"epoch":
|
| 349 |
-
"eval_accuracy": 0.
|
| 350 |
-
"eval_loss": 0.
|
| 351 |
-
"eval_runtime":
|
| 352 |
-
"eval_samples_per_second":
|
| 353 |
-
"eval_steps_per_second":
|
| 354 |
"step": 500
|
| 355 |
},
|
| 356 |
{
|
| 357 |
-
"epoch":
|
| 358 |
-
"learning_rate": 8.
|
| 359 |
-
"loss": 0.
|
| 360 |
"step": 510
|
| 361 |
},
|
| 362 |
{
|
| 363 |
-
"epoch":
|
| 364 |
-
"learning_rate": 8.
|
| 365 |
-
"loss": 0.
|
| 366 |
"step": 520
|
| 367 |
},
|
| 368 |
{
|
| 369 |
-
"epoch":
|
| 370 |
-
"learning_rate": 8.
|
| 371 |
-
"loss": 0.
|
| 372 |
"step": 530
|
| 373 |
},
|
| 374 |
{
|
| 375 |
-
"epoch":
|
| 376 |
-
"learning_rate":
|
| 377 |
-
"loss": 0.
|
| 378 |
"step": 540
|
| 379 |
},
|
| 380 |
{
|
| 381 |
-
"epoch":
|
| 382 |
-
"learning_rate": 7.
|
| 383 |
-
"loss": 0.
|
| 384 |
"step": 550
|
| 385 |
},
|
| 386 |
{
|
| 387 |
-
"epoch":
|
| 388 |
-
"learning_rate": 7.
|
| 389 |
-
"loss": 0.
|
| 390 |
"step": 560
|
| 391 |
},
|
| 392 |
{
|
| 393 |
-
"epoch":
|
| 394 |
-
"learning_rate": 7.
|
| 395 |
-
"loss": 0.
|
| 396 |
"step": 570
|
| 397 |
},
|
| 398 |
{
|
| 399 |
-
"epoch":
|
| 400 |
-
"learning_rate": 7.
|
| 401 |
-
"loss": 0.
|
| 402 |
"step": 580
|
| 403 |
},
|
| 404 |
{
|
| 405 |
-
"epoch":
|
| 406 |
-
"learning_rate":
|
| 407 |
-
"loss": 0.
|
| 408 |
"step": 590
|
| 409 |
},
|
| 410 |
{
|
| 411 |
-
"epoch":
|
| 412 |
-
"learning_rate": 6.
|
| 413 |
-
"loss": 0.
|
| 414 |
"step": 600
|
| 415 |
},
|
| 416 |
{
|
| 417 |
-
"epoch":
|
| 418 |
-
"eval_accuracy": 0.
|
| 419 |
-
"eval_loss":
|
| 420 |
-
"eval_runtime":
|
| 421 |
-
"eval_samples_per_second":
|
| 422 |
-
"eval_steps_per_second":
|
| 423 |
"step": 600
|
| 424 |
},
|
| 425 |
{
|
| 426 |
-
"epoch":
|
| 427 |
-
"learning_rate": 6.
|
| 428 |
-
"loss": 0.
|
| 429 |
"step": 610
|
| 430 |
},
|
| 431 |
{
|
| 432 |
-
"epoch":
|
| 433 |
-
"learning_rate": 6.
|
| 434 |
-
"loss": 0.
|
| 435 |
"step": 620
|
| 436 |
},
|
| 437 |
{
|
| 438 |
-
"epoch":
|
| 439 |
-
"learning_rate":
|
| 440 |
-
"loss": 0.
|
| 441 |
"step": 630
|
| 442 |
},
|
| 443 |
{
|
| 444 |
-
"epoch":
|
| 445 |
-
"learning_rate":
|
| 446 |
-
"loss": 0.
|
| 447 |
"step": 640
|
| 448 |
},
|
| 449 |
{
|
| 450 |
-
"epoch":
|
| 451 |
-
"learning_rate": 5.
|
| 452 |
-
"loss": 0.
|
| 453 |
"step": 650
|
| 454 |
},
|
| 455 |
{
|
| 456 |
-
"epoch":
|
| 457 |
-
"learning_rate": 5.
|
| 458 |
-
"loss": 0.
|
| 459 |
"step": 660
|
| 460 |
},
|
| 461 |
{
|
| 462 |
-
"epoch":
|
| 463 |
-
"learning_rate": 5.
|
| 464 |
-
"loss": 0.
|
| 465 |
"step": 670
|
| 466 |
},
|
| 467 |
{
|
| 468 |
-
"epoch":
|
| 469 |
-
"learning_rate":
|
| 470 |
-
"loss": 0.
|
| 471 |
"step": 680
|
| 472 |
},
|
| 473 |
{
|
| 474 |
-
"epoch":
|
| 475 |
-
"learning_rate": 4.
|
| 476 |
-
"loss": 0.
|
| 477 |
"step": 690
|
| 478 |
},
|
| 479 |
{
|
| 480 |
-
"epoch":
|
| 481 |
-
"learning_rate": 4.
|
| 482 |
-
"loss": 0.
|
| 483 |
"step": 700
|
| 484 |
},
|
| 485 |
{
|
| 486 |
-
"epoch":
|
| 487 |
-
"eval_accuracy": 0.
|
| 488 |
-
"eval_loss":
|
| 489 |
-
"eval_runtime":
|
| 490 |
-
"eval_samples_per_second":
|
| 491 |
-
"eval_steps_per_second":
|
| 492 |
"step": 700
|
| 493 |
},
|
| 494 |
{
|
| 495 |
-
"epoch":
|
| 496 |
-
"learning_rate": 4.
|
| 497 |
-
"loss": 0.
|
| 498 |
"step": 710
|
| 499 |
},
|
| 500 |
{
|
| 501 |
-
"epoch":
|
| 502 |
-
"learning_rate":
|
| 503 |
-
"loss": 0.
|
| 504 |
"step": 720
|
| 505 |
},
|
| 506 |
{
|
| 507 |
-
"epoch":
|
| 508 |
-
"learning_rate":
|
| 509 |
-
"loss": 0.
|
| 510 |
"step": 730
|
| 511 |
},
|
| 512 |
{
|
| 513 |
-
"epoch":
|
| 514 |
-
"learning_rate": 3.
|
| 515 |
-
"loss": 0.
|
| 516 |
"step": 740
|
| 517 |
},
|
| 518 |
{
|
| 519 |
-
"epoch":
|
| 520 |
-
"learning_rate": 3.
|
| 521 |
-
"loss": 0.
|
| 522 |
"step": 750
|
| 523 |
},
|
| 524 |
{
|
| 525 |
-
"epoch":
|
| 526 |
-
"learning_rate": 3.
|
| 527 |
-
"loss": 0.
|
| 528 |
"step": 760
|
| 529 |
},
|
| 530 |
{
|
| 531 |
-
"epoch":
|
| 532 |
-
"learning_rate":
|
| 533 |
-
"loss": 0.
|
| 534 |
"step": 770
|
| 535 |
},
|
| 536 |
{
|
| 537 |
-
"epoch":
|
| 538 |
-
"learning_rate": 2.
|
| 539 |
-
"loss": 0.
|
| 540 |
"step": 780
|
| 541 |
},
|
| 542 |
{
|
| 543 |
-
"epoch":
|
| 544 |
-
"learning_rate": 2.
|
| 545 |
-
"loss": 0.
|
| 546 |
"step": 790
|
| 547 |
},
|
| 548 |
{
|
| 549 |
-
"epoch":
|
| 550 |
-
"learning_rate": 2.
|
| 551 |
-
"loss": 0.
|
| 552 |
"step": 800
|
| 553 |
},
|
| 554 |
{
|
| 555 |
-
"epoch":
|
| 556 |
-
"eval_accuracy": 0.
|
| 557 |
-
"eval_loss": 0.
|
| 558 |
-
"eval_runtime":
|
| 559 |
-
"eval_samples_per_second":
|
| 560 |
-
"eval_steps_per_second": 9.
|
| 561 |
"step": 800
|
| 562 |
},
|
| 563 |
{
|
| 564 |
-
"epoch":
|
| 565 |
-
"learning_rate":
|
| 566 |
-
"loss": 0.
|
| 567 |
"step": 810
|
| 568 |
},
|
| 569 |
{
|
| 570 |
-
"epoch":
|
| 571 |
-
"learning_rate":
|
| 572 |
-
"loss": 0.
|
| 573 |
"step": 820
|
| 574 |
},
|
| 575 |
{
|
| 576 |
-
"epoch":
|
| 577 |
-
"learning_rate": 1.
|
| 578 |
-
"loss": 0.
|
| 579 |
"step": 830
|
| 580 |
},
|
| 581 |
{
|
| 582 |
-
"epoch":
|
| 583 |
-
"learning_rate": 1.
|
| 584 |
-
"loss": 0.
|
| 585 |
"step": 840
|
| 586 |
},
|
| 587 |
{
|
| 588 |
-
"epoch":
|
| 589 |
-
"learning_rate": 1.
|
| 590 |
-
"loss": 0.
|
| 591 |
"step": 850
|
| 592 |
},
|
| 593 |
{
|
| 594 |
-
"epoch":
|
| 595 |
-
"learning_rate":
|
| 596 |
-
"loss": 0.
|
| 597 |
"step": 860
|
| 598 |
},
|
| 599 |
{
|
| 600 |
-
"epoch":
|
| 601 |
-
"learning_rate":
|
| 602 |
"loss": 0.0096,
|
| 603 |
"step": 870
|
| 604 |
},
|
| 605 |
{
|
| 606 |
-
"epoch":
|
| 607 |
-
"learning_rate":
|
| 608 |
-
"loss": 0.
|
| 609 |
"step": 880
|
| 610 |
},
|
| 611 |
{
|
| 612 |
-
"epoch":
|
| 613 |
-
"learning_rate":
|
| 614 |
-
"loss": 0.
|
| 615 |
"step": 890
|
| 616 |
},
|
| 617 |
{
|
| 618 |
-
"epoch":
|
| 619 |
-
"learning_rate":
|
| 620 |
-
"loss": 0.
|
| 621 |
"step": 900
|
| 622 |
},
|
| 623 |
{
|
| 624 |
-
"epoch":
|
| 625 |
-
"eval_accuracy": 0.
|
| 626 |
-
"eval_loss": 0.
|
| 627 |
-
"eval_runtime":
|
| 628 |
-
"eval_samples_per_second":
|
| 629 |
-
"eval_steps_per_second":
|
| 630 |
"step": 900
|
| 631 |
},
|
| 632 |
{
|
| 633 |
-
"epoch":
|
| 634 |
-
"
|
| 635 |
-
"
|
| 636 |
-
"
|
| 637 |
-
|
| 638 |
-
|
| 639 |
-
"
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 640 |
}
|
| 641 |
],
|
| 642 |
"logging_steps": 10,
|
| 643 |
-
"max_steps":
|
| 644 |
-
"num_train_epochs":
|
| 645 |
"save_steps": 1000,
|
| 646 |
-
"total_flos": 3.
|
| 647 |
"trial_name": null,
|
| 648 |
"trial_params": null
|
| 649 |
}
|
|
|
|
| 1 |
{
|
| 2 |
"best_metric": null,
|
| 3 |
"best_model_checkpoint": null,
|
| 4 |
+
"epoch": 5.0,
|
| 5 |
"eval_steps": 100,
|
| 6 |
+
"global_step": 915,
|
| 7 |
"is_hyper_param_search": false,
|
| 8 |
"is_local_process_zero": true,
|
| 9 |
"is_world_process_zero": true,
|
| 10 |
"log_history": [
|
| 11 |
{
|
| 12 |
+
"epoch": 0.05,
|
| 13 |
+
"learning_rate": 0.00019781420765027324,
|
| 14 |
+
"loss": 2.1723,
|
| 15 |
"step": 10
|
| 16 |
},
|
| 17 |
{
|
| 18 |
+
"epoch": 0.11,
|
| 19 |
+
"learning_rate": 0.00019562841530054644,
|
| 20 |
+
"loss": 1.7534,
|
| 21 |
"step": 20
|
| 22 |
},
|
| 23 |
{
|
| 24 |
+
"epoch": 0.16,
|
| 25 |
+
"learning_rate": 0.00019344262295081967,
|
| 26 |
+
"loss": 1.4381,
|
| 27 |
"step": 30
|
| 28 |
},
|
| 29 |
{
|
| 30 |
+
"epoch": 0.22,
|
| 31 |
+
"learning_rate": 0.0001912568306010929,
|
| 32 |
+
"loss": 1.3181,
|
| 33 |
"step": 40
|
| 34 |
},
|
| 35 |
{
|
| 36 |
+
"epoch": 0.27,
|
| 37 |
+
"learning_rate": 0.00018907103825136615,
|
| 38 |
+
"loss": 1.2203,
|
| 39 |
"step": 50
|
| 40 |
},
|
| 41 |
{
|
| 42 |
+
"epoch": 0.33,
|
| 43 |
+
"learning_rate": 0.00018688524590163935,
|
| 44 |
+
"loss": 1.0679,
|
| 45 |
"step": 60
|
| 46 |
},
|
| 47 |
{
|
| 48 |
+
"epoch": 0.38,
|
| 49 |
+
"learning_rate": 0.00018469945355191258,
|
| 50 |
+
"loss": 1.0329,
|
| 51 |
"step": 70
|
| 52 |
},
|
| 53 |
{
|
| 54 |
+
"epoch": 0.44,
|
| 55 |
+
"learning_rate": 0.0001825136612021858,
|
| 56 |
+
"loss": 0.9817,
|
| 57 |
"step": 80
|
| 58 |
},
|
| 59 |
{
|
| 60 |
+
"epoch": 0.49,
|
| 61 |
+
"learning_rate": 0.00018032786885245904,
|
| 62 |
+
"loss": 0.866,
|
| 63 |
"step": 90
|
| 64 |
},
|
| 65 |
{
|
| 66 |
+
"epoch": 0.55,
|
| 67 |
+
"learning_rate": 0.00017814207650273224,
|
| 68 |
+
"loss": 0.8634,
|
| 69 |
"step": 100
|
| 70 |
},
|
| 71 |
{
|
| 72 |
+
"epoch": 0.55,
|
| 73 |
+
"eval_accuracy": 0.6975,
|
| 74 |
+
"eval_loss": 0.9265598654747009,
|
| 75 |
+
"eval_runtime": 17.6341,
|
| 76 |
+
"eval_samples_per_second": 68.05,
|
| 77 |
+
"eval_steps_per_second": 8.506,
|
| 78 |
"step": 100
|
| 79 |
},
|
| 80 |
{
|
| 81 |
+
"epoch": 0.6,
|
| 82 |
+
"learning_rate": 0.00017595628415300547,
|
| 83 |
+
"loss": 0.6979,
|
| 84 |
"step": 110
|
| 85 |
},
|
| 86 |
{
|
| 87 |
+
"epoch": 0.66,
|
| 88 |
+
"learning_rate": 0.0001737704918032787,
|
| 89 |
+
"loss": 0.7132,
|
| 90 |
"step": 120
|
| 91 |
},
|
| 92 |
{
|
| 93 |
+
"epoch": 0.71,
|
| 94 |
+
"learning_rate": 0.00017158469945355192,
|
| 95 |
+
"loss": 0.6692,
|
| 96 |
"step": 130
|
| 97 |
},
|
| 98 |
{
|
| 99 |
+
"epoch": 0.77,
|
| 100 |
+
"learning_rate": 0.00016939890710382515,
|
| 101 |
+
"loss": 0.6,
|
| 102 |
"step": 140
|
| 103 |
},
|
| 104 |
{
|
| 105 |
+
"epoch": 0.82,
|
| 106 |
+
"learning_rate": 0.00016721311475409838,
|
| 107 |
+
"loss": 0.5497,
|
| 108 |
"step": 150
|
| 109 |
},
|
| 110 |
{
|
| 111 |
+
"epoch": 0.87,
|
| 112 |
+
"learning_rate": 0.0001650273224043716,
|
| 113 |
+
"loss": 0.4835,
|
| 114 |
"step": 160
|
| 115 |
},
|
| 116 |
{
|
| 117 |
+
"epoch": 0.93,
|
| 118 |
+
"learning_rate": 0.0001628415300546448,
|
| 119 |
+
"loss": 0.4699,
|
| 120 |
"step": 170
|
| 121 |
},
|
| 122 |
{
|
| 123 |
+
"epoch": 0.98,
|
| 124 |
+
"learning_rate": 0.00016065573770491804,
|
| 125 |
+
"loss": 0.4311,
|
| 126 |
"step": 180
|
| 127 |
},
|
| 128 |
{
|
| 129 |
+
"epoch": 1.04,
|
| 130 |
+
"learning_rate": 0.00015846994535519127,
|
| 131 |
+
"loss": 0.4363,
|
| 132 |
"step": 190
|
| 133 |
},
|
| 134 |
{
|
| 135 |
+
"epoch": 1.09,
|
| 136 |
+
"learning_rate": 0.0001562841530054645,
|
| 137 |
+
"loss": 0.3225,
|
| 138 |
"step": 200
|
| 139 |
},
|
| 140 |
{
|
| 141 |
+
"epoch": 1.09,
|
| 142 |
+
"eval_accuracy": 0.7325,
|
| 143 |
+
"eval_loss": 0.8994238376617432,
|
| 144 |
+
"eval_runtime": 16.2421,
|
| 145 |
+
"eval_samples_per_second": 73.882,
|
| 146 |
+
"eval_steps_per_second": 9.235,
|
| 147 |
"step": 200
|
| 148 |
},
|
| 149 |
{
|
| 150 |
+
"epoch": 1.15,
|
| 151 |
+
"learning_rate": 0.0001540983606557377,
|
| 152 |
+
"loss": 0.3329,
|
| 153 |
"step": 210
|
| 154 |
},
|
| 155 |
{
|
| 156 |
+
"epoch": 1.2,
|
| 157 |
+
"learning_rate": 0.00015191256830601093,
|
| 158 |
+
"loss": 0.3262,
|
| 159 |
"step": 220
|
| 160 |
},
|
| 161 |
{
|
| 162 |
+
"epoch": 1.26,
|
| 163 |
+
"learning_rate": 0.00014972677595628418,
|
| 164 |
+
"loss": 0.3394,
|
| 165 |
"step": 230
|
| 166 |
},
|
| 167 |
{
|
| 168 |
+
"epoch": 1.31,
|
| 169 |
+
"learning_rate": 0.00014754098360655738,
|
| 170 |
+
"loss": 0.3319,
|
| 171 |
"step": 240
|
| 172 |
},
|
| 173 |
{
|
| 174 |
+
"epoch": 1.37,
|
| 175 |
+
"learning_rate": 0.0001453551912568306,
|
| 176 |
+
"loss": 0.2554,
|
| 177 |
"step": 250
|
| 178 |
},
|
| 179 |
{
|
| 180 |
+
"epoch": 1.42,
|
| 181 |
+
"learning_rate": 0.00014316939890710384,
|
| 182 |
+
"loss": 0.2912,
|
| 183 |
"step": 260
|
| 184 |
},
|
| 185 |
{
|
| 186 |
+
"epoch": 1.48,
|
| 187 |
+
"learning_rate": 0.00014098360655737707,
|
| 188 |
+
"loss": 0.2801,
|
| 189 |
"step": 270
|
| 190 |
},
|
| 191 |
{
|
| 192 |
+
"epoch": 1.53,
|
| 193 |
+
"learning_rate": 0.00013879781420765027,
|
| 194 |
+
"loss": 0.2443,
|
| 195 |
"step": 280
|
| 196 |
},
|
| 197 |
{
|
| 198 |
+
"epoch": 1.58,
|
| 199 |
+
"learning_rate": 0.0001366120218579235,
|
| 200 |
+
"loss": 0.278,
|
| 201 |
"step": 290
|
| 202 |
},
|
| 203 |
{
|
| 204 |
+
"epoch": 1.64,
|
| 205 |
+
"learning_rate": 0.00013442622950819673,
|
| 206 |
+
"loss": 0.2353,
|
| 207 |
"step": 300
|
| 208 |
},
|
| 209 |
{
|
| 210 |
+
"epoch": 1.64,
|
| 211 |
+
"eval_accuracy": 0.73,
|
| 212 |
+
"eval_loss": 0.968270480632782,
|
| 213 |
+
"eval_runtime": 17.23,
|
| 214 |
+
"eval_samples_per_second": 69.646,
|
| 215 |
+
"eval_steps_per_second": 8.706,
|
| 216 |
"step": 300
|
| 217 |
},
|
| 218 |
{
|
| 219 |
+
"epoch": 1.69,
|
| 220 |
+
"learning_rate": 0.00013224043715846995,
|
| 221 |
+
"loss": 0.2092,
|
| 222 |
"step": 310
|
| 223 |
},
|
| 224 |
{
|
| 225 |
+
"epoch": 1.75,
|
| 226 |
+
"learning_rate": 0.00013005464480874316,
|
| 227 |
+
"loss": 0.1959,
|
| 228 |
"step": 320
|
| 229 |
},
|
| 230 |
{
|
| 231 |
+
"epoch": 1.8,
|
| 232 |
+
"learning_rate": 0.0001278688524590164,
|
| 233 |
+
"loss": 0.1841,
|
| 234 |
"step": 330
|
| 235 |
},
|
| 236 |
{
|
| 237 |
+
"epoch": 1.86,
|
| 238 |
+
"learning_rate": 0.00012568306010928964,
|
| 239 |
+
"loss": 0.2678,
|
| 240 |
"step": 340
|
| 241 |
},
|
| 242 |
{
|
| 243 |
+
"epoch": 1.91,
|
| 244 |
+
"learning_rate": 0.00012349726775956284,
|
| 245 |
+
"loss": 0.2366,
|
| 246 |
"step": 350
|
| 247 |
},
|
| 248 |
{
|
| 249 |
+
"epoch": 1.97,
|
| 250 |
+
"learning_rate": 0.00012131147540983607,
|
| 251 |
+
"loss": 0.2198,
|
| 252 |
"step": 360
|
| 253 |
},
|
| 254 |
{
|
| 255 |
+
"epoch": 2.02,
|
| 256 |
+
"learning_rate": 0.0001191256830601093,
|
| 257 |
+
"loss": 0.1459,
|
| 258 |
"step": 370
|
| 259 |
},
|
| 260 |
{
|
| 261 |
+
"epoch": 2.08,
|
| 262 |
+
"learning_rate": 0.00011693989071038251,
|
| 263 |
+
"loss": 0.0904,
|
| 264 |
"step": 380
|
| 265 |
},
|
| 266 |
{
|
| 267 |
+
"epoch": 2.13,
|
| 268 |
+
"learning_rate": 0.00011475409836065574,
|
| 269 |
+
"loss": 0.1093,
|
| 270 |
"step": 390
|
| 271 |
},
|
| 272 |
{
|
| 273 |
+
"epoch": 2.19,
|
| 274 |
+
"learning_rate": 0.00011256830601092896,
|
| 275 |
+
"loss": 0.1119,
|
| 276 |
"step": 400
|
| 277 |
},
|
| 278 |
{
|
| 279 |
+
"epoch": 2.19,
|
| 280 |
+
"eval_accuracy": 0.7491666666666666,
|
| 281 |
+
"eval_loss": 0.9247021675109863,
|
| 282 |
+
"eval_runtime": 16.092,
|
| 283 |
+
"eval_samples_per_second": 74.571,
|
| 284 |
+
"eval_steps_per_second": 9.321,
|
| 285 |
"step": 400
|
| 286 |
},
|
| 287 |
{
|
| 288 |
+
"epoch": 2.24,
|
| 289 |
+
"learning_rate": 0.00011038251366120218,
|
| 290 |
+
"loss": 0.1183,
|
| 291 |
"step": 410
|
| 292 |
},
|
| 293 |
{
|
| 294 |
+
"epoch": 2.3,
|
| 295 |
+
"learning_rate": 0.00010819672131147543,
|
| 296 |
+
"loss": 0.1155,
|
| 297 |
"step": 420
|
| 298 |
},
|
| 299 |
{
|
| 300 |
+
"epoch": 2.35,
|
| 301 |
+
"learning_rate": 0.00010601092896174864,
|
| 302 |
+
"loss": 0.138,
|
| 303 |
"step": 430
|
| 304 |
},
|
| 305 |
{
|
| 306 |
+
"epoch": 2.4,
|
| 307 |
+
"learning_rate": 0.00010382513661202187,
|
| 308 |
+
"loss": 0.1124,
|
| 309 |
"step": 440
|
| 310 |
},
|
| 311 |
{
|
| 312 |
+
"epoch": 2.46,
|
| 313 |
+
"learning_rate": 0.00010163934426229508,
|
| 314 |
+
"loss": 0.1535,
|
| 315 |
"step": 450
|
| 316 |
},
|
| 317 |
{
|
| 318 |
+
"epoch": 2.51,
|
| 319 |
+
"learning_rate": 9.945355191256831e-05,
|
| 320 |
+
"loss": 0.1895,
|
| 321 |
"step": 460
|
| 322 |
},
|
| 323 |
{
|
| 324 |
+
"epoch": 2.57,
|
| 325 |
+
"learning_rate": 9.726775956284153e-05,
|
| 326 |
+
"loss": 0.128,
|
| 327 |
"step": 470
|
| 328 |
},
|
| 329 |
{
|
| 330 |
+
"epoch": 2.62,
|
| 331 |
+
"learning_rate": 9.508196721311476e-05,
|
| 332 |
+
"loss": 0.0633,
|
| 333 |
"step": 480
|
| 334 |
},
|
| 335 |
{
|
| 336 |
+
"epoch": 2.68,
|
| 337 |
+
"learning_rate": 9.289617486338798e-05,
|
| 338 |
+
"loss": 0.0788,
|
| 339 |
"step": 490
|
| 340 |
},
|
| 341 |
{
|
| 342 |
+
"epoch": 2.73,
|
| 343 |
+
"learning_rate": 9.071038251366121e-05,
|
| 344 |
+
"loss": 0.049,
|
| 345 |
"step": 500
|
| 346 |
},
|
| 347 |
{
|
| 348 |
+
"epoch": 2.73,
|
| 349 |
+
"eval_accuracy": 0.7566666666666667,
|
| 350 |
+
"eval_loss": 0.9662973880767822,
|
| 351 |
+
"eval_runtime": 15.9004,
|
| 352 |
+
"eval_samples_per_second": 75.47,
|
| 353 |
+
"eval_steps_per_second": 9.434,
|
| 354 |
"step": 500
|
| 355 |
},
|
| 356 |
{
|
| 357 |
+
"epoch": 2.79,
|
| 358 |
+
"learning_rate": 8.852459016393443e-05,
|
| 359 |
+
"loss": 0.0829,
|
| 360 |
"step": 510
|
| 361 |
},
|
| 362 |
{
|
| 363 |
+
"epoch": 2.84,
|
| 364 |
+
"learning_rate": 8.633879781420766e-05,
|
| 365 |
+
"loss": 0.0821,
|
| 366 |
"step": 520
|
| 367 |
},
|
| 368 |
{
|
| 369 |
+
"epoch": 2.9,
|
| 370 |
+
"learning_rate": 8.415300546448088e-05,
|
| 371 |
+
"loss": 0.0554,
|
| 372 |
"step": 530
|
| 373 |
},
|
| 374 |
{
|
| 375 |
+
"epoch": 2.95,
|
| 376 |
+
"learning_rate": 8.19672131147541e-05,
|
| 377 |
+
"loss": 0.0756,
|
| 378 |
"step": 540
|
| 379 |
},
|
| 380 |
{
|
| 381 |
+
"epoch": 3.01,
|
| 382 |
+
"learning_rate": 7.978142076502733e-05,
|
| 383 |
+
"loss": 0.136,
|
| 384 |
"step": 550
|
| 385 |
},
|
| 386 |
{
|
| 387 |
+
"epoch": 3.06,
|
| 388 |
+
"learning_rate": 7.759562841530054e-05,
|
| 389 |
+
"loss": 0.0903,
|
| 390 |
"step": 560
|
| 391 |
},
|
| 392 |
{
|
| 393 |
+
"epoch": 3.11,
|
| 394 |
+
"learning_rate": 7.540983606557377e-05,
|
| 395 |
+
"loss": 0.0868,
|
| 396 |
"step": 570
|
| 397 |
},
|
| 398 |
{
|
| 399 |
+
"epoch": 3.17,
|
| 400 |
+
"learning_rate": 7.3224043715847e-05,
|
| 401 |
+
"loss": 0.0523,
|
| 402 |
"step": 580
|
| 403 |
},
|
| 404 |
{
|
| 405 |
+
"epoch": 3.22,
|
| 406 |
+
"learning_rate": 7.103825136612023e-05,
|
| 407 |
+
"loss": 0.0396,
|
| 408 |
"step": 590
|
| 409 |
},
|
| 410 |
{
|
| 411 |
+
"epoch": 3.28,
|
| 412 |
+
"learning_rate": 6.885245901639344e-05,
|
| 413 |
+
"loss": 0.0537,
|
| 414 |
"step": 600
|
| 415 |
},
|
| 416 |
{
|
| 417 |
+
"epoch": 3.28,
|
| 418 |
+
"eval_accuracy": 0.7566666666666667,
|
| 419 |
+
"eval_loss": 1.0557572841644287,
|
| 420 |
+
"eval_runtime": 16.1696,
|
| 421 |
+
"eval_samples_per_second": 74.213,
|
| 422 |
+
"eval_steps_per_second": 9.277,
|
| 423 |
"step": 600
|
| 424 |
},
|
| 425 |
{
|
| 426 |
+
"epoch": 3.33,
|
| 427 |
+
"learning_rate": 6.666666666666667e-05,
|
| 428 |
+
"loss": 0.064,
|
| 429 |
"step": 610
|
| 430 |
},
|
| 431 |
{
|
| 432 |
+
"epoch": 3.39,
|
| 433 |
+
"learning_rate": 6.44808743169399e-05,
|
| 434 |
+
"loss": 0.0206,
|
| 435 |
"step": 620
|
| 436 |
},
|
| 437 |
{
|
| 438 |
+
"epoch": 3.44,
|
| 439 |
+
"learning_rate": 6.229508196721313e-05,
|
| 440 |
+
"loss": 0.028,
|
| 441 |
"step": 630
|
| 442 |
},
|
| 443 |
{
|
| 444 |
+
"epoch": 3.5,
|
| 445 |
+
"learning_rate": 6.010928961748634e-05,
|
| 446 |
+
"loss": 0.026,
|
| 447 |
"step": 640
|
| 448 |
},
|
| 449 |
{
|
| 450 |
+
"epoch": 3.55,
|
| 451 |
+
"learning_rate": 5.792349726775956e-05,
|
| 452 |
+
"loss": 0.0229,
|
| 453 |
"step": 650
|
| 454 |
},
|
| 455 |
{
|
| 456 |
+
"epoch": 3.61,
|
| 457 |
+
"learning_rate": 5.5737704918032785e-05,
|
| 458 |
+
"loss": 0.032,
|
| 459 |
"step": 660
|
| 460 |
},
|
| 461 |
{
|
| 462 |
+
"epoch": 3.66,
|
| 463 |
+
"learning_rate": 5.355191256830602e-05,
|
| 464 |
+
"loss": 0.023,
|
| 465 |
"step": 670
|
| 466 |
},
|
| 467 |
{
|
| 468 |
+
"epoch": 3.72,
|
| 469 |
+
"learning_rate": 5.136612021857924e-05,
|
| 470 |
+
"loss": 0.0269,
|
| 471 |
"step": 680
|
| 472 |
},
|
| 473 |
{
|
| 474 |
+
"epoch": 3.77,
|
| 475 |
+
"learning_rate": 4.918032786885246e-05,
|
| 476 |
+
"loss": 0.023,
|
| 477 |
"step": 690
|
| 478 |
},
|
| 479 |
{
|
| 480 |
+
"epoch": 3.83,
|
| 481 |
+
"learning_rate": 4.6994535519125685e-05,
|
| 482 |
+
"loss": 0.0274,
|
| 483 |
"step": 700
|
| 484 |
},
|
| 485 |
{
|
| 486 |
+
"epoch": 3.83,
|
| 487 |
+
"eval_accuracy": 0.7691666666666667,
|
| 488 |
+
"eval_loss": 1.0343540906906128,
|
| 489 |
+
"eval_runtime": 16.2987,
|
| 490 |
+
"eval_samples_per_second": 73.626,
|
| 491 |
+
"eval_steps_per_second": 9.203,
|
| 492 |
"step": 700
|
| 493 |
},
|
| 494 |
{
|
| 495 |
+
"epoch": 3.88,
|
| 496 |
+
"learning_rate": 4.4808743169398906e-05,
|
| 497 |
+
"loss": 0.0197,
|
| 498 |
"step": 710
|
| 499 |
},
|
| 500 |
{
|
| 501 |
+
"epoch": 3.93,
|
| 502 |
+
"learning_rate": 4.262295081967213e-05,
|
| 503 |
+
"loss": 0.0195,
|
| 504 |
"step": 720
|
| 505 |
},
|
| 506 |
{
|
| 507 |
+
"epoch": 3.99,
|
| 508 |
+
"learning_rate": 4.0437158469945356e-05,
|
| 509 |
+
"loss": 0.0122,
|
| 510 |
"step": 730
|
| 511 |
},
|
| 512 |
{
|
| 513 |
+
"epoch": 4.04,
|
| 514 |
+
"learning_rate": 3.825136612021858e-05,
|
| 515 |
+
"loss": 0.0156,
|
| 516 |
"step": 740
|
| 517 |
},
|
| 518 |
{
|
| 519 |
+
"epoch": 4.1,
|
| 520 |
+
"learning_rate": 3.6065573770491806e-05,
|
| 521 |
+
"loss": 0.0115,
|
| 522 |
"step": 750
|
| 523 |
},
|
| 524 |
{
|
| 525 |
+
"epoch": 4.15,
|
| 526 |
+
"learning_rate": 3.387978142076503e-05,
|
| 527 |
+
"loss": 0.0118,
|
| 528 |
"step": 760
|
| 529 |
},
|
| 530 |
{
|
| 531 |
+
"epoch": 4.21,
|
| 532 |
+
"learning_rate": 3.1693989071038256e-05,
|
| 533 |
+
"loss": 0.0123,
|
| 534 |
"step": 770
|
| 535 |
},
|
| 536 |
{
|
| 537 |
+
"epoch": 4.26,
|
| 538 |
+
"learning_rate": 2.9508196721311478e-05,
|
| 539 |
+
"loss": 0.0108,
|
| 540 |
"step": 780
|
| 541 |
},
|
| 542 |
{
|
| 543 |
+
"epoch": 4.32,
|
| 544 |
+
"learning_rate": 2.7322404371584703e-05,
|
| 545 |
+
"loss": 0.0112,
|
| 546 |
"step": 790
|
| 547 |
},
|
| 548 |
{
|
| 549 |
+
"epoch": 4.37,
|
| 550 |
+
"learning_rate": 2.5136612021857924e-05,
|
| 551 |
+
"loss": 0.0102,
|
| 552 |
"step": 800
|
| 553 |
},
|
| 554 |
{
|
| 555 |
+
"epoch": 4.37,
|
| 556 |
+
"eval_accuracy": 0.7941666666666667,
|
| 557 |
+
"eval_loss": 0.9259100556373596,
|
| 558 |
+
"eval_runtime": 16.5205,
|
| 559 |
+
"eval_samples_per_second": 72.637,
|
| 560 |
+
"eval_steps_per_second": 9.08,
|
| 561 |
"step": 800
|
| 562 |
},
|
| 563 |
{
|
| 564 |
+
"epoch": 4.43,
|
| 565 |
+
"learning_rate": 2.295081967213115e-05,
|
| 566 |
+
"loss": 0.0101,
|
| 567 |
"step": 810
|
| 568 |
},
|
| 569 |
{
|
| 570 |
+
"epoch": 4.48,
|
| 571 |
+
"learning_rate": 2.0765027322404374e-05,
|
| 572 |
+
"loss": 0.0099,
|
| 573 |
"step": 820
|
| 574 |
},
|
| 575 |
{
|
| 576 |
+
"epoch": 4.54,
|
| 577 |
+
"learning_rate": 1.85792349726776e-05,
|
| 578 |
+
"loss": 0.0097,
|
| 579 |
"step": 830
|
| 580 |
},
|
| 581 |
{
|
| 582 |
+
"epoch": 4.59,
|
| 583 |
+
"learning_rate": 1.6393442622950818e-05,
|
| 584 |
+
"loss": 0.011,
|
| 585 |
"step": 840
|
| 586 |
},
|
| 587 |
{
|
| 588 |
+
"epoch": 4.64,
|
| 589 |
+
"learning_rate": 1.4207650273224044e-05,
|
| 590 |
+
"loss": 0.0094,
|
| 591 |
"step": 850
|
| 592 |
},
|
| 593 |
{
|
| 594 |
+
"epoch": 4.7,
|
| 595 |
+
"learning_rate": 1.2021857923497268e-05,
|
| 596 |
+
"loss": 0.01,
|
| 597 |
"step": 860
|
| 598 |
},
|
| 599 |
{
|
| 600 |
+
"epoch": 4.75,
|
| 601 |
+
"learning_rate": 9.836065573770493e-06,
|
| 602 |
"loss": 0.0096,
|
| 603 |
"step": 870
|
| 604 |
},
|
| 605 |
{
|
| 606 |
+
"epoch": 4.81,
|
| 607 |
+
"learning_rate": 7.650273224043716e-06,
|
| 608 |
+
"loss": 0.0098,
|
| 609 |
"step": 880
|
| 610 |
},
|
| 611 |
{
|
| 612 |
+
"epoch": 4.86,
|
| 613 |
+
"learning_rate": 5.46448087431694e-06,
|
| 614 |
+
"loss": 0.0163,
|
| 615 |
"step": 890
|
| 616 |
},
|
| 617 |
{
|
| 618 |
+
"epoch": 4.92,
|
| 619 |
+
"learning_rate": 3.278688524590164e-06,
|
| 620 |
+
"loss": 0.0095,
|
| 621 |
"step": 900
|
| 622 |
},
|
| 623 |
{
|
| 624 |
+
"epoch": 4.92,
|
| 625 |
+
"eval_accuracy": 0.785,
|
| 626 |
+
"eval_loss": 0.9604464769363403,
|
| 627 |
+
"eval_runtime": 17.3811,
|
| 628 |
+
"eval_samples_per_second": 69.04,
|
| 629 |
+
"eval_steps_per_second": 8.63,
|
| 630 |
"step": 900
|
| 631 |
},
|
| 632 |
{
|
| 633 |
+
"epoch": 4.97,
|
| 634 |
+
"learning_rate": 1.092896174863388e-06,
|
| 635 |
+
"loss": 0.0092,
|
| 636 |
+
"step": 910
|
| 637 |
+
},
|
| 638 |
+
{
|
| 639 |
+
"epoch": 5.0,
|
| 640 |
+
"step": 915,
|
| 641 |
+
"total_flos": 3.4029172406502605e+18,
|
| 642 |
+
"train_loss": 0.2756463099698551,
|
| 643 |
+
"train_runtime": 974.1891,
|
| 644 |
+
"train_samples_per_second": 45.073,
|
| 645 |
+
"train_steps_per_second": 0.939
|
| 646 |
}
|
| 647 |
],
|
| 648 |
"logging_steps": 10,
|
| 649 |
+
"max_steps": 915,
|
| 650 |
+
"num_train_epochs": 5,
|
| 651 |
"save_steps": 1000,
|
| 652 |
+
"total_flos": 3.4029172406502605e+18,
|
| 653 |
"trial_name": null,
|
| 654 |
"trial_params": null
|
| 655 |
}
|
training_args.bin
CHANGED
|
@@ -1,3 +1,3 @@
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
-
oid sha256:
|
| 3 |
size 4536
|
|
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:e3c6c5a1c7e0e3827f5dff1aefd6f08bc6fd5d04c2387bb90f3b5b019703d59b
|
| 3 |
size 4536
|