Training in progress, step 500, checkpoint
Browse files- checkpoint-500/optimizer.pt +1 -1
- checkpoint-500/pytorch_model.bin +1 -1
- checkpoint-500/rng_state.pth +1 -1
- checkpoint-500/trainer_state.json +310 -175
- checkpoint-500/training_args.bin +1 -1
checkpoint-500/optimizer.pt
CHANGED
|
@@ -1,3 +1,3 @@
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
-
oid sha256:
|
| 3 |
size 997351674
|
|
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:d9b5331a049bf007e3fe60e6b2ee0194b68959b298abbf3a9c2ee94807a4dd2d
|
| 3 |
size 997351674
|
checkpoint-500/pytorch_model.bin
CHANGED
|
@@ -1,3 +1,3 @@
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
-
oid sha256:
|
| 3 |
size 498661166
|
|
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:58f17f1f744378c30c2b3f5533b9d7bedf71b3d71f7f094f101e878c5d4bf595
|
| 3 |
size 498661166
|
checkpoint-500/rng_state.pth
CHANGED
|
@@ -1,3 +1,3 @@
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
-
oid sha256:
|
| 3 |
size 14244
|
|
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:fd8ed62c04c41b6e0c29c4bf4ad972a544ebceea21938ff9f3a06432ffd506cd
|
| 3 |
size 14244
|
checkpoint-500/trainer_state.json
CHANGED
|
@@ -1,409 +1,544 @@
|
|
| 1 |
{
|
| 2 |
"best_metric": null,
|
| 3 |
"best_model_checkpoint": null,
|
| 4 |
-
"epoch":
|
| 5 |
-
"eval_steps":
|
| 6 |
"global_step": 500,
|
| 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": 5e-07,
|
| 14 |
-
"loss": 1.
|
| 15 |
"step": 10
|
| 16 |
},
|
| 17 |
{
|
| 18 |
-
"epoch": 0.
|
| 19 |
"learning_rate": 1e-06,
|
| 20 |
-
"loss": 1.
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 21 |
"step": 20
|
| 22 |
},
|
| 23 |
{
|
| 24 |
-
"epoch": 0.
|
| 25 |
"learning_rate": 1.5e-06,
|
| 26 |
-
"loss": 1.
|
| 27 |
"step": 30
|
| 28 |
},
|
| 29 |
{
|
| 30 |
-
"epoch": 0.
|
| 31 |
"learning_rate": 2e-06,
|
| 32 |
-
"loss": 1.
|
| 33 |
"step": 40
|
| 34 |
},
|
| 35 |
{
|
| 36 |
-
"epoch": 0.
|
| 37 |
-
"
|
| 38 |
-
"
|
| 39 |
-
"
|
|
|
|
|
|
|
|
|
|
| 40 |
},
|
| 41 |
{
|
| 42 |
-
"epoch": 0.
|
| 43 |
-
"
|
| 44 |
-
"
|
| 45 |
-
"eval_runtime": 49.5848,
|
| 46 |
-
"eval_samples_per_second": 605.307,
|
| 47 |
-
"eval_steps_per_second": 1.19,
|
| 48 |
"step": 50
|
| 49 |
},
|
| 50 |
{
|
| 51 |
-
"epoch": 0.
|
| 52 |
"learning_rate": 3e-06,
|
| 53 |
-
"loss": 1.
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 54 |
"step": 60
|
| 55 |
},
|
| 56 |
{
|
| 57 |
-
"epoch": 0.
|
| 58 |
"learning_rate": 3.5e-06,
|
| 59 |
-
"loss": 1.
|
| 60 |
"step": 70
|
| 61 |
},
|
| 62 |
{
|
| 63 |
-
"epoch": 0.
|
| 64 |
"learning_rate": 4e-06,
|
| 65 |
-
"loss": 1.
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 66 |
"step": 80
|
| 67 |
},
|
| 68 |
{
|
| 69 |
-
"epoch": 0.
|
| 70 |
"learning_rate": 4.5e-06,
|
| 71 |
-
"loss":
|
| 72 |
"step": 90
|
| 73 |
},
|
| 74 |
{
|
| 75 |
-
"epoch": 0.
|
| 76 |
"learning_rate": 5e-06,
|
| 77 |
-
"loss":
|
| 78 |
"step": 100
|
| 79 |
},
|
| 80 |
{
|
| 81 |
-
"epoch": 0.
|
| 82 |
-
"eval_accuracy": 0.
|
| 83 |
-
"eval_loss": 0.
|
| 84 |
-
"eval_runtime": 49.
|
| 85 |
-
"eval_samples_per_second":
|
| 86 |
-
"eval_steps_per_second": 1.
|
| 87 |
"step": 100
|
| 88 |
},
|
| 89 |
{
|
| 90 |
-
"epoch": 0.
|
| 91 |
"learning_rate": 5.5e-06,
|
| 92 |
-
"loss":
|
| 93 |
"step": 110
|
| 94 |
},
|
| 95 |
{
|
| 96 |
-
"epoch": 0.
|
| 97 |
"learning_rate": 6e-06,
|
| 98 |
-
"loss":
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 99 |
"step": 120
|
| 100 |
},
|
| 101 |
{
|
| 102 |
-
"epoch": 0.
|
| 103 |
"learning_rate": 6.5e-06,
|
| 104 |
-
"loss":
|
| 105 |
"step": 130
|
| 106 |
},
|
| 107 |
{
|
| 108 |
-
"epoch": 0.
|
| 109 |
"learning_rate": 7e-06,
|
| 110 |
-
"loss": 0.
|
| 111 |
"step": 140
|
| 112 |
},
|
| 113 |
{
|
| 114 |
-
"epoch": 0.
|
| 115 |
-
"
|
| 116 |
-
"
|
| 117 |
-
"
|
|
|
|
|
|
|
|
|
|
| 118 |
},
|
| 119 |
{
|
| 120 |
-
"epoch": 0.
|
| 121 |
-
"
|
| 122 |
-
"
|
| 123 |
-
"eval_runtime": 49.4568,
|
| 124 |
-
"eval_samples_per_second": 606.874,
|
| 125 |
-
"eval_steps_per_second": 1.193,
|
| 126 |
"step": 150
|
| 127 |
},
|
| 128 |
{
|
| 129 |
-
"epoch": 0.
|
| 130 |
"learning_rate": 8e-06,
|
| 131 |
-
"loss": 0.
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 132 |
"step": 160
|
| 133 |
},
|
| 134 |
{
|
| 135 |
-
"epoch": 0.
|
| 136 |
"learning_rate": 8.500000000000002e-06,
|
| 137 |
-
"loss": 0.
|
| 138 |
"step": 170
|
| 139 |
},
|
| 140 |
{
|
| 141 |
-
"epoch": 0.
|
| 142 |
"learning_rate": 9e-06,
|
| 143 |
-
"loss": 0.
|
| 144 |
"step": 180
|
| 145 |
},
|
| 146 |
{
|
| 147 |
-
"epoch": 0.
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 148 |
"learning_rate": 9.5e-06,
|
| 149 |
-
"loss": 0.
|
| 150 |
"step": 190
|
| 151 |
},
|
| 152 |
{
|
| 153 |
-
"epoch": 0.
|
| 154 |
"learning_rate": 1e-05,
|
| 155 |
-
"loss": 0.
|
| 156 |
"step": 200
|
| 157 |
},
|
| 158 |
{
|
| 159 |
-
"epoch": 0.
|
| 160 |
-
"eval_accuracy": 0.
|
| 161 |
-
"eval_loss": 0.
|
| 162 |
-
"eval_runtime": 49.
|
| 163 |
-
"eval_samples_per_second":
|
| 164 |
-
"eval_steps_per_second": 1.
|
| 165 |
"step": 200
|
| 166 |
},
|
| 167 |
{
|
| 168 |
-
"epoch": 0.
|
| 169 |
"learning_rate": 1.0500000000000001e-05,
|
| 170 |
-
"loss": 0.
|
| 171 |
"step": 210
|
| 172 |
},
|
| 173 |
{
|
| 174 |
-
"epoch": 0.
|
| 175 |
"learning_rate": 1.1e-05,
|
| 176 |
-
"loss": 0.
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 177 |
"step": 220
|
| 178 |
},
|
| 179 |
{
|
| 180 |
-
"epoch": 0.
|
| 181 |
"learning_rate": 1.15e-05,
|
| 182 |
-
"loss": 0.
|
| 183 |
"step": 230
|
| 184 |
},
|
| 185 |
{
|
| 186 |
-
"epoch": 0.
|
| 187 |
"learning_rate": 1.2e-05,
|
| 188 |
-
"loss": 0.
|
| 189 |
"step": 240
|
| 190 |
},
|
| 191 |
{
|
| 192 |
-
"epoch": 0.
|
| 193 |
-
"
|
| 194 |
-
"
|
| 195 |
-
"
|
|
|
|
|
|
|
|
|
|
| 196 |
},
|
| 197 |
{
|
| 198 |
-
"epoch": 0.
|
| 199 |
-
"
|
| 200 |
-
"
|
| 201 |
-
"eval_runtime": 49.5108,
|
| 202 |
-
"eval_samples_per_second": 606.212,
|
| 203 |
-
"eval_steps_per_second": 1.192,
|
| 204 |
"step": 250
|
| 205 |
},
|
| 206 |
{
|
| 207 |
-
"epoch": 0.
|
| 208 |
"learning_rate": 1.3e-05,
|
| 209 |
-
"loss": 0.
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 210 |
"step": 260
|
| 211 |
},
|
| 212 |
{
|
| 213 |
-
"epoch": 0.
|
| 214 |
"learning_rate": 1.35e-05,
|
| 215 |
-
"loss": 0.
|
| 216 |
"step": 270
|
| 217 |
},
|
| 218 |
{
|
| 219 |
-
"epoch": 0.
|
| 220 |
"learning_rate": 1.4e-05,
|
| 221 |
-
"loss": 0.
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 222 |
"step": 280
|
| 223 |
},
|
| 224 |
{
|
| 225 |
-
"epoch": 0.
|
| 226 |
"learning_rate": 1.4500000000000002e-05,
|
| 227 |
-
"loss": 0.
|
| 228 |
"step": 290
|
| 229 |
},
|
| 230 |
{
|
| 231 |
-
"epoch": 0.
|
| 232 |
"learning_rate": 1.5e-05,
|
| 233 |
-
"loss": 0.
|
| 234 |
"step": 300
|
| 235 |
},
|
| 236 |
{
|
| 237 |
-
"epoch": 0.
|
| 238 |
-
"eval_accuracy": 0.
|
| 239 |
-
"eval_loss": 0.
|
| 240 |
-
"eval_runtime": 49.
|
| 241 |
-
"eval_samples_per_second":
|
| 242 |
-
"eval_steps_per_second": 1.
|
| 243 |
"step": 300
|
| 244 |
},
|
| 245 |
{
|
| 246 |
-
"epoch": 0.
|
| 247 |
"learning_rate": 1.55e-05,
|
| 248 |
-
"loss": 0.
|
| 249 |
"step": 310
|
| 250 |
},
|
| 251 |
{
|
| 252 |
-
"epoch": 0.
|
| 253 |
"learning_rate": 1.6e-05,
|
| 254 |
-
"loss": 0.
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 255 |
"step": 320
|
| 256 |
},
|
| 257 |
{
|
| 258 |
-
"epoch": 0.
|
| 259 |
"learning_rate": 1.65e-05,
|
| 260 |
-
"loss": 0.
|
| 261 |
"step": 330
|
| 262 |
},
|
| 263 |
{
|
| 264 |
-
"epoch":
|
| 265 |
"learning_rate": 1.7000000000000003e-05,
|
| 266 |
-
"loss": 0.
|
| 267 |
"step": 340
|
| 268 |
},
|
| 269 |
{
|
| 270 |
-
"epoch":
|
| 271 |
-
"
|
| 272 |
-
"
|
| 273 |
-
"
|
|
|
|
|
|
|
|
|
|
| 274 |
},
|
| 275 |
{
|
| 276 |
-
"epoch":
|
| 277 |
-
"
|
| 278 |
-
"
|
| 279 |
-
"eval_runtime": 49.5126,
|
| 280 |
-
"eval_samples_per_second": 606.189,
|
| 281 |
-
"eval_steps_per_second": 1.192,
|
| 282 |
"step": 350
|
| 283 |
},
|
| 284 |
{
|
| 285 |
-
"epoch":
|
| 286 |
"learning_rate": 1.8e-05,
|
| 287 |
-
"loss": 0.
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 288 |
"step": 360
|
| 289 |
},
|
| 290 |
{
|
| 291 |
-
"epoch":
|
| 292 |
"learning_rate": 1.85e-05,
|
| 293 |
-
"loss": 0.
|
| 294 |
"step": 370
|
| 295 |
},
|
| 296 |
{
|
| 297 |
-
"epoch":
|
| 298 |
"learning_rate": 1.9e-05,
|
| 299 |
-
"loss": 0.
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 300 |
"step": 380
|
| 301 |
},
|
| 302 |
{
|
| 303 |
-
"epoch":
|
| 304 |
"learning_rate": 1.95e-05,
|
| 305 |
-
"loss": 0.
|
| 306 |
"step": 390
|
| 307 |
},
|
| 308 |
{
|
| 309 |
-
"epoch":
|
| 310 |
"learning_rate": 2e-05,
|
| 311 |
-
"loss": 0.
|
| 312 |
"step": 400
|
| 313 |
},
|
| 314 |
{
|
| 315 |
-
"epoch":
|
| 316 |
-
"eval_accuracy": 0.
|
| 317 |
-
"eval_loss": 0.
|
| 318 |
-
"eval_runtime": 49.
|
| 319 |
-
"eval_samples_per_second": 606.
|
| 320 |
-
"eval_steps_per_second": 1.
|
| 321 |
"step": 400
|
| 322 |
},
|
| 323 |
{
|
| 324 |
-
"epoch":
|
| 325 |
"learning_rate": 2.05e-05,
|
| 326 |
-
"loss": 0.
|
| 327 |
"step": 410
|
| 328 |
},
|
| 329 |
{
|
| 330 |
-
"epoch":
|
| 331 |
"learning_rate": 2.1000000000000002e-05,
|
| 332 |
-
"loss": 0.
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 333 |
"step": 420
|
| 334 |
},
|
| 335 |
{
|
| 336 |
-
"epoch":
|
| 337 |
"learning_rate": 2.1499999999999997e-05,
|
| 338 |
-
"loss": 0.
|
| 339 |
"step": 430
|
| 340 |
},
|
| 341 |
{
|
| 342 |
-
"epoch":
|
| 343 |
"learning_rate": 2.2e-05,
|
| 344 |
-
"loss": 0.
|
| 345 |
"step": 440
|
| 346 |
},
|
| 347 |
{
|
| 348 |
-
"epoch":
|
| 349 |
-
"
|
| 350 |
-
"
|
| 351 |
-
"
|
|
|
|
|
|
|
|
|
|
| 352 |
},
|
| 353 |
{
|
| 354 |
-
"epoch":
|
| 355 |
-
"
|
| 356 |
-
"
|
| 357 |
-
"eval_runtime": 49.5226,
|
| 358 |
-
"eval_samples_per_second": 606.066,
|
| 359 |
-
"eval_steps_per_second": 1.191,
|
| 360 |
"step": 450
|
| 361 |
},
|
| 362 |
{
|
| 363 |
-
"epoch":
|
| 364 |
"learning_rate": 2.3e-05,
|
| 365 |
-
"loss": 0.
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 366 |
"step": 460
|
| 367 |
},
|
| 368 |
{
|
| 369 |
-
"epoch":
|
| 370 |
"learning_rate": 2.3500000000000002e-05,
|
| 371 |
-
"loss": 0.
|
| 372 |
"step": 470
|
| 373 |
},
|
| 374 |
{
|
| 375 |
-
"epoch":
|
| 376 |
"learning_rate": 2.4e-05,
|
| 377 |
-
"loss": 0.
|
| 378 |
"step": 480
|
| 379 |
},
|
| 380 |
{
|
| 381 |
-
"epoch":
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 382 |
"learning_rate": 2.4500000000000003e-05,
|
| 383 |
-
"loss": 0.
|
| 384 |
"step": 490
|
| 385 |
},
|
| 386 |
{
|
| 387 |
-
"epoch":
|
| 388 |
"learning_rate": 2.5e-05,
|
| 389 |
-
"loss": 0.
|
| 390 |
"step": 500
|
| 391 |
},
|
| 392 |
{
|
| 393 |
-
"epoch":
|
| 394 |
-
"eval_accuracy": 0.
|
| 395 |
-
"eval_loss": 0.
|
| 396 |
-
"eval_runtime": 49.
|
| 397 |
-
"eval_samples_per_second":
|
| 398 |
-
"eval_steps_per_second": 1.
|
| 399 |
"step": 500
|
| 400 |
}
|
| 401 |
],
|
| 402 |
"logging_steps": 10,
|
| 403 |
-
"max_steps":
|
| 404 |
"num_train_epochs": 3,
|
| 405 |
-
"save_steps":
|
| 406 |
-
"total_flos":
|
| 407 |
"trial_name": null,
|
| 408 |
"trial_params": null
|
| 409 |
}
|
|
|
|
| 1 |
{
|
| 2 |
"best_metric": null,
|
| 3 |
"best_model_checkpoint": null,
|
| 4 |
+
"epoch": 1.5048908954100828,
|
| 5 |
+
"eval_steps": 20,
|
| 6 |
"global_step": 500,
|
| 7 |
"is_hyper_param_search": false,
|
| 8 |
"is_local_process_zero": true,
|
| 9 |
"is_world_process_zero": true,
|
| 10 |
"log_history": [
|
| 11 |
{
|
| 12 |
+
"epoch": 0.03,
|
| 13 |
"learning_rate": 5e-07,
|
| 14 |
+
"loss": 1.1046,
|
| 15 |
"step": 10
|
| 16 |
},
|
| 17 |
{
|
| 18 |
+
"epoch": 0.06,
|
| 19 |
"learning_rate": 1e-06,
|
| 20 |
+
"loss": 1.0664,
|
| 21 |
+
"step": 20
|
| 22 |
+
},
|
| 23 |
+
{
|
| 24 |
+
"epoch": 0.06,
|
| 25 |
+
"eval_accuracy": 0.44479243019924036,
|
| 26 |
+
"eval_loss": 1.065091609954834,
|
| 27 |
+
"eval_runtime": 50.348,
|
| 28 |
+
"eval_samples_per_second": 596.131,
|
| 29 |
+
"eval_steps_per_second": 1.172,
|
| 30 |
"step": 20
|
| 31 |
},
|
| 32 |
{
|
| 33 |
+
"epoch": 0.09,
|
| 34 |
"learning_rate": 1.5e-06,
|
| 35 |
+
"loss": 1.0672,
|
| 36 |
"step": 30
|
| 37 |
},
|
| 38 |
{
|
| 39 |
+
"epoch": 0.12,
|
| 40 |
"learning_rate": 2e-06,
|
| 41 |
+
"loss": 1.0423,
|
| 42 |
"step": 40
|
| 43 |
},
|
| 44 |
{
|
| 45 |
+
"epoch": 0.12,
|
| 46 |
+
"eval_accuracy": 0.5033650962883988,
|
| 47 |
+
"eval_loss": 1.0188277959823608,
|
| 48 |
+
"eval_runtime": 49.614,
|
| 49 |
+
"eval_samples_per_second": 604.95,
|
| 50 |
+
"eval_steps_per_second": 1.189,
|
| 51 |
+
"step": 40
|
| 52 |
},
|
| 53 |
{
|
| 54 |
+
"epoch": 0.15,
|
| 55 |
+
"learning_rate": 2.5e-06,
|
| 56 |
+
"loss": 1.027,
|
|
|
|
|
|
|
|
|
|
| 57 |
"step": 50
|
| 58 |
},
|
| 59 |
{
|
| 60 |
+
"epoch": 0.18,
|
| 61 |
"learning_rate": 3e-06,
|
| 62 |
+
"loss": 1.0137,
|
| 63 |
+
"step": 60
|
| 64 |
+
},
|
| 65 |
+
{
|
| 66 |
+
"epoch": 0.18,
|
| 67 |
+
"eval_accuracy": 0.5279203038581995,
|
| 68 |
+
"eval_loss": 0.9871189594268799,
|
| 69 |
+
"eval_runtime": 49.5201,
|
| 70 |
+
"eval_samples_per_second": 606.098,
|
| 71 |
+
"eval_steps_per_second": 1.191,
|
| 72 |
"step": 60
|
| 73 |
},
|
| 74 |
{
|
| 75 |
+
"epoch": 0.21,
|
| 76 |
"learning_rate": 3.5e-06,
|
| 77 |
+
"loss": 1.0005,
|
| 78 |
"step": 70
|
| 79 |
},
|
| 80 |
{
|
| 81 |
+
"epoch": 0.24,
|
| 82 |
"learning_rate": 4e-06,
|
| 83 |
+
"loss": 1.0027,
|
| 84 |
+
"step": 80
|
| 85 |
+
},
|
| 86 |
+
{
|
| 87 |
+
"epoch": 0.24,
|
| 88 |
+
"eval_accuracy": 0.5308189511561271,
|
| 89 |
+
"eval_loss": 0.9888613224029541,
|
| 90 |
+
"eval_runtime": 49.4469,
|
| 91 |
+
"eval_samples_per_second": 606.994,
|
| 92 |
+
"eval_steps_per_second": 1.193,
|
| 93 |
"step": 80
|
| 94 |
},
|
| 95 |
{
|
| 96 |
+
"epoch": 0.27,
|
| 97 |
"learning_rate": 4.5e-06,
|
| 98 |
+
"loss": 0.9937,
|
| 99 |
"step": 90
|
| 100 |
},
|
| 101 |
{
|
| 102 |
+
"epoch": 0.3,
|
| 103 |
"learning_rate": 5e-06,
|
| 104 |
+
"loss": 0.9914,
|
| 105 |
"step": 100
|
| 106 |
},
|
| 107 |
{
|
| 108 |
+
"epoch": 0.3,
|
| 109 |
+
"eval_accuracy": 0.5307856333710935,
|
| 110 |
+
"eval_loss": 0.9762536883354187,
|
| 111 |
+
"eval_runtime": 49.4804,
|
| 112 |
+
"eval_samples_per_second": 606.583,
|
| 113 |
+
"eval_steps_per_second": 1.192,
|
| 114 |
"step": 100
|
| 115 |
},
|
| 116 |
{
|
| 117 |
+
"epoch": 0.33,
|
| 118 |
"learning_rate": 5.5e-06,
|
| 119 |
+
"loss": 0.9884,
|
| 120 |
"step": 110
|
| 121 |
},
|
| 122 |
{
|
| 123 |
+
"epoch": 0.36,
|
| 124 |
"learning_rate": 6e-06,
|
| 125 |
+
"loss": 0.9826,
|
| 126 |
+
"step": 120
|
| 127 |
+
},
|
| 128 |
+
{
|
| 129 |
+
"epoch": 0.36,
|
| 130 |
+
"eval_accuracy": 0.5387819017791697,
|
| 131 |
+
"eval_loss": 0.9713281989097595,
|
| 132 |
+
"eval_runtime": 49.4983,
|
| 133 |
+
"eval_samples_per_second": 606.364,
|
| 134 |
+
"eval_steps_per_second": 1.192,
|
| 135 |
"step": 120
|
| 136 |
},
|
| 137 |
{
|
| 138 |
+
"epoch": 0.39,
|
| 139 |
"learning_rate": 6.5e-06,
|
| 140 |
+
"loss": 0.9882,
|
| 141 |
"step": 130
|
| 142 |
},
|
| 143 |
{
|
| 144 |
+
"epoch": 0.42,
|
| 145 |
"learning_rate": 7e-06,
|
| 146 |
+
"loss": 0.9788,
|
| 147 |
"step": 140
|
| 148 |
},
|
| 149 |
{
|
| 150 |
+
"epoch": 0.42,
|
| 151 |
+
"eval_accuracy": 0.5312520823615646,
|
| 152 |
+
"eval_loss": 0.9766249656677246,
|
| 153 |
+
"eval_runtime": 49.5213,
|
| 154 |
+
"eval_samples_per_second": 606.083,
|
| 155 |
+
"eval_steps_per_second": 1.191,
|
| 156 |
+
"step": 140
|
| 157 |
},
|
| 158 |
{
|
| 159 |
+
"epoch": 0.45,
|
| 160 |
+
"learning_rate": 7.5e-06,
|
| 161 |
+
"loss": 0.9854,
|
|
|
|
|
|
|
|
|
|
| 162 |
"step": 150
|
| 163 |
},
|
| 164 |
{
|
| 165 |
+
"epoch": 0.48,
|
| 166 |
"learning_rate": 8e-06,
|
| 167 |
+
"loss": 0.984,
|
| 168 |
+
"step": 160
|
| 169 |
+
},
|
| 170 |
+
{
|
| 171 |
+
"epoch": 0.48,
|
| 172 |
+
"eval_accuracy": 0.5398147531152129,
|
| 173 |
+
"eval_loss": 0.9589501619338989,
|
| 174 |
+
"eval_runtime": 49.4977,
|
| 175 |
+
"eval_samples_per_second": 606.372,
|
| 176 |
+
"eval_steps_per_second": 1.192,
|
| 177 |
"step": 160
|
| 178 |
},
|
| 179 |
{
|
| 180 |
+
"epoch": 0.51,
|
| 181 |
"learning_rate": 8.500000000000002e-06,
|
| 182 |
+
"loss": 0.9795,
|
| 183 |
"step": 170
|
| 184 |
},
|
| 185 |
{
|
| 186 |
+
"epoch": 0.54,
|
| 187 |
"learning_rate": 9e-06,
|
| 188 |
+
"loss": 0.9694,
|
| 189 |
"step": 180
|
| 190 |
},
|
| 191 |
{
|
| 192 |
+
"epoch": 0.54,
|
| 193 |
+
"eval_accuracy": 0.5423469047777704,
|
| 194 |
+
"eval_loss": 0.953514814376831,
|
| 195 |
+
"eval_runtime": 49.4511,
|
| 196 |
+
"eval_samples_per_second": 606.943,
|
| 197 |
+
"eval_steps_per_second": 1.193,
|
| 198 |
+
"step": 180
|
| 199 |
+
},
|
| 200 |
+
{
|
| 201 |
+
"epoch": 0.57,
|
| 202 |
"learning_rate": 9.5e-06,
|
| 203 |
+
"loss": 0.9757,
|
| 204 |
"step": 190
|
| 205 |
},
|
| 206 |
{
|
| 207 |
+
"epoch": 0.6,
|
| 208 |
"learning_rate": 1e-05,
|
| 209 |
+
"loss": 0.9676,
|
| 210 |
"step": 200
|
| 211 |
},
|
| 212 |
{
|
| 213 |
+
"epoch": 0.6,
|
| 214 |
+
"eval_accuracy": 0.567201972412874,
|
| 215 |
+
"eval_loss": 0.9273685812950134,
|
| 216 |
+
"eval_runtime": 49.4122,
|
| 217 |
+
"eval_samples_per_second": 607.421,
|
| 218 |
+
"eval_steps_per_second": 1.194,
|
| 219 |
"step": 200
|
| 220 |
},
|
| 221 |
{
|
| 222 |
+
"epoch": 0.63,
|
| 223 |
"learning_rate": 1.0500000000000001e-05,
|
| 224 |
+
"loss": 0.9708,
|
| 225 |
"step": 210
|
| 226 |
},
|
| 227 |
{
|
| 228 |
+
"epoch": 0.66,
|
| 229 |
"learning_rate": 1.1e-05,
|
| 230 |
+
"loss": 0.9753,
|
| 231 |
+
"step": 220
|
| 232 |
+
},
|
| 233 |
+
{
|
| 234 |
+
"epoch": 0.66,
|
| 235 |
+
"eval_accuracy": 0.573598987139335,
|
| 236 |
+
"eval_loss": 0.912590503692627,
|
| 237 |
+
"eval_runtime": 49.2827,
|
| 238 |
+
"eval_samples_per_second": 609.017,
|
| 239 |
+
"eval_steps_per_second": 1.197,
|
| 240 |
"step": 220
|
| 241 |
},
|
| 242 |
{
|
| 243 |
+
"epoch": 0.69,
|
| 244 |
"learning_rate": 1.15e-05,
|
| 245 |
+
"loss": 0.9586,
|
| 246 |
"step": 230
|
| 247 |
},
|
| 248 |
{
|
| 249 |
+
"epoch": 0.72,
|
| 250 |
"learning_rate": 1.2e-05,
|
| 251 |
+
"loss": 0.9557,
|
| 252 |
"step": 240
|
| 253 |
},
|
| 254 |
{
|
| 255 |
+
"epoch": 0.72,
|
| 256 |
+
"eval_accuracy": 0.5759645498767242,
|
| 257 |
+
"eval_loss": 0.9052607417106628,
|
| 258 |
+
"eval_runtime": 49.4457,
|
| 259 |
+
"eval_samples_per_second": 607.01,
|
| 260 |
+
"eval_steps_per_second": 1.193,
|
| 261 |
+
"step": 240
|
| 262 |
},
|
| 263 |
{
|
| 264 |
+
"epoch": 0.75,
|
| 265 |
+
"learning_rate": 1.25e-05,
|
| 266 |
+
"loss": 0.9518,
|
|
|
|
|
|
|
|
|
|
| 267 |
"step": 250
|
| 268 |
},
|
| 269 |
{
|
| 270 |
+
"epoch": 0.78,
|
| 271 |
"learning_rate": 1.3e-05,
|
| 272 |
+
"loss": 0.9508,
|
| 273 |
+
"step": 260
|
| 274 |
+
},
|
| 275 |
+
{
|
| 276 |
+
"epoch": 0.78,
|
| 277 |
+
"eval_accuracy": 0.5766975411474645,
|
| 278 |
+
"eval_loss": 0.9178985953330994,
|
| 279 |
+
"eval_runtime": 49.3132,
|
| 280 |
+
"eval_samples_per_second": 608.641,
|
| 281 |
+
"eval_steps_per_second": 1.196,
|
| 282 |
"step": 260
|
| 283 |
},
|
| 284 |
{
|
| 285 |
+
"epoch": 0.81,
|
| 286 |
"learning_rate": 1.35e-05,
|
| 287 |
+
"loss": 0.9405,
|
| 288 |
"step": 270
|
| 289 |
},
|
| 290 |
{
|
| 291 |
+
"epoch": 0.84,
|
| 292 |
"learning_rate": 1.4e-05,
|
| 293 |
+
"loss": 0.9355,
|
| 294 |
+
"step": 280
|
| 295 |
+
},
|
| 296 |
+
{
|
| 297 |
+
"epoch": 0.84,
|
| 298 |
+
"eval_accuracy": 0.58915839275005,
|
| 299 |
+
"eval_loss": 0.8937407732009888,
|
| 300 |
+
"eval_runtime": 49.4342,
|
| 301 |
+
"eval_samples_per_second": 607.15,
|
| 302 |
+
"eval_steps_per_second": 1.194,
|
| 303 |
"step": 280
|
| 304 |
},
|
| 305 |
{
|
| 306 |
+
"epoch": 0.87,
|
| 307 |
"learning_rate": 1.4500000000000002e-05,
|
| 308 |
+
"loss": 0.9235,
|
| 309 |
"step": 290
|
| 310 |
},
|
| 311 |
{
|
| 312 |
+
"epoch": 0.9,
|
| 313 |
"learning_rate": 1.5e-05,
|
| 314 |
+
"loss": 0.9,
|
| 315 |
"step": 300
|
| 316 |
},
|
| 317 |
{
|
| 318 |
+
"epoch": 0.9,
|
| 319 |
+
"eval_accuracy": 0.613013926834144,
|
| 320 |
+
"eval_loss": 0.8468813300132751,
|
| 321 |
+
"eval_runtime": 49.5046,
|
| 322 |
+
"eval_samples_per_second": 606.287,
|
| 323 |
+
"eval_steps_per_second": 1.192,
|
| 324 |
"step": 300
|
| 325 |
},
|
| 326 |
{
|
| 327 |
+
"epoch": 0.93,
|
| 328 |
"learning_rate": 1.55e-05,
|
| 329 |
+
"loss": 0.8857,
|
| 330 |
"step": 310
|
| 331 |
},
|
| 332 |
{
|
| 333 |
+
"epoch": 0.96,
|
| 334 |
"learning_rate": 1.6e-05,
|
| 335 |
+
"loss": 0.993,
|
| 336 |
+
"step": 320
|
| 337 |
+
},
|
| 338 |
+
{
|
| 339 |
+
"epoch": 0.96,
|
| 340 |
+
"eval_accuracy": 0.6046511627906976,
|
| 341 |
+
"eval_loss": 0.8615403771400452,
|
| 342 |
+
"eval_runtime": 49.5041,
|
| 343 |
+
"eval_samples_per_second": 606.294,
|
| 344 |
+
"eval_steps_per_second": 1.192,
|
| 345 |
"step": 320
|
| 346 |
},
|
| 347 |
{
|
| 348 |
+
"epoch": 0.99,
|
| 349 |
"learning_rate": 1.65e-05,
|
| 350 |
+
"loss": 0.8418,
|
| 351 |
"step": 330
|
| 352 |
},
|
| 353 |
{
|
| 354 |
+
"epoch": 1.02,
|
| 355 |
"learning_rate": 1.7000000000000003e-05,
|
| 356 |
+
"loss": 0.8527,
|
| 357 |
"step": 340
|
| 358 |
},
|
| 359 |
{
|
| 360 |
+
"epoch": 1.02,
|
| 361 |
+
"eval_accuracy": 0.6439328313453722,
|
| 362 |
+
"eval_loss": 0.7896137237548828,
|
| 363 |
+
"eval_runtime": 49.4448,
|
| 364 |
+
"eval_samples_per_second": 607.02,
|
| 365 |
+
"eval_steps_per_second": 1.193,
|
| 366 |
+
"step": 340
|
| 367 |
},
|
| 368 |
{
|
| 369 |
+
"epoch": 1.05,
|
| 370 |
+
"learning_rate": 1.7500000000000002e-05,
|
| 371 |
+
"loss": 0.8982,
|
|
|
|
|
|
|
|
|
|
| 372 |
"step": 350
|
| 373 |
},
|
| 374 |
{
|
| 375 |
+
"epoch": 1.08,
|
| 376 |
"learning_rate": 1.8e-05,
|
| 377 |
+
"loss": 0.966,
|
| 378 |
+
"step": 360
|
| 379 |
+
},
|
| 380 |
+
{
|
| 381 |
+
"epoch": 1.08,
|
| 382 |
+
"eval_accuracy": 0.5315852602119011,
|
| 383 |
+
"eval_loss": 1.0123510360717773,
|
| 384 |
+
"eval_runtime": 49.4838,
|
| 385 |
+
"eval_samples_per_second": 606.542,
|
| 386 |
+
"eval_steps_per_second": 1.192,
|
| 387 |
"step": 360
|
| 388 |
},
|
| 389 |
{
|
| 390 |
+
"epoch": 1.11,
|
| 391 |
"learning_rate": 1.85e-05,
|
| 392 |
+
"loss": 0.9413,
|
| 393 |
"step": 370
|
| 394 |
},
|
| 395 |
{
|
| 396 |
+
"epoch": 1.14,
|
| 397 |
"learning_rate": 1.9e-05,
|
| 398 |
+
"loss": 0.8441,
|
| 399 |
+
"step": 380
|
| 400 |
+
},
|
| 401 |
+
{
|
| 402 |
+
"epoch": 1.14,
|
| 403 |
+
"eval_accuracy": 0.6488638635303525,
|
| 404 |
+
"eval_loss": 0.791083574295044,
|
| 405 |
+
"eval_runtime": 49.4701,
|
| 406 |
+
"eval_samples_per_second": 606.71,
|
| 407 |
+
"eval_steps_per_second": 1.193,
|
| 408 |
"step": 380
|
| 409 |
},
|
| 410 |
{
|
| 411 |
+
"epoch": 1.17,
|
| 412 |
"learning_rate": 1.95e-05,
|
| 413 |
+
"loss": 0.8223,
|
| 414 |
"step": 390
|
| 415 |
},
|
| 416 |
{
|
| 417 |
+
"epoch": 1.2,
|
| 418 |
"learning_rate": 2e-05,
|
| 419 |
+
"loss": 0.8226,
|
| 420 |
"step": 400
|
| 421 |
},
|
| 422 |
{
|
| 423 |
+
"epoch": 1.2,
|
| 424 |
+
"eval_accuracy": 0.6699873392416872,
|
| 425 |
+
"eval_loss": 0.7472424507141113,
|
| 426 |
+
"eval_runtime": 49.4938,
|
| 427 |
+
"eval_samples_per_second": 606.419,
|
| 428 |
+
"eval_steps_per_second": 1.192,
|
| 429 |
"step": 400
|
| 430 |
},
|
| 431 |
{
|
| 432 |
+
"epoch": 1.23,
|
| 433 |
"learning_rate": 2.05e-05,
|
| 434 |
+
"loss": 0.7924,
|
| 435 |
"step": 410
|
| 436 |
},
|
| 437 |
{
|
| 438 |
+
"epoch": 1.26,
|
| 439 |
"learning_rate": 2.1000000000000002e-05,
|
| 440 |
+
"loss": 0.7948,
|
| 441 |
+
"step": 420
|
| 442 |
+
},
|
| 443 |
+
{
|
| 444 |
+
"epoch": 1.26,
|
| 445 |
+
"eval_accuracy": 0.6580595721996402,
|
| 446 |
+
"eval_loss": 0.7663838863372803,
|
| 447 |
+
"eval_runtime": 49.5064,
|
| 448 |
+
"eval_samples_per_second": 606.265,
|
| 449 |
+
"eval_steps_per_second": 1.192,
|
| 450 |
"step": 420
|
| 451 |
},
|
| 452 |
{
|
| 453 |
+
"epoch": 1.29,
|
| 454 |
"learning_rate": 2.1499999999999997e-05,
|
| 455 |
+
"loss": 0.776,
|
| 456 |
"step": 430
|
| 457 |
},
|
| 458 |
{
|
| 459 |
+
"epoch": 1.32,
|
| 460 |
"learning_rate": 2.2e-05,
|
| 461 |
+
"loss": 0.7428,
|
| 462 |
"step": 440
|
| 463 |
},
|
| 464 |
{
|
| 465 |
+
"epoch": 1.32,
|
| 466 |
+
"eval_accuracy": 0.6991737189311654,
|
| 467 |
+
"eval_loss": 0.6993714570999146,
|
| 468 |
+
"eval_runtime": 49.4888,
|
| 469 |
+
"eval_samples_per_second": 606.481,
|
| 470 |
+
"eval_steps_per_second": 1.192,
|
| 471 |
+
"step": 440
|
| 472 |
},
|
| 473 |
{
|
| 474 |
+
"epoch": 1.35,
|
| 475 |
+
"learning_rate": 2.2499999999999998e-05,
|
| 476 |
+
"loss": 0.7512,
|
|
|
|
|
|
|
|
|
|
| 477 |
"step": 450
|
| 478 |
},
|
| 479 |
{
|
| 480 |
+
"epoch": 1.38,
|
| 481 |
"learning_rate": 2.3e-05,
|
| 482 |
+
"loss": 0.7109,
|
| 483 |
+
"step": 460
|
| 484 |
+
},
|
| 485 |
+
{
|
| 486 |
+
"epoch": 1.38,
|
| 487 |
+
"eval_accuracy": 0.7283600986206437,
|
| 488 |
+
"eval_loss": 0.6510820984840393,
|
| 489 |
+
"eval_runtime": 49.5999,
|
| 490 |
+
"eval_samples_per_second": 605.122,
|
| 491 |
+
"eval_steps_per_second": 1.19,
|
| 492 |
"step": 460
|
| 493 |
},
|
| 494 |
{
|
| 495 |
+
"epoch": 1.41,
|
| 496 |
"learning_rate": 2.3500000000000002e-05,
|
| 497 |
+
"loss": 0.701,
|
| 498 |
"step": 470
|
| 499 |
},
|
| 500 |
{
|
| 501 |
+
"epoch": 1.44,
|
| 502 |
"learning_rate": 2.4e-05,
|
| 503 |
+
"loss": 0.6882,
|
| 504 |
"step": 480
|
| 505 |
},
|
| 506 |
{
|
| 507 |
+
"epoch": 1.44,
|
| 508 |
+
"eval_accuracy": 0.7576797494502565,
|
| 509 |
+
"eval_loss": 0.5987845063209534,
|
| 510 |
+
"eval_runtime": 49.5123,
|
| 511 |
+
"eval_samples_per_second": 606.192,
|
| 512 |
+
"eval_steps_per_second": 1.192,
|
| 513 |
+
"step": 480
|
| 514 |
+
},
|
| 515 |
+
{
|
| 516 |
+
"epoch": 1.47,
|
| 517 |
"learning_rate": 2.4500000000000003e-05,
|
| 518 |
+
"loss": 0.651,
|
| 519 |
"step": 490
|
| 520 |
},
|
| 521 |
{
|
| 522 |
+
"epoch": 1.5,
|
| 523 |
"learning_rate": 2.5e-05,
|
| 524 |
+
"loss": 0.7296,
|
| 525 |
"step": 500
|
| 526 |
},
|
| 527 |
{
|
| 528 |
+
"epoch": 1.5,
|
| 529 |
+
"eval_accuracy": 0.7564136736189778,
|
| 530 |
+
"eval_loss": 0.5992804765701294,
|
| 531 |
+
"eval_runtime": 49.5641,
|
| 532 |
+
"eval_samples_per_second": 605.56,
|
| 533 |
+
"eval_steps_per_second": 1.19,
|
| 534 |
"step": 500
|
| 535 |
}
|
| 536 |
],
|
| 537 |
"logging_steps": 10,
|
| 538 |
+
"max_steps": 996,
|
| 539 |
"num_train_epochs": 3,
|
| 540 |
+
"save_steps": 100,
|
| 541 |
+
"total_flos": 6.73549277766615e+16,
|
| 542 |
"trial_name": null,
|
| 543 |
"trial_params": null
|
| 544 |
}
|
checkpoint-500/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 sha256:1f1be80ebe52f6e43af0b8aa087e72fad77310d5998b6e0b8f66a6a1d53be7b7
|
| 3 |
size 4536
|