Commit
•
8cc4659
1
Parent(s):
cc9cd91
End of training
Browse files- README.md +3 -1
- all_results.json +10 -10
- eval_results.json +5 -5
- train_results.json +5 -5
- trainer_state.json +255 -549
README.md
CHANGED
|
@@ -3,6 +3,8 @@ license: apache-2.0
|
|
| 3 |
base_model: google/mt5-base
|
| 4 |
tags:
|
| 5 |
- generated_from_trainer
|
|
|
|
|
|
|
| 6 |
model-index:
|
| 7 |
- name: mt5_base-qg-ap-oficial
|
| 8 |
results: []
|
|
@@ -13,7 +15,7 @@ should probably proofread and complete it, then remove this comment. -->
|
|
| 13 |
|
| 14 |
# mt5_base-qg-ap-oficial
|
| 15 |
|
| 16 |
-
This model is a fine-tuned version of [google/mt5-base](https://huggingface.co/google/mt5-base) on
|
| 17 |
It achieves the following results on the evaluation set:
|
| 18 |
- Loss: 1.0951
|
| 19 |
|
|
|
|
| 3 |
base_model: google/mt5-base
|
| 4 |
tags:
|
| 5 |
- generated_from_trainer
|
| 6 |
+
datasets:
|
| 7 |
+
- tiagoblima/preprocessed-du-qg-squadv1_pt
|
| 8 |
model-index:
|
| 9 |
- name: mt5_base-qg-ap-oficial
|
| 10 |
results: []
|
|
|
|
| 15 |
|
| 16 |
# mt5_base-qg-ap-oficial
|
| 17 |
|
| 18 |
+
This model is a fine-tuned version of [google/mt5-base](https://huggingface.co/google/mt5-base) on the tiagoblima/preprocessed-du-qg-squadv1_pt dataset.
|
| 19 |
It achieves the following results on the evaluation set:
|
| 20 |
- Loss: 1.0951
|
| 21 |
|
all_results.json
CHANGED
|
@@ -1,13 +1,13 @@
|
|
| 1 |
{
|
| 2 |
"epoch": 5.0,
|
| 3 |
-
"eval_loss": 1.
|
| 4 |
-
"eval_runtime":
|
| 5 |
-
"eval_samples":
|
| 6 |
-
"eval_samples_per_second":
|
| 7 |
-
"eval_steps_per_second": 3.
|
| 8 |
-
"train_loss": 1.
|
| 9 |
-
"train_runtime":
|
| 10 |
-
"train_samples":
|
| 11 |
-
"train_samples_per_second":
|
| 12 |
-
"train_steps_per_second": 1.
|
| 13 |
}
|
|
|
|
| 1 |
{
|
| 2 |
"epoch": 5.0,
|
| 3 |
+
"eval_loss": 1.095082402229309,
|
| 4 |
+
"eval_runtime": 231.9975,
|
| 5 |
+
"eval_samples": 45107,
|
| 6 |
+
"eval_samples_per_second": 194.429,
|
| 7 |
+
"eval_steps_per_second": 3.039,
|
| 8 |
+
"train_loss": 1.650609011189074,
|
| 9 |
+
"train_runtime": 5663.0471,
|
| 10 |
+
"train_samples": 44334,
|
| 11 |
+
"train_samples_per_second": 39.143,
|
| 12 |
+
"train_steps_per_second": 1.224
|
| 13 |
}
|
eval_results.json
CHANGED
|
@@ -1,8 +1,8 @@
|
|
| 1 |
{
|
| 2 |
"epoch": 5.0,
|
| 3 |
-
"eval_loss": 1.
|
| 4 |
-
"eval_runtime":
|
| 5 |
-
"eval_samples":
|
| 6 |
-
"eval_samples_per_second":
|
| 7 |
-
"eval_steps_per_second": 3.
|
| 8 |
}
|
|
|
|
| 1 |
{
|
| 2 |
"epoch": 5.0,
|
| 3 |
+
"eval_loss": 1.095082402229309,
|
| 4 |
+
"eval_runtime": 231.9975,
|
| 5 |
+
"eval_samples": 45107,
|
| 6 |
+
"eval_samples_per_second": 194.429,
|
| 7 |
+
"eval_steps_per_second": 3.039
|
| 8 |
}
|
train_results.json
CHANGED
|
@@ -1,8 +1,8 @@
|
|
| 1 |
{
|
| 2 |
"epoch": 5.0,
|
| 3 |
-
"train_loss": 1.
|
| 4 |
-
"train_runtime":
|
| 5 |
-
"train_samples":
|
| 6 |
-
"train_samples_per_second":
|
| 7 |
-
"train_steps_per_second": 1.
|
| 8 |
}
|
|
|
|
| 1 |
{
|
| 2 |
"epoch": 5.0,
|
| 3 |
+
"train_loss": 1.650609011189074,
|
| 4 |
+
"train_runtime": 5663.0471,
|
| 5 |
+
"train_samples": 44334,
|
| 6 |
+
"train_samples_per_second": 39.143,
|
| 7 |
+
"train_steps_per_second": 1.224
|
| 8 |
}
|
trainer_state.json
CHANGED
|
@@ -1,776 +1,482 @@
|
|
| 1 |
{
|
| 2 |
-
"best_metric": 1.
|
| 3 |
-
"best_model_checkpoint": "/temp/mt5_base-qg-ap-oficial/checkpoint-
|
| 4 |
"epoch": 5.0,
|
| 5 |
"eval_steps": 500,
|
| 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": 9.
|
| 14 |
-
"loss":
|
| 15 |
"step": 100
|
| 16 |
},
|
| 17 |
{
|
| 18 |
-
"epoch": 0.
|
| 19 |
-
"learning_rate": 9.
|
| 20 |
-
"loss": 2.
|
| 21 |
"step": 200
|
| 22 |
},
|
| 23 |
{
|
| 24 |
-
"epoch": 0.
|
| 25 |
-
"learning_rate": 9.
|
| 26 |
-
"loss": 2.
|
| 27 |
"step": 300
|
| 28 |
},
|
| 29 |
{
|
| 30 |
-
"epoch": 0.
|
| 31 |
-
"learning_rate": 9.
|
| 32 |
-
"loss": 2.
|
| 33 |
"step": 400
|
| 34 |
},
|
| 35 |
{
|
| 36 |
-
"epoch": 0.
|
| 37 |
-
"learning_rate": 9.
|
| 38 |
-
"loss": 1.
|
| 39 |
"step": 500
|
| 40 |
},
|
| 41 |
{
|
| 42 |
-
"epoch": 0.
|
| 43 |
-
"learning_rate": 9.
|
| 44 |
-
"loss": 1.
|
| 45 |
"step": 600
|
| 46 |
},
|
| 47 |
{
|
| 48 |
-
"epoch": 0.
|
| 49 |
-
"learning_rate":
|
| 50 |
-
"loss": 1.
|
| 51 |
"step": 700
|
| 52 |
},
|
| 53 |
{
|
| 54 |
-
"epoch": 0.
|
| 55 |
-
"learning_rate":
|
| 56 |
-
"loss": 1.
|
| 57 |
"step": 800
|
| 58 |
},
|
| 59 |
{
|
| 60 |
-
"epoch": 0.
|
| 61 |
-
"learning_rate":
|
| 62 |
-
"loss": 1.
|
| 63 |
"step": 900
|
| 64 |
},
|
| 65 |
{
|
| 66 |
-
"epoch": 0.
|
| 67 |
-
"learning_rate":
|
| 68 |
-
"loss": 1.
|
| 69 |
"step": 1000
|
| 70 |
},
|
| 71 |
{
|
| 72 |
-
"epoch": 0.
|
| 73 |
-
"learning_rate":
|
| 74 |
-
"loss": 1.
|
| 75 |
"step": 1100
|
| 76 |
},
|
| 77 |
{
|
| 78 |
-
"epoch": 0.
|
| 79 |
-
"learning_rate": 8.
|
| 80 |
-
"loss": 1.
|
| 81 |
"step": 1200
|
| 82 |
},
|
| 83 |
{
|
| 84 |
-
"epoch": 0.
|
| 85 |
-
"learning_rate": 8.
|
| 86 |
-
"loss": 1.
|
| 87 |
"step": 1300
|
| 88 |
},
|
| 89 |
{
|
| 90 |
-
"epoch": 0
|
| 91 |
-
"
|
| 92 |
-
"
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 93 |
"step": 1400
|
| 94 |
},
|
| 95 |
{
|
| 96 |
-
"epoch":
|
| 97 |
-
"learning_rate":
|
| 98 |
-
"loss": 1.
|
| 99 |
"step": 1500
|
| 100 |
},
|
| 101 |
{
|
| 102 |
-
"epoch":
|
| 103 |
-
"learning_rate":
|
| 104 |
-
"loss": 1.
|
| 105 |
"step": 1600
|
| 106 |
},
|
| 107 |
{
|
| 108 |
-
"epoch":
|
| 109 |
-
"learning_rate":
|
| 110 |
-
"loss": 1.
|
| 111 |
"step": 1700
|
| 112 |
},
|
| 113 |
{
|
| 114 |
-
"epoch":
|
| 115 |
-
"learning_rate":
|
| 116 |
-
"loss": 1.
|
| 117 |
"step": 1800
|
| 118 |
},
|
| 119 |
{
|
| 120 |
-
"epoch":
|
| 121 |
-
"learning_rate":
|
| 122 |
-
"loss": 1.
|
| 123 |
"step": 1900
|
| 124 |
},
|
| 125 |
{
|
| 126 |
-
"epoch":
|
| 127 |
-
"learning_rate":
|
| 128 |
-
"loss": 1.
|
| 129 |
"step": 2000
|
| 130 |
},
|
| 131 |
{
|
| 132 |
-
"epoch":
|
| 133 |
-
"learning_rate":
|
| 134 |
-
"loss": 1.
|
| 135 |
"step": 2100
|
| 136 |
},
|
| 137 |
{
|
| 138 |
-
"epoch":
|
| 139 |
-
"learning_rate":
|
| 140 |
-
"loss": 1.
|
| 141 |
"step": 2200
|
| 142 |
},
|
| 143 |
{
|
| 144 |
-
"epoch":
|
| 145 |
-
"learning_rate":
|
| 146 |
-
"loss": 1.
|
| 147 |
"step": 2300
|
| 148 |
},
|
| 149 |
{
|
| 150 |
-
"epoch": 1.
|
| 151 |
-
"
|
| 152 |
-
"
|
| 153 |
-
"eval_samples_per_second": 196.538,
|
| 154 |
-
"eval_steps_per_second": 3.086,
|
| 155 |
-
"step": 2367
|
| 156 |
-
},
|
| 157 |
-
{
|
| 158 |
-
"epoch": 1.01,
|
| 159 |
-
"learning_rate": 7.972116603295311e-05,
|
| 160 |
-
"loss": 1.628,
|
| 161 |
"step": 2400
|
| 162 |
},
|
| 163 |
{
|
| 164 |
-
"epoch": 1.
|
| 165 |
-
"learning_rate":
|
| 166 |
-
"loss": 1.
|
| 167 |
"step": 2500
|
| 168 |
},
|
| 169 |
{
|
| 170 |
-
"epoch": 1.
|
| 171 |
-
"learning_rate":
|
| 172 |
-
"loss": 1.
|
| 173 |
"step": 2600
|
| 174 |
},
|
| 175 |
{
|
| 176 |
-
"epoch": 1.
|
| 177 |
-
"learning_rate":
|
| 178 |
-
"loss": 1.
|
| 179 |
"step": 2700
|
| 180 |
},
|
| 181 |
{
|
| 182 |
-
"epoch":
|
| 183 |
-
"
|
| 184 |
-
"
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 185 |
"step": 2800
|
| 186 |
},
|
| 187 |
{
|
| 188 |
-
"epoch":
|
| 189 |
-
"learning_rate":
|
| 190 |
-
"loss": 1.
|
| 191 |
"step": 2900
|
| 192 |
},
|
| 193 |
{
|
| 194 |
-
"epoch":
|
| 195 |
-
"learning_rate":
|
| 196 |
-
"loss": 1.
|
| 197 |
"step": 3000
|
| 198 |
},
|
| 199 |
{
|
| 200 |
-
"epoch":
|
| 201 |
-
"learning_rate":
|
| 202 |
-
"loss": 1.
|
| 203 |
"step": 3100
|
| 204 |
},
|
| 205 |
{
|
| 206 |
-
"epoch":
|
| 207 |
-
"learning_rate":
|
| 208 |
-
"loss": 1.
|
| 209 |
"step": 3200
|
| 210 |
},
|
| 211 |
{
|
| 212 |
-
"epoch":
|
| 213 |
-
"learning_rate":
|
| 214 |
-
"loss": 1.
|
| 215 |
"step": 3300
|
| 216 |
},
|
| 217 |
{
|
| 218 |
-
"epoch":
|
| 219 |
-
"learning_rate":
|
| 220 |
-
"loss": 1.
|
| 221 |
"step": 3400
|
| 222 |
},
|
| 223 |
{
|
| 224 |
-
"epoch":
|
| 225 |
-
"learning_rate":
|
| 226 |
-
"loss": 1.
|
| 227 |
"step": 3500
|
| 228 |
},
|
| 229 |
{
|
| 230 |
-
"epoch":
|
| 231 |
-
"learning_rate":
|
| 232 |
-
"loss": 1.
|
| 233 |
"step": 3600
|
| 234 |
},
|
| 235 |
{
|
| 236 |
-
"epoch":
|
| 237 |
-
"learning_rate":
|
| 238 |
-
"loss": 1.
|
| 239 |
"step": 3700
|
| 240 |
},
|
| 241 |
{
|
| 242 |
-
"epoch":
|
| 243 |
-
"learning_rate":
|
| 244 |
-
"loss": 1.
|
| 245 |
"step": 3800
|
| 246 |
},
|
| 247 |
{
|
| 248 |
-
"epoch":
|
| 249 |
-
"learning_rate":
|
| 250 |
-
"loss": 1.
|
| 251 |
"step": 3900
|
| 252 |
},
|
| 253 |
{
|
| 254 |
-
"epoch":
|
| 255 |
-
"learning_rate":
|
| 256 |
-
"loss": 1.
|
| 257 |
"step": 4000
|
| 258 |
},
|
| 259 |
{
|
| 260 |
-
"epoch":
|
| 261 |
-
"learning_rate":
|
| 262 |
-
"loss": 1.
|
| 263 |
"step": 4100
|
| 264 |
},
|
| 265 |
{
|
| 266 |
-
"epoch":
|
| 267 |
-
"
|
| 268 |
-
"
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 269 |
"step": 4200
|
| 270 |
},
|
| 271 |
{
|
| 272 |
-
"epoch": 1
|
| 273 |
-
"learning_rate":
|
| 274 |
-
"loss": 1.
|
| 275 |
"step": 4300
|
| 276 |
},
|
| 277 |
{
|
| 278 |
-
"epoch":
|
| 279 |
-
"learning_rate":
|
| 280 |
-
"loss": 1.
|
| 281 |
"step": 4400
|
| 282 |
},
|
| 283 |
{
|
| 284 |
-
"epoch":
|
| 285 |
-
"learning_rate":
|
| 286 |
-
"loss": 1.
|
| 287 |
"step": 4500
|
| 288 |
},
|
| 289 |
{
|
| 290 |
-
"epoch":
|
| 291 |
-
"learning_rate":
|
| 292 |
-
"loss": 1.
|
| 293 |
"step": 4600
|
| 294 |
},
|
| 295 |
{
|
| 296 |
-
"epoch":
|
| 297 |
-
"learning_rate":
|
| 298 |
-
"loss": 1.
|
| 299 |
"step": 4700
|
| 300 |
},
|
| 301 |
{
|
| 302 |
-
"epoch":
|
| 303 |
-
"
|
| 304 |
-
"
|
| 305 |
-
"eval_samples_per_second": 196.472,
|
| 306 |
-
"eval_steps_per_second": 3.085,
|
| 307 |
-
"step": 4734
|
| 308 |
-
},
|
| 309 |
-
{
|
| 310 |
-
"epoch": 2.03,
|
| 311 |
-
"learning_rate": 5.944233206590622e-05,
|
| 312 |
-
"loss": 1.4431,
|
| 313 |
"step": 4800
|
| 314 |
},
|
| 315 |
{
|
| 316 |
-
"epoch":
|
| 317 |
-
"learning_rate":
|
| 318 |
-
"loss": 1.
|
| 319 |
"step": 4900
|
| 320 |
},
|
| 321 |
{
|
| 322 |
-
"epoch":
|
| 323 |
-
"learning_rate":
|
| 324 |
-
"loss": 1.
|
| 325 |
"step": 5000
|
| 326 |
},
|
| 327 |
{
|
| 328 |
-
"epoch":
|
| 329 |
-
"learning_rate":
|
| 330 |
-
"loss": 1.
|
| 331 |
"step": 5100
|
| 332 |
},
|
| 333 |
{
|
| 334 |
-
"epoch":
|
| 335 |
-
"learning_rate":
|
| 336 |
-
"loss": 1.
|
| 337 |
"step": 5200
|
| 338 |
},
|
| 339 |
{
|
| 340 |
-
"epoch":
|
| 341 |
-
"learning_rate":
|
| 342 |
-
"loss": 1.
|
| 343 |
"step": 5300
|
| 344 |
},
|
| 345 |
{
|
| 346 |
-
"epoch":
|
| 347 |
-
"learning_rate":
|
| 348 |
-
"loss": 1.
|
| 349 |
"step": 5400
|
| 350 |
},
|
| 351 |
{
|
| 352 |
-
"epoch":
|
| 353 |
-
"learning_rate":
|
| 354 |
-
"loss": 1.
|
| 355 |
"step": 5500
|
| 356 |
},
|
| 357 |
{
|
| 358 |
-
"epoch":
|
| 359 |
-
"
|
| 360 |
-
"
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 361 |
"step": 5600
|
| 362 |
},
|
| 363 |
{
|
| 364 |
-
"epoch":
|
| 365 |
-
"learning_rate":
|
| 366 |
-
"loss": 1.
|
| 367 |
"step": 5700
|
| 368 |
},
|
| 369 |
{
|
| 370 |
-
"epoch":
|
| 371 |
-
"learning_rate":
|
| 372 |
-
"loss": 1.
|
| 373 |
"step": 5800
|
| 374 |
},
|
| 375 |
{
|
| 376 |
-
"epoch":
|
| 377 |
-
"learning_rate":
|
| 378 |
-
"loss": 1.
|
| 379 |
"step": 5900
|
| 380 |
},
|
| 381 |
{
|
| 382 |
-
"epoch":
|
| 383 |
-
"learning_rate":
|
| 384 |
-
"loss": 1.
|
| 385 |
"step": 6000
|
| 386 |
},
|
| 387 |
{
|
| 388 |
-
"epoch":
|
| 389 |
-
"learning_rate":
|
| 390 |
-
"loss": 1.
|
| 391 |
"step": 6100
|
| 392 |
},
|
| 393 |
{
|
| 394 |
-
"epoch":
|
| 395 |
-
"learning_rate":
|
| 396 |
-
"loss": 1.
|
| 397 |
"step": 6200
|
| 398 |
},
|
| 399 |
{
|
| 400 |
-
"epoch":
|
| 401 |
-
"learning_rate":
|
| 402 |
-
"loss": 1.
|
| 403 |
"step": 6300
|
| 404 |
},
|
| 405 |
{
|
| 406 |
-
"epoch":
|
| 407 |
-
"learning_rate":
|
| 408 |
-
"loss": 1.
|
| 409 |
"step": 6400
|
| 410 |
},
|
| 411 |
{
|
| 412 |
-
"epoch":
|
| 413 |
-
"learning_rate":
|
| 414 |
-
"loss": 1.
|
| 415 |
"step": 6500
|
| 416 |
},
|
| 417 |
{
|
| 418 |
-
"epoch":
|
| 419 |
-
"learning_rate": 4.
|
| 420 |
-
"loss": 1.
|
| 421 |
"step": 6600
|
| 422 |
},
|
| 423 |
{
|
| 424 |
-
"epoch":
|
| 425 |
-
"learning_rate":
|
| 426 |
-
"loss": 1.
|
| 427 |
"step": 6700
|
| 428 |
},
|
| 429 |
{
|
| 430 |
-
"epoch":
|
| 431 |
-
"learning_rate":
|
| 432 |
-
"loss": 1.
|
| 433 |
"step": 6800
|
| 434 |
},
|
| 435 |
{
|
| 436 |
-
"epoch":
|
| 437 |
-
"learning_rate": 4.
|
| 438 |
-
"loss": 1.
|
| 439 |
"step": 6900
|
| 440 |
},
|
| 441 |
-
{
|
| 442 |
-
"epoch": 2.96,
|
| 443 |
-
"learning_rate": 4.085340092944656e-05,
|
| 444 |
-
"loss": 1.4325,
|
| 445 |
-
"step": 7000
|
| 446 |
-
},
|
| 447 |
-
{
|
| 448 |
-
"epoch": 3.0,
|
| 449 |
-
"learning_rate": 4.000844951415294e-05,
|
| 450 |
-
"loss": 1.3982,
|
| 451 |
-
"step": 7100
|
| 452 |
-
},
|
| 453 |
-
{
|
| 454 |
-
"epoch": 3.0,
|
| 455 |
-
"eval_loss": 1.3959996700286865,
|
| 456 |
-
"eval_runtime": 55.4627,
|
| 457 |
-
"eval_samples_per_second": 196.366,
|
| 458 |
-
"eval_steps_per_second": 3.083,
|
| 459 |
-
"step": 7101
|
| 460 |
-
},
|
| 461 |
-
{
|
| 462 |
-
"epoch": 3.04,
|
| 463 |
-
"learning_rate": 3.916349809885932e-05,
|
| 464 |
-
"loss": 1.3626,
|
| 465 |
-
"step": 7200
|
| 466 |
-
},
|
| 467 |
-
{
|
| 468 |
-
"epoch": 3.08,
|
| 469 |
-
"learning_rate": 3.83185466835657e-05,
|
| 470 |
-
"loss": 1.3606,
|
| 471 |
-
"step": 7300
|
| 472 |
-
},
|
| 473 |
-
{
|
| 474 |
-
"epoch": 3.13,
|
| 475 |
-
"learning_rate": 3.747359526827208e-05,
|
| 476 |
-
"loss": 1.341,
|
| 477 |
-
"step": 7400
|
| 478 |
-
},
|
| 479 |
-
{
|
| 480 |
-
"epoch": 3.17,
|
| 481 |
-
"learning_rate": 3.6628643852978456e-05,
|
| 482 |
-
"loss": 1.368,
|
| 483 |
-
"step": 7500
|
| 484 |
-
},
|
| 485 |
-
{
|
| 486 |
-
"epoch": 3.21,
|
| 487 |
-
"learning_rate": 3.5783692437684833e-05,
|
| 488 |
-
"loss": 1.3629,
|
| 489 |
-
"step": 7600
|
| 490 |
-
},
|
| 491 |
-
{
|
| 492 |
-
"epoch": 3.25,
|
| 493 |
-
"learning_rate": 3.493874102239121e-05,
|
| 494 |
-
"loss": 1.3424,
|
| 495 |
-
"step": 7700
|
| 496 |
-
},
|
| 497 |
-
{
|
| 498 |
-
"epoch": 3.3,
|
| 499 |
-
"learning_rate": 3.4093789607097596e-05,
|
| 500 |
-
"loss": 1.3423,
|
| 501 |
-
"step": 7800
|
| 502 |
-
},
|
| 503 |
-
{
|
| 504 |
-
"epoch": 3.34,
|
| 505 |
-
"learning_rate": 3.324883819180397e-05,
|
| 506 |
-
"loss": 1.3487,
|
| 507 |
-
"step": 7900
|
| 508 |
-
},
|
| 509 |
-
{
|
| 510 |
-
"epoch": 3.38,
|
| 511 |
-
"learning_rate": 3.240388677651035e-05,
|
| 512 |
-
"loss": 1.3473,
|
| 513 |
-
"step": 8000
|
| 514 |
-
},
|
| 515 |
-
{
|
| 516 |
-
"epoch": 3.42,
|
| 517 |
-
"learning_rate": 3.155893536121673e-05,
|
| 518 |
-
"loss": 1.3626,
|
| 519 |
-
"step": 8100
|
| 520 |
-
},
|
| 521 |
-
{
|
| 522 |
-
"epoch": 3.46,
|
| 523 |
-
"learning_rate": 3.0713983945923107e-05,
|
| 524 |
-
"loss": 1.3643,
|
| 525 |
-
"step": 8200
|
| 526 |
-
},
|
| 527 |
-
{
|
| 528 |
-
"epoch": 3.51,
|
| 529 |
-
"learning_rate": 2.986903253062949e-05,
|
| 530 |
-
"loss": 1.3493,
|
| 531 |
-
"step": 8300
|
| 532 |
-
},
|
| 533 |
-
{
|
| 534 |
-
"epoch": 3.55,
|
| 535 |
-
"learning_rate": 2.9024081115335872e-05,
|
| 536 |
-
"loss": 1.3346,
|
| 537 |
-
"step": 8400
|
| 538 |
-
},
|
| 539 |
-
{
|
| 540 |
-
"epoch": 3.59,
|
| 541 |
-
"learning_rate": 2.817912970004225e-05,
|
| 542 |
-
"loss": 1.3516,
|
| 543 |
-
"step": 8500
|
| 544 |
-
},
|
| 545 |
-
{
|
| 546 |
-
"epoch": 3.63,
|
| 547 |
-
"learning_rate": 2.733417828474863e-05,
|
| 548 |
-
"loss": 1.3428,
|
| 549 |
-
"step": 8600
|
| 550 |
-
},
|
| 551 |
-
{
|
| 552 |
-
"epoch": 3.68,
|
| 553 |
-
"learning_rate": 2.648922686945501e-05,
|
| 554 |
-
"loss": 1.3295,
|
| 555 |
-
"step": 8700
|
| 556 |
-
},
|
| 557 |
-
{
|
| 558 |
-
"epoch": 3.72,
|
| 559 |
-
"learning_rate": 2.5644275454161386e-05,
|
| 560 |
-
"loss": 1.3409,
|
| 561 |
-
"step": 8800
|
| 562 |
-
},
|
| 563 |
-
{
|
| 564 |
-
"epoch": 3.76,
|
| 565 |
-
"learning_rate": 2.4799324038867768e-05,
|
| 566 |
-
"loss": 1.3548,
|
| 567 |
-
"step": 8900
|
| 568 |
-
},
|
| 569 |
-
{
|
| 570 |
-
"epoch": 3.8,
|
| 571 |
-
"learning_rate": 2.3954372623574145e-05,
|
| 572 |
-
"loss": 1.3524,
|
| 573 |
-
"step": 9000
|
| 574 |
-
},
|
| 575 |
-
{
|
| 576 |
-
"epoch": 3.84,
|
| 577 |
-
"learning_rate": 2.3109421208280523e-05,
|
| 578 |
-
"loss": 1.3256,
|
| 579 |
-
"step": 9100
|
| 580 |
-
},
|
| 581 |
-
{
|
| 582 |
-
"epoch": 3.89,
|
| 583 |
-
"learning_rate": 2.2264469792986904e-05,
|
| 584 |
-
"loss": 1.3438,
|
| 585 |
-
"step": 9200
|
| 586 |
-
},
|
| 587 |
-
{
|
| 588 |
-
"epoch": 3.93,
|
| 589 |
-
"learning_rate": 2.1419518377693285e-05,
|
| 590 |
-
"loss": 1.3524,
|
| 591 |
-
"step": 9300
|
| 592 |
-
},
|
| 593 |
-
{
|
| 594 |
-
"epoch": 3.97,
|
| 595 |
-
"learning_rate": 2.0574566962399663e-05,
|
| 596 |
-
"loss": 1.3304,
|
| 597 |
-
"step": 9400
|
| 598 |
-
},
|
| 599 |
-
{
|
| 600 |
-
"epoch": 4.0,
|
| 601 |
-
"eval_loss": 1.385326623916626,
|
| 602 |
-
"eval_runtime": 55.4341,
|
| 603 |
-
"eval_samples_per_second": 196.468,
|
| 604 |
-
"eval_steps_per_second": 3.085,
|
| 605 |
-
"step": 9468
|
| 606 |
-
},
|
| 607 |
-
{
|
| 608 |
-
"epoch": 4.01,
|
| 609 |
-
"learning_rate": 1.9729615547106044e-05,
|
| 610 |
-
"loss": 1.3273,
|
| 611 |
-
"step": 9500
|
| 612 |
-
},
|
| 613 |
-
{
|
| 614 |
-
"epoch": 4.06,
|
| 615 |
-
"learning_rate": 1.8884664131812422e-05,
|
| 616 |
-
"loss": 1.2927,
|
| 617 |
-
"step": 9600
|
| 618 |
-
},
|
| 619 |
-
{
|
| 620 |
-
"epoch": 4.1,
|
| 621 |
-
"learning_rate": 1.80397127165188e-05,
|
| 622 |
-
"loss": 1.3069,
|
| 623 |
-
"step": 9700
|
| 624 |
-
},
|
| 625 |
-
{
|
| 626 |
-
"epoch": 4.14,
|
| 627 |
-
"learning_rate": 1.719476130122518e-05,
|
| 628 |
-
"loss": 1.3173,
|
| 629 |
-
"step": 9800
|
| 630 |
-
},
|
| 631 |
-
{
|
| 632 |
-
"epoch": 4.18,
|
| 633 |
-
"learning_rate": 1.634980988593156e-05,
|
| 634 |
-
"loss": 1.3201,
|
| 635 |
-
"step": 9900
|
| 636 |
-
},
|
| 637 |
-
{
|
| 638 |
-
"epoch": 4.22,
|
| 639 |
-
"learning_rate": 1.550485847063794e-05,
|
| 640 |
-
"loss": 1.3126,
|
| 641 |
-
"step": 10000
|
| 642 |
-
},
|
| 643 |
-
{
|
| 644 |
-
"epoch": 4.27,
|
| 645 |
-
"learning_rate": 1.4659907055344319e-05,
|
| 646 |
-
"loss": 1.3047,
|
| 647 |
-
"step": 10100
|
| 648 |
-
},
|
| 649 |
-
{
|
| 650 |
-
"epoch": 4.31,
|
| 651 |
-
"learning_rate": 1.3814955640050698e-05,
|
| 652 |
-
"loss": 1.3137,
|
| 653 |
-
"step": 10200
|
| 654 |
-
},
|
| 655 |
-
{
|
| 656 |
-
"epoch": 4.35,
|
| 657 |
-
"learning_rate": 1.2970004224757076e-05,
|
| 658 |
-
"loss": 1.3029,
|
| 659 |
-
"step": 10300
|
| 660 |
-
},
|
| 661 |
-
{
|
| 662 |
-
"epoch": 4.39,
|
| 663 |
-
"learning_rate": 1.2125052809463457e-05,
|
| 664 |
-
"loss": 1.3095,
|
| 665 |
-
"step": 10400
|
| 666 |
-
},
|
| 667 |
-
{
|
| 668 |
-
"epoch": 4.44,
|
| 669 |
-
"learning_rate": 1.1280101394169835e-05,
|
| 670 |
-
"loss": 1.3075,
|
| 671 |
-
"step": 10500
|
| 672 |
-
},
|
| 673 |
-
{
|
| 674 |
-
"epoch": 4.48,
|
| 675 |
-
"learning_rate": 1.0435149978876214e-05,
|
| 676 |
-
"loss": 1.3002,
|
| 677 |
-
"step": 10600
|
| 678 |
-
},
|
| 679 |
-
{
|
| 680 |
-
"epoch": 4.52,
|
| 681 |
-
"learning_rate": 9.590198563582595e-06,
|
| 682 |
-
"loss": 1.3297,
|
| 683 |
-
"step": 10700
|
| 684 |
-
},
|
| 685 |
-
{
|
| 686 |
-
"epoch": 4.56,
|
| 687 |
-
"learning_rate": 8.745247148288973e-06,
|
| 688 |
-
"loss": 1.3083,
|
| 689 |
-
"step": 10800
|
| 690 |
-
},
|
| 691 |
-
{
|
| 692 |
-
"epoch": 4.6,
|
| 693 |
-
"learning_rate": 7.900295732995352e-06,
|
| 694 |
-
"loss": 1.3077,
|
| 695 |
-
"step": 10900
|
| 696 |
-
},
|
| 697 |
-
{
|
| 698 |
-
"epoch": 4.65,
|
| 699 |
-
"learning_rate": 7.055344317701733e-06,
|
| 700 |
-
"loss": 1.3037,
|
| 701 |
-
"step": 11000
|
| 702 |
-
},
|
| 703 |
-
{
|
| 704 |
-
"epoch": 4.69,
|
| 705 |
-
"learning_rate": 6.210392902408112e-06,
|
| 706 |
-
"loss": 1.3199,
|
| 707 |
-
"step": 11100
|
| 708 |
-
},
|
| 709 |
-
{
|
| 710 |
-
"epoch": 4.73,
|
| 711 |
-
"learning_rate": 5.3654414871144915e-06,
|
| 712 |
-
"loss": 1.2947,
|
| 713 |
-
"step": 11200
|
| 714 |
-
},
|
| 715 |
-
{
|
| 716 |
-
"epoch": 4.77,
|
| 717 |
-
"learning_rate": 4.52049007182087e-06,
|
| 718 |
-
"loss": 1.3035,
|
| 719 |
-
"step": 11300
|
| 720 |
-
},
|
| 721 |
-
{
|
| 722 |
-
"epoch": 4.82,
|
| 723 |
-
"learning_rate": 3.67553865652725e-06,
|
| 724 |
-
"loss": 1.3147,
|
| 725 |
-
"step": 11400
|
| 726 |
-
},
|
| 727 |
-
{
|
| 728 |
-
"epoch": 4.86,
|
| 729 |
-
"learning_rate": 2.830587241233629e-06,
|
| 730 |
-
"loss": 1.304,
|
| 731 |
-
"step": 11500
|
| 732 |
-
},
|
| 733 |
-
{
|
| 734 |
-
"epoch": 4.9,
|
| 735 |
-
"learning_rate": 1.9856358259400087e-06,
|
| 736 |
-
"loss": 1.307,
|
| 737 |
-
"step": 11600
|
| 738 |
-
},
|
| 739 |
-
{
|
| 740 |
-
"epoch": 4.94,
|
| 741 |
-
"learning_rate": 1.140684410646388e-06,
|
| 742 |
-
"loss": 1.2814,
|
| 743 |
-
"step": 11700
|
| 744 |
-
},
|
| 745 |
-
{
|
| 746 |
-
"epoch": 4.99,
|
| 747 |
-
"learning_rate": 2.957329953527672e-07,
|
| 748 |
-
"loss": 1.2909,
|
| 749 |
-
"step": 11800
|
| 750 |
-
},
|
| 751 |
{
|
| 752 |
"epoch": 5.0,
|
| 753 |
-
"eval_loss": 1.
|
| 754 |
-
"eval_runtime":
|
| 755 |
-
"eval_samples_per_second":
|
| 756 |
-
"eval_steps_per_second": 3.
|
| 757 |
-
"step":
|
| 758 |
},
|
| 759 |
{
|
| 760 |
"epoch": 5.0,
|
| 761 |
-
"step":
|
| 762 |
-
"total_flos":
|
| 763 |
-
"train_loss": 1.
|
| 764 |
-
"train_runtime":
|
| 765 |
-
"train_samples_per_second":
|
| 766 |
-
"train_steps_per_second": 1.
|
| 767 |
}
|
| 768 |
],
|
| 769 |
"logging_steps": 100,
|
| 770 |
-
"max_steps":
|
| 771 |
"num_train_epochs": 5,
|
| 772 |
"save_steps": 500,
|
| 773 |
-
"total_flos":
|
| 774 |
"trial_name": null,
|
| 775 |
"trial_params": null
|
| 776 |
}
|
|
|
|
| 1 |
{
|
| 2 |
+
"best_metric": 1.095082402229309,
|
| 3 |
+
"best_model_checkpoint": "/temp/mt5_base-qg-ap-oficial/checkpoint-6930",
|
| 4 |
"epoch": 5.0,
|
| 5 |
"eval_steps": 500,
|
| 6 |
+
"global_step": 6930,
|
| 7 |
"is_hyper_param_search": false,
|
| 8 |
"is_local_process_zero": true,
|
| 9 |
"is_world_process_zero": true,
|
| 10 |
"log_history": [
|
| 11 |
{
|
| 12 |
+
"epoch": 0.07,
|
| 13 |
+
"learning_rate": 9.855699855699856e-05,
|
| 14 |
+
"loss": 9.102,
|
| 15 |
"step": 100
|
| 16 |
},
|
| 17 |
{
|
| 18 |
+
"epoch": 0.14,
|
| 19 |
+
"learning_rate": 9.711399711399713e-05,
|
| 20 |
+
"loss": 2.6848,
|
| 21 |
"step": 200
|
| 22 |
},
|
| 23 |
{
|
| 24 |
+
"epoch": 0.22,
|
| 25 |
+
"learning_rate": 9.567099567099568e-05,
|
| 26 |
+
"loss": 2.1899,
|
| 27 |
"step": 300
|
| 28 |
},
|
| 29 |
{
|
| 30 |
+
"epoch": 0.29,
|
| 31 |
+
"learning_rate": 9.422799422799424e-05,
|
| 32 |
+
"loss": 2.0546,
|
| 33 |
"step": 400
|
| 34 |
},
|
| 35 |
{
|
| 36 |
+
"epoch": 0.36,
|
| 37 |
+
"learning_rate": 9.278499278499279e-05,
|
| 38 |
+
"loss": 1.9973,
|
| 39 |
"step": 500
|
| 40 |
},
|
| 41 |
{
|
| 42 |
+
"epoch": 0.43,
|
| 43 |
+
"learning_rate": 9.134199134199136e-05,
|
| 44 |
+
"loss": 1.953,
|
| 45 |
"step": 600
|
| 46 |
},
|
| 47 |
{
|
| 48 |
+
"epoch": 0.51,
|
| 49 |
+
"learning_rate": 8.98989898989899e-05,
|
| 50 |
+
"loss": 1.9029,
|
| 51 |
"step": 700
|
| 52 |
},
|
| 53 |
{
|
| 54 |
+
"epoch": 0.58,
|
| 55 |
+
"learning_rate": 8.845598845598845e-05,
|
| 56 |
+
"loss": 1.8914,
|
| 57 |
"step": 800
|
| 58 |
},
|
| 59 |
{
|
| 60 |
+
"epoch": 0.65,
|
| 61 |
+
"learning_rate": 8.701298701298701e-05,
|
| 62 |
+
"loss": 1.8197,
|
| 63 |
"step": 900
|
| 64 |
},
|
| 65 |
{
|
| 66 |
+
"epoch": 0.72,
|
| 67 |
+
"learning_rate": 8.556998556998557e-05,
|
| 68 |
+
"loss": 1.7947,
|
| 69 |
"step": 1000
|
| 70 |
},
|
| 71 |
{
|
| 72 |
+
"epoch": 0.79,
|
| 73 |
+
"learning_rate": 8.412698412698413e-05,
|
| 74 |
+
"loss": 1.8097,
|
| 75 |
"step": 1100
|
| 76 |
},
|
| 77 |
{
|
| 78 |
+
"epoch": 0.87,
|
| 79 |
+
"learning_rate": 8.268398268398268e-05,
|
| 80 |
+
"loss": 1.7513,
|
| 81 |
"step": 1200
|
| 82 |
},
|
| 83 |
{
|
| 84 |
+
"epoch": 0.94,
|
| 85 |
+
"learning_rate": 8.124098124098124e-05,
|
| 86 |
+
"loss": 1.7276,
|
| 87 |
"step": 1300
|
| 88 |
},
|
| 89 |
{
|
| 90 |
+
"epoch": 1.0,
|
| 91 |
+
"eval_loss": 1.348934531211853,
|
| 92 |
+
"eval_runtime": 231.7834,
|
| 93 |
+
"eval_samples_per_second": 194.608,
|
| 94 |
+
"eval_steps_per_second": 3.042,
|
| 95 |
+
"step": 1386
|
| 96 |
+
},
|
| 97 |
+
{
|
| 98 |
+
"epoch": 1.01,
|
| 99 |
+
"learning_rate": 7.97979797979798e-05,
|
| 100 |
+
"loss": 1.7093,
|
| 101 |
"step": 1400
|
| 102 |
},
|
| 103 |
{
|
| 104 |
+
"epoch": 1.08,
|
| 105 |
+
"learning_rate": 7.835497835497836e-05,
|
| 106 |
+
"loss": 1.6345,
|
| 107 |
"step": 1500
|
| 108 |
},
|
| 109 |
{
|
| 110 |
+
"epoch": 1.15,
|
| 111 |
+
"learning_rate": 7.691197691197691e-05,
|
| 112 |
+
"loss": 1.6314,
|
| 113 |
"step": 1600
|
| 114 |
},
|
| 115 |
{
|
| 116 |
+
"epoch": 1.23,
|
| 117 |
+
"learning_rate": 7.546897546897548e-05,
|
| 118 |
+
"loss": 1.6207,
|
| 119 |
"step": 1700
|
| 120 |
},
|
| 121 |
{
|
| 122 |
+
"epoch": 1.3,
|
| 123 |
+
"learning_rate": 7.402597402597404e-05,
|
| 124 |
+
"loss": 1.589,
|
| 125 |
"step": 1800
|
| 126 |
},
|
| 127 |
{
|
| 128 |
+
"epoch": 1.37,
|
| 129 |
+
"learning_rate": 7.258297258297259e-05,
|
| 130 |
+
"loss": 1.612,
|
| 131 |
"step": 1900
|
| 132 |
},
|
| 133 |
{
|
| 134 |
+
"epoch": 1.44,
|
| 135 |
+
"learning_rate": 7.113997113997114e-05,
|
| 136 |
+
"loss": 1.5812,
|
| 137 |
"step": 2000
|
| 138 |
},
|
| 139 |
{
|
| 140 |
+
"epoch": 1.52,
|
| 141 |
+
"learning_rate": 6.96969696969697e-05,
|
| 142 |
+
"loss": 1.6024,
|
| 143 |
"step": 2100
|
| 144 |
},
|
| 145 |
{
|
| 146 |
+
"epoch": 1.59,
|
| 147 |
+
"learning_rate": 6.825396825396825e-05,
|
| 148 |
+
"loss": 1.5719,
|
| 149 |
"step": 2200
|
| 150 |
},
|
| 151 |
{
|
| 152 |
+
"epoch": 1.66,
|
| 153 |
+
"learning_rate": 6.681096681096681e-05,
|
| 154 |
+
"loss": 1.5706,
|
| 155 |
"step": 2300
|
| 156 |
},
|
| 157 |
{
|
| 158 |
+
"epoch": 1.73,
|
| 159 |
+
"learning_rate": 6.536796536796536e-05,
|
| 160 |
+
"loss": 1.5704,
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 161 |
"step": 2400
|
| 162 |
},
|
| 163 |
{
|
| 164 |
+
"epoch": 1.8,
|
| 165 |
+
"learning_rate": 6.392496392496393e-05,
|
| 166 |
+
"loss": 1.5733,
|
| 167 |
"step": 2500
|
| 168 |
},
|
| 169 |
{
|
| 170 |
+
"epoch": 1.88,
|
| 171 |
+
"learning_rate": 6.248196248196248e-05,
|
| 172 |
+
"loss": 1.5852,
|
| 173 |
"step": 2600
|
| 174 |
},
|
| 175 |
{
|
| 176 |
+
"epoch": 1.95,
|
| 177 |
+
"learning_rate": 6.103896103896104e-05,
|
| 178 |
+
"loss": 1.5698,
|
| 179 |
"step": 2700
|
| 180 |
},
|
| 181 |
{
|
| 182 |
+
"epoch": 2.0,
|
| 183 |
+
"eval_loss": 1.2226001024246216,
|
| 184 |
+
"eval_runtime": 231.7119,
|
| 185 |
+
"eval_samples_per_second": 194.668,
|
| 186 |
+
"eval_steps_per_second": 3.043,
|
| 187 |
+
"step": 2772
|
| 188 |
+
},
|
| 189 |
+
{
|
| 190 |
+
"epoch": 2.02,
|
| 191 |
+
"learning_rate": 5.959595959595959e-05,
|
| 192 |
+
"loss": 1.5239,
|
| 193 |
"step": 2800
|
| 194 |
},
|
| 195 |
{
|
| 196 |
+
"epoch": 2.09,
|
| 197 |
+
"learning_rate": 5.815295815295816e-05,
|
| 198 |
+
"loss": 1.4899,
|
| 199 |
"step": 2900
|
| 200 |
},
|
| 201 |
{
|
| 202 |
+
"epoch": 2.16,
|
| 203 |
+
"learning_rate": 5.6709956709956715e-05,
|
| 204 |
+
"loss": 1.461,
|
| 205 |
"step": 3000
|
| 206 |
},
|
| 207 |
{
|
| 208 |
+
"epoch": 2.24,
|
| 209 |
+
"learning_rate": 5.526695526695527e-05,
|
| 210 |
+
"loss": 1.4747,
|
| 211 |
"step": 3100
|
| 212 |
},
|
| 213 |
{
|
| 214 |
+
"epoch": 2.31,
|
| 215 |
+
"learning_rate": 5.382395382395382e-05,
|
| 216 |
+
"loss": 1.481,
|
| 217 |
"step": 3200
|
| 218 |
},
|
| 219 |
{
|
| 220 |
+
"epoch": 2.38,
|
| 221 |
+
"learning_rate": 5.2380952380952384e-05,
|
| 222 |
+
"loss": 1.4496,
|
| 223 |
"step": 3300
|
| 224 |
},
|
| 225 |
{
|
| 226 |
+
"epoch": 2.45,
|
| 227 |
+
"learning_rate": 5.093795093795094e-05,
|
| 228 |
+
"loss": 1.4665,
|
| 229 |
"step": 3400
|
| 230 |
},
|
| 231 |
{
|
| 232 |
+
"epoch": 2.53,
|
| 233 |
+
"learning_rate": 4.94949494949495e-05,
|
| 234 |
+
"loss": 1.4397,
|
| 235 |
"step": 3500
|
| 236 |
},
|
| 237 |
{
|
| 238 |
+
"epoch": 2.6,
|
| 239 |
+
"learning_rate": 4.8051948051948054e-05,
|
| 240 |
+
"loss": 1.4711,
|
| 241 |
"step": 3600
|
| 242 |
},
|
| 243 |
{
|
| 244 |
+
"epoch": 2.67,
|
| 245 |
+
"learning_rate": 4.6608946608946615e-05,
|
| 246 |
+
"loss": 1.4528,
|
| 247 |
"step": 3700
|
| 248 |
},
|
| 249 |
{
|
| 250 |
+
"epoch": 2.74,
|
| 251 |
+
"learning_rate": 4.516594516594517e-05,
|
| 252 |
+
"loss": 1.4661,
|
| 253 |
"step": 3800
|
| 254 |
},
|
| 255 |
{
|
| 256 |
+
"epoch": 2.81,
|
| 257 |
+
"learning_rate": 4.3722943722943724e-05,
|
| 258 |
+
"loss": 1.4583,
|
| 259 |
"step": 3900
|
| 260 |
},
|
| 261 |
{
|
| 262 |
+
"epoch": 2.89,
|
| 263 |
+
"learning_rate": 4.227994227994228e-05,
|
| 264 |
+
"loss": 1.4645,
|
| 265 |
"step": 4000
|
| 266 |
},
|
| 267 |
{
|
| 268 |
+
"epoch": 2.96,
|
| 269 |
+
"learning_rate": 4.083694083694084e-05,
|
| 270 |
+
"loss": 1.4547,
|
| 271 |
"step": 4100
|
| 272 |
},
|
| 273 |
{
|
| 274 |
+
"epoch": 3.0,
|
| 275 |
+
"eval_loss": 1.1470191478729248,
|
| 276 |
+
"eval_runtime": 231.6851,
|
| 277 |
+
"eval_samples_per_second": 194.691,
|
| 278 |
+
"eval_steps_per_second": 3.043,
|
| 279 |
+
"step": 4158
|
| 280 |
+
},
|
| 281 |
+
{
|
| 282 |
+
"epoch": 3.03,
|
| 283 |
+
"learning_rate": 3.939393939393939e-05,
|
| 284 |
+
"loss": 1.4251,
|
| 285 |
"step": 4200
|
| 286 |
},
|
| 287 |
{
|
| 288 |
+
"epoch": 3.1,
|
| 289 |
+
"learning_rate": 3.7950937950937954e-05,
|
| 290 |
+
"loss": 1.3732,
|
| 291 |
"step": 4300
|
| 292 |
},
|
| 293 |
{
|
| 294 |
+
"epoch": 3.17,
|
| 295 |
+
"learning_rate": 3.650793650793651e-05,
|
| 296 |
+
"loss": 1.3843,
|
| 297 |
"step": 4400
|
| 298 |
},
|
| 299 |
{
|
| 300 |
+
"epoch": 3.25,
|
| 301 |
+
"learning_rate": 3.506493506493507e-05,
|
| 302 |
+
"loss": 1.3884,
|
| 303 |
"step": 4500
|
| 304 |
},
|
| 305 |
{
|
| 306 |
+
"epoch": 3.32,
|
| 307 |
+
"learning_rate": 3.3621933621933624e-05,
|
| 308 |
+
"loss": 1.3902,
|
| 309 |
"step": 4600
|
| 310 |
},
|
| 311 |
{
|
| 312 |
+
"epoch": 3.39,
|
| 313 |
+
"learning_rate": 3.217893217893218e-05,
|
| 314 |
+
"loss": 1.4071,
|
| 315 |
"step": 4700
|
| 316 |
},
|
| 317 |
{
|
| 318 |
+
"epoch": 3.46,
|
| 319 |
+
"learning_rate": 3.073593073593073e-05,
|
| 320 |
+
"loss": 1.3857,
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 321 |
"step": 4800
|
| 322 |
},
|
| 323 |
{
|
| 324 |
+
"epoch": 3.54,
|
| 325 |
+
"learning_rate": 2.9292929292929294e-05,
|
| 326 |
+
"loss": 1.3623,
|
| 327 |
"step": 4900
|
| 328 |
},
|
| 329 |
{
|
| 330 |
+
"epoch": 3.61,
|
| 331 |
+
"learning_rate": 2.7849927849927855e-05,
|
| 332 |
+
"loss": 1.3954,
|
| 333 |
"step": 5000
|
| 334 |
},
|
| 335 |
{
|
| 336 |
+
"epoch": 3.68,
|
| 337 |
+
"learning_rate": 2.640692640692641e-05,
|
| 338 |
+
"loss": 1.39,
|
| 339 |
"step": 5100
|
| 340 |
},
|
| 341 |
{
|
| 342 |
+
"epoch": 3.75,
|
| 343 |
+
"learning_rate": 2.4963924963924963e-05,
|
| 344 |
+
"loss": 1.3866,
|
| 345 |
"step": 5200
|
| 346 |
},
|
| 347 |
{
|
| 348 |
+
"epoch": 3.82,
|
| 349 |
+
"learning_rate": 2.352092352092352e-05,
|
| 350 |
+
"loss": 1.382,
|
| 351 |
"step": 5300
|
| 352 |
},
|
| 353 |
{
|
| 354 |
+
"epoch": 3.9,
|
| 355 |
+
"learning_rate": 2.207792207792208e-05,
|
| 356 |
+
"loss": 1.3959,
|
| 357 |
"step": 5400
|
| 358 |
},
|
| 359 |
{
|
| 360 |
+
"epoch": 3.97,
|
| 361 |
+
"learning_rate": 2.0634920634920636e-05,
|
| 362 |
+
"loss": 1.3969,
|
| 363 |
"step": 5500
|
| 364 |
},
|
| 365 |
{
|
| 366 |
+
"epoch": 4.0,
|
| 367 |
+
"eval_loss": 1.1056883335113525,
|
| 368 |
+
"eval_runtime": 231.7985,
|
| 369 |
+
"eval_samples_per_second": 194.596,
|
| 370 |
+
"eval_steps_per_second": 3.041,
|
| 371 |
+
"step": 5544
|
| 372 |
+
},
|
| 373 |
+
{
|
| 374 |
+
"epoch": 4.04,
|
| 375 |
+
"learning_rate": 1.919191919191919e-05,
|
| 376 |
+
"loss": 1.3549,
|
| 377 |
"step": 5600
|
| 378 |
},
|
| 379 |
{
|
| 380 |
+
"epoch": 4.11,
|
| 381 |
+
"learning_rate": 1.7748917748917752e-05,
|
| 382 |
+
"loss": 1.3454,
|
| 383 |
"step": 5700
|
| 384 |
},
|
| 385 |
{
|
| 386 |
+
"epoch": 4.18,
|
| 387 |
+
"learning_rate": 1.630591630591631e-05,
|
| 388 |
+
"loss": 1.3436,
|
| 389 |
"step": 5800
|
| 390 |
},
|
| 391 |
{
|
| 392 |
+
"epoch": 4.26,
|
| 393 |
+
"learning_rate": 1.4862914862914865e-05,
|
| 394 |
+
"loss": 1.3465,
|
| 395 |
"step": 5900
|
| 396 |
},
|
| 397 |
{
|
| 398 |
+
"epoch": 4.33,
|
| 399 |
+
"learning_rate": 1.3419913419913421e-05,
|
| 400 |
+
"loss": 1.3498,
|
| 401 |
"step": 6000
|
| 402 |
},
|
| 403 |
{
|
| 404 |
+
"epoch": 4.4,
|
| 405 |
+
"learning_rate": 1.1976911976911977e-05,
|
| 406 |
+
"loss": 1.3535,
|
| 407 |
"step": 6100
|
| 408 |
},
|
| 409 |
{
|
| 410 |
+
"epoch": 4.47,
|
| 411 |
+
"learning_rate": 1.0533910533910535e-05,
|
| 412 |
+
"loss": 1.3471,
|
| 413 |
"step": 6200
|
| 414 |
},
|
| 415 |
{
|
| 416 |
+
"epoch": 4.55,
|
| 417 |
+
"learning_rate": 9.090909090909091e-06,
|
| 418 |
+
"loss": 1.3427,
|
| 419 |
"step": 6300
|
| 420 |
},
|
| 421 |
{
|
| 422 |
+
"epoch": 4.62,
|
| 423 |
+
"learning_rate": 7.647907647907649e-06,
|
| 424 |
+
"loss": 1.3496,
|
| 425 |
"step": 6400
|
| 426 |
},
|
| 427 |
{
|
| 428 |
+
"epoch": 4.69,
|
| 429 |
+
"learning_rate": 6.204906204906205e-06,
|
| 430 |
+
"loss": 1.3428,
|
| 431 |
"step": 6500
|
| 432 |
},
|
| 433 |
{
|
| 434 |
+
"epoch": 4.76,
|
| 435 |
+
"learning_rate": 4.7619047619047615e-06,
|
| 436 |
+
"loss": 1.3362,
|
| 437 |
"step": 6600
|
| 438 |
},
|
| 439 |
{
|
| 440 |
+
"epoch": 4.83,
|
| 441 |
+
"learning_rate": 3.318903318903319e-06,
|
| 442 |
+
"loss": 1.3424,
|
| 443 |
"step": 6700
|
| 444 |
},
|
| 445 |
{
|
| 446 |
+
"epoch": 4.91,
|
| 447 |
+
"learning_rate": 1.875901875901876e-06,
|
| 448 |
+
"loss": 1.3475,
|
| 449 |
"step": 6800
|
| 450 |
},
|
| 451 |
{
|
| 452 |
+
"epoch": 4.98,
|
| 453 |
+
"learning_rate": 4.329004329004329e-07,
|
| 454 |
+
"loss": 1.3748,
|
| 455 |
"step": 6900
|
| 456 |
},
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 457 |
{
|
| 458 |
"epoch": 5.0,
|
| 459 |
+
"eval_loss": 1.095082402229309,
|
| 460 |
+
"eval_runtime": 232.1115,
|
| 461 |
+
"eval_samples_per_second": 194.333,
|
| 462 |
+
"eval_steps_per_second": 3.037,
|
| 463 |
+
"step": 6930
|
| 464 |
},
|
| 465 |
{
|
| 466 |
"epoch": 5.0,
|
| 467 |
+
"step": 6930,
|
| 468 |
+
"total_flos": 2.6579255601659904e+17,
|
| 469 |
+
"train_loss": 1.650609011189074,
|
| 470 |
+
"train_runtime": 5663.0471,
|
| 471 |
+
"train_samples_per_second": 39.143,
|
| 472 |
+
"train_steps_per_second": 1.224
|
| 473 |
}
|
| 474 |
],
|
| 475 |
"logging_steps": 100,
|
| 476 |
+
"max_steps": 6930,
|
| 477 |
"num_train_epochs": 5,
|
| 478 |
"save_steps": 500,
|
| 479 |
+
"total_flos": 2.6579255601659904e+17,
|
| 480 |
"trial_name": null,
|
| 481 |
"trial_params": null
|
| 482 |
}
|