Model save
Browse files- README.md +1 -1
- all_results.json +5 -10
- train_results.json +5 -5
- trainer_state.json +373 -323
README.md
CHANGED
|
@@ -27,7 +27,7 @@ print(output["generated_text"])
|
|
| 27 |
|
| 28 |
## Training procedure
|
| 29 |
|
| 30 |
-
[<img src="https://raw.githubusercontent.com/wandb/assets/main/wandb-github-badge-28.svg" alt="Visualize in Weights & Biases" width="150" height="24"/>](https://wandb.ai/fandou-team/huggingface/runs/
|
| 31 |
|
| 32 |
|
| 33 |
This model was trained with SFT.
|
|
|
|
| 27 |
|
| 28 |
## Training procedure
|
| 29 |
|
| 30 |
+
[<img src="https://raw.githubusercontent.com/wandb/assets/main/wandb-github-badge-28.svg" alt="Visualize in Weights & Biases" width="150" height="24"/>](https://wandb.ai/fandou-team/huggingface/runs/m3e8fia3)
|
| 31 |
|
| 32 |
|
| 33 |
This model was trained with SFT.
|
all_results.json
CHANGED
|
@@ -1,13 +1,8 @@
|
|
| 1 |
{
|
| 2 |
-
"
|
| 3 |
-
"
|
| 4 |
-
"
|
| 5 |
-
"eval_samples_per_second": 5.004,
|
| 6 |
-
"eval_steps_per_second": 1.28,
|
| 7 |
-
"total_flos": 7.883277455484518e+16,
|
| 8 |
-
"train_loss": 1.23619465266957,
|
| 9 |
-
"train_runtime": 1783.7305,
|
| 10 |
"train_samples": 1000,
|
| 11 |
-
"train_samples_per_second":
|
| 12 |
-
"train_steps_per_second": 0.
|
| 13 |
}
|
|
|
|
| 1 |
{
|
| 2 |
+
"total_flos": 43871077662720.0,
|
| 3 |
+
"train_loss": 0.9090226414915803,
|
| 4 |
+
"train_runtime": 2127.6426,
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 5 |
"train_samples": 1000,
|
| 6 |
+
"train_samples_per_second": 23.091,
|
| 7 |
+
"train_steps_per_second": 0.181
|
| 8 |
}
|
train_results.json
CHANGED
|
@@ -1,8 +1,8 @@
|
|
| 1 |
{
|
| 2 |
-
"total_flos":
|
| 3 |
-
"train_loss":
|
| 4 |
-
"train_runtime":
|
| 5 |
"train_samples": 1000,
|
| 6 |
-
"train_samples_per_second":
|
| 7 |
-
"train_steps_per_second": 0.
|
| 8 |
}
|
|
|
|
| 1 |
{
|
| 2 |
+
"total_flos": 43871077662720.0,
|
| 3 |
+
"train_loss": 0.9090226414915803,
|
| 4 |
+
"train_runtime": 2127.6426,
|
| 5 |
"train_samples": 1000,
|
| 6 |
+
"train_samples_per_second": 23.091,
|
| 7 |
+
"train_steps_per_second": 0.181
|
| 8 |
}
|
trainer_state.json
CHANGED
|
@@ -1,531 +1,581 @@
|
|
| 1 |
{
|
| 2 |
"best_metric": null,
|
| 3 |
"best_model_checkpoint": null,
|
| 4 |
-
"epoch": 0
|
| 5 |
-
"eval_steps":
|
| 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 |
-
"grad_norm": 2.
|
| 14 |
-
"learning_rate":
|
| 15 |
-
"loss": 1.
|
| 16 |
-
"mean_token_accuracy": 0.6584272754958135,
|
| 17 |
"step": 5
|
| 18 |
},
|
| 19 |
{
|
| 20 |
-
"epoch": 0.
|
| 21 |
-
"grad_norm": 2.
|
| 22 |
-
"learning_rate":
|
| 23 |
-
"loss": 1.
|
| 24 |
-
"mean_token_accuracy": 0.662085565849035,
|
| 25 |
"step": 10
|
| 26 |
},
|
| 27 |
{
|
| 28 |
-
"epoch": 0.
|
| 29 |
-
"grad_norm":
|
| 30 |
-
"learning_rate":
|
| 31 |
-
"loss": 1.
|
| 32 |
-
"mean_token_accuracy": 0.6418076518243186,
|
| 33 |
"step": 15
|
| 34 |
},
|
| 35 |
{
|
| 36 |
-
"epoch": 0.
|
| 37 |
-
"grad_norm": 1.
|
| 38 |
-
"learning_rate":
|
| 39 |
-
"loss": 1.
|
| 40 |
-
"mean_token_accuracy": 0.6534564204767823,
|
| 41 |
"step": 20
|
| 42 |
},
|
| 43 |
{
|
| 44 |
-
"epoch": 0.
|
| 45 |
-
"grad_norm":
|
| 46 |
-
"learning_rate":
|
| 47 |
-
"loss": 1.
|
| 48 |
-
"mean_token_accuracy": 0.6651619066075792,
|
| 49 |
"step": 25
|
| 50 |
},
|
| 51 |
{
|
| 52 |
-
"epoch": 0.
|
| 53 |
-
"grad_norm":
|
| 54 |
-
"learning_rate":
|
| 55 |
-
"loss": 1.
|
| 56 |
-
"mean_token_accuracy": 0.6717782021288032,
|
| 57 |
"step": 30
|
| 58 |
},
|
| 59 |
{
|
| 60 |
-
"epoch": 0.
|
| 61 |
-
"grad_norm": 1.
|
| 62 |
-
"learning_rate":
|
| 63 |
-
"loss": 1.
|
| 64 |
-
"mean_token_accuracy": 0.694698959999242,
|
| 65 |
"step": 35
|
| 66 |
},
|
| 67 |
{
|
| 68 |
-
"epoch": 0.
|
| 69 |
-
"grad_norm": 1.
|
| 70 |
-
"learning_rate":
|
| 71 |
-
"loss": 1.
|
| 72 |
-
"mean_token_accuracy": 0.6780847732470013,
|
| 73 |
"step": 40
|
| 74 |
},
|
| 75 |
{
|
| 76 |
-
"epoch": 0.
|
| 77 |
-
"grad_norm":
|
| 78 |
-
"learning_rate":
|
| 79 |
-
"loss": 1.
|
| 80 |
-
"mean_token_accuracy": 0.6968185226866065,
|
| 81 |
"step": 45
|
| 82 |
},
|
| 83 |
{
|
| 84 |
-
"epoch": 0.
|
| 85 |
-
"grad_norm":
|
| 86 |
-
"learning_rate":
|
| 87 |
-
"loss": 1.
|
| 88 |
-
"mean_token_accuracy": 0.672758400550632,
|
| 89 |
"step": 50
|
| 90 |
},
|
| 91 |
{
|
| 92 |
-
"epoch": 0.
|
| 93 |
-
"grad_norm": 1.
|
| 94 |
-
"learning_rate":
|
| 95 |
-
"loss": 1.
|
| 96 |
-
"mean_token_accuracy": 0.6891739777463799,
|
| 97 |
"step": 55
|
| 98 |
},
|
| 99 |
{
|
| 100 |
-
"epoch": 0.
|
| 101 |
-
"grad_norm": 1.
|
| 102 |
-
"learning_rate":
|
| 103 |
-
"loss": 1.
|
| 104 |
-
"mean_token_accuracy": 0.7052165754765654,
|
| 105 |
"step": 60
|
| 106 |
},
|
| 107 |
{
|
| 108 |
-
"epoch": 0.
|
| 109 |
-
"grad_norm":
|
| 110 |
-
"learning_rate":
|
| 111 |
-
"loss": 1.
|
| 112 |
-
"mean_token_accuracy": 0.6921520639873796,
|
| 113 |
"step": 65
|
| 114 |
},
|
| 115 |
{
|
| 116 |
-
"epoch": 0.
|
| 117 |
-
"grad_norm":
|
| 118 |
-
"learning_rate":
|
| 119 |
-
"loss": 1.
|
| 120 |
-
"mean_token_accuracy": 0.6918742825268815,
|
| 121 |
"step": 70
|
| 122 |
},
|
| 123 |
{
|
| 124 |
-
"epoch": 0.
|
| 125 |
-
"grad_norm": 1.
|
| 126 |
-
"learning_rate":
|
| 127 |
-
"loss": 1.
|
| 128 |
-
"mean_token_accuracy": 0.6744394009627726,
|
| 129 |
"step": 75
|
| 130 |
},
|
| 131 |
{
|
| 132 |
-
"epoch":
|
| 133 |
-
"grad_norm":
|
| 134 |
-
"learning_rate":
|
| 135 |
-
"loss": 1.
|
| 136 |
-
"mean_token_accuracy": 0.6919606949918741,
|
| 137 |
"step": 80
|
| 138 |
},
|
| 139 |
{
|
| 140 |
-
"epoch":
|
| 141 |
-
"grad_norm": 1.
|
| 142 |
-
"learning_rate":
|
| 143 |
-
"loss": 1.
|
| 144 |
-
"mean_token_accuracy": 0.6730983288833745,
|
| 145 |
"step": 85
|
| 146 |
},
|
| 147 |
{
|
| 148 |
-
"epoch":
|
| 149 |
-
"grad_norm": 1.
|
| 150 |
-
"learning_rate":
|
| 151 |
-
"loss": 1.
|
| 152 |
-
"mean_token_accuracy": 0.677167603530814,
|
| 153 |
"step": 90
|
| 154 |
},
|
| 155 |
{
|
| 156 |
-
"epoch":
|
| 157 |
-
"grad_norm":
|
| 158 |
-
"learning_rate":
|
| 159 |
-
"loss": 1.
|
| 160 |
-
"mean_token_accuracy": 0.6805341821135611,
|
| 161 |
"step": 95
|
| 162 |
},
|
| 163 |
{
|
| 164 |
-
"epoch":
|
| 165 |
-
"grad_norm":
|
| 166 |
-
"learning_rate":
|
| 167 |
-
"loss": 1.
|
| 168 |
-
"mean_token_accuracy": 0.6857057946710582,
|
| 169 |
"step": 100
|
| 170 |
},
|
| 171 |
{
|
| 172 |
-
"epoch":
|
| 173 |
-
"grad_norm": 1.
|
| 174 |
-
"learning_rate":
|
| 175 |
-
"loss": 1.
|
| 176 |
-
"mean_token_accuracy": 0.687469468283924,
|
| 177 |
"step": 105
|
| 178 |
},
|
| 179 |
{
|
| 180 |
-
"epoch":
|
| 181 |
-
"grad_norm":
|
| 182 |
-
"learning_rate":
|
| 183 |
-
"loss": 1.
|
| 184 |
-
"mean_token_accuracy": 0.6902886780811677,
|
| 185 |
"step": 110
|
| 186 |
},
|
| 187 |
{
|
| 188 |
-
"epoch":
|
| 189 |
-
"grad_norm": 1.
|
| 190 |
-
"learning_rate":
|
| 191 |
-
"loss": 1.
|
| 192 |
-
"mean_token_accuracy": 0.6857961919369919,
|
| 193 |
"step": 115
|
| 194 |
},
|
| 195 |
{
|
| 196 |
-
"epoch":
|
| 197 |
-
"grad_norm":
|
| 198 |
-
"learning_rate":
|
| 199 |
-
"loss": 1.
|
| 200 |
-
"mean_token_accuracy": 0.7004586354413661,
|
| 201 |
"step": 120
|
| 202 |
},
|
| 203 |
{
|
| 204 |
-
"epoch":
|
| 205 |
-
"grad_norm":
|
| 206 |
-
"learning_rate":
|
| 207 |
-
"loss": 1.
|
| 208 |
-
"mean_token_accuracy": 0.6946914374684998,
|
| 209 |
"step": 125
|
| 210 |
},
|
| 211 |
{
|
| 212 |
-
"epoch":
|
| 213 |
-
"grad_norm":
|
| 214 |
-
"learning_rate":
|
| 215 |
-
"loss": 1.
|
| 216 |
-
"mean_token_accuracy": 0.6917642628755305,
|
| 217 |
"step": 130
|
| 218 |
},
|
| 219 |
{
|
| 220 |
-
"epoch":
|
| 221 |
-
"grad_norm":
|
| 222 |
-
"learning_rate":
|
| 223 |
-
"loss": 1.
|
| 224 |
-
"mean_token_accuracy": 0.6803009834502376,
|
| 225 |
"step": 135
|
| 226 |
},
|
| 227 |
{
|
| 228 |
-
"epoch":
|
| 229 |
-
"grad_norm": 1.
|
| 230 |
-
"learning_rate":
|
| 231 |
-
"loss": 1.
|
| 232 |
-
"mean_token_accuracy": 0.66906340898559,
|
| 233 |
"step": 140
|
| 234 |
},
|
| 235 |
{
|
| 236 |
-
"epoch":
|
| 237 |
-
"grad_norm":
|
| 238 |
-
"learning_rate":
|
| 239 |
-
"loss": 1.
|
| 240 |
-
"mean_token_accuracy": 0.6748017583716992,
|
| 241 |
"step": 145
|
| 242 |
},
|
| 243 |
{
|
| 244 |
-
"epoch":
|
| 245 |
-
"grad_norm":
|
| 246 |
-
"learning_rate":
|
| 247 |
-
"loss": 1.
|
| 248 |
-
"mean_token_accuracy": 0.691278874465356,
|
| 249 |
"step": 150
|
| 250 |
},
|
| 251 |
{
|
| 252 |
-
"epoch":
|
| 253 |
-
"grad_norm": 1.
|
| 254 |
-
"learning_rate":
|
| 255 |
-
"loss": 1.
|
| 256 |
-
"mean_token_accuracy": 0.6915639418258935,
|
| 257 |
"step": 155
|
| 258 |
},
|
| 259 |
{
|
| 260 |
-
"epoch":
|
| 261 |
-
"grad_norm": 1.
|
| 262 |
-
"learning_rate":
|
| 263 |
-
"loss":
|
| 264 |
-
"mean_token_accuracy": 0.6795721711260161,
|
| 265 |
"step": 160
|
| 266 |
},
|
| 267 |
{
|
| 268 |
-
"epoch":
|
| 269 |
-
"grad_norm":
|
| 270 |
-
"learning_rate":
|
| 271 |
-
"loss":
|
| 272 |
-
"mean_token_accuracy": 0.6911618495404431,
|
| 273 |
"step": 165
|
| 274 |
},
|
| 275 |
{
|
| 276 |
-
"epoch":
|
| 277 |
-
"grad_norm":
|
| 278 |
-
"learning_rate":
|
| 279 |
-
"loss":
|
| 280 |
-
"mean_token_accuracy": 0.6691213045203754,
|
| 281 |
"step": 170
|
| 282 |
},
|
| 283 |
{
|
| 284 |
-
"epoch":
|
| 285 |
-
"grad_norm":
|
| 286 |
-
"learning_rate":
|
| 287 |
-
"loss":
|
| 288 |
-
"mean_token_accuracy": 0.6846231446931508,
|
| 289 |
"step": 175
|
| 290 |
},
|
| 291 |
{
|
| 292 |
-
"epoch":
|
| 293 |
-
"grad_norm":
|
| 294 |
-
"learning_rate":
|
| 295 |
-
"loss":
|
| 296 |
-
"mean_token_accuracy": 0.6980786468955738,
|
| 297 |
"step": 180
|
| 298 |
},
|
| 299 |
{
|
| 300 |
-
"epoch":
|
| 301 |
-
"grad_norm":
|
| 302 |
-
"learning_rate":
|
| 303 |
-
"loss":
|
| 304 |
-
"mean_token_accuracy": 0.698524151500448,
|
| 305 |
"step": 185
|
| 306 |
},
|
| 307 |
{
|
| 308 |
-
"epoch":
|
| 309 |
-
"grad_norm":
|
| 310 |
-
"learning_rate":
|
| 311 |
-
"loss":
|
| 312 |
-
"mean_token_accuracy": 0.7087621864221246,
|
| 313 |
"step": 190
|
| 314 |
},
|
| 315 |
{
|
| 316 |
-
"epoch":
|
| 317 |
-
"grad_norm":
|
| 318 |
-
"learning_rate":
|
| 319 |
-
"loss":
|
| 320 |
-
"mean_token_accuracy": 0.6694032774839085,
|
| 321 |
"step": 195
|
| 322 |
},
|
| 323 |
{
|
| 324 |
-
"epoch":
|
| 325 |
-
"grad_norm":
|
| 326 |
-
"learning_rate":
|
| 327 |
-
"loss":
|
| 328 |
-
"mean_token_accuracy": 0.6659829648734708,
|
| 329 |
"step": 200
|
| 330 |
},
|
| 331 |
{
|
| 332 |
-
"epoch":
|
| 333 |
-
"grad_norm":
|
| 334 |
-
"learning_rate":
|
| 335 |
-
"loss":
|
| 336 |
-
"mean_token_accuracy": 0.6830356432134138,
|
| 337 |
"step": 205
|
| 338 |
},
|
| 339 |
{
|
| 340 |
-
"epoch":
|
| 341 |
-
"grad_norm":
|
| 342 |
-
"learning_rate":
|
| 343 |
-
"loss":
|
| 344 |
-
"mean_token_accuracy": 0.6673393247330371,
|
| 345 |
"step": 210
|
| 346 |
},
|
| 347 |
{
|
| 348 |
-
"epoch":
|
| 349 |
-
"grad_norm":
|
| 350 |
-
"learning_rate":
|
| 351 |
-
"loss":
|
| 352 |
-
"mean_token_accuracy": 0.6839774101528373,
|
| 353 |
"step": 215
|
| 354 |
},
|
| 355 |
{
|
| 356 |
-
"epoch":
|
| 357 |
-
"grad_norm":
|
| 358 |
-
"learning_rate":
|
| 359 |
-
"loss":
|
| 360 |
-
"mean_token_accuracy": 0.6862650424423485,
|
| 361 |
"step": 220
|
| 362 |
},
|
| 363 |
{
|
| 364 |
-
"epoch":
|
| 365 |
-
"grad_norm":
|
| 366 |
-
"learning_rate":
|
| 367 |
-
"loss":
|
| 368 |
-
"mean_token_accuracy": 0.6849734049971641,
|
| 369 |
"step": 225
|
| 370 |
},
|
| 371 |
{
|
| 372 |
-
"epoch":
|
| 373 |
-
"grad_norm":
|
| 374 |
-
"learning_rate":
|
| 375 |
-
"loss":
|
| 376 |
-
"mean_token_accuracy": 0.6994565541878519,
|
| 377 |
"step": 230
|
| 378 |
},
|
| 379 |
{
|
| 380 |
-
"epoch":
|
| 381 |
-
"grad_norm": 1.
|
| 382 |
-
"learning_rate":
|
| 383 |
-
"loss":
|
| 384 |
-
"mean_token_accuracy": 0.6945823398256653,
|
| 385 |
"step": 235
|
| 386 |
},
|
| 387 |
{
|
| 388 |
-
"epoch":
|
| 389 |
-
"grad_norm": 1.
|
| 390 |
-
"learning_rate": 2.
|
| 391 |
-
"loss":
|
| 392 |
-
"mean_token_accuracy": 0.7025941539981695,
|
| 393 |
"step": 240
|
| 394 |
},
|
| 395 |
{
|
| 396 |
-
"epoch":
|
| 397 |
-
"grad_norm":
|
| 398 |
-
"learning_rate":
|
| 399 |
-
"loss":
|
| 400 |
-
"mean_token_accuracy": 0.6670428901128733,
|
| 401 |
"step": 245
|
| 402 |
},
|
| 403 |
{
|
| 404 |
-
"epoch":
|
| 405 |
-
"grad_norm":
|
| 406 |
-
"learning_rate": 1.
|
| 407 |
-
"loss":
|
| 408 |
-
"mean_token_accuracy": 0.6966172545481755,
|
| 409 |
"step": 250
|
| 410 |
},
|
| 411 |
{
|
| 412 |
-
"epoch":
|
| 413 |
-
"grad_norm":
|
| 414 |
-
"learning_rate": 1.
|
| 415 |
-
"loss":
|
| 416 |
-
"mean_token_accuracy": 0.7020263962587155,
|
| 417 |
"step": 255
|
| 418 |
},
|
| 419 |
{
|
| 420 |
-
"epoch":
|
| 421 |
-
"grad_norm":
|
| 422 |
-
"learning_rate": 1.
|
| 423 |
-
"loss":
|
| 424 |
-
"mean_token_accuracy": 0.7035694430910295,
|
| 425 |
"step": 260
|
| 426 |
},
|
| 427 |
{
|
| 428 |
-
"epoch":
|
| 429 |
-
"grad_norm":
|
| 430 |
-
"learning_rate": 1.
|
| 431 |
-
"loss":
|
| 432 |
-
"mean_token_accuracy": 0.6922763738951797,
|
| 433 |
"step": 265
|
| 434 |
},
|
| 435 |
{
|
| 436 |
-
"epoch":
|
| 437 |
-
"grad_norm":
|
| 438 |
-
"learning_rate":
|
| 439 |
-
"loss":
|
| 440 |
-
"mean_token_accuracy": 0.6892363770014897,
|
| 441 |
"step": 270
|
| 442 |
},
|
| 443 |
{
|
| 444 |
-
"epoch":
|
| 445 |
-
"grad_norm":
|
| 446 |
-
"learning_rate":
|
| 447 |
-
"loss":
|
| 448 |
-
"mean_token_accuracy": 0.6834054369061775,
|
| 449 |
"step": 275
|
| 450 |
},
|
| 451 |
{
|
| 452 |
-
"epoch":
|
| 453 |
-
"grad_norm":
|
| 454 |
-
"learning_rate":
|
| 455 |
-
"loss":
|
| 456 |
-
"mean_token_accuracy": 0.6909252205130498,
|
| 457 |
"step": 280
|
| 458 |
},
|
| 459 |
{
|
| 460 |
-
"epoch":
|
| 461 |
-
"grad_norm":
|
| 462 |
-
"learning_rate":
|
| 463 |
-
"loss":
|
| 464 |
-
"mean_token_accuracy": 0.69716578948841,
|
| 465 |
"step": 285
|
| 466 |
},
|
| 467 |
{
|
| 468 |
-
"epoch":
|
| 469 |
-
"grad_norm":
|
| 470 |
-
"learning_rate": 1.
|
| 471 |
-
"loss":
|
| 472 |
-
"mean_token_accuracy": 0.6928531967188305,
|
| 473 |
"step": 290
|
| 474 |
},
|
| 475 |
{
|
| 476 |
-
"epoch":
|
| 477 |
-
"grad_norm":
|
| 478 |
-
"learning_rate":
|
| 479 |
-
"loss":
|
| 480 |
-
"mean_token_accuracy": 0.694887794722856,
|
| 481 |
"step": 295
|
| 482 |
},
|
| 483 |
{
|
| 484 |
-
"epoch":
|
| 485 |
-
"grad_norm":
|
| 486 |
-
"learning_rate":
|
| 487 |
-
"loss":
|
| 488 |
-
"mean_token_accuracy": 0.6835920887329172,
|
| 489 |
"step": 300
|
| 490 |
},
|
| 491 |
{
|
| 492 |
-
"epoch":
|
| 493 |
-
"grad_norm":
|
| 494 |
-
"learning_rate":
|
| 495 |
-
"loss":
|
| 496 |
-
"mean_token_accuracy": 0.7035911209586428,
|
| 497 |
"step": 305
|
| 498 |
},
|
| 499 |
{
|
| 500 |
-
"epoch":
|
| 501 |
-
"
|
| 502 |
-
"
|
| 503 |
-
"
|
| 504 |
-
"
|
| 505 |
-
|
| 506 |
-
|
| 507 |
-
"
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 508 |
}
|
| 509 |
],
|
| 510 |
"logging_steps": 5,
|
| 511 |
-
"max_steps":
|
| 512 |
"num_input_tokens_seen": 0,
|
| 513 |
-
"num_train_epochs":
|
| 514 |
-
"save_steps":
|
| 515 |
"stateful_callbacks": {
|
| 516 |
"TrainerControl": {
|
| 517 |
"args": {
|
| 518 |
"should_epoch_stop": false,
|
| 519 |
"should_evaluate": false,
|
| 520 |
"should_log": false,
|
| 521 |
-
"should_save":
|
| 522 |
-
"should_training_stop":
|
| 523 |
},
|
| 524 |
"attributes": {}
|
| 525 |
}
|
| 526 |
},
|
| 527 |
-
"total_flos":
|
| 528 |
-
"train_batch_size":
|
| 529 |
"trial_name": null,
|
| 530 |
"trial_params": null
|
| 531 |
}
|
|
|
|
| 1 |
{
|
| 2 |
"best_metric": null,
|
| 3 |
"best_model_checkpoint": null,
|
| 4 |
+
"epoch": 5.0,
|
| 5 |
+
"eval_steps": 500,
|
| 6 |
+
"global_step": 385,
|
| 7 |
"is_hyper_param_search": false,
|
| 8 |
"is_local_process_zero": true,
|
| 9 |
"is_world_process_zero": true,
|
| 10 |
"log_history": [
|
| 11 |
{
|
| 12 |
+
"epoch": 0.06493506493506493,
|
| 13 |
+
"grad_norm": 2.600665284354585,
|
| 14 |
+
"learning_rate": 1.25e-05,
|
| 15 |
+
"loss": 1.6075,
|
|
|
|
| 16 |
"step": 5
|
| 17 |
},
|
| 18 |
{
|
| 19 |
+
"epoch": 0.12987012987012986,
|
| 20 |
+
"grad_norm": 2.3704504908479493,
|
| 21 |
+
"learning_rate": 2.5e-05,
|
| 22 |
+
"loss": 1.4792,
|
|
|
|
| 23 |
"step": 10
|
| 24 |
},
|
| 25 |
{
|
| 26 |
+
"epoch": 0.19480519480519481,
|
| 27 |
+
"grad_norm": 1.2779795738968422,
|
| 28 |
+
"learning_rate": 3.7500000000000003e-05,
|
| 29 |
+
"loss": 1.4115,
|
|
|
|
| 30 |
"step": 15
|
| 31 |
},
|
| 32 |
{
|
| 33 |
+
"epoch": 0.2597402597402597,
|
| 34 |
+
"grad_norm": 1.267787561996043,
|
| 35 |
+
"learning_rate": 5e-05,
|
| 36 |
+
"loss": 1.4018,
|
|
|
|
| 37 |
"step": 20
|
| 38 |
},
|
| 39 |
{
|
| 40 |
+
"epoch": 0.3246753246753247,
|
| 41 |
+
"grad_norm": 0.9468473252185556,
|
| 42 |
+
"learning_rate": 4.9979167589800175e-05,
|
| 43 |
+
"loss": 1.3367,
|
|
|
|
| 44 |
"step": 25
|
| 45 |
},
|
| 46 |
{
|
| 47 |
+
"epoch": 0.38961038961038963,
|
| 48 |
+
"grad_norm": 0.9893253972313399,
|
| 49 |
+
"learning_rate": 4.991670893602868e-05,
|
| 50 |
+
"loss": 1.3354,
|
|
|
|
| 51 |
"step": 30
|
| 52 |
},
|
| 53 |
{
|
| 54 |
+
"epoch": 0.45454545454545453,
|
| 55 |
+
"grad_norm": 1.0881069970614872,
|
| 56 |
+
"learning_rate": 4.9812739697734024e-05,
|
| 57 |
+
"loss": 1.3007,
|
|
|
|
| 58 |
"step": 35
|
| 59 |
},
|
| 60 |
{
|
| 61 |
+
"epoch": 0.5194805194805194,
|
| 62 |
+
"grad_norm": 1.0194980858529235,
|
| 63 |
+
"learning_rate": 4.9667452402011365e-05,
|
| 64 |
+
"loss": 1.2934,
|
|
|
|
| 65 |
"step": 40
|
| 66 |
},
|
| 67 |
{
|
| 68 |
+
"epoch": 0.5844155844155844,
|
| 69 |
+
"grad_norm": 0.9223444885083613,
|
| 70 |
+
"learning_rate": 4.94811160874866e-05,
|
| 71 |
+
"loss": 1.3045,
|
|
|
|
| 72 |
"step": 45
|
| 73 |
},
|
| 74 |
{
|
| 75 |
+
"epoch": 0.6493506493506493,
|
| 76 |
+
"grad_norm": 0.923065081102753,
|
| 77 |
+
"learning_rate": 4.925407580611875e-05,
|
| 78 |
+
"loss": 1.2753,
|
|
|
|
| 79 |
"step": 50
|
| 80 |
},
|
| 81 |
{
|
| 82 |
+
"epoch": 0.7142857142857143,
|
| 83 |
+
"grad_norm": 1.0718722181581042,
|
| 84 |
+
"learning_rate": 4.898675198424325e-05,
|
| 85 |
+
"loss": 1.2756,
|
|
|
|
| 86 |
"step": 55
|
| 87 |
},
|
| 88 |
{
|
| 89 |
+
"epoch": 0.7792207792207793,
|
| 90 |
+
"grad_norm": 1.052542976749598,
|
| 91 |
+
"learning_rate": 4.867963964403906e-05,
|
| 92 |
+
"loss": 1.2911,
|
|
|
|
| 93 |
"step": 60
|
| 94 |
},
|
| 95 |
{
|
| 96 |
+
"epoch": 0.8441558441558441,
|
| 97 |
+
"grad_norm": 0.8128323214762116,
|
| 98 |
+
"learning_rate": 4.833330748686162e-05,
|
| 99 |
+
"loss": 1.2779,
|
|
|
|
| 100 |
"step": 65
|
| 101 |
},
|
| 102 |
{
|
| 103 |
+
"epoch": 0.9090909090909091,
|
| 104 |
+
"grad_norm": 0.8676636174920215,
|
| 105 |
+
"learning_rate": 4.794839684013882e-05,
|
| 106 |
+
"loss": 1.2848,
|
|
|
|
| 107 |
"step": 70
|
| 108 |
},
|
| 109 |
{
|
| 110 |
+
"epoch": 0.974025974025974,
|
| 111 |
+
"grad_norm": 1.042065531092415,
|
| 112 |
+
"learning_rate": 4.7525620469780227e-05,
|
| 113 |
+
"loss": 1.2567,
|
|
|
|
| 114 |
"step": 75
|
| 115 |
},
|
| 116 |
{
|
| 117 |
+
"epoch": 1.0389610389610389,
|
| 118 |
+
"grad_norm": 0.972909624095194,
|
| 119 |
+
"learning_rate": 4.7065761260298747e-05,
|
| 120 |
+
"loss": 1.1721,
|
|
|
|
| 121 |
"step": 80
|
| 122 |
},
|
| 123 |
{
|
| 124 |
+
"epoch": 1.103896103896104,
|
| 125 |
+
"grad_norm": 1.032654503817413,
|
| 126 |
+
"learning_rate": 4.6569670765088703e-05,
|
| 127 |
+
"loss": 1.0906,
|
|
|
|
| 128 |
"step": 85
|
| 129 |
},
|
| 130 |
{
|
| 131 |
+
"epoch": 1.1688311688311688,
|
| 132 |
+
"grad_norm": 1.0351195035386815,
|
| 133 |
+
"learning_rate": 4.603826762954497e-05,
|
| 134 |
+
"loss": 1.1172,
|
|
|
|
| 135 |
"step": 90
|
| 136 |
},
|
| 137 |
{
|
| 138 |
+
"epoch": 1.2337662337662338,
|
| 139 |
+
"grad_norm": 0.96162795934013,
|
| 140 |
+
"learning_rate": 4.5472535889943214e-05,
|
| 141 |
+
"loss": 1.0908,
|
|
|
|
| 142 |
"step": 95
|
| 143 |
},
|
| 144 |
{
|
| 145 |
+
"epoch": 1.2987012987012987,
|
| 146 |
+
"grad_norm": 0.8351069694923322,
|
| 147 |
+
"learning_rate": 4.487352315123119e-05,
|
| 148 |
+
"loss": 1.1043,
|
|
|
|
| 149 |
"step": 100
|
| 150 |
},
|
| 151 |
{
|
| 152 |
+
"epoch": 1.3636363636363638,
|
| 153 |
+
"grad_norm": 1.0009708201516048,
|
| 154 |
+
"learning_rate": 4.424233864710562e-05,
|
| 155 |
+
"loss": 1.0782,
|
|
|
|
| 156 |
"step": 105
|
| 157 |
},
|
| 158 |
{
|
| 159 |
+
"epoch": 1.4285714285714286,
|
| 160 |
+
"grad_norm": 0.8512417675384085,
|
| 161 |
+
"learning_rate": 4.3580151185966625e-05,
|
| 162 |
+
"loss": 1.0982,
|
|
|
|
| 163 |
"step": 110
|
| 164 |
},
|
| 165 |
{
|
| 166 |
+
"epoch": 1.4935064935064934,
|
| 167 |
+
"grad_norm": 1.0103004383513716,
|
| 168 |
+
"learning_rate": 4.288818698655374e-05,
|
| 169 |
+
"loss": 1.0942,
|
|
|
|
| 170 |
"step": 115
|
| 171 |
},
|
| 172 |
{
|
| 173 |
+
"epoch": 1.5584415584415585,
|
| 174 |
+
"grad_norm": 0.8573478770737569,
|
| 175 |
+
"learning_rate": 4.216772740727103e-05,
|
| 176 |
+
"loss": 1.0897,
|
|
|
|
| 177 |
"step": 120
|
| 178 |
},
|
| 179 |
{
|
| 180 |
+
"epoch": 1.6233766233766234,
|
| 181 |
+
"grad_norm": 0.781876249476916,
|
| 182 |
+
"learning_rate": 4.142010657340632e-05,
|
| 183 |
+
"loss": 1.0763,
|
|
|
|
| 184 |
"step": 125
|
| 185 |
},
|
| 186 |
{
|
| 187 |
+
"epoch": 1.6883116883116882,
|
| 188 |
+
"grad_norm": 0.9579078710565259,
|
| 189 |
+
"learning_rate": 4.064670890663829e-05,
|
| 190 |
+
"loss": 1.1087,
|
|
|
|
| 191 |
"step": 130
|
| 192 |
},
|
| 193 |
{
|
| 194 |
+
"epoch": 1.7532467532467533,
|
| 195 |
+
"grad_norm": 0.7939190983394226,
|
| 196 |
+
"learning_rate": 3.9848966561406185e-05,
|
| 197 |
+
"loss": 1.0746,
|
|
|
|
| 198 |
"step": 135
|
| 199 |
},
|
| 200 |
{
|
| 201 |
+
"epoch": 1.8181818181818183,
|
| 202 |
+
"grad_norm": 1.0342389011696922,
|
| 203 |
+
"learning_rate": 3.902835677288954e-05,
|
| 204 |
+
"loss": 1.0677,
|
|
|
|
| 205 |
"step": 140
|
| 206 |
},
|
| 207 |
{
|
| 208 |
+
"epoch": 1.883116883116883,
|
| 209 |
+
"grad_norm": 0.7587581170618278,
|
| 210 |
+
"learning_rate": 3.818639912150864e-05,
|
| 211 |
+
"loss": 1.0916,
|
|
|
|
| 212 |
"step": 145
|
| 213 |
},
|
| 214 |
{
|
| 215 |
+
"epoch": 1.948051948051948,
|
| 216 |
+
"grad_norm": 0.7978087410640827,
|
| 217 |
+
"learning_rate": 3.7324652719011446e-05,
|
| 218 |
+
"loss": 1.0789,
|
|
|
|
| 219 |
"step": 150
|
| 220 |
},
|
| 221 |
{
|
| 222 |
+
"epoch": 2.012987012987013,
|
| 223 |
+
"grad_norm": 1.1656631051781983,
|
| 224 |
+
"learning_rate": 3.644471332135751e-05,
|
| 225 |
+
"loss": 1.0629,
|
|
|
|
| 226 |
"step": 155
|
| 227 |
},
|
| 228 |
{
|
| 229 |
+
"epoch": 2.0779220779220777,
|
| 230 |
+
"grad_norm": 1.0903515201097143,
|
| 231 |
+
"learning_rate": 3.554821037374533e-05,
|
| 232 |
+
"loss": 0.8856,
|
|
|
|
| 233 |
"step": 160
|
| 234 |
},
|
| 235 |
{
|
| 236 |
+
"epoch": 2.142857142857143,
|
| 237 |
+
"grad_norm": 0.7595990161921076,
|
| 238 |
+
"learning_rate": 3.463680399325489e-05,
|
| 239 |
+
"loss": 0.8855,
|
|
|
|
| 240 |
"step": 165
|
| 241 |
},
|
| 242 |
{
|
| 243 |
+
"epoch": 2.207792207792208,
|
| 244 |
+
"grad_norm": 0.7774624026109681,
|
| 245 |
+
"learning_rate": 3.371218189469306e-05,
|
| 246 |
+
"loss": 0.8659,
|
|
|
|
| 247 |
"step": 170
|
| 248 |
},
|
| 249 |
{
|
| 250 |
+
"epoch": 2.2727272727272725,
|
| 251 |
+
"grad_norm": 0.7461072755462708,
|
| 252 |
+
"learning_rate": 3.277605626533422e-05,
|
| 253 |
+
"loss": 0.8772,
|
|
|
|
| 254 |
"step": 175
|
| 255 |
},
|
| 256 |
{
|
| 257 |
+
"epoch": 2.3376623376623376,
|
| 258 |
+
"grad_norm": 0.7432368434791046,
|
| 259 |
+
"learning_rate": 3.183016059434367e-05,
|
| 260 |
+
"loss": 0.8525,
|
|
|
|
| 261 |
"step": 180
|
| 262 |
},
|
| 263 |
{
|
| 264 |
+
"epoch": 2.4025974025974026,
|
| 265 |
+
"grad_norm": 0.7590366429814798,
|
| 266 |
+
"learning_rate": 3.0876246462754685e-05,
|
| 267 |
+
"loss": 0.8742,
|
|
|
|
| 268 |
"step": 185
|
| 269 |
},
|
| 270 |
{
|
| 271 |
+
"epoch": 2.4675324675324677,
|
| 272 |
+
"grad_norm": 0.7247155849317677,
|
| 273 |
+
"learning_rate": 2.9916080299943672e-05,
|
| 274 |
+
"loss": 0.9026,
|
|
|
|
| 275 |
"step": 190
|
| 276 |
},
|
| 277 |
{
|
| 278 |
+
"epoch": 2.5324675324675323,
|
| 279 |
+
"grad_norm": 0.7310534548444702,
|
| 280 |
+
"learning_rate": 2.8951440112609623e-05,
|
| 281 |
+
"loss": 0.8883,
|
|
|
|
| 282 |
"step": 195
|
| 283 |
},
|
| 284 |
{
|
| 285 |
+
"epoch": 2.5974025974025974,
|
| 286 |
+
"grad_norm": 0.7129005263727524,
|
| 287 |
+
"learning_rate": 2.7984112192315004e-05,
|
| 288 |
+
"loss": 0.8736,
|
|
|
|
| 289 |
"step": 200
|
| 290 |
},
|
| 291 |
{
|
| 292 |
+
"epoch": 2.6623376623376624,
|
| 293 |
+
"grad_norm": 0.7437690616597665,
|
| 294 |
+
"learning_rate": 2.7015887807685002e-05,
|
| 295 |
+
"loss": 0.8981,
|
|
|
|
| 296 |
"step": 205
|
| 297 |
},
|
| 298 |
{
|
| 299 |
+
"epoch": 2.7272727272727275,
|
| 300 |
+
"grad_norm": 0.7385255276801924,
|
| 301 |
+
"learning_rate": 2.604855988739039e-05,
|
| 302 |
+
"loss": 0.8895,
|
|
|
|
| 303 |
"step": 210
|
| 304 |
},
|
| 305 |
{
|
| 306 |
+
"epoch": 2.792207792207792,
|
| 307 |
+
"grad_norm": 0.7144722424573203,
|
| 308 |
+
"learning_rate": 2.5083919700056337e-05,
|
| 309 |
+
"loss": 0.8744,
|
|
|
|
| 310 |
"step": 215
|
| 311 |
},
|
| 312 |
{
|
| 313 |
+
"epoch": 2.857142857142857,
|
| 314 |
+
"grad_norm": 0.7010146967901848,
|
| 315 |
+
"learning_rate": 2.412375353724532e-05,
|
| 316 |
+
"loss": 0.896,
|
|
|
|
| 317 |
"step": 220
|
| 318 |
},
|
| 319 |
{
|
| 320 |
+
"epoch": 2.9220779220779223,
|
| 321 |
+
"grad_norm": 0.7504002323519033,
|
| 322 |
+
"learning_rate": 2.316983940565633e-05,
|
| 323 |
+
"loss": 0.8868,
|
|
|
|
| 324 |
"step": 225
|
| 325 |
},
|
| 326 |
{
|
| 327 |
+
"epoch": 2.987012987012987,
|
| 328 |
+
"grad_norm": 0.7865383039242797,
|
| 329 |
+
"learning_rate": 2.2223943734665787e-05,
|
| 330 |
+
"loss": 0.8745,
|
|
|
|
| 331 |
"step": 230
|
| 332 |
},
|
| 333 |
{
|
| 334 |
+
"epoch": 3.051948051948052,
|
| 335 |
+
"grad_norm": 1.0159335857839107,
|
| 336 |
+
"learning_rate": 2.128781810530695e-05,
|
| 337 |
+
"loss": 0.7511,
|
|
|
|
| 338 |
"step": 235
|
| 339 |
},
|
| 340 |
{
|
| 341 |
+
"epoch": 3.116883116883117,
|
| 342 |
+
"grad_norm": 1.2002393858292175,
|
| 343 |
+
"learning_rate": 2.0363196006745117e-05,
|
| 344 |
+
"loss": 0.6809,
|
|
|
|
| 345 |
"step": 240
|
| 346 |
},
|
| 347 |
{
|
| 348 |
+
"epoch": 3.1818181818181817,
|
| 349 |
+
"grad_norm": 0.7878609155092301,
|
| 350 |
+
"learning_rate": 1.9451789626254672e-05,
|
| 351 |
+
"loss": 0.6976,
|
|
|
|
| 352 |
"step": 245
|
| 353 |
},
|
| 354 |
{
|
| 355 |
+
"epoch": 3.2467532467532467,
|
| 356 |
+
"grad_norm": 0.8016037192685418,
|
| 357 |
+
"learning_rate": 1.8555286678642496e-05,
|
| 358 |
+
"loss": 0.6747,
|
|
|
|
| 359 |
"step": 250
|
| 360 |
},
|
| 361 |
{
|
| 362 |
+
"epoch": 3.311688311688312,
|
| 363 |
+
"grad_norm": 0.7587512144299318,
|
| 364 |
+
"learning_rate": 1.7675347280988562e-05,
|
| 365 |
+
"loss": 0.6964,
|
|
|
|
| 366 |
"step": 255
|
| 367 |
},
|
| 368 |
{
|
| 369 |
+
"epoch": 3.3766233766233764,
|
| 370 |
+
"grad_norm": 0.7356037246911953,
|
| 371 |
+
"learning_rate": 1.6813600878491376e-05,
|
| 372 |
+
"loss": 0.6865,
|
|
|
|
| 373 |
"step": 260
|
| 374 |
},
|
| 375 |
{
|
| 376 |
+
"epoch": 3.4415584415584415,
|
| 377 |
+
"grad_norm": 0.7631434364636613,
|
| 378 |
+
"learning_rate": 1.597164322711047e-05,
|
| 379 |
+
"loss": 0.6909,
|
|
|
|
| 380 |
"step": 265
|
| 381 |
},
|
| 382 |
{
|
| 383 |
+
"epoch": 3.5064935064935066,
|
| 384 |
+
"grad_norm": 0.7131404046904652,
|
| 385 |
+
"learning_rate": 1.5151033438593826e-05,
|
| 386 |
+
"loss": 0.6934,
|
|
|
|
| 387 |
"step": 270
|
| 388 |
},
|
| 389 |
{
|
| 390 |
+
"epoch": 3.571428571428571,
|
| 391 |
+
"grad_norm": 0.711551265262125,
|
| 392 |
+
"learning_rate": 1.4353291093361709e-05,
|
| 393 |
+
"loss": 0.6973,
|
|
|
|
| 394 |
"step": 275
|
| 395 |
},
|
| 396 |
{
|
| 397 |
+
"epoch": 3.6363636363636362,
|
| 398 |
+
"grad_norm": 0.705346898473505,
|
| 399 |
+
"learning_rate": 1.3579893426593681e-05,
|
| 400 |
+
"loss": 0.68,
|
|
|
|
| 401 |
"step": 280
|
| 402 |
},
|
| 403 |
{
|
| 404 |
+
"epoch": 3.7012987012987013,
|
| 405 |
+
"grad_norm": 0.7118443145611075,
|
| 406 |
+
"learning_rate": 1.2832272592728966e-05,
|
| 407 |
+
"loss": 0.6904,
|
|
|
|
| 408 |
"step": 285
|
| 409 |
},
|
| 410 |
{
|
| 411 |
+
"epoch": 3.7662337662337664,
|
| 412 |
+
"grad_norm": 0.7427739908839855,
|
| 413 |
+
"learning_rate": 1.211181301344627e-05,
|
| 414 |
+
"loss": 0.6793,
|
|
|
|
| 415 |
"step": 290
|
| 416 |
},
|
| 417 |
{
|
| 418 |
+
"epoch": 3.8311688311688314,
|
| 419 |
+
"grad_norm": 0.6999298394629496,
|
| 420 |
+
"learning_rate": 1.141984881403338e-05,
|
| 421 |
+
"loss": 0.6876,
|
|
|
|
| 422 |
"step": 295
|
| 423 |
},
|
| 424 |
{
|
| 425 |
+
"epoch": 3.896103896103896,
|
| 426 |
+
"grad_norm": 0.7120263608160156,
|
| 427 |
+
"learning_rate": 1.0757661352894394e-05,
|
| 428 |
+
"loss": 0.6893,
|
|
|
|
| 429 |
"step": 300
|
| 430 |
},
|
| 431 |
{
|
| 432 |
+
"epoch": 3.961038961038961,
|
| 433 |
+
"grad_norm": 0.7144317661772965,
|
| 434 |
+
"learning_rate": 1.0126476848768805e-05,
|
| 435 |
+
"loss": 0.6789,
|
|
|
|
| 436 |
"step": 305
|
| 437 |
},
|
| 438 |
{
|
| 439 |
+
"epoch": 4.025974025974026,
|
| 440 |
+
"grad_norm": 1.063865329311315,
|
| 441 |
+
"learning_rate": 9.527464110056795e-06,
|
| 442 |
+
"loss": 0.6269,
|
| 443 |
+
"step": 310
|
| 444 |
+
},
|
| 445 |
+
{
|
| 446 |
+
"epoch": 4.090909090909091,
|
| 447 |
+
"grad_norm": 1.2451272279185113,
|
| 448 |
+
"learning_rate": 8.961732370455032e-06,
|
| 449 |
+
"loss": 0.5587,
|
| 450 |
+
"step": 315
|
| 451 |
+
},
|
| 452 |
+
{
|
| 453 |
+
"epoch": 4.1558441558441555,
|
| 454 |
+
"grad_norm": 0.8874131926210624,
|
| 455 |
+
"learning_rate": 8.430329234911305e-06,
|
| 456 |
+
"loss": 0.5358,
|
| 457 |
+
"step": 320
|
| 458 |
+
},
|
| 459 |
+
{
|
| 460 |
+
"epoch": 4.220779220779221,
|
| 461 |
+
"grad_norm": 0.8464865813880585,
|
| 462 |
+
"learning_rate": 7.934238739701252e-06,
|
| 463 |
+
"loss": 0.5414,
|
| 464 |
+
"step": 325
|
| 465 |
+
},
|
| 466 |
+
{
|
| 467 |
+
"epoch": 4.285714285714286,
|
| 468 |
+
"grad_norm": 0.8425143948713799,
|
| 469 |
+
"learning_rate": 7.4743795302197754e-06,
|
| 470 |
+
"loss": 0.5494,
|
| 471 |
+
"step": 330
|
| 472 |
+
},
|
| 473 |
+
{
|
| 474 |
+
"epoch": 4.35064935064935,
|
| 475 |
+
"grad_norm": 0.7956269077359945,
|
| 476 |
+
"learning_rate": 7.051603159861185e-06,
|
| 477 |
+
"loss": 0.5434,
|
| 478 |
+
"step": 335
|
| 479 |
+
},
|
| 480 |
+
{
|
| 481 |
+
"epoch": 4.415584415584416,
|
| 482 |
+
"grad_norm": 0.7869754406789251,
|
| 483 |
+
"learning_rate": 6.66669251313838e-06,
|
| 484 |
+
"loss": 0.5545,
|
| 485 |
+
"step": 340
|
| 486 |
+
},
|
| 487 |
+
{
|
| 488 |
+
"epoch": 4.48051948051948,
|
| 489 |
+
"grad_norm": 0.7360290448864131,
|
| 490 |
+
"learning_rate": 6.320360355960941e-06,
|
| 491 |
+
"loss": 0.5292,
|
| 492 |
+
"step": 345
|
| 493 |
+
},
|
| 494 |
+
{
|
| 495 |
+
"epoch": 4.545454545454545,
|
| 496 |
+
"grad_norm": 0.7623311010292727,
|
| 497 |
+
"learning_rate": 6.013248015756759e-06,
|
| 498 |
+
"loss": 0.547,
|
| 499 |
+
"step": 350
|
| 500 |
+
},
|
| 501 |
+
{
|
| 502 |
+
"epoch": 4.6103896103896105,
|
| 503 |
+
"grad_norm": 0.7371275230206066,
|
| 504 |
+
"learning_rate": 5.745924193881257e-06,
|
| 505 |
+
"loss": 0.5501,
|
| 506 |
+
"step": 355
|
| 507 |
+
},
|
| 508 |
+
{
|
| 509 |
+
"epoch": 4.675324675324675,
|
| 510 |
+
"grad_norm": 0.7480895176553981,
|
| 511 |
+
"learning_rate": 5.518883912513413e-06,
|
| 512 |
+
"loss": 0.5357,
|
| 513 |
+
"step": 360
|
| 514 |
+
},
|
| 515 |
+
{
|
| 516 |
+
"epoch": 4.740259740259741,
|
| 517 |
+
"grad_norm": 0.7362818934957748,
|
| 518 |
+
"learning_rate": 5.332547597988636e-06,
|
| 519 |
+
"loss": 0.5451,
|
| 520 |
+
"step": 365
|
| 521 |
+
},
|
| 522 |
+
{
|
| 523 |
+
"epoch": 4.805194805194805,
|
| 524 |
+
"grad_norm": 0.7678966619355423,
|
| 525 |
+
"learning_rate": 5.1872603022659765e-06,
|
| 526 |
+
"loss": 0.5297,
|
| 527 |
+
"step": 370
|
| 528 |
+
},
|
| 529 |
+
{
|
| 530 |
+
"epoch": 4.87012987012987,
|
| 531 |
+
"grad_norm": 0.7535710804010259,
|
| 532 |
+
"learning_rate": 5.083291063971324e-06,
|
| 533 |
+
"loss": 0.5356,
|
| 534 |
+
"step": 375
|
| 535 |
+
},
|
| 536 |
+
{
|
| 537 |
+
"epoch": 4.935064935064935,
|
| 538 |
+
"grad_norm": 0.7578922717450597,
|
| 539 |
+
"learning_rate": 5.020832410199826e-06,
|
| 540 |
+
"loss": 0.5347,
|
| 541 |
+
"step": 380
|
| 542 |
+
},
|
| 543 |
+
{
|
| 544 |
+
"epoch": 5.0,
|
| 545 |
+
"grad_norm": 0.7276287792812202,
|
| 546 |
+
"learning_rate": 5e-06,
|
| 547 |
+
"loss": 0.5505,
|
| 548 |
+
"step": 385
|
| 549 |
+
},
|
| 550 |
+
{
|
| 551 |
+
"epoch": 5.0,
|
| 552 |
+
"step": 385,
|
| 553 |
+
"total_flos": 43871077662720.0,
|
| 554 |
+
"train_loss": 0.9090226414915803,
|
| 555 |
+
"train_runtime": 2127.6426,
|
| 556 |
+
"train_samples_per_second": 23.091,
|
| 557 |
+
"train_steps_per_second": 0.181
|
| 558 |
}
|
| 559 |
],
|
| 560 |
"logging_steps": 5,
|
| 561 |
+
"max_steps": 385,
|
| 562 |
"num_input_tokens_seen": 0,
|
| 563 |
+
"num_train_epochs": 5,
|
| 564 |
+
"save_steps": 100,
|
| 565 |
"stateful_callbacks": {
|
| 566 |
"TrainerControl": {
|
| 567 |
"args": {
|
| 568 |
"should_epoch_stop": false,
|
| 569 |
"should_evaluate": false,
|
| 570 |
"should_log": false,
|
| 571 |
+
"should_save": true,
|
| 572 |
+
"should_training_stop": true
|
| 573 |
},
|
| 574 |
"attributes": {}
|
| 575 |
}
|
| 576 |
},
|
| 577 |
+
"total_flos": 43871077662720.0,
|
| 578 |
+
"train_batch_size": 16,
|
| 579 |
"trial_name": null,
|
| 580 |
"trial_params": null
|
| 581 |
}
|