Training in progress, step 500, checkpoint
Browse files
last-checkpoint/model.safetensors
CHANGED
|
@@ -1,3 +1,3 @@
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
-
oid sha256:
|
| 3 |
size 1304192904
|
|
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:2ea69c0b38350140bde6045847eb3aca83bc4627c14f15f7070c9ce872395853
|
| 3 |
size 1304192904
|
last-checkpoint/optimizer.pt
CHANGED
|
@@ -1,3 +1,3 @@
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
-
oid sha256:
|
| 3 |
size 2608620781
|
|
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:549062bd70f44cd72a6e19af8c66126dbfad4947544c75c576ccc4ad90e84829
|
| 3 |
size 2608620781
|
last-checkpoint/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 sha256:90887233ddb5f52291b6e282e00383caf7e27642579529b0c9362e1fd1793cd4
|
| 3 |
size 14244
|
last-checkpoint/scheduler.pt
CHANGED
|
@@ -1,3 +1,3 @@
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
-
oid sha256:
|
| 3 |
size 1064
|
|
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:7d2328d0ebba1a4f73f30ce024a73fa11c5a0fee7ecde6e6f0d609cf0ddc5dcd
|
| 3 |
size 1064
|
last-checkpoint/trainer_state.json
CHANGED
|
@@ -1,1312 +1,416 @@
|
|
| 1 |
{
|
| 2 |
-
"best_metric": 0.
|
| 3 |
-
"best_model_checkpoint": "./results/checkpoint-
|
| 4 |
-
"epoch":
|
| 5 |
"eval_steps": 20,
|
| 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":
|
| 14 |
-
"learning_rate": 1.
|
| 15 |
-
"loss": 0.
|
| 16 |
"step": 20
|
| 17 |
},
|
| 18 |
{
|
| 19 |
-
"epoch": 0.
|
| 20 |
-
"eval_accuracy": 0.
|
| 21 |
-
"eval_loss": 0.
|
| 22 |
-
"eval_runtime":
|
| 23 |
-
"eval_samples_per_second": 10.
|
| 24 |
"eval_steps_per_second": 2.599,
|
| 25 |
"step": 20
|
| 26 |
},
|
| 27 |
{
|
| 28 |
-
"epoch": 0.
|
| 29 |
-
"grad_norm":
|
| 30 |
-
"learning_rate": 1.
|
| 31 |
-
"loss": 0.
|
| 32 |
"step": 40
|
| 33 |
},
|
| 34 |
{
|
| 35 |
-
"epoch": 0.
|
| 36 |
-
"eval_accuracy": 0.
|
| 37 |
-
"eval_loss": 0.
|
| 38 |
-
"eval_runtime":
|
| 39 |
-
"eval_samples_per_second":
|
| 40 |
-
"eval_steps_per_second": 2.
|
| 41 |
"step": 40
|
| 42 |
},
|
| 43 |
{
|
| 44 |
-
"epoch": 0.
|
| 45 |
-
"grad_norm":
|
| 46 |
-
"learning_rate": 1.
|
| 47 |
-
"loss": 0.
|
| 48 |
"step": 60
|
| 49 |
},
|
| 50 |
{
|
| 51 |
-
"epoch": 0.
|
| 52 |
-
"eval_accuracy": 0.
|
| 53 |
-
"eval_loss": 0.
|
| 54 |
-
"eval_runtime":
|
| 55 |
-
"eval_samples_per_second": 10.
|
| 56 |
-
"eval_steps_per_second": 2.
|
| 57 |
"step": 60
|
| 58 |
},
|
| 59 |
{
|
| 60 |
-
"epoch": 0.
|
| 61 |
-
"grad_norm":
|
| 62 |
-
"learning_rate": 1.
|
| 63 |
-
"loss": 0.
|
| 64 |
"step": 80
|
| 65 |
},
|
| 66 |
{
|
| 67 |
-
"epoch": 0.
|
| 68 |
-
"eval_accuracy": 0.
|
| 69 |
-
"eval_loss": 0.
|
| 70 |
-
"eval_runtime":
|
| 71 |
-
"eval_samples_per_second":
|
| 72 |
-
"eval_steps_per_second": 2.
|
| 73 |
"step": 80
|
| 74 |
},
|
| 75 |
{
|
| 76 |
-
"epoch": 0.
|
| 77 |
-
"grad_norm":
|
| 78 |
-
"learning_rate": 1.
|
| 79 |
-
"loss": 0.
|
| 80 |
"step": 100
|
| 81 |
},
|
| 82 |
{
|
| 83 |
-
"epoch": 0.
|
| 84 |
-
"eval_accuracy": 0.
|
| 85 |
-
"eval_loss": 0.
|
| 86 |
-
"eval_runtime":
|
| 87 |
-
"eval_samples_per_second":
|
| 88 |
-
"eval_steps_per_second": 2.
|
| 89 |
"step": 100
|
| 90 |
},
|
| 91 |
{
|
| 92 |
-
"epoch": 0.
|
| 93 |
-
"grad_norm":
|
| 94 |
-
"learning_rate": 1.
|
| 95 |
-
"loss": 0.
|
| 96 |
"step": 120
|
| 97 |
},
|
| 98 |
{
|
| 99 |
-
"epoch": 0.
|
| 100 |
-
"eval_accuracy": 0.
|
| 101 |
-
"eval_loss": 0.
|
| 102 |
-
"eval_runtime":
|
| 103 |
-
"eval_samples_per_second":
|
| 104 |
-
"eval_steps_per_second": 2.
|
| 105 |
"step": 120
|
| 106 |
},
|
| 107 |
{
|
| 108 |
-
"epoch": 0.
|
| 109 |
-
"grad_norm":
|
| 110 |
-
"learning_rate": 1.
|
| 111 |
-
"loss": 0.
|
| 112 |
"step": 140
|
| 113 |
},
|
| 114 |
{
|
| 115 |
-
"epoch": 0.
|
| 116 |
-
"eval_accuracy": 0.
|
| 117 |
-
"eval_loss": 0.
|
| 118 |
-
"eval_runtime":
|
| 119 |
-
"eval_samples_per_second":
|
| 120 |
-
"eval_steps_per_second": 2.
|
| 121 |
"step": 140
|
| 122 |
},
|
| 123 |
{
|
| 124 |
-
"epoch": 0.
|
| 125 |
-
"grad_norm":
|
| 126 |
-
"learning_rate": 1.
|
| 127 |
-
"loss": 0.
|
| 128 |
"step": 160
|
| 129 |
},
|
| 130 |
{
|
| 131 |
-
"epoch": 0.
|
| 132 |
-
"eval_accuracy": 0.
|
| 133 |
-
"eval_loss": 0.
|
| 134 |
-
"eval_runtime":
|
| 135 |
-
"eval_samples_per_second":
|
| 136 |
-
"eval_steps_per_second": 2.
|
| 137 |
"step": 160
|
| 138 |
},
|
| 139 |
{
|
| 140 |
-
"epoch": 0.
|
| 141 |
-
"grad_norm":
|
| 142 |
-
"learning_rate": 1.
|
| 143 |
-
"loss": 0.
|
| 144 |
"step": 180
|
| 145 |
},
|
| 146 |
{
|
| 147 |
-
"epoch": 0.
|
| 148 |
-
"eval_accuracy": 0.
|
| 149 |
-
"eval_loss": 0.
|
| 150 |
-
"eval_runtime":
|
| 151 |
-
"eval_samples_per_second":
|
| 152 |
-
"eval_steps_per_second": 2.
|
| 153 |
"step": 180
|
| 154 |
},
|
| 155 |
{
|
| 156 |
-
"epoch": 0.
|
| 157 |
-
"grad_norm":
|
| 158 |
-
"learning_rate": 1.
|
| 159 |
-
"loss": 0.
|
| 160 |
"step": 200
|
| 161 |
},
|
| 162 |
{
|
| 163 |
-
"epoch": 0.
|
| 164 |
-
"eval_accuracy": 0.
|
| 165 |
-
"eval_loss": 0.
|
| 166 |
-
"eval_runtime":
|
| 167 |
-
"eval_samples_per_second":
|
| 168 |
-
"eval_steps_per_second": 2.
|
| 169 |
"step": 200
|
| 170 |
},
|
| 171 |
{
|
| 172 |
-
"epoch": 0.
|
| 173 |
-
"grad_norm":
|
| 174 |
-
"learning_rate": 1.
|
| 175 |
-
"loss": 0.
|
| 176 |
"step": 220
|
| 177 |
},
|
| 178 |
{
|
| 179 |
-
"epoch": 0.
|
| 180 |
-
"eval_accuracy": 0.
|
| 181 |
-
"eval_loss": 0.
|
| 182 |
-
"eval_runtime":
|
| 183 |
-
"eval_samples_per_second":
|
| 184 |
-
"eval_steps_per_second": 2.
|
| 185 |
"step": 220
|
| 186 |
},
|
| 187 |
{
|
| 188 |
-
"epoch": 0.
|
| 189 |
-
"grad_norm":
|
| 190 |
-
"learning_rate": 1.
|
| 191 |
-
"loss": 0.
|
| 192 |
"step": 240
|
| 193 |
},
|
| 194 |
{
|
| 195 |
-
"epoch": 0.
|
| 196 |
-
"eval_accuracy": 0.
|
| 197 |
-
"eval_loss": 0.
|
| 198 |
-
"eval_runtime":
|
| 199 |
-
"eval_samples_per_second":
|
| 200 |
-
"eval_steps_per_second": 2.
|
| 201 |
"step": 240
|
| 202 |
},
|
| 203 |
{
|
| 204 |
-
"epoch": 0.
|
| 205 |
-
"grad_norm":
|
| 206 |
-
"learning_rate": 1.
|
| 207 |
-
"loss": 0.
|
| 208 |
"step": 260
|
| 209 |
},
|
| 210 |
{
|
| 211 |
-
"epoch": 0.
|
| 212 |
-
"eval_accuracy": 0.
|
| 213 |
-
"eval_loss": 0.
|
| 214 |
-
"eval_runtime":
|
| 215 |
-
"eval_samples_per_second":
|
| 216 |
-
"eval_steps_per_second": 2.
|
| 217 |
"step": 260
|
| 218 |
},
|
| 219 |
{
|
| 220 |
-
"epoch": 0.
|
| 221 |
-
"grad_norm":
|
| 222 |
-
"learning_rate": 1.
|
| 223 |
-
"loss": 0.
|
| 224 |
"step": 280
|
| 225 |
},
|
| 226 |
{
|
| 227 |
-
"epoch": 0.
|
| 228 |
-
"eval_accuracy": 0.
|
| 229 |
-
"eval_loss": 0.
|
| 230 |
-
"eval_runtime":
|
| 231 |
-
"eval_samples_per_second":
|
| 232 |
-
"eval_steps_per_second": 2.
|
| 233 |
"step": 280
|
| 234 |
},
|
| 235 |
{
|
| 236 |
-
"epoch": 0.
|
| 237 |
-
"grad_norm":
|
| 238 |
-
"learning_rate": 1.
|
| 239 |
-
"loss": 0.
|
| 240 |
"step": 300
|
| 241 |
},
|
| 242 |
{
|
| 243 |
-
"epoch": 0.
|
| 244 |
-
"eval_accuracy": 0.
|
| 245 |
-
"eval_loss": 0.
|
| 246 |
-
"eval_runtime":
|
| 247 |
-
"eval_samples_per_second":
|
| 248 |
-
"eval_steps_per_second": 2.
|
| 249 |
"step": 300
|
| 250 |
},
|
| 251 |
{
|
| 252 |
-
"epoch": 0.
|
| 253 |
-
"grad_norm":
|
| 254 |
-
"learning_rate": 1.
|
| 255 |
-
"loss": 0.
|
| 256 |
"step": 320
|
| 257 |
},
|
| 258 |
{
|
| 259 |
-
"epoch": 0.
|
| 260 |
-
"eval_accuracy": 0.
|
| 261 |
-
"eval_loss": 0.
|
| 262 |
-
"eval_runtime":
|
| 263 |
-
"eval_samples_per_second":
|
| 264 |
-
"eval_steps_per_second": 2.
|
| 265 |
"step": 320
|
| 266 |
},
|
| 267 |
{
|
| 268 |
-
"epoch": 1.
|
| 269 |
-
"grad_norm": 0.
|
| 270 |
-
"learning_rate": 1.
|
| 271 |
-
"loss": 0.
|
| 272 |
"step": 340
|
| 273 |
},
|
| 274 |
{
|
| 275 |
-
"epoch": 1.
|
| 276 |
-
"eval_accuracy": 0.
|
| 277 |
-
"eval_loss": 0.
|
| 278 |
-
"eval_runtime":
|
| 279 |
-
"eval_samples_per_second":
|
| 280 |
-
"eval_steps_per_second": 2.
|
| 281 |
"step": 340
|
| 282 |
},
|
| 283 |
{
|
| 284 |
-
"epoch": 1.
|
| 285 |
-
"grad_norm":
|
| 286 |
-
"learning_rate": 1.
|
| 287 |
-
"loss": 0.
|
| 288 |
"step": 360
|
| 289 |
},
|
| 290 |
{
|
| 291 |
-
"epoch": 1.
|
| 292 |
-
"eval_accuracy": 0.
|
| 293 |
-
"eval_loss": 0.
|
| 294 |
-
"eval_runtime":
|
| 295 |
-
"eval_samples_per_second":
|
| 296 |
-
"eval_steps_per_second": 2.
|
| 297 |
"step": 360
|
| 298 |
},
|
| 299 |
{
|
| 300 |
-
"epoch": 1.
|
| 301 |
-
"grad_norm":
|
| 302 |
-
"learning_rate": 1.
|
| 303 |
-
"loss": 0.
|
| 304 |
"step": 380
|
| 305 |
},
|
| 306 |
{
|
| 307 |
-
"epoch": 1.
|
| 308 |
-
"eval_accuracy": 0.
|
| 309 |
-
"eval_loss": 0.
|
| 310 |
-
"eval_runtime":
|
| 311 |
-
"eval_samples_per_second":
|
| 312 |
-
"eval_steps_per_second": 2.
|
| 313 |
"step": 380
|
| 314 |
},
|
| 315 |
{
|
| 316 |
-
"epoch": 1.
|
| 317 |
-
"grad_norm": 0.
|
| 318 |
-
"learning_rate": 1.
|
| 319 |
-
"loss": 0.
|
| 320 |
"step": 400
|
| 321 |
},
|
| 322 |
{
|
| 323 |
-
"epoch": 1.
|
| 324 |
-
"eval_accuracy": 0.
|
| 325 |
-
"eval_loss": 0.
|
| 326 |
-
"eval_runtime":
|
| 327 |
-
"eval_samples_per_second":
|
| 328 |
-
"eval_steps_per_second": 2.
|
| 329 |
"step": 400
|
| 330 |
},
|
| 331 |
{
|
| 332 |
-
"epoch": 1.
|
| 333 |
-
"grad_norm": 0.
|
| 334 |
-
"learning_rate": 1.
|
| 335 |
-
"loss": 0.
|
| 336 |
"step": 420
|
| 337 |
},
|
| 338 |
{
|
| 339 |
-
"epoch": 1.
|
| 340 |
-
"eval_accuracy": 0.
|
| 341 |
-
"eval_loss": 0.
|
| 342 |
-
"eval_runtime":
|
| 343 |
-
"eval_samples_per_second":
|
| 344 |
-
"eval_steps_per_second": 2.
|
| 345 |
"step": 420
|
| 346 |
},
|
| 347 |
{
|
| 348 |
-
"epoch": 1.
|
| 349 |
-
"grad_norm":
|
| 350 |
-
"learning_rate": 1.
|
| 351 |
-
"loss": 0.
|
| 352 |
"step": 440
|
| 353 |
},
|
| 354 |
{
|
| 355 |
-
"epoch": 1.
|
| 356 |
-
"eval_accuracy": 0.
|
| 357 |
-
"eval_loss": 0.
|
| 358 |
-
"eval_runtime":
|
| 359 |
-
"eval_samples_per_second":
|
| 360 |
-
"eval_steps_per_second": 2.
|
| 361 |
"step": 440
|
| 362 |
},
|
| 363 |
{
|
| 364 |
-
"epoch": 1.
|
| 365 |
-
"grad_norm":
|
| 366 |
-
"learning_rate": 1.
|
| 367 |
-
"loss": 0.
|
| 368 |
"step": 460
|
| 369 |
},
|
| 370 |
{
|
| 371 |
-
"epoch": 1.
|
| 372 |
-
"eval_accuracy": 0.
|
| 373 |
-
"eval_loss": 0.
|
| 374 |
-
"eval_runtime":
|
| 375 |
-
"eval_samples_per_second":
|
| 376 |
-
"eval_steps_per_second": 2.
|
| 377 |
"step": 460
|
| 378 |
},
|
| 379 |
{
|
| 380 |
-
"epoch": 1.
|
| 381 |
-
"grad_norm": 0.
|
| 382 |
-
"learning_rate": 1.
|
| 383 |
-
"loss": 0.
|
| 384 |
"step": 480
|
| 385 |
},
|
| 386 |
{
|
| 387 |
-
"epoch": 1.
|
| 388 |
-
"eval_accuracy": 0.
|
| 389 |
-
"eval_loss": 0.
|
| 390 |
-
"eval_runtime":
|
| 391 |
-
"eval_samples_per_second":
|
| 392 |
-
"eval_steps_per_second": 2.
|
| 393 |
"step": 480
|
| 394 |
},
|
| 395 |
{
|
| 396 |
-
"epoch": 1.
|
| 397 |
-
"grad_norm": 0.
|
| 398 |
-
"learning_rate": 1.
|
| 399 |
-
"loss": 0.
|
| 400 |
"step": 500
|
| 401 |
},
|
| 402 |
{
|
| 403 |
-
"epoch": 1.
|
| 404 |
-
"eval_accuracy": 0.
|
| 405 |
-
"eval_loss": 0.
|
| 406 |
-
"eval_runtime":
|
| 407 |
-
"eval_samples_per_second":
|
| 408 |
-
"eval_steps_per_second": 2.
|
| 409 |
"step": 500
|
| 410 |
-
},
|
| 411 |
-
{
|
| 412 |
-
"epoch": 1.5950920245398774,
|
| 413 |
-
"grad_norm": 0.12642166018486023,
|
| 414 |
-
"learning_rate": 1.3619631901840491e-05,
|
| 415 |
-
"loss": 0.3565,
|
| 416 |
-
"step": 520
|
| 417 |
-
},
|
| 418 |
-
{
|
| 419 |
-
"epoch": 1.5950920245398774,
|
| 420 |
-
"eval_accuracy": 0.9,
|
| 421 |
-
"eval_loss": 0.36820870637893677,
|
| 422 |
-
"eval_runtime": 14.3754,
|
| 423 |
-
"eval_samples_per_second": 10.434,
|
| 424 |
-
"eval_steps_per_second": 2.643,
|
| 425 |
-
"step": 520
|
| 426 |
-
},
|
| 427 |
-
{
|
| 428 |
-
"epoch": 1.656441717791411,
|
| 429 |
-
"grad_norm": 3.1343636512756348,
|
| 430 |
-
"learning_rate": 1.3374233128834357e-05,
|
| 431 |
-
"loss": 0.3111,
|
| 432 |
-
"step": 540
|
| 433 |
-
},
|
| 434 |
-
{
|
| 435 |
-
"epoch": 1.656441717791411,
|
| 436 |
-
"eval_accuracy": 0.9,
|
| 437 |
-
"eval_loss": 0.32980188727378845,
|
| 438 |
-
"eval_runtime": 14.0656,
|
| 439 |
-
"eval_samples_per_second": 10.664,
|
| 440 |
-
"eval_steps_per_second": 2.702,
|
| 441 |
-
"step": 540
|
| 442 |
-
},
|
| 443 |
-
{
|
| 444 |
-
"epoch": 1.7177914110429446,
|
| 445 |
-
"grad_norm": 0.2962506115436554,
|
| 446 |
-
"learning_rate": 1.3128834355828221e-05,
|
| 447 |
-
"loss": 0.4191,
|
| 448 |
-
"step": 560
|
| 449 |
-
},
|
| 450 |
-
{
|
| 451 |
-
"epoch": 1.7177914110429446,
|
| 452 |
-
"eval_accuracy": 0.8733333333333333,
|
| 453 |
-
"eval_loss": 0.4493354856967926,
|
| 454 |
-
"eval_runtime": 14.0611,
|
| 455 |
-
"eval_samples_per_second": 10.668,
|
| 456 |
-
"eval_steps_per_second": 2.702,
|
| 457 |
-
"step": 560
|
| 458 |
-
},
|
| 459 |
-
{
|
| 460 |
-
"epoch": 1.7791411042944785,
|
| 461 |
-
"grad_norm": 0.14546217024326324,
|
| 462 |
-
"learning_rate": 1.2883435582822085e-05,
|
| 463 |
-
"loss": 0.2731,
|
| 464 |
-
"step": 580
|
| 465 |
-
},
|
| 466 |
-
{
|
| 467 |
-
"epoch": 1.7791411042944785,
|
| 468 |
-
"eval_accuracy": 0.9066666666666666,
|
| 469 |
-
"eval_loss": 0.3831864297389984,
|
| 470 |
-
"eval_runtime": 13.9539,
|
| 471 |
-
"eval_samples_per_second": 10.75,
|
| 472 |
-
"eval_steps_per_second": 2.723,
|
| 473 |
-
"step": 580
|
| 474 |
-
},
|
| 475 |
-
{
|
| 476 |
-
"epoch": 1.8404907975460123,
|
| 477 |
-
"grad_norm": 0.022235451266169548,
|
| 478 |
-
"learning_rate": 1.2638036809815953e-05,
|
| 479 |
-
"loss": 0.1803,
|
| 480 |
-
"step": 600
|
| 481 |
-
},
|
| 482 |
-
{
|
| 483 |
-
"epoch": 1.8404907975460123,
|
| 484 |
-
"eval_accuracy": 0.8933333333333333,
|
| 485 |
-
"eval_loss": 0.4403243958950043,
|
| 486 |
-
"eval_runtime": 13.8915,
|
| 487 |
-
"eval_samples_per_second": 10.798,
|
| 488 |
-
"eval_steps_per_second": 2.735,
|
| 489 |
-
"step": 600
|
| 490 |
-
},
|
| 491 |
-
{
|
| 492 |
-
"epoch": 1.9018404907975461,
|
| 493 |
-
"grad_norm": 0.10121641308069229,
|
| 494 |
-
"learning_rate": 1.2392638036809817e-05,
|
| 495 |
-
"loss": 0.4462,
|
| 496 |
-
"step": 620
|
| 497 |
-
},
|
| 498 |
-
{
|
| 499 |
-
"epoch": 1.9018404907975461,
|
| 500 |
-
"eval_accuracy": 0.9066666666666666,
|
| 501 |
-
"eval_loss": 0.38436785340309143,
|
| 502 |
-
"eval_runtime": 14.0,
|
| 503 |
-
"eval_samples_per_second": 10.714,
|
| 504 |
-
"eval_steps_per_second": 2.714,
|
| 505 |
-
"step": 620
|
| 506 |
-
},
|
| 507 |
-
{
|
| 508 |
-
"epoch": 1.9631901840490797,
|
| 509 |
-
"grad_norm": 0.05252746492624283,
|
| 510 |
-
"learning_rate": 1.2147239263803683e-05,
|
| 511 |
-
"loss": 0.0025,
|
| 512 |
-
"step": 640
|
| 513 |
-
},
|
| 514 |
-
{
|
| 515 |
-
"epoch": 1.9631901840490797,
|
| 516 |
-
"eval_accuracy": 0.9066666666666666,
|
| 517 |
-
"eval_loss": 0.4562518894672394,
|
| 518 |
-
"eval_runtime": 13.9613,
|
| 519 |
-
"eval_samples_per_second": 10.744,
|
| 520 |
-
"eval_steps_per_second": 2.722,
|
| 521 |
-
"step": 640
|
| 522 |
-
},
|
| 523 |
-
{
|
| 524 |
-
"epoch": 2.0245398773006134,
|
| 525 |
-
"grad_norm": 0.007440468296408653,
|
| 526 |
-
"learning_rate": 1.1901840490797547e-05,
|
| 527 |
-
"loss": 0.1574,
|
| 528 |
-
"step": 660
|
| 529 |
-
},
|
| 530 |
-
{
|
| 531 |
-
"epoch": 2.0245398773006134,
|
| 532 |
-
"eval_accuracy": 0.8933333333333333,
|
| 533 |
-
"eval_loss": 0.5508309006690979,
|
| 534 |
-
"eval_runtime": 13.9141,
|
| 535 |
-
"eval_samples_per_second": 10.78,
|
| 536 |
-
"eval_steps_per_second": 2.731,
|
| 537 |
-
"step": 660
|
| 538 |
-
},
|
| 539 |
-
{
|
| 540 |
-
"epoch": 2.085889570552147,
|
| 541 |
-
"grad_norm": 0.11675461381673813,
|
| 542 |
-
"learning_rate": 1.1656441717791411e-05,
|
| 543 |
-
"loss": 0.0927,
|
| 544 |
-
"step": 680
|
| 545 |
-
},
|
| 546 |
-
{
|
| 547 |
-
"epoch": 2.085889570552147,
|
| 548 |
-
"eval_accuracy": 0.9066666666666666,
|
| 549 |
-
"eval_loss": 0.5528630018234253,
|
| 550 |
-
"eval_runtime": 13.8779,
|
| 551 |
-
"eval_samples_per_second": 10.809,
|
| 552 |
-
"eval_steps_per_second": 2.738,
|
| 553 |
-
"step": 680
|
| 554 |
-
},
|
| 555 |
-
{
|
| 556 |
-
"epoch": 2.147239263803681,
|
| 557 |
-
"grad_norm": 185.11602783203125,
|
| 558 |
-
"learning_rate": 1.1411042944785277e-05,
|
| 559 |
-
"loss": 0.184,
|
| 560 |
-
"step": 700
|
| 561 |
-
},
|
| 562 |
-
{
|
| 563 |
-
"epoch": 2.147239263803681,
|
| 564 |
-
"eval_accuracy": 0.9,
|
| 565 |
-
"eval_loss": 0.5161141753196716,
|
| 566 |
-
"eval_runtime": 13.8694,
|
| 567 |
-
"eval_samples_per_second": 10.815,
|
| 568 |
-
"eval_steps_per_second": 2.74,
|
| 569 |
-
"step": 700
|
| 570 |
-
},
|
| 571 |
-
{
|
| 572 |
-
"epoch": 2.208588957055215,
|
| 573 |
-
"grad_norm": 110.4225082397461,
|
| 574 |
-
"learning_rate": 1.1165644171779141e-05,
|
| 575 |
-
"loss": 0.2446,
|
| 576 |
-
"step": 720
|
| 577 |
-
},
|
| 578 |
-
{
|
| 579 |
-
"epoch": 2.208588957055215,
|
| 580 |
-
"eval_accuracy": 0.8933333333333333,
|
| 581 |
-
"eval_loss": 0.5063968896865845,
|
| 582 |
-
"eval_runtime": 13.9289,
|
| 583 |
-
"eval_samples_per_second": 10.769,
|
| 584 |
-
"eval_steps_per_second": 2.728,
|
| 585 |
-
"step": 720
|
| 586 |
-
},
|
| 587 |
-
{
|
| 588 |
-
"epoch": 2.2699386503067487,
|
| 589 |
-
"grad_norm": 1.8425345420837402,
|
| 590 |
-
"learning_rate": 1.0920245398773005e-05,
|
| 591 |
-
"loss": 0.2498,
|
| 592 |
-
"step": 740
|
| 593 |
-
},
|
| 594 |
-
{
|
| 595 |
-
"epoch": 2.2699386503067487,
|
| 596 |
-
"eval_accuracy": 0.92,
|
| 597 |
-
"eval_loss": 0.40342533588409424,
|
| 598 |
-
"eval_runtime": 13.923,
|
| 599 |
-
"eval_samples_per_second": 10.774,
|
| 600 |
-
"eval_steps_per_second": 2.729,
|
| 601 |
-
"step": 740
|
| 602 |
-
},
|
| 603 |
-
{
|
| 604 |
-
"epoch": 2.331288343558282,
|
| 605 |
-
"grad_norm": 1.4138000011444092,
|
| 606 |
-
"learning_rate": 1.0674846625766873e-05,
|
| 607 |
-
"loss": 0.2217,
|
| 608 |
-
"step": 760
|
| 609 |
-
},
|
| 610 |
-
{
|
| 611 |
-
"epoch": 2.331288343558282,
|
| 612 |
-
"eval_accuracy": 0.8733333333333333,
|
| 613 |
-
"eval_loss": 0.5094826221466064,
|
| 614 |
-
"eval_runtime": 13.9433,
|
| 615 |
-
"eval_samples_per_second": 10.758,
|
| 616 |
-
"eval_steps_per_second": 2.725,
|
| 617 |
-
"step": 760
|
| 618 |
-
},
|
| 619 |
-
{
|
| 620 |
-
"epoch": 2.392638036809816,
|
| 621 |
-
"grad_norm": 0.02712080627679825,
|
| 622 |
-
"learning_rate": 1.0429447852760737e-05,
|
| 623 |
-
"loss": 0.2938,
|
| 624 |
-
"step": 780
|
| 625 |
-
},
|
| 626 |
-
{
|
| 627 |
-
"epoch": 2.392638036809816,
|
| 628 |
-
"eval_accuracy": 0.9066666666666666,
|
| 629 |
-
"eval_loss": 0.37535837292671204,
|
| 630 |
-
"eval_runtime": 13.9067,
|
| 631 |
-
"eval_samples_per_second": 10.786,
|
| 632 |
-
"eval_steps_per_second": 2.732,
|
| 633 |
-
"step": 780
|
| 634 |
-
},
|
| 635 |
-
{
|
| 636 |
-
"epoch": 2.4539877300613497,
|
| 637 |
-
"grad_norm": 10.157035827636719,
|
| 638 |
-
"learning_rate": 1.0184049079754601e-05,
|
| 639 |
-
"loss": 0.109,
|
| 640 |
-
"step": 800
|
| 641 |
-
},
|
| 642 |
-
{
|
| 643 |
-
"epoch": 2.4539877300613497,
|
| 644 |
-
"eval_accuracy": 0.8933333333333333,
|
| 645 |
-
"eval_loss": 0.4770869314670563,
|
| 646 |
-
"eval_runtime": 13.9502,
|
| 647 |
-
"eval_samples_per_second": 10.753,
|
| 648 |
-
"eval_steps_per_second": 2.724,
|
| 649 |
-
"step": 800
|
| 650 |
-
},
|
| 651 |
-
{
|
| 652 |
-
"epoch": 2.5153374233128836,
|
| 653 |
-
"grad_norm": 0.006353612057864666,
|
| 654 |
-
"learning_rate": 9.938650306748467e-06,
|
| 655 |
-
"loss": 0.0282,
|
| 656 |
-
"step": 820
|
| 657 |
-
},
|
| 658 |
-
{
|
| 659 |
-
"epoch": 2.5153374233128836,
|
| 660 |
-
"eval_accuracy": 0.8933333333333333,
|
| 661 |
-
"eval_loss": 0.5535228252410889,
|
| 662 |
-
"eval_runtime": 13.9084,
|
| 663 |
-
"eval_samples_per_second": 10.785,
|
| 664 |
-
"eval_steps_per_second": 2.732,
|
| 665 |
-
"step": 820
|
| 666 |
-
},
|
| 667 |
-
{
|
| 668 |
-
"epoch": 2.5766871165644174,
|
| 669 |
-
"grad_norm": 0.050232015550136566,
|
| 670 |
-
"learning_rate": 9.693251533742331e-06,
|
| 671 |
-
"loss": 0.2455,
|
| 672 |
-
"step": 840
|
| 673 |
-
},
|
| 674 |
-
{
|
| 675 |
-
"epoch": 2.5766871165644174,
|
| 676 |
-
"eval_accuracy": 0.9066666666666666,
|
| 677 |
-
"eval_loss": 0.42059311270713806,
|
| 678 |
-
"eval_runtime": 13.942,
|
| 679 |
-
"eval_samples_per_second": 10.759,
|
| 680 |
-
"eval_steps_per_second": 2.726,
|
| 681 |
-
"step": 840
|
| 682 |
-
},
|
| 683 |
-
{
|
| 684 |
-
"epoch": 2.638036809815951,
|
| 685 |
-
"grad_norm": 11.762324333190918,
|
| 686 |
-
"learning_rate": 9.447852760736197e-06,
|
| 687 |
-
"loss": 0.4728,
|
| 688 |
-
"step": 860
|
| 689 |
-
},
|
| 690 |
-
{
|
| 691 |
-
"epoch": 2.638036809815951,
|
| 692 |
-
"eval_accuracy": 0.9066666666666666,
|
| 693 |
-
"eval_loss": 0.3018172085285187,
|
| 694 |
-
"eval_runtime": 13.9449,
|
| 695 |
-
"eval_samples_per_second": 10.757,
|
| 696 |
-
"eval_steps_per_second": 2.725,
|
| 697 |
-
"step": 860
|
| 698 |
-
},
|
| 699 |
-
{
|
| 700 |
-
"epoch": 2.6993865030674846,
|
| 701 |
-
"grad_norm": 0.3914521038532257,
|
| 702 |
-
"learning_rate": 9.202453987730062e-06,
|
| 703 |
-
"loss": 0.1145,
|
| 704 |
-
"step": 880
|
| 705 |
-
},
|
| 706 |
-
{
|
| 707 |
-
"epoch": 2.6993865030674846,
|
| 708 |
-
"eval_accuracy": 0.9066666666666666,
|
| 709 |
-
"eval_loss": 0.30533984303474426,
|
| 710 |
-
"eval_runtime": 13.9197,
|
| 711 |
-
"eval_samples_per_second": 10.776,
|
| 712 |
-
"eval_steps_per_second": 2.73,
|
| 713 |
-
"step": 880
|
| 714 |
-
},
|
| 715 |
-
{
|
| 716 |
-
"epoch": 2.7607361963190185,
|
| 717 |
-
"grad_norm": 0.5647504329681396,
|
| 718 |
-
"learning_rate": 8.957055214723927e-06,
|
| 719 |
-
"loss": 0.1045,
|
| 720 |
-
"step": 900
|
| 721 |
-
},
|
| 722 |
-
{
|
| 723 |
-
"epoch": 2.7607361963190185,
|
| 724 |
-
"eval_accuracy": 0.9066666666666666,
|
| 725 |
-
"eval_loss": 0.3431243598461151,
|
| 726 |
-
"eval_runtime": 13.8904,
|
| 727 |
-
"eval_samples_per_second": 10.799,
|
| 728 |
-
"eval_steps_per_second": 2.736,
|
| 729 |
-
"step": 900
|
| 730 |
-
},
|
| 731 |
-
{
|
| 732 |
-
"epoch": 2.8220858895705523,
|
| 733 |
-
"grad_norm": 156.81553649902344,
|
| 734 |
-
"learning_rate": 8.711656441717792e-06,
|
| 735 |
-
"loss": 0.2207,
|
| 736 |
-
"step": 920
|
| 737 |
-
},
|
| 738 |
-
{
|
| 739 |
-
"epoch": 2.8220858895705523,
|
| 740 |
-
"eval_accuracy": 0.86,
|
| 741 |
-
"eval_loss": 0.648215651512146,
|
| 742 |
-
"eval_runtime": 13.9076,
|
| 743 |
-
"eval_samples_per_second": 10.785,
|
| 744 |
-
"eval_steps_per_second": 2.732,
|
| 745 |
-
"step": 920
|
| 746 |
-
},
|
| 747 |
-
{
|
| 748 |
-
"epoch": 2.883435582822086,
|
| 749 |
-
"grad_norm": 0.02706349454820156,
|
| 750 |
-
"learning_rate": 8.466257668711658e-06,
|
| 751 |
-
"loss": 0.427,
|
| 752 |
-
"step": 940
|
| 753 |
-
},
|
| 754 |
-
{
|
| 755 |
-
"epoch": 2.883435582822086,
|
| 756 |
-
"eval_accuracy": 0.9133333333333333,
|
| 757 |
-
"eval_loss": 0.43959808349609375,
|
| 758 |
-
"eval_runtime": 13.9415,
|
| 759 |
-
"eval_samples_per_second": 10.759,
|
| 760 |
-
"eval_steps_per_second": 2.726,
|
| 761 |
-
"step": 940
|
| 762 |
-
},
|
| 763 |
-
{
|
| 764 |
-
"epoch": 2.9447852760736195,
|
| 765 |
-
"grad_norm": 61.51915740966797,
|
| 766 |
-
"learning_rate": 8.220858895705522e-06,
|
| 767 |
-
"loss": 0.1898,
|
| 768 |
-
"step": 960
|
| 769 |
-
},
|
| 770 |
-
{
|
| 771 |
-
"epoch": 2.9447852760736195,
|
| 772 |
-
"eval_accuracy": 0.92,
|
| 773 |
-
"eval_loss": 0.3327239453792572,
|
| 774 |
-
"eval_runtime": 13.9378,
|
| 775 |
-
"eval_samples_per_second": 10.762,
|
| 776 |
-
"eval_steps_per_second": 2.726,
|
| 777 |
-
"step": 960
|
| 778 |
-
},
|
| 779 |
-
{
|
| 780 |
-
"epoch": 3.0061349693251533,
|
| 781 |
-
"grad_norm": 0.01911596581339836,
|
| 782 |
-
"learning_rate": 7.975460122699386e-06,
|
| 783 |
-
"loss": 0.0019,
|
| 784 |
-
"step": 980
|
| 785 |
-
},
|
| 786 |
-
{
|
| 787 |
-
"epoch": 3.0061349693251533,
|
| 788 |
-
"eval_accuracy": 0.92,
|
| 789 |
-
"eval_loss": 0.39932945370674133,
|
| 790 |
-
"eval_runtime": 13.9414,
|
| 791 |
-
"eval_samples_per_second": 10.759,
|
| 792 |
-
"eval_steps_per_second": 2.726,
|
| 793 |
-
"step": 980
|
| 794 |
-
},
|
| 795 |
-
{
|
| 796 |
-
"epoch": 3.067484662576687,
|
| 797 |
-
"grad_norm": 0.016511617228388786,
|
| 798 |
-
"learning_rate": 7.730061349693252e-06,
|
| 799 |
-
"loss": 0.0842,
|
| 800 |
-
"step": 1000
|
| 801 |
-
},
|
| 802 |
-
{
|
| 803 |
-
"epoch": 3.067484662576687,
|
| 804 |
-
"eval_accuracy": 0.9266666666666666,
|
| 805 |
-
"eval_loss": 0.41659799218177795,
|
| 806 |
-
"eval_runtime": 13.9699,
|
| 807 |
-
"eval_samples_per_second": 10.737,
|
| 808 |
-
"eval_steps_per_second": 2.72,
|
| 809 |
-
"step": 1000
|
| 810 |
-
},
|
| 811 |
-
{
|
| 812 |
-
"epoch": 3.128834355828221,
|
| 813 |
-
"grad_norm": 0.021293368190526962,
|
| 814 |
-
"learning_rate": 7.484662576687118e-06,
|
| 815 |
-
"loss": 0.1619,
|
| 816 |
-
"step": 1020
|
| 817 |
-
},
|
| 818 |
-
{
|
| 819 |
-
"epoch": 3.128834355828221,
|
| 820 |
-
"eval_accuracy": 0.9133333333333333,
|
| 821 |
-
"eval_loss": 0.4180934429168701,
|
| 822 |
-
"eval_runtime": 14.3308,
|
| 823 |
-
"eval_samples_per_second": 10.467,
|
| 824 |
-
"eval_steps_per_second": 2.652,
|
| 825 |
-
"step": 1020
|
| 826 |
-
},
|
| 827 |
-
{
|
| 828 |
-
"epoch": 3.190184049079755,
|
| 829 |
-
"grad_norm": 14.229204177856445,
|
| 830 |
-
"learning_rate": 7.239263803680983e-06,
|
| 831 |
-
"loss": 0.1849,
|
| 832 |
-
"step": 1040
|
| 833 |
-
},
|
| 834 |
-
{
|
| 835 |
-
"epoch": 3.190184049079755,
|
| 836 |
-
"eval_accuracy": 0.92,
|
| 837 |
-
"eval_loss": 0.47268736362457275,
|
| 838 |
-
"eval_runtime": 13.9315,
|
| 839 |
-
"eval_samples_per_second": 10.767,
|
| 840 |
-
"eval_steps_per_second": 2.728,
|
| 841 |
-
"step": 1040
|
| 842 |
-
},
|
| 843 |
-
{
|
| 844 |
-
"epoch": 3.2515337423312882,
|
| 845 |
-
"grad_norm": 0.017998775467276573,
|
| 846 |
-
"learning_rate": 6.993865030674847e-06,
|
| 847 |
-
"loss": 0.1949,
|
| 848 |
-
"step": 1060
|
| 849 |
-
},
|
| 850 |
-
{
|
| 851 |
-
"epoch": 3.2515337423312882,
|
| 852 |
-
"eval_accuracy": 0.8933333333333333,
|
| 853 |
-
"eval_loss": 0.3345566689968109,
|
| 854 |
-
"eval_runtime": 14.0413,
|
| 855 |
-
"eval_samples_per_second": 10.683,
|
| 856 |
-
"eval_steps_per_second": 2.706,
|
| 857 |
-
"step": 1060
|
| 858 |
-
},
|
| 859 |
-
{
|
| 860 |
-
"epoch": 3.312883435582822,
|
| 861 |
-
"grad_norm": 0.020114585757255554,
|
| 862 |
-
"learning_rate": 6.748466257668712e-06,
|
| 863 |
-
"loss": 0.1796,
|
| 864 |
-
"step": 1080
|
| 865 |
-
},
|
| 866 |
-
{
|
| 867 |
-
"epoch": 3.312883435582822,
|
| 868 |
-
"eval_accuracy": 0.9266666666666666,
|
| 869 |
-
"eval_loss": 0.3471291661262512,
|
| 870 |
-
"eval_runtime": 13.9741,
|
| 871 |
-
"eval_samples_per_second": 10.734,
|
| 872 |
-
"eval_steps_per_second": 2.719,
|
| 873 |
-
"step": 1080
|
| 874 |
-
},
|
| 875 |
-
{
|
| 876 |
-
"epoch": 3.374233128834356,
|
| 877 |
-
"grad_norm": 0.013905039988458157,
|
| 878 |
-
"learning_rate": 6.503067484662578e-06,
|
| 879 |
-
"loss": 0.086,
|
| 880 |
-
"step": 1100
|
| 881 |
-
},
|
| 882 |
-
{
|
| 883 |
-
"epoch": 3.374233128834356,
|
| 884 |
-
"eval_accuracy": 0.8866666666666667,
|
| 885 |
-
"eval_loss": 0.4089130759239197,
|
| 886 |
-
"eval_runtime": 13.9473,
|
| 887 |
-
"eval_samples_per_second": 10.755,
|
| 888 |
-
"eval_steps_per_second": 2.725,
|
| 889 |
-
"step": 1100
|
| 890 |
-
},
|
| 891 |
-
{
|
| 892 |
-
"epoch": 3.4355828220858897,
|
| 893 |
-
"grad_norm": 0.009700474329292774,
|
| 894 |
-
"learning_rate": 6.257668711656443e-06,
|
| 895 |
-
"loss": 0.0187,
|
| 896 |
-
"step": 1120
|
| 897 |
-
},
|
| 898 |
-
{
|
| 899 |
-
"epoch": 3.4355828220858897,
|
| 900 |
-
"eval_accuracy": 0.92,
|
| 901 |
-
"eval_loss": 0.3867844045162201,
|
| 902 |
-
"eval_runtime": 14.013,
|
| 903 |
-
"eval_samples_per_second": 10.704,
|
| 904 |
-
"eval_steps_per_second": 2.712,
|
| 905 |
-
"step": 1120
|
| 906 |
-
},
|
| 907 |
-
{
|
| 908 |
-
"epoch": 3.4969325153374236,
|
| 909 |
-
"grad_norm": 0.003781616687774658,
|
| 910 |
-
"learning_rate": 6.012269938650307e-06,
|
| 911 |
-
"loss": 0.0768,
|
| 912 |
-
"step": 1140
|
| 913 |
-
},
|
| 914 |
-
{
|
| 915 |
-
"epoch": 3.4969325153374236,
|
| 916 |
-
"eval_accuracy": 0.9266666666666666,
|
| 917 |
-
"eval_loss": 0.4095223546028137,
|
| 918 |
-
"eval_runtime": 13.9902,
|
| 919 |
-
"eval_samples_per_second": 10.722,
|
| 920 |
-
"eval_steps_per_second": 2.716,
|
| 921 |
-
"step": 1140
|
| 922 |
-
},
|
| 923 |
-
{
|
| 924 |
-
"epoch": 3.558282208588957,
|
| 925 |
-
"grad_norm": 0.21029236912727356,
|
| 926 |
-
"learning_rate": 5.766871165644172e-06,
|
| 927 |
-
"loss": 0.0008,
|
| 928 |
-
"step": 1160
|
| 929 |
-
},
|
| 930 |
-
{
|
| 931 |
-
"epoch": 3.558282208588957,
|
| 932 |
-
"eval_accuracy": 0.9066666666666666,
|
| 933 |
-
"eval_loss": 0.3779890835285187,
|
| 934 |
-
"eval_runtime": 13.9578,
|
| 935 |
-
"eval_samples_per_second": 10.747,
|
| 936 |
-
"eval_steps_per_second": 2.722,
|
| 937 |
-
"step": 1160
|
| 938 |
-
},
|
| 939 |
-
{
|
| 940 |
-
"epoch": 3.6196319018404908,
|
| 941 |
-
"grad_norm": 0.032502181828022,
|
| 942 |
-
"learning_rate": 5.521472392638038e-06,
|
| 943 |
-
"loss": 0.183,
|
| 944 |
-
"step": 1180
|
| 945 |
-
},
|
| 946 |
-
{
|
| 947 |
-
"epoch": 3.6196319018404908,
|
| 948 |
-
"eval_accuracy": 0.9,
|
| 949 |
-
"eval_loss": 0.3827475905418396,
|
| 950 |
-
"eval_runtime": 13.933,
|
| 951 |
-
"eval_samples_per_second": 10.766,
|
| 952 |
-
"eval_steps_per_second": 2.727,
|
| 953 |
-
"step": 1180
|
| 954 |
-
},
|
| 955 |
-
{
|
| 956 |
-
"epoch": 3.6809815950920246,
|
| 957 |
-
"grad_norm": 0.0289248526096344,
|
| 958 |
-
"learning_rate": 5.276073619631902e-06,
|
| 959 |
-
"loss": 0.204,
|
| 960 |
-
"step": 1200
|
| 961 |
-
},
|
| 962 |
-
{
|
| 963 |
-
"epoch": 3.6809815950920246,
|
| 964 |
-
"eval_accuracy": 0.9,
|
| 965 |
-
"eval_loss": 0.5132840871810913,
|
| 966 |
-
"eval_runtime": 13.9107,
|
| 967 |
-
"eval_samples_per_second": 10.783,
|
| 968 |
-
"eval_steps_per_second": 2.732,
|
| 969 |
-
"step": 1200
|
| 970 |
-
},
|
| 971 |
-
{
|
| 972 |
-
"epoch": 3.7423312883435584,
|
| 973 |
-
"grad_norm": 1.0068583488464355,
|
| 974 |
-
"learning_rate": 5.030674846625767e-06,
|
| 975 |
-
"loss": 0.0758,
|
| 976 |
-
"step": 1220
|
| 977 |
-
},
|
| 978 |
-
{
|
| 979 |
-
"epoch": 3.7423312883435584,
|
| 980 |
-
"eval_accuracy": 0.9133333333333333,
|
| 981 |
-
"eval_loss": 0.4279702305793762,
|
| 982 |
-
"eval_runtime": 13.8908,
|
| 983 |
-
"eval_samples_per_second": 10.799,
|
| 984 |
-
"eval_steps_per_second": 2.736,
|
| 985 |
-
"step": 1220
|
| 986 |
-
},
|
| 987 |
-
{
|
| 988 |
-
"epoch": 3.8036809815950923,
|
| 989 |
-
"grad_norm": 0.0101453373208642,
|
| 990 |
-
"learning_rate": 4.785276073619632e-06,
|
| 991 |
-
"loss": 0.0237,
|
| 992 |
-
"step": 1240
|
| 993 |
-
},
|
| 994 |
-
{
|
| 995 |
-
"epoch": 3.8036809815950923,
|
| 996 |
-
"eval_accuracy": 0.92,
|
| 997 |
-
"eval_loss": 0.3941916823387146,
|
| 998 |
-
"eval_runtime": 13.8887,
|
| 999 |
-
"eval_samples_per_second": 10.8,
|
| 1000 |
-
"eval_steps_per_second": 2.736,
|
| 1001 |
-
"step": 1240
|
| 1002 |
-
},
|
| 1003 |
-
{
|
| 1004 |
-
"epoch": 3.8650306748466257,
|
| 1005 |
-
"grad_norm": 182.59510803222656,
|
| 1006 |
-
"learning_rate": 4.539877300613497e-06,
|
| 1007 |
-
"loss": 0.2143,
|
| 1008 |
-
"step": 1260
|
| 1009 |
-
},
|
| 1010 |
-
{
|
| 1011 |
-
"epoch": 3.8650306748466257,
|
| 1012 |
-
"eval_accuracy": 0.9066666666666666,
|
| 1013 |
-
"eval_loss": 0.36801090836524963,
|
| 1014 |
-
"eval_runtime": 13.8946,
|
| 1015 |
-
"eval_samples_per_second": 10.796,
|
| 1016 |
-
"eval_steps_per_second": 2.735,
|
| 1017 |
-
"step": 1260
|
| 1018 |
-
},
|
| 1019 |
-
{
|
| 1020 |
-
"epoch": 3.9263803680981595,
|
| 1021 |
-
"grad_norm": 0.03958132117986679,
|
| 1022 |
-
"learning_rate": 4.294478527607362e-06,
|
| 1023 |
-
"loss": 0.0106,
|
| 1024 |
-
"step": 1280
|
| 1025 |
-
},
|
| 1026 |
-
{
|
| 1027 |
-
"epoch": 3.9263803680981595,
|
| 1028 |
-
"eval_accuracy": 0.8866666666666667,
|
| 1029 |
-
"eval_loss": 0.5633125901222229,
|
| 1030 |
-
"eval_runtime": 13.9747,
|
| 1031 |
-
"eval_samples_per_second": 10.734,
|
| 1032 |
-
"eval_steps_per_second": 2.719,
|
| 1033 |
-
"step": 1280
|
| 1034 |
-
},
|
| 1035 |
-
{
|
| 1036 |
-
"epoch": 3.9877300613496933,
|
| 1037 |
-
"grad_norm": 140.984375,
|
| 1038 |
-
"learning_rate": 4.049079754601227e-06,
|
| 1039 |
-
"loss": 0.2221,
|
| 1040 |
-
"step": 1300
|
| 1041 |
-
},
|
| 1042 |
-
{
|
| 1043 |
-
"epoch": 3.9877300613496933,
|
| 1044 |
-
"eval_accuracy": 0.92,
|
| 1045 |
-
"eval_loss": 0.38154712319374084,
|
| 1046 |
-
"eval_runtime": 13.996,
|
| 1047 |
-
"eval_samples_per_second": 10.717,
|
| 1048 |
-
"eval_steps_per_second": 2.715,
|
| 1049 |
-
"step": 1300
|
| 1050 |
-
},
|
| 1051 |
-
{
|
| 1052 |
-
"epoch": 4.049079754601227,
|
| 1053 |
-
"grad_norm": 0.0037327792961150408,
|
| 1054 |
-
"learning_rate": 3.8036809815950928e-06,
|
| 1055 |
-
"loss": 0.0212,
|
| 1056 |
-
"step": 1320
|
| 1057 |
-
},
|
| 1058 |
-
{
|
| 1059 |
-
"epoch": 4.049079754601227,
|
| 1060 |
-
"eval_accuracy": 0.9266666666666666,
|
| 1061 |
-
"eval_loss": 0.4598991274833679,
|
| 1062 |
-
"eval_runtime": 13.9374,
|
| 1063 |
-
"eval_samples_per_second": 10.762,
|
| 1064 |
-
"eval_steps_per_second": 2.726,
|
| 1065 |
-
"step": 1320
|
| 1066 |
-
},
|
| 1067 |
-
{
|
| 1068 |
-
"epoch": 4.110429447852761,
|
| 1069 |
-
"grad_norm": 0.010530122555792332,
|
| 1070 |
-
"learning_rate": 3.5582822085889574e-06,
|
| 1071 |
-
"loss": 0.1678,
|
| 1072 |
-
"step": 1340
|
| 1073 |
-
},
|
| 1074 |
-
{
|
| 1075 |
-
"epoch": 4.110429447852761,
|
| 1076 |
-
"eval_accuracy": 0.92,
|
| 1077 |
-
"eval_loss": 0.34579145908355713,
|
| 1078 |
-
"eval_runtime": 13.9042,
|
| 1079 |
-
"eval_samples_per_second": 10.788,
|
| 1080 |
-
"eval_steps_per_second": 2.733,
|
| 1081 |
-
"step": 1340
|
| 1082 |
-
},
|
| 1083 |
-
{
|
| 1084 |
-
"epoch": 4.171779141104294,
|
| 1085 |
-
"grad_norm": 0.015701429918408394,
|
| 1086 |
-
"learning_rate": 3.312883435582822e-06,
|
| 1087 |
-
"loss": 0.1153,
|
| 1088 |
-
"step": 1360
|
| 1089 |
-
},
|
| 1090 |
-
{
|
| 1091 |
-
"epoch": 4.171779141104294,
|
| 1092 |
-
"eval_accuracy": 0.92,
|
| 1093 |
-
"eval_loss": 0.3261447250843048,
|
| 1094 |
-
"eval_runtime": 13.9236,
|
| 1095 |
-
"eval_samples_per_second": 10.773,
|
| 1096 |
-
"eval_steps_per_second": 2.729,
|
| 1097 |
-
"step": 1360
|
| 1098 |
-
},
|
| 1099 |
-
{
|
| 1100 |
-
"epoch": 4.233128834355828,
|
| 1101 |
-
"grad_norm": 0.011384344659745693,
|
| 1102 |
-
"learning_rate": 3.0674846625766875e-06,
|
| 1103 |
-
"loss": 0.0006,
|
| 1104 |
-
"step": 1380
|
| 1105 |
-
},
|
| 1106 |
-
{
|
| 1107 |
-
"epoch": 4.233128834355828,
|
| 1108 |
-
"eval_accuracy": 0.9133333333333333,
|
| 1109 |
-
"eval_loss": 0.3404422998428345,
|
| 1110 |
-
"eval_runtime": 13.9552,
|
| 1111 |
-
"eval_samples_per_second": 10.749,
|
| 1112 |
-
"eval_steps_per_second": 2.723,
|
| 1113 |
-
"step": 1380
|
| 1114 |
-
},
|
| 1115 |
-
{
|
| 1116 |
-
"epoch": 4.294478527607362,
|
| 1117 |
-
"grad_norm": 0.08929850906133652,
|
| 1118 |
-
"learning_rate": 2.822085889570552e-06,
|
| 1119 |
-
"loss": 0.0193,
|
| 1120 |
-
"step": 1400
|
| 1121 |
-
},
|
| 1122 |
-
{
|
| 1123 |
-
"epoch": 4.294478527607362,
|
| 1124 |
-
"eval_accuracy": 0.92,
|
| 1125 |
-
"eval_loss": 0.3601679503917694,
|
| 1126 |
-
"eval_runtime": 13.9574,
|
| 1127 |
-
"eval_samples_per_second": 10.747,
|
| 1128 |
-
"eval_steps_per_second": 2.723,
|
| 1129 |
-
"step": 1400
|
| 1130 |
-
},
|
| 1131 |
-
{
|
| 1132 |
-
"epoch": 4.355828220858895,
|
| 1133 |
-
"grad_norm": 129.2814483642578,
|
| 1134 |
-
"learning_rate": 2.5766871165644175e-06,
|
| 1135 |
-
"loss": 0.0994,
|
| 1136 |
-
"step": 1420
|
| 1137 |
-
},
|
| 1138 |
-
{
|
| 1139 |
-
"epoch": 4.355828220858895,
|
| 1140 |
-
"eval_accuracy": 0.94,
|
| 1141 |
-
"eval_loss": 0.33025145530700684,
|
| 1142 |
-
"eval_runtime": 13.9627,
|
| 1143 |
-
"eval_samples_per_second": 10.743,
|
| 1144 |
-
"eval_steps_per_second": 2.722,
|
| 1145 |
-
"step": 1420
|
| 1146 |
-
},
|
| 1147 |
-
{
|
| 1148 |
-
"epoch": 4.41717791411043,
|
| 1149 |
-
"grad_norm": 0.024497592821717262,
|
| 1150 |
-
"learning_rate": 2.331288343558282e-06,
|
| 1151 |
-
"loss": 0.0032,
|
| 1152 |
-
"step": 1440
|
| 1153 |
-
},
|
| 1154 |
-
{
|
| 1155 |
-
"epoch": 4.41717791411043,
|
| 1156 |
-
"eval_accuracy": 0.94,
|
| 1157 |
-
"eval_loss": 0.2884907126426697,
|
| 1158 |
-
"eval_runtime": 13.9317,
|
| 1159 |
-
"eval_samples_per_second": 10.767,
|
| 1160 |
-
"eval_steps_per_second": 2.728,
|
| 1161 |
-
"step": 1440
|
| 1162 |
-
},
|
| 1163 |
-
{
|
| 1164 |
-
"epoch": 4.478527607361963,
|
| 1165 |
-
"grad_norm": 0.009228991344571114,
|
| 1166 |
-
"learning_rate": 2.085889570552147e-06,
|
| 1167 |
-
"loss": 0.0008,
|
| 1168 |
-
"step": 1460
|
| 1169 |
-
},
|
| 1170 |
-
{
|
| 1171 |
-
"epoch": 4.478527607361963,
|
| 1172 |
-
"eval_accuracy": 0.92,
|
| 1173 |
-
"eval_loss": 0.31121641397476196,
|
| 1174 |
-
"eval_runtime": 13.9417,
|
| 1175 |
-
"eval_samples_per_second": 10.759,
|
| 1176 |
-
"eval_steps_per_second": 2.726,
|
| 1177 |
-
"step": 1460
|
| 1178 |
-
},
|
| 1179 |
-
{
|
| 1180 |
-
"epoch": 4.539877300613497,
|
| 1181 |
-
"grad_norm": 0.22681556642055511,
|
| 1182 |
-
"learning_rate": 1.8404907975460124e-06,
|
| 1183 |
-
"loss": 0.0823,
|
| 1184 |
-
"step": 1480
|
| 1185 |
-
},
|
| 1186 |
-
{
|
| 1187 |
-
"epoch": 4.539877300613497,
|
| 1188 |
-
"eval_accuracy": 0.9266666666666666,
|
| 1189 |
-
"eval_loss": 0.3145321011543274,
|
| 1190 |
-
"eval_runtime": 13.9075,
|
| 1191 |
-
"eval_samples_per_second": 10.786,
|
| 1192 |
-
"eval_steps_per_second": 2.732,
|
| 1193 |
-
"step": 1480
|
| 1194 |
-
},
|
| 1195 |
-
{
|
| 1196 |
-
"epoch": 4.601226993865031,
|
| 1197 |
-
"grad_norm": 0.010360241867601871,
|
| 1198 |
-
"learning_rate": 1.5950920245398775e-06,
|
| 1199 |
-
"loss": 0.0086,
|
| 1200 |
-
"step": 1500
|
| 1201 |
-
},
|
| 1202 |
-
{
|
| 1203 |
-
"epoch": 4.601226993865031,
|
| 1204 |
-
"eval_accuracy": 0.94,
|
| 1205 |
-
"eval_loss": 0.2954442799091339,
|
| 1206 |
-
"eval_runtime": 14.0025,
|
| 1207 |
-
"eval_samples_per_second": 10.712,
|
| 1208 |
-
"eval_steps_per_second": 2.714,
|
| 1209 |
-
"step": 1500
|
| 1210 |
-
},
|
| 1211 |
-
{
|
| 1212 |
-
"epoch": 4.662576687116564,
|
| 1213 |
-
"grad_norm": 0.00811721384525299,
|
| 1214 |
-
"learning_rate": 1.3496932515337425e-06,
|
| 1215 |
-
"loss": 0.0009,
|
| 1216 |
-
"step": 1520
|
| 1217 |
-
},
|
| 1218 |
-
{
|
| 1219 |
-
"epoch": 4.662576687116564,
|
| 1220 |
-
"eval_accuracy": 0.94,
|
| 1221 |
-
"eval_loss": 0.30819201469421387,
|
| 1222 |
-
"eval_runtime": 14.1712,
|
| 1223 |
-
"eval_samples_per_second": 10.585,
|
| 1224 |
-
"eval_steps_per_second": 2.681,
|
| 1225 |
-
"step": 1520
|
| 1226 |
-
},
|
| 1227 |
-
{
|
| 1228 |
-
"epoch": 4.723926380368098,
|
| 1229 |
-
"grad_norm": 0.028587836772203445,
|
| 1230 |
-
"learning_rate": 1.1042944785276075e-06,
|
| 1231 |
-
"loss": 0.1619,
|
| 1232 |
-
"step": 1540
|
| 1233 |
-
},
|
| 1234 |
-
{
|
| 1235 |
-
"epoch": 4.723926380368098,
|
| 1236 |
-
"eval_accuracy": 0.94,
|
| 1237 |
-
"eval_loss": 0.29277047514915466,
|
| 1238 |
-
"eval_runtime": 14.0163,
|
| 1239 |
-
"eval_samples_per_second": 10.702,
|
| 1240 |
-
"eval_steps_per_second": 2.711,
|
| 1241 |
-
"step": 1540
|
| 1242 |
-
},
|
| 1243 |
-
{
|
| 1244 |
-
"epoch": 4.785276073619632,
|
| 1245 |
-
"grad_norm": 0.009555824100971222,
|
| 1246 |
-
"learning_rate": 8.588957055214725e-07,
|
| 1247 |
-
"loss": 0.0004,
|
| 1248 |
-
"step": 1560
|
| 1249 |
-
},
|
| 1250 |
-
{
|
| 1251 |
-
"epoch": 4.785276073619632,
|
| 1252 |
-
"eval_accuracy": 0.9333333333333333,
|
| 1253 |
-
"eval_loss": 0.2908971905708313,
|
| 1254 |
-
"eval_runtime": 14.0147,
|
| 1255 |
-
"eval_samples_per_second": 10.703,
|
| 1256 |
-
"eval_steps_per_second": 2.711,
|
| 1257 |
-
"step": 1560
|
| 1258 |
-
},
|
| 1259 |
-
{
|
| 1260 |
-
"epoch": 4.846625766871165,
|
| 1261 |
-
"grad_norm": 0.031364671885967255,
|
| 1262 |
-
"learning_rate": 6.134969325153375e-07,
|
| 1263 |
-
"loss": 0.0006,
|
| 1264 |
-
"step": 1580
|
| 1265 |
-
},
|
| 1266 |
-
{
|
| 1267 |
-
"epoch": 4.846625766871165,
|
| 1268 |
-
"eval_accuracy": 0.9466666666666667,
|
| 1269 |
-
"eval_loss": 0.28794920444488525,
|
| 1270 |
-
"eval_runtime": 14.0237,
|
| 1271 |
-
"eval_samples_per_second": 10.696,
|
| 1272 |
-
"eval_steps_per_second": 2.71,
|
| 1273 |
-
"step": 1580
|
| 1274 |
-
},
|
| 1275 |
-
{
|
| 1276 |
-
"epoch": 4.9079754601226995,
|
| 1277 |
-
"grad_norm": 0.005726322531700134,
|
| 1278 |
-
"learning_rate": 3.680981595092025e-07,
|
| 1279 |
-
"loss": 0.0005,
|
| 1280 |
-
"step": 1600
|
| 1281 |
-
},
|
| 1282 |
-
{
|
| 1283 |
-
"epoch": 4.9079754601226995,
|
| 1284 |
-
"eval_accuracy": 0.9466666666666667,
|
| 1285 |
-
"eval_loss": 0.28943532705307007,
|
| 1286 |
-
"eval_runtime": 13.9811,
|
| 1287 |
-
"eval_samples_per_second": 10.729,
|
| 1288 |
-
"eval_steps_per_second": 2.718,
|
| 1289 |
-
"step": 1600
|
| 1290 |
-
},
|
| 1291 |
-
{
|
| 1292 |
-
"epoch": 4.969325153374233,
|
| 1293 |
-
"grad_norm": 0.019030340015888214,
|
| 1294 |
-
"learning_rate": 1.226993865030675e-07,
|
| 1295 |
-
"loss": 0.0559,
|
| 1296 |
-
"step": 1620
|
| 1297 |
-
},
|
| 1298 |
-
{
|
| 1299 |
-
"epoch": 4.969325153374233,
|
| 1300 |
-
"eval_accuracy": 0.9466666666666667,
|
| 1301 |
-
"eval_loss": 0.2903057932853699,
|
| 1302 |
-
"eval_runtime": 14.0064,
|
| 1303 |
-
"eval_samples_per_second": 10.709,
|
| 1304 |
-
"eval_steps_per_second": 2.713,
|
| 1305 |
-
"step": 1620
|
| 1306 |
}
|
| 1307 |
],
|
| 1308 |
"logging_steps": 20,
|
| 1309 |
-
"max_steps":
|
| 1310 |
"num_input_tokens_seen": 0,
|
| 1311 |
"num_train_epochs": 5,
|
| 1312 |
"save_steps": 500,
|
|
@@ -1317,12 +421,12 @@
|
|
| 1317 |
"should_evaluate": false,
|
| 1318 |
"should_log": false,
|
| 1319 |
"should_save": true,
|
| 1320 |
-
"should_training_stop":
|
| 1321 |
},
|
| 1322 |
"attributes": {}
|
| 1323 |
}
|
| 1324 |
},
|
| 1325 |
-
"total_flos":
|
| 1326 |
"train_batch_size": 4,
|
| 1327 |
"trial_name": null,
|
| 1328 |
"trial_params": null
|
|
|
|
| 1 |
{
|
| 2 |
+
"best_metric": 0.2986587882041931,
|
| 3 |
+
"best_model_checkpoint": "./results/checkpoint-180",
|
| 4 |
+
"epoch": 1.510574018126888,
|
| 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": 0.06042296072507553,
|
| 13 |
+
"grad_norm": 0.5974797606468201,
|
| 14 |
+
"learning_rate": 1.97583081570997e-05,
|
| 15 |
+
"loss": 0.5426,
|
| 16 |
"step": 20
|
| 17 |
},
|
| 18 |
{
|
| 19 |
+
"epoch": 0.06042296072507553,
|
| 20 |
+
"eval_accuracy": 0.7450980392156863,
|
| 21 |
+
"eval_loss": 0.7869178652763367,
|
| 22 |
+
"eval_runtime": 15.0039,
|
| 23 |
+
"eval_samples_per_second": 10.197,
|
| 24 |
"eval_steps_per_second": 2.599,
|
| 25 |
"step": 20
|
| 26 |
},
|
| 27 |
{
|
| 28 |
+
"epoch": 0.12084592145015106,
|
| 29 |
+
"grad_norm": 13.038370132446289,
|
| 30 |
+
"learning_rate": 1.9516616314199397e-05,
|
| 31 |
+
"loss": 0.5984,
|
| 32 |
"step": 40
|
| 33 |
},
|
| 34 |
{
|
| 35 |
+
"epoch": 0.12084592145015106,
|
| 36 |
+
"eval_accuracy": 0.7908496732026143,
|
| 37 |
+
"eval_loss": 0.39430469274520874,
|
| 38 |
+
"eval_runtime": 15.9435,
|
| 39 |
+
"eval_samples_per_second": 9.596,
|
| 40 |
+
"eval_steps_per_second": 2.446,
|
| 41 |
"step": 40
|
| 42 |
},
|
| 43 |
{
|
| 44 |
+
"epoch": 0.18126888217522658,
|
| 45 |
+
"grad_norm": 0.1351632922887802,
|
| 46 |
+
"learning_rate": 1.9274924471299096e-05,
|
| 47 |
+
"loss": 0.4864,
|
| 48 |
"step": 60
|
| 49 |
},
|
| 50 |
{
|
| 51 |
+
"epoch": 0.18126888217522658,
|
| 52 |
+
"eval_accuracy": 0.7843137254901961,
|
| 53 |
+
"eval_loss": 0.9364686608314514,
|
| 54 |
+
"eval_runtime": 15.2263,
|
| 55 |
+
"eval_samples_per_second": 10.048,
|
| 56 |
+
"eval_steps_per_second": 2.561,
|
| 57 |
"step": 60
|
| 58 |
},
|
| 59 |
{
|
| 60 |
+
"epoch": 0.24169184290030213,
|
| 61 |
+
"grad_norm": 1.4119517803192139,
|
| 62 |
+
"learning_rate": 1.9033232628398792e-05,
|
| 63 |
+
"loss": 0.6039,
|
| 64 |
"step": 80
|
| 65 |
},
|
| 66 |
{
|
| 67 |
+
"epoch": 0.24169184290030213,
|
| 68 |
+
"eval_accuracy": 0.7712418300653595,
|
| 69 |
+
"eval_loss": 0.6580381989479065,
|
| 70 |
+
"eval_runtime": 15.5472,
|
| 71 |
+
"eval_samples_per_second": 9.841,
|
| 72 |
+
"eval_steps_per_second": 2.508,
|
| 73 |
"step": 80
|
| 74 |
},
|
| 75 |
{
|
| 76 |
+
"epoch": 0.3021148036253776,
|
| 77 |
+
"grad_norm": 13.04010009765625,
|
| 78 |
+
"learning_rate": 1.879154078549849e-05,
|
| 79 |
+
"loss": 0.5741,
|
| 80 |
"step": 100
|
| 81 |
},
|
| 82 |
{
|
| 83 |
+
"epoch": 0.3021148036253776,
|
| 84 |
+
"eval_accuracy": 0.8235294117647058,
|
| 85 |
+
"eval_loss": 0.3454345464706421,
|
| 86 |
+
"eval_runtime": 15.5082,
|
| 87 |
+
"eval_samples_per_second": 9.866,
|
| 88 |
+
"eval_steps_per_second": 2.515,
|
| 89 |
"step": 100
|
| 90 |
},
|
| 91 |
{
|
| 92 |
+
"epoch": 0.36253776435045315,
|
| 93 |
+
"grad_norm": 61.882415771484375,
|
| 94 |
+
"learning_rate": 1.854984894259819e-05,
|
| 95 |
+
"loss": 0.4276,
|
| 96 |
"step": 120
|
| 97 |
},
|
| 98 |
{
|
| 99 |
+
"epoch": 0.36253776435045315,
|
| 100 |
+
"eval_accuracy": 0.8169934640522876,
|
| 101 |
+
"eval_loss": 0.5421260595321655,
|
| 102 |
+
"eval_runtime": 15.4627,
|
| 103 |
+
"eval_samples_per_second": 9.895,
|
| 104 |
+
"eval_steps_per_second": 2.522,
|
| 105 |
"step": 120
|
| 106 |
},
|
| 107 |
{
|
| 108 |
+
"epoch": 0.4229607250755287,
|
| 109 |
+
"grad_norm": 8.9187593460083,
|
| 110 |
+
"learning_rate": 1.8308157099697886e-05,
|
| 111 |
+
"loss": 0.4342,
|
| 112 |
"step": 140
|
| 113 |
},
|
| 114 |
{
|
| 115 |
+
"epoch": 0.4229607250755287,
|
| 116 |
+
"eval_accuracy": 0.8562091503267973,
|
| 117 |
+
"eval_loss": 0.4258342981338501,
|
| 118 |
+
"eval_runtime": 15.4753,
|
| 119 |
+
"eval_samples_per_second": 9.887,
|
| 120 |
+
"eval_steps_per_second": 2.52,
|
| 121 |
"step": 140
|
| 122 |
},
|
| 123 |
{
|
| 124 |
+
"epoch": 0.48338368580060426,
|
| 125 |
+
"grad_norm": 40.476078033447266,
|
| 126 |
+
"learning_rate": 1.8066465256797586e-05,
|
| 127 |
+
"loss": 0.4915,
|
| 128 |
"step": 160
|
| 129 |
},
|
| 130 |
{
|
| 131 |
+
"epoch": 0.48338368580060426,
|
| 132 |
+
"eval_accuracy": 0.8300653594771242,
|
| 133 |
+
"eval_loss": 0.5960604548454285,
|
| 134 |
+
"eval_runtime": 15.4631,
|
| 135 |
+
"eval_samples_per_second": 9.895,
|
| 136 |
+
"eval_steps_per_second": 2.522,
|
| 137 |
"step": 160
|
| 138 |
},
|
| 139 |
{
|
| 140 |
+
"epoch": 0.5438066465256798,
|
| 141 |
+
"grad_norm": 3.8627891540527344,
|
| 142 |
+
"learning_rate": 1.782477341389728e-05,
|
| 143 |
+
"loss": 0.4127,
|
| 144 |
"step": 180
|
| 145 |
},
|
| 146 |
{
|
| 147 |
+
"epoch": 0.5438066465256798,
|
| 148 |
+
"eval_accuracy": 0.869281045751634,
|
| 149 |
+
"eval_loss": 0.2986587882041931,
|
| 150 |
+
"eval_runtime": 15.5147,
|
| 151 |
+
"eval_samples_per_second": 9.862,
|
| 152 |
+
"eval_steps_per_second": 2.514,
|
| 153 |
"step": 180
|
| 154 |
},
|
| 155 |
{
|
| 156 |
+
"epoch": 0.6042296072507553,
|
| 157 |
+
"grad_norm": 16.525339126586914,
|
| 158 |
+
"learning_rate": 1.758308157099698e-05,
|
| 159 |
+
"loss": 0.3166,
|
| 160 |
"step": 200
|
| 161 |
},
|
| 162 |
{
|
| 163 |
+
"epoch": 0.6042296072507553,
|
| 164 |
+
"eval_accuracy": 0.869281045751634,
|
| 165 |
+
"eval_loss": 0.33075031638145447,
|
| 166 |
+
"eval_runtime": 15.5417,
|
| 167 |
+
"eval_samples_per_second": 9.845,
|
| 168 |
+
"eval_steps_per_second": 2.509,
|
| 169 |
"step": 200
|
| 170 |
},
|
| 171 |
{
|
| 172 |
+
"epoch": 0.6646525679758308,
|
| 173 |
+
"grad_norm": 1.5080480575561523,
|
| 174 |
+
"learning_rate": 1.7341389728096677e-05,
|
| 175 |
+
"loss": 0.4018,
|
| 176 |
"step": 220
|
| 177 |
},
|
| 178 |
{
|
| 179 |
+
"epoch": 0.6646525679758308,
|
| 180 |
+
"eval_accuracy": 0.803921568627451,
|
| 181 |
+
"eval_loss": 0.5285586714744568,
|
| 182 |
+
"eval_runtime": 15.4908,
|
| 183 |
+
"eval_samples_per_second": 9.877,
|
| 184 |
+
"eval_steps_per_second": 2.518,
|
| 185 |
"step": 220
|
| 186 |
},
|
| 187 |
{
|
| 188 |
+
"epoch": 0.7250755287009063,
|
| 189 |
+
"grad_norm": 21.941341400146484,
|
| 190 |
+
"learning_rate": 1.7099697885196376e-05,
|
| 191 |
+
"loss": 0.3007,
|
| 192 |
"step": 240
|
| 193 |
},
|
| 194 |
{
|
| 195 |
+
"epoch": 0.7250755287009063,
|
| 196 |
+
"eval_accuracy": 0.8627450980392157,
|
| 197 |
+
"eval_loss": 0.584548830986023,
|
| 198 |
+
"eval_runtime": 15.5027,
|
| 199 |
+
"eval_samples_per_second": 9.869,
|
| 200 |
+
"eval_steps_per_second": 2.516,
|
| 201 |
"step": 240
|
| 202 |
},
|
| 203 |
{
|
| 204 |
+
"epoch": 0.7854984894259819,
|
| 205 |
+
"grad_norm": 7.442047119140625,
|
| 206 |
+
"learning_rate": 1.6858006042296072e-05,
|
| 207 |
+
"loss": 0.4893,
|
| 208 |
"step": 260
|
| 209 |
},
|
| 210 |
{
|
| 211 |
+
"epoch": 0.7854984894259819,
|
| 212 |
+
"eval_accuracy": 0.8627450980392157,
|
| 213 |
+
"eval_loss": 0.36624589562416077,
|
| 214 |
+
"eval_runtime": 15.471,
|
| 215 |
+
"eval_samples_per_second": 9.889,
|
| 216 |
+
"eval_steps_per_second": 2.521,
|
| 217 |
"step": 260
|
| 218 |
},
|
| 219 |
{
|
| 220 |
+
"epoch": 0.8459214501510574,
|
| 221 |
+
"grad_norm": 27.062305450439453,
|
| 222 |
+
"learning_rate": 1.661631419939577e-05,
|
| 223 |
+
"loss": 0.274,
|
| 224 |
"step": 280
|
| 225 |
},
|
| 226 |
{
|
| 227 |
+
"epoch": 0.8459214501510574,
|
| 228 |
+
"eval_accuracy": 0.869281045751634,
|
| 229 |
+
"eval_loss": 0.34829556941986084,
|
| 230 |
+
"eval_runtime": 15.4719,
|
| 231 |
+
"eval_samples_per_second": 9.889,
|
| 232 |
+
"eval_steps_per_second": 2.521,
|
| 233 |
"step": 280
|
| 234 |
},
|
| 235 |
{
|
| 236 |
+
"epoch": 0.9063444108761329,
|
| 237 |
+
"grad_norm": 9.517237663269043,
|
| 238 |
+
"learning_rate": 1.637462235649547e-05,
|
| 239 |
+
"loss": 0.5741,
|
| 240 |
"step": 300
|
| 241 |
},
|
| 242 |
{
|
| 243 |
+
"epoch": 0.9063444108761329,
|
| 244 |
+
"eval_accuracy": 0.8823529411764706,
|
| 245 |
+
"eval_loss": 0.32800009846687317,
|
| 246 |
+
"eval_runtime": 15.4828,
|
| 247 |
+
"eval_samples_per_second": 9.882,
|
| 248 |
+
"eval_steps_per_second": 2.519,
|
| 249 |
"step": 300
|
| 250 |
},
|
| 251 |
{
|
| 252 |
+
"epoch": 0.9667673716012085,
|
| 253 |
+
"grad_norm": 40.94089889526367,
|
| 254 |
+
"learning_rate": 1.6132930513595166e-05,
|
| 255 |
+
"loss": 0.3752,
|
| 256 |
"step": 320
|
| 257 |
},
|
| 258 |
{
|
| 259 |
+
"epoch": 0.9667673716012085,
|
| 260 |
+
"eval_accuracy": 0.8888888888888888,
|
| 261 |
+
"eval_loss": 0.5250552296638489,
|
| 262 |
+
"eval_runtime": 15.485,
|
| 263 |
+
"eval_samples_per_second": 9.881,
|
| 264 |
+
"eval_steps_per_second": 2.519,
|
| 265 |
"step": 320
|
| 266 |
},
|
| 267 |
{
|
| 268 |
+
"epoch": 1.027190332326284,
|
| 269 |
+
"grad_norm": 0.2767094373703003,
|
| 270 |
+
"learning_rate": 1.5891238670694865e-05,
|
| 271 |
+
"loss": 0.2711,
|
| 272 |
"step": 340
|
| 273 |
},
|
| 274 |
{
|
| 275 |
+
"epoch": 1.027190332326284,
|
| 276 |
+
"eval_accuracy": 0.8562091503267973,
|
| 277 |
+
"eval_loss": 0.6096686720848083,
|
| 278 |
+
"eval_runtime": 15.4954,
|
| 279 |
+
"eval_samples_per_second": 9.874,
|
| 280 |
+
"eval_steps_per_second": 2.517,
|
| 281 |
"step": 340
|
| 282 |
},
|
| 283 |
{
|
| 284 |
+
"epoch": 1.0876132930513596,
|
| 285 |
+
"grad_norm": 1.7755597829818726,
|
| 286 |
+
"learning_rate": 1.5649546827794565e-05,
|
| 287 |
+
"loss": 0.2369,
|
| 288 |
"step": 360
|
| 289 |
},
|
| 290 |
{
|
| 291 |
+
"epoch": 1.0876132930513596,
|
| 292 |
+
"eval_accuracy": 0.869281045751634,
|
| 293 |
+
"eval_loss": 0.5457373857498169,
|
| 294 |
+
"eval_runtime": 15.479,
|
| 295 |
+
"eval_samples_per_second": 9.884,
|
| 296 |
+
"eval_steps_per_second": 2.52,
|
| 297 |
"step": 360
|
| 298 |
},
|
| 299 |
{
|
| 300 |
+
"epoch": 1.148036253776435,
|
| 301 |
+
"grad_norm": 197.30111694335938,
|
| 302 |
+
"learning_rate": 1.540785498489426e-05,
|
| 303 |
+
"loss": 0.3756,
|
| 304 |
"step": 380
|
| 305 |
},
|
| 306 |
{
|
| 307 |
+
"epoch": 1.148036253776435,
|
| 308 |
+
"eval_accuracy": 0.8758169934640523,
|
| 309 |
+
"eval_loss": 0.6890403628349304,
|
| 310 |
+
"eval_runtime": 15.448,
|
| 311 |
+
"eval_samples_per_second": 9.904,
|
| 312 |
+
"eval_steps_per_second": 2.525,
|
| 313 |
"step": 380
|
| 314 |
},
|
| 315 |
{
|
| 316 |
+
"epoch": 1.2084592145015105,
|
| 317 |
+
"grad_norm": 0.16098028421401978,
|
| 318 |
+
"learning_rate": 1.516616314199396e-05,
|
| 319 |
+
"loss": 0.6575,
|
| 320 |
"step": 400
|
| 321 |
},
|
| 322 |
{
|
| 323 |
+
"epoch": 1.2084592145015105,
|
| 324 |
+
"eval_accuracy": 0.869281045751634,
|
| 325 |
+
"eval_loss": 0.47085481882095337,
|
| 326 |
+
"eval_runtime": 15.4824,
|
| 327 |
+
"eval_samples_per_second": 9.882,
|
| 328 |
+
"eval_steps_per_second": 2.519,
|
| 329 |
"step": 400
|
| 330 |
},
|
| 331 |
{
|
| 332 |
+
"epoch": 1.2688821752265862,
|
| 333 |
+
"grad_norm": 0.3975774943828583,
|
| 334 |
+
"learning_rate": 1.4924471299093657e-05,
|
| 335 |
+
"loss": 0.3268,
|
| 336 |
"step": 420
|
| 337 |
},
|
| 338 |
{
|
| 339 |
+
"epoch": 1.2688821752265862,
|
| 340 |
+
"eval_accuracy": 0.8496732026143791,
|
| 341 |
+
"eval_loss": 0.5218892693519592,
|
| 342 |
+
"eval_runtime": 15.4867,
|
| 343 |
+
"eval_samples_per_second": 9.879,
|
| 344 |
+
"eval_steps_per_second": 2.518,
|
| 345 |
"step": 420
|
| 346 |
},
|
| 347 |
{
|
| 348 |
+
"epoch": 1.3293051359516617,
|
| 349 |
+
"grad_norm": 8.90912914276123,
|
| 350 |
+
"learning_rate": 1.4682779456193355e-05,
|
| 351 |
+
"loss": 0.3994,
|
| 352 |
"step": 440
|
| 353 |
},
|
| 354 |
{
|
| 355 |
+
"epoch": 1.3293051359516617,
|
| 356 |
+
"eval_accuracy": 0.869281045751634,
|
| 357 |
+
"eval_loss": 0.42816421389579773,
|
| 358 |
+
"eval_runtime": 15.4691,
|
| 359 |
+
"eval_samples_per_second": 9.891,
|
| 360 |
+
"eval_steps_per_second": 2.521,
|
| 361 |
"step": 440
|
| 362 |
},
|
| 363 |
{
|
| 364 |
+
"epoch": 1.3897280966767371,
|
| 365 |
+
"grad_norm": 93.57767486572266,
|
| 366 |
+
"learning_rate": 1.4441087613293053e-05,
|
| 367 |
+
"loss": 0.0879,
|
| 368 |
"step": 460
|
| 369 |
},
|
| 370 |
{
|
| 371 |
+
"epoch": 1.3897280966767371,
|
| 372 |
+
"eval_accuracy": 0.8758169934640523,
|
| 373 |
+
"eval_loss": 0.6293966174125671,
|
| 374 |
+
"eval_runtime": 15.4702,
|
| 375 |
+
"eval_samples_per_second": 9.89,
|
| 376 |
+
"eval_steps_per_second": 2.521,
|
| 377 |
"step": 460
|
| 378 |
},
|
| 379 |
{
|
| 380 |
+
"epoch": 1.4501510574018126,
|
| 381 |
+
"grad_norm": 0.061734456568956375,
|
| 382 |
+
"learning_rate": 1.419939577039275e-05,
|
| 383 |
+
"loss": 0.2566,
|
| 384 |
"step": 480
|
| 385 |
},
|
| 386 |
{
|
| 387 |
+
"epoch": 1.4501510574018126,
|
| 388 |
+
"eval_accuracy": 0.8627450980392157,
|
| 389 |
+
"eval_loss": 0.7143150568008423,
|
| 390 |
+
"eval_runtime": 15.449,
|
| 391 |
+
"eval_samples_per_second": 9.904,
|
| 392 |
+
"eval_steps_per_second": 2.524,
|
| 393 |
"step": 480
|
| 394 |
},
|
| 395 |
{
|
| 396 |
+
"epoch": 1.510574018126888,
|
| 397 |
+
"grad_norm": 0.11357846111059189,
|
| 398 |
+
"learning_rate": 1.3957703927492448e-05,
|
| 399 |
+
"loss": 0.2897,
|
| 400 |
"step": 500
|
| 401 |
},
|
| 402 |
{
|
| 403 |
+
"epoch": 1.510574018126888,
|
| 404 |
+
"eval_accuracy": 0.869281045751634,
|
| 405 |
+
"eval_loss": 0.6120204329490662,
|
| 406 |
+
"eval_runtime": 15.463,
|
| 407 |
+
"eval_samples_per_second": 9.895,
|
| 408 |
+
"eval_steps_per_second": 2.522,
|
| 409 |
"step": 500
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 410 |
}
|
| 411 |
],
|
| 412 |
"logging_steps": 20,
|
| 413 |
+
"max_steps": 1655,
|
| 414 |
"num_input_tokens_seen": 0,
|
| 415 |
"num_train_epochs": 5,
|
| 416 |
"save_steps": 500,
|
|
|
|
| 421 |
"should_evaluate": false,
|
| 422 |
"should_log": false,
|
| 423 |
"should_save": true,
|
| 424 |
+
"should_training_stop": false
|
| 425 |
},
|
| 426 |
"attributes": {}
|
| 427 |
}
|
| 428 |
},
|
| 429 |
+
"total_flos": 1604904119249676.0,
|
| 430 |
"train_batch_size": 4,
|
| 431 |
"trial_name": null,
|
| 432 |
"trial_params": null
|
last-checkpoint/training_args.bin
CHANGED
|
@@ -1,3 +1,3 @@
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
-
oid sha256:
|
| 3 |
size 5368
|
|
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:86f09b33190e89a5f43d415554f33c8d75ab6485f60f2b9e3aec61203e52e1d9
|
| 3 |
size 5368
|