Model save
Browse files- all_results.json +5 -5
- train_results.json +5 -5
- trainer_state.json +51 -419
all_results.json
CHANGED
|
@@ -1,8 +1,8 @@
|
|
| 1 |
{
|
| 2 |
-
"epoch":
|
| 3 |
-
"train_loss": 0.
|
| 4 |
-
"train_runtime":
|
| 5 |
"train_samples": 6599,
|
| 6 |
-
"train_samples_per_second":
|
| 7 |
-
"train_steps_per_second": 0.
|
| 8 |
}
|
|
|
|
| 1 |
{
|
| 2 |
+
"epoch": 1.0,
|
| 3 |
+
"train_loss": 0.908859632783017,
|
| 4 |
+
"train_runtime": 1626.0912,
|
| 5 |
"train_samples": 6599,
|
| 6 |
+
"train_samples_per_second": 4.625,
|
| 7 |
+
"train_steps_per_second": 0.036
|
| 8 |
}
|
train_results.json
CHANGED
|
@@ -1,8 +1,8 @@
|
|
| 1 |
{
|
| 2 |
-
"epoch":
|
| 3 |
-
"train_loss": 0.
|
| 4 |
-
"train_runtime":
|
| 5 |
"train_samples": 6599,
|
| 6 |
-
"train_samples_per_second":
|
| 7 |
-
"train_steps_per_second": 0.
|
| 8 |
}
|
|
|
|
| 1 |
{
|
| 2 |
+
"epoch": 1.0,
|
| 3 |
+
"train_loss": 0.908859632783017,
|
| 4 |
+
"train_runtime": 1626.0912,
|
| 5 |
"train_samples": 6599,
|
| 6 |
+
"train_samples_per_second": 4.625,
|
| 7 |
+
"train_steps_per_second": 0.036
|
| 8 |
}
|
trainer_state.json
CHANGED
|
@@ -1,489 +1,121 @@
|
|
| 1 |
{
|
| 2 |
"best_metric": null,
|
| 3 |
"best_model_checkpoint": null,
|
| 4 |
-
"epoch":
|
| 5 |
"eval_steps": 500,
|
| 6 |
-
"global_step":
|
| 7 |
"is_hyper_param_search": false,
|
| 8 |
"is_local_process_zero": true,
|
| 9 |
"is_world_process_zero": true,
|
| 10 |
"log_history": [
|
| 11 |
{
|
| 12 |
"epoch": 0.02,
|
| 13 |
-
"grad_norm": 4.
|
| 14 |
-
"learning_rate":
|
| 15 |
"loss": 0.9545,
|
| 16 |
"step": 1
|
| 17 |
},
|
| 18 |
{
|
| 19 |
"epoch": 0.08,
|
| 20 |
-
"grad_norm":
|
| 21 |
-
"learning_rate":
|
| 22 |
-
"loss": 0.
|
| 23 |
"step": 5
|
| 24 |
},
|
| 25 |
{
|
| 26 |
"epoch": 0.17,
|
| 27 |
-
"grad_norm":
|
| 28 |
-
"learning_rate":
|
| 29 |
-
"loss":
|
| 30 |
"step": 10
|
| 31 |
},
|
| 32 |
{
|
| 33 |
"epoch": 0.25,
|
| 34 |
-
"grad_norm":
|
| 35 |
-
"learning_rate":
|
| 36 |
-
"loss":
|
| 37 |
"step": 15
|
| 38 |
},
|
| 39 |
{
|
| 40 |
"epoch": 0.34,
|
| 41 |
-
"grad_norm":
|
| 42 |
-
"learning_rate": 1.
|
| 43 |
-
"loss": 0.
|
| 44 |
"step": 20
|
| 45 |
},
|
| 46 |
{
|
| 47 |
"epoch": 0.42,
|
| 48 |
-
"grad_norm": 2.
|
| 49 |
-
"learning_rate": 1.
|
| 50 |
-
"loss": 0.
|
| 51 |
"step": 25
|
| 52 |
},
|
| 53 |
{
|
| 54 |
"epoch": 0.51,
|
| 55 |
-
"grad_norm": 2.
|
| 56 |
-
"learning_rate":
|
| 57 |
-
"loss": 0.
|
| 58 |
"step": 30
|
| 59 |
},
|
| 60 |
{
|
| 61 |
"epoch": 0.59,
|
| 62 |
-
"grad_norm": 2.
|
| 63 |
-
"learning_rate":
|
| 64 |
-
"loss": 0.
|
| 65 |
"step": 35
|
| 66 |
},
|
| 67 |
{
|
| 68 |
"epoch": 0.68,
|
| 69 |
-
"grad_norm":
|
| 70 |
-
"learning_rate":
|
| 71 |
-
"loss": 0.
|
| 72 |
"step": 40
|
| 73 |
},
|
| 74 |
{
|
| 75 |
"epoch": 0.76,
|
| 76 |
-
"grad_norm":
|
| 77 |
-
"learning_rate":
|
| 78 |
-
"loss": 0.
|
| 79 |
"step": 45
|
| 80 |
},
|
| 81 |
{
|
| 82 |
"epoch": 0.85,
|
| 83 |
-
"grad_norm":
|
| 84 |
-
"learning_rate": 1.
|
| 85 |
-
"loss": 0.
|
| 86 |
"step": 50
|
| 87 |
},
|
| 88 |
{
|
| 89 |
"epoch": 0.93,
|
| 90 |
-
"grad_norm":
|
| 91 |
-
"learning_rate":
|
| 92 |
-
"loss": 0.
|
| 93 |
"step": 55
|
| 94 |
},
|
| 95 |
{
|
| 96 |
"epoch": 1.0,
|
| 97 |
-
"eval_loss": 0.
|
| 98 |
-
"eval_runtime":
|
| 99 |
-
"eval_samples_per_second": 22.
|
| 100 |
-
"eval_steps_per_second": 0.
|
| 101 |
"step": 59
|
| 102 |
},
|
| 103 |
{
|
| 104 |
-
"epoch": 1.
|
| 105 |
-
"
|
| 106 |
-
"
|
| 107 |
-
"
|
| 108 |
-
"
|
| 109 |
-
|
| 110 |
-
|
| 111 |
-
"epoch": 1.1,
|
| 112 |
-
"grad_norm": 3.1379021048274596,
|
| 113 |
-
"learning_rate": 1.9151456172430186e-05,
|
| 114 |
-
"loss": 0.4188,
|
| 115 |
-
"step": 65
|
| 116 |
-
},
|
| 117 |
-
{
|
| 118 |
-
"epoch": 1.19,
|
| 119 |
-
"grad_norm": 2.478413663736477,
|
| 120 |
-
"learning_rate": 1.8896570909947477e-05,
|
| 121 |
-
"loss": 0.4197,
|
| 122 |
-
"step": 70
|
| 123 |
-
},
|
| 124 |
-
{
|
| 125 |
-
"epoch": 1.27,
|
| 126 |
-
"grad_norm": 3.1965773364967114,
|
| 127 |
-
"learning_rate": 1.8610436117673557e-05,
|
| 128 |
-
"loss": 0.3996,
|
| 129 |
-
"step": 75
|
| 130 |
-
},
|
| 131 |
-
{
|
| 132 |
-
"epoch": 1.36,
|
| 133 |
-
"grad_norm": 1.8635605447479031,
|
| 134 |
-
"learning_rate": 1.829405685450202e-05,
|
| 135 |
-
"loss": 0.4087,
|
| 136 |
-
"step": 80
|
| 137 |
-
},
|
| 138 |
-
{
|
| 139 |
-
"epoch": 1.44,
|
| 140 |
-
"grad_norm": 4.049894316503265,
|
| 141 |
-
"learning_rate": 1.7948544414133534e-05,
|
| 142 |
-
"loss": 0.4113,
|
| 143 |
-
"step": 85
|
| 144 |
-
},
|
| 145 |
-
{
|
| 146 |
-
"epoch": 1.53,
|
| 147 |
-
"grad_norm": 175.82031352565957,
|
| 148 |
-
"learning_rate": 1.7575112421616203e-05,
|
| 149 |
-
"loss": 0.7264,
|
| 150 |
-
"step": 90
|
| 151 |
-
},
|
| 152 |
-
{
|
| 153 |
-
"epoch": 1.61,
|
| 154 |
-
"grad_norm": 97.72323752808755,
|
| 155 |
-
"learning_rate": 1.717507257044331e-05,
|
| 156 |
-
"loss": 0.7019,
|
| 157 |
-
"step": 95
|
| 158 |
-
},
|
| 159 |
-
{
|
| 160 |
-
"epoch": 1.69,
|
| 161 |
-
"grad_norm": 10.439473368823062,
|
| 162 |
-
"learning_rate": 1.6749830015182106e-05,
|
| 163 |
-
"loss": 0.8288,
|
| 164 |
-
"step": 100
|
| 165 |
-
},
|
| 166 |
-
{
|
| 167 |
-
"epoch": 1.78,
|
| 168 |
-
"grad_norm": 14.65204562703183,
|
| 169 |
-
"learning_rate": 1.6300878435817115e-05,
|
| 170 |
-
"loss": 0.622,
|
| 171 |
-
"step": 105
|
| 172 |
-
},
|
| 173 |
-
{
|
| 174 |
-
"epoch": 1.86,
|
| 175 |
-
"grad_norm": 4.935916944271238,
|
| 176 |
-
"learning_rate": 1.5829794791144723e-05,
|
| 177 |
-
"loss": 0.5399,
|
| 178 |
-
"step": 110
|
| 179 |
-
},
|
| 180 |
-
{
|
| 181 |
-
"epoch": 1.95,
|
| 182 |
-
"grad_norm": 3.1671885548837544,
|
| 183 |
-
"learning_rate": 1.533823377964791e-05,
|
| 184 |
-
"loss": 0.4747,
|
| 185 |
-
"step": 115
|
| 186 |
-
},
|
| 187 |
-
{
|
| 188 |
-
"epoch": 2.0,
|
| 189 |
-
"eval_loss": 0.31807941198349,
|
| 190 |
-
"eval_runtime": 335.1235,
|
| 191 |
-
"eval_samples_per_second": 22.442,
|
| 192 |
-
"eval_steps_per_second": 0.352,
|
| 193 |
-
"step": 118
|
| 194 |
-
},
|
| 195 |
-
{
|
| 196 |
-
"epoch": 2.03,
|
| 197 |
-
"grad_norm": 2.7978128026277864,
|
| 198 |
-
"learning_rate": 1.482792202730745e-05,
|
| 199 |
-
"loss": 0.3959,
|
| 200 |
-
"step": 120
|
| 201 |
-
},
|
| 202 |
-
{
|
| 203 |
-
"epoch": 2.12,
|
| 204 |
-
"grad_norm": 2.1385098508541693,
|
| 205 |
-
"learning_rate": 1.4300652022765207e-05,
|
| 206 |
-
"loss": 0.2884,
|
| 207 |
-
"step": 125
|
| 208 |
-
},
|
| 209 |
-
{
|
| 210 |
-
"epoch": 2.2,
|
| 211 |
-
"grad_norm": 1.9703828411806736,
|
| 212 |
-
"learning_rate": 1.3758275821142382e-05,
|
| 213 |
-
"loss": 0.2745,
|
| 214 |
-
"step": 130
|
| 215 |
-
},
|
| 216 |
-
{
|
| 217 |
-
"epoch": 2.29,
|
| 218 |
-
"grad_norm": 1.870032223898805,
|
| 219 |
-
"learning_rate": 1.3202698538628376e-05,
|
| 220 |
-
"loss": 0.2539,
|
| 221 |
-
"step": 135
|
| 222 |
-
},
|
| 223 |
-
{
|
| 224 |
-
"epoch": 2.37,
|
| 225 |
-
"grad_norm": 7.255082480038806,
|
| 226 |
-
"learning_rate": 1.2635871660690677e-05,
|
| 227 |
-
"loss": 0.2588,
|
| 228 |
-
"step": 140
|
| 229 |
-
},
|
| 230 |
-
{
|
| 231 |
-
"epoch": 2.46,
|
| 232 |
-
"grad_norm": 1.7733158231347383,
|
| 233 |
-
"learning_rate": 1.2059786187410984e-05,
|
| 234 |
-
"loss": 0.2564,
|
| 235 |
-
"step": 145
|
| 236 |
-
},
|
| 237 |
-
{
|
| 238 |
-
"epoch": 2.54,
|
| 239 |
-
"grad_norm": 1.9182852956652028,
|
| 240 |
-
"learning_rate": 1.1476465640024814e-05,
|
| 241 |
-
"loss": 0.2282,
|
| 242 |
-
"step": 150
|
| 243 |
-
},
|
| 244 |
-
{
|
| 245 |
-
"epoch": 2.63,
|
| 246 |
-
"grad_norm": 1.6601552566584281,
|
| 247 |
-
"learning_rate": 1.0887958953229349e-05,
|
| 248 |
-
"loss": 0.2323,
|
| 249 |
-
"step": 155
|
| 250 |
-
},
|
| 251 |
-
{
|
| 252 |
-
"epoch": 2.71,
|
| 253 |
-
"grad_norm": 1.4919318406362796,
|
| 254 |
-
"learning_rate": 1.0296333278225599e-05,
|
| 255 |
-
"loss": 0.2354,
|
| 256 |
-
"step": 160
|
| 257 |
-
},
|
| 258 |
-
{
|
| 259 |
-
"epoch": 2.8,
|
| 260 |
-
"grad_norm": 1.5076718836100451,
|
| 261 |
-
"learning_rate": 9.703666721774403e-06,
|
| 262 |
-
"loss": 0.23,
|
| 263 |
-
"step": 165
|
| 264 |
-
},
|
| 265 |
-
{
|
| 266 |
-
"epoch": 2.88,
|
| 267 |
-
"grad_norm": 1.703882565208849,
|
| 268 |
-
"learning_rate": 9.112041046770653e-06,
|
| 269 |
-
"loss": 0.2312,
|
| 270 |
-
"step": 170
|
| 271 |
-
},
|
| 272 |
-
{
|
| 273 |
-
"epoch": 2.97,
|
| 274 |
-
"grad_norm": 1.3815318382548532,
|
| 275 |
-
"learning_rate": 8.52353435997519e-06,
|
| 276 |
-
"loss": 0.2246,
|
| 277 |
-
"step": 175
|
| 278 |
-
},
|
| 279 |
-
{
|
| 280 |
-
"epoch": 3.0,
|
| 281 |
-
"eval_loss": 0.12802040576934814,
|
| 282 |
-
"eval_runtime": 335.0335,
|
| 283 |
-
"eval_samples_per_second": 22.449,
|
| 284 |
-
"eval_steps_per_second": 0.352,
|
| 285 |
-
"step": 177
|
| 286 |
-
},
|
| 287 |
-
{
|
| 288 |
-
"epoch": 3.05,
|
| 289 |
-
"grad_norm": 1.340084619420041,
|
| 290 |
-
"learning_rate": 7.940213812589018e-06,
|
| 291 |
-
"loss": 0.1568,
|
| 292 |
-
"step": 180
|
| 293 |
-
},
|
| 294 |
-
{
|
| 295 |
-
"epoch": 3.14,
|
| 296 |
-
"grad_norm": 1.4363530293138602,
|
| 297 |
-
"learning_rate": 7.364128339309326e-06,
|
| 298 |
-
"loss": 0.1171,
|
| 299 |
-
"step": 185
|
| 300 |
-
},
|
| 301 |
-
{
|
| 302 |
-
"epoch": 3.22,
|
| 303 |
-
"grad_norm": 7.055735555930357,
|
| 304 |
-
"learning_rate": 6.797301461371626e-06,
|
| 305 |
-
"loss": 0.129,
|
| 306 |
-
"step": 190
|
| 307 |
-
},
|
| 308 |
-
{
|
| 309 |
-
"epoch": 3.31,
|
| 310 |
-
"grad_norm": 3.3324655424159606,
|
| 311 |
-
"learning_rate": 6.241724178857621e-06,
|
| 312 |
-
"loss": 0.15,
|
| 313 |
-
"step": 195
|
| 314 |
-
},
|
| 315 |
-
{
|
| 316 |
-
"epoch": 3.39,
|
| 317 |
-
"grad_norm": 2.4413759201237917,
|
| 318 |
-
"learning_rate": 5.699347977234799e-06,
|
| 319 |
-
"loss": 0.137,
|
| 320 |
-
"step": 200
|
| 321 |
-
},
|
| 322 |
-
{
|
| 323 |
-
"epoch": 3.47,
|
| 324 |
-
"grad_norm": 1.6325745641721867,
|
| 325 |
-
"learning_rate": 5.172077972692553e-06,
|
| 326 |
-
"loss": 0.1218,
|
| 327 |
-
"step": 205
|
| 328 |
-
},
|
| 329 |
-
{
|
| 330 |
-
"epoch": 3.56,
|
| 331 |
-
"grad_norm": 1.5192997389176583,
|
| 332 |
-
"learning_rate": 4.661766220352098e-06,
|
| 333 |
-
"loss": 0.1241,
|
| 334 |
-
"step": 210
|
| 335 |
-
},
|
| 336 |
-
{
|
| 337 |
-
"epoch": 3.64,
|
| 338 |
-
"grad_norm": 1.416532576646694,
|
| 339 |
-
"learning_rate": 4.170205208855281e-06,
|
| 340 |
-
"loss": 0.1163,
|
| 341 |
-
"step": 215
|
| 342 |
-
},
|
| 343 |
-
{
|
| 344 |
-
"epoch": 3.73,
|
| 345 |
-
"grad_norm": 1.1859416249764543,
|
| 346 |
-
"learning_rate": 3.6991215641828903e-06,
|
| 347 |
-
"loss": 0.1212,
|
| 348 |
-
"step": 220
|
| 349 |
-
},
|
| 350 |
-
{
|
| 351 |
-
"epoch": 3.81,
|
| 352 |
-
"grad_norm": 1.2425425377133077,
|
| 353 |
-
"learning_rate": 3.250169984817897e-06,
|
| 354 |
-
"loss": 0.1261,
|
| 355 |
-
"step": 225
|
| 356 |
-
},
|
| 357 |
-
{
|
| 358 |
-
"epoch": 3.9,
|
| 359 |
-
"grad_norm": 1.197573551306019,
|
| 360 |
-
"learning_rate": 2.8249274295566863e-06,
|
| 361 |
-
"loss": 0.1206,
|
| 362 |
-
"step": 230
|
| 363 |
-
},
|
| 364 |
-
{
|
| 365 |
-
"epoch": 3.98,
|
| 366 |
-
"grad_norm": 1.2203991841318311,
|
| 367 |
-
"learning_rate": 2.424887578383799e-06,
|
| 368 |
-
"loss": 0.1179,
|
| 369 |
-
"step": 235
|
| 370 |
-
},
|
| 371 |
-
{
|
| 372 |
-
"epoch": 4.0,
|
| 373 |
-
"eval_loss": 0.08678202331066132,
|
| 374 |
-
"eval_runtime": 334.8445,
|
| 375 |
-
"eval_samples_per_second": 22.461,
|
| 376 |
-
"eval_steps_per_second": 0.352,
|
| 377 |
-
"step": 236
|
| 378 |
-
},
|
| 379 |
-
{
|
| 380 |
-
"epoch": 4.07,
|
| 381 |
-
"grad_norm": 1.3225710226725387,
|
| 382 |
-
"learning_rate": 2.0514555858664663e-06,
|
| 383 |
-
"loss": 0.0919,
|
| 384 |
-
"step": 240
|
| 385 |
-
},
|
| 386 |
-
{
|
| 387 |
-
"epoch": 4.15,
|
| 388 |
-
"grad_norm": 1.2828578734194267,
|
| 389 |
-
"learning_rate": 1.7059431454979825e-06,
|
| 390 |
-
"loss": 0.0833,
|
| 391 |
-
"step": 245
|
| 392 |
-
},
|
| 393 |
-
{
|
| 394 |
-
"epoch": 4.24,
|
| 395 |
-
"grad_norm": 1.1833858250329508,
|
| 396 |
-
"learning_rate": 1.3895638823264447e-06,
|
| 397 |
-
"loss": 0.0861,
|
| 398 |
-
"step": 250
|
| 399 |
-
},
|
| 400 |
-
{
|
| 401 |
-
"epoch": 4.32,
|
| 402 |
-
"grad_norm": 0.9459675579164174,
|
| 403 |
-
"learning_rate": 1.1034290900525279e-06,
|
| 404 |
-
"loss": 0.0749,
|
| 405 |
-
"step": 255
|
| 406 |
-
},
|
| 407 |
-
{
|
| 408 |
-
"epoch": 4.41,
|
| 409 |
-
"grad_norm": 0.968509379642604,
|
| 410 |
-
"learning_rate": 8.485438275698154e-07,
|
| 411 |
-
"loss": 0.0847,
|
| 412 |
-
"step": 260
|
| 413 |
-
},
|
| 414 |
-
{
|
| 415 |
-
"epoch": 4.49,
|
| 416 |
-
"grad_norm": 1.021637819526785,
|
| 417 |
-
"learning_rate": 6.258033886587911e-07,
|
| 418 |
-
"loss": 0.0832,
|
| 419 |
-
"step": 265
|
| 420 |
-
},
|
| 421 |
-
{
|
| 422 |
-
"epoch": 4.58,
|
| 423 |
-
"grad_norm": 0.9348358669170516,
|
| 424 |
-
"learning_rate": 4.359901572347758e-07,
|
| 425 |
-
"loss": 0.0819,
|
| 426 |
-
"step": 270
|
| 427 |
-
},
|
| 428 |
-
{
|
| 429 |
-
"epoch": 4.66,
|
| 430 |
-
"grad_norm": 0.9018557213311438,
|
| 431 |
-
"learning_rate": 2.7977085919589253e-07,
|
| 432 |
-
"loss": 0.081,
|
| 433 |
-
"step": 275
|
| 434 |
-
},
|
| 435 |
-
{
|
| 436 |
-
"epoch": 4.75,
|
| 437 |
-
"grad_norm": 0.9143406331670494,
|
| 438 |
-
"learning_rate": 1.5769422052403172e-07,
|
| 439 |
-
"loss": 0.0777,
|
| 440 |
-
"step": 280
|
| 441 |
-
},
|
| 442 |
-
{
|
| 443 |
-
"epoch": 4.83,
|
| 444 |
-
"grad_norm": 0.965550232388536,
|
| 445 |
-
"learning_rate": 7.018903986483083e-08,
|
| 446 |
-
"loss": 0.0804,
|
| 447 |
-
"step": 285
|
| 448 |
-
},
|
| 449 |
-
{
|
| 450 |
-
"epoch": 4.92,
|
| 451 |
-
"grad_norm": 0.982878932025226,
|
| 452 |
-
"learning_rate": 1.7562682356786488e-08,
|
| 453 |
-
"loss": 0.0842,
|
| 454 |
-
"step": 290
|
| 455 |
-
},
|
| 456 |
-
{
|
| 457 |
-
"epoch": 5.0,
|
| 458 |
-
"grad_norm": 0.8492670680881892,
|
| 459 |
-
"learning_rate": 0.0,
|
| 460 |
-
"loss": 0.0755,
|
| 461 |
-
"step": 295
|
| 462 |
-
},
|
| 463 |
-
{
|
| 464 |
-
"epoch": 5.0,
|
| 465 |
-
"eval_loss": 0.07571765035390854,
|
| 466 |
-
"eval_runtime": 334.5414,
|
| 467 |
-
"eval_samples_per_second": 22.482,
|
| 468 |
-
"eval_steps_per_second": 0.353,
|
| 469 |
-
"step": 295
|
| 470 |
-
},
|
| 471 |
-
{
|
| 472 |
-
"epoch": 5.0,
|
| 473 |
-
"step": 295,
|
| 474 |
-
"total_flos": 247067993702400.0,
|
| 475 |
-
"train_loss": 0.03106498374777325,
|
| 476 |
-
"train_runtime": 2739.2502,
|
| 477 |
-
"train_samples_per_second": 13.728,
|
| 478 |
-
"train_steps_per_second": 0.108
|
| 479 |
}
|
| 480 |
],
|
| 481 |
"logging_steps": 5,
|
| 482 |
-
"max_steps":
|
| 483 |
"num_input_tokens_seen": 0,
|
| 484 |
-
"num_train_epochs":
|
| 485 |
"save_steps": 100,
|
| 486 |
-
"total_flos":
|
| 487 |
"train_batch_size": 16,
|
| 488 |
"trial_name": null,
|
| 489 |
"trial_params": null
|
|
|
|
| 1 |
{
|
| 2 |
"best_metric": null,
|
| 3 |
"best_model_checkpoint": null,
|
| 4 |
+
"epoch": 1.0,
|
| 5 |
"eval_steps": 500,
|
| 6 |
+
"global_step": 59,
|
| 7 |
"is_hyper_param_search": false,
|
| 8 |
"is_local_process_zero": true,
|
| 9 |
"is_world_process_zero": true,
|
| 10 |
"log_history": [
|
| 11 |
{
|
| 12 |
"epoch": 0.02,
|
| 13 |
+
"grad_norm": 4.72473337945939,
|
| 14 |
+
"learning_rate": 3.3333333333333333e-06,
|
| 15 |
"loss": 0.9545,
|
| 16 |
"step": 1
|
| 17 |
},
|
| 18 |
{
|
| 19 |
"epoch": 0.08,
|
| 20 |
+
"grad_norm": 15.606619949426117,
|
| 21 |
+
"learning_rate": 1.6666666666666667e-05,
|
| 22 |
+
"loss": 0.9459,
|
| 23 |
"step": 5
|
| 24 |
},
|
| 25 |
{
|
| 26 |
"epoch": 0.17,
|
| 27 |
+
"grad_norm": 97.5546177933576,
|
| 28 |
+
"learning_rate": 1.972022914080411e-05,
|
| 29 |
+
"loss": 1.309,
|
| 30 |
"step": 10
|
| 31 |
},
|
| 32 |
{
|
| 33 |
"epoch": 0.25,
|
| 34 |
+
"grad_norm": 7.451729536881689,
|
| 35 |
+
"learning_rate": 1.8610436117673557e-05,
|
| 36 |
+
"loss": 1.1022,
|
| 37 |
"step": 15
|
| 38 |
},
|
| 39 |
{
|
| 40 |
"epoch": 0.34,
|
| 41 |
+
"grad_norm": 4.013413307542302,
|
| 42 |
+
"learning_rate": 1.6749830015182106e-05,
|
| 43 |
+
"loss": 0.9536,
|
| 44 |
"step": 20
|
| 45 |
},
|
| 46 |
{
|
| 47 |
"epoch": 0.42,
|
| 48 |
+
"grad_norm": 2.5281389594304153,
|
| 49 |
+
"learning_rate": 1.4300652022765207e-05,
|
| 50 |
+
"loss": 0.9122,
|
| 51 |
"step": 25
|
| 52 |
},
|
| 53 |
{
|
| 54 |
"epoch": 0.51,
|
| 55 |
+
"grad_norm": 2.4087072108196184,
|
| 56 |
+
"learning_rate": 1.1476465640024814e-05,
|
| 57 |
+
"loss": 0.8548,
|
| 58 |
"step": 30
|
| 59 |
},
|
| 60 |
{
|
| 61 |
"epoch": 0.59,
|
| 62 |
+
"grad_norm": 2.127180353057553,
|
| 63 |
+
"learning_rate": 8.52353435997519e-06,
|
| 64 |
+
"loss": 0.8386,
|
| 65 |
"step": 35
|
| 66 |
},
|
| 67 |
{
|
| 68 |
"epoch": 0.68,
|
| 69 |
+
"grad_norm": 1.9083395084428687,
|
| 70 |
+
"learning_rate": 5.699347977234799e-06,
|
| 71 |
+
"loss": 0.8324,
|
| 72 |
"step": 40
|
| 73 |
},
|
| 74 |
{
|
| 75 |
"epoch": 0.76,
|
| 76 |
+
"grad_norm": 1.7712149272954687,
|
| 77 |
+
"learning_rate": 3.250169984817897e-06,
|
| 78 |
+
"loss": 0.7792,
|
| 79 |
"step": 45
|
| 80 |
},
|
| 81 |
{
|
| 82 |
"epoch": 0.85,
|
| 83 |
+
"grad_norm": 1.6965019256204015,
|
| 84 |
+
"learning_rate": 1.3895638823264447e-06,
|
| 85 |
+
"loss": 0.77,
|
| 86 |
"step": 50
|
| 87 |
},
|
| 88 |
{
|
| 89 |
"epoch": 0.93,
|
| 90 |
+
"grad_norm": 1.6261522519462899,
|
| 91 |
+
"learning_rate": 2.7977085919589253e-07,
|
| 92 |
+
"loss": 0.7609,
|
| 93 |
"step": 55
|
| 94 |
},
|
| 95 |
{
|
| 96 |
"epoch": 1.0,
|
| 97 |
+
"eval_loss": 0.7076106071472168,
|
| 98 |
+
"eval_runtime": 338.673,
|
| 99 |
+
"eval_samples_per_second": 22.207,
|
| 100 |
+
"eval_steps_per_second": 0.348,
|
| 101 |
"step": 59
|
| 102 |
},
|
| 103 |
{
|
| 104 |
+
"epoch": 1.0,
|
| 105 |
+
"step": 59,
|
| 106 |
+
"total_flos": 49413598740480.0,
|
| 107 |
+
"train_loss": 0.908859632783017,
|
| 108 |
+
"train_runtime": 1626.0912,
|
| 109 |
+
"train_samples_per_second": 4.625,
|
| 110 |
+
"train_steps_per_second": 0.036
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 111 |
}
|
| 112 |
],
|
| 113 |
"logging_steps": 5,
|
| 114 |
+
"max_steps": 59,
|
| 115 |
"num_input_tokens_seen": 0,
|
| 116 |
+
"num_train_epochs": 1,
|
| 117 |
"save_steps": 100,
|
| 118 |
+
"total_flos": 49413598740480.0,
|
| 119 |
"train_batch_size": 16,
|
| 120 |
"trial_name": null,
|
| 121 |
"trial_params": null
|