End of training
Browse files- all_results.json +9 -9
- eval_results.json +4 -4
- train_results.json +5 -5
- trainer_state.json +135 -885
all_results.json
CHANGED
|
@@ -1,12 +1,12 @@
|
|
| 1 |
{
|
| 2 |
"epoch": 2.0,
|
| 3 |
-
"eval_loss":
|
| 4 |
-
"eval_runtime":
|
| 5 |
-
"eval_samples_per_second":
|
| 6 |
-
"eval_steps_per_second": 0.
|
| 7 |
-
"total_flos":
|
| 8 |
-
"train_loss": 1.
|
| 9 |
-
"train_runtime":
|
| 10 |
-
"train_samples_per_second": 0.
|
| 11 |
-
"train_steps_per_second": 0.
|
| 12 |
}
|
|
|
|
| 1 |
{
|
| 2 |
"epoch": 2.0,
|
| 3 |
+
"eval_loss": 0.5955778956413269,
|
| 4 |
+
"eval_runtime": 6.1056,
|
| 5 |
+
"eval_samples_per_second": 1.638,
|
| 6 |
+
"eval_steps_per_second": 0.819,
|
| 7 |
+
"total_flos": 5816699796600.0,
|
| 8 |
+
"train_loss": 1.3652541448495217,
|
| 9 |
+
"train_runtime": 216.8469,
|
| 10 |
+
"train_samples_per_second": 0.461,
|
| 11 |
+
"train_steps_per_second": 0.065
|
| 12 |
}
|
eval_results.json
CHANGED
|
@@ -1,7 +1,7 @@
|
|
| 1 |
{
|
| 2 |
"epoch": 2.0,
|
| 3 |
-
"eval_loss":
|
| 4 |
-
"eval_runtime":
|
| 5 |
-
"eval_samples_per_second":
|
| 6 |
-
"eval_steps_per_second": 0.
|
| 7 |
}
|
|
|
|
| 1 |
{
|
| 2 |
"epoch": 2.0,
|
| 3 |
+
"eval_loss": 0.5955778956413269,
|
| 4 |
+
"eval_runtime": 6.1056,
|
| 5 |
+
"eval_samples_per_second": 1.638,
|
| 6 |
+
"eval_steps_per_second": 0.819
|
| 7 |
}
|
train_results.json
CHANGED
|
@@ -1,8 +1,8 @@
|
|
| 1 |
{
|
| 2 |
"epoch": 2.0,
|
| 3 |
-
"total_flos":
|
| 4 |
-
"train_loss": 1.
|
| 5 |
-
"train_runtime":
|
| 6 |
-
"train_samples_per_second": 0.
|
| 7 |
-
"train_steps_per_second": 0.
|
| 8 |
}
|
|
|
|
| 1 |
{
|
| 2 |
"epoch": 2.0,
|
| 3 |
+
"total_flos": 5816699796600.0,
|
| 4 |
+
"train_loss": 1.3652541448495217,
|
| 5 |
+
"train_runtime": 216.8469,
|
| 6 |
+
"train_samples_per_second": 0.461,
|
| 7 |
+
"train_steps_per_second": 0.065
|
| 8 |
}
|
trainer_state.json
CHANGED
|
@@ -3,983 +3,233 @@
|
|
| 3 |
"best_model_checkpoint": null,
|
| 4 |
"epoch": 2.0,
|
| 5 |
"eval_steps": 1.0,
|
| 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": 4.
|
| 15 |
-
"loss":
|
| 16 |
"step": 1
|
| 17 |
},
|
| 18 |
{
|
| 19 |
-
"epoch": 0.
|
| 20 |
-
"eval_loss": 1.
|
| 21 |
-
"eval_runtime":
|
| 22 |
-
"eval_samples_per_second":
|
| 23 |
-
"eval_steps_per_second":
|
| 24 |
"step": 1
|
| 25 |
},
|
| 26 |
{
|
| 27 |
-
"epoch": 0.
|
| 28 |
-
"grad_norm":
|
| 29 |
-
"learning_rate": 4.
|
| 30 |
-
"loss": 1.
|
| 31 |
"step": 2
|
| 32 |
},
|
| 33 |
{
|
| 34 |
-
"epoch": 0.
|
| 35 |
-
"eval_loss":
|
| 36 |
-
"eval_runtime":
|
| 37 |
-
"eval_samples_per_second": 3.
|
| 38 |
-
"eval_steps_per_second":
|
| 39 |
"step": 2
|
| 40 |
},
|
| 41 |
{
|
| 42 |
-
"epoch": 0.
|
| 43 |
-
"grad_norm":
|
| 44 |
-
"learning_rate":
|
| 45 |
-
"loss":
|
| 46 |
"step": 3
|
| 47 |
},
|
| 48 |
{
|
| 49 |
-
"epoch": 0.
|
| 50 |
-
"eval_loss":
|
| 51 |
-
"eval_runtime":
|
| 52 |
-
"eval_samples_per_second":
|
| 53 |
-
"eval_steps_per_second":
|
| 54 |
"step": 3
|
| 55 |
},
|
| 56 |
{
|
| 57 |
-
"epoch": 0.
|
| 58 |
-
"grad_norm":
|
| 59 |
-
"learning_rate":
|
| 60 |
-
"loss":
|
| 61 |
"step": 4
|
| 62 |
},
|
| 63 |
{
|
| 64 |
-
"epoch": 0.
|
| 65 |
-
"eval_loss":
|
| 66 |
-
"eval_runtime":
|
| 67 |
-
"eval_samples_per_second": 3.
|
| 68 |
-
"eval_steps_per_second":
|
| 69 |
"step": 4
|
| 70 |
},
|
| 71 |
{
|
| 72 |
-
"epoch": 0.
|
| 73 |
-
"grad_norm":
|
| 74 |
-
"learning_rate":
|
| 75 |
-
"loss":
|
| 76 |
"step": 5
|
| 77 |
},
|
| 78 |
{
|
| 79 |
-
"epoch": 0.
|
| 80 |
-
"eval_loss":
|
| 81 |
-
"eval_runtime":
|
| 82 |
-
"eval_samples_per_second":
|
| 83 |
-
"eval_steps_per_second":
|
| 84 |
"step": 5
|
| 85 |
},
|
| 86 |
{
|
| 87 |
-
"epoch": 0.
|
| 88 |
-
"grad_norm":
|
| 89 |
-
"learning_rate":
|
| 90 |
-
"loss":
|
| 91 |
"step": 6
|
| 92 |
},
|
| 93 |
{
|
| 94 |
-
"epoch": 0.
|
| 95 |
-
"eval_loss":
|
| 96 |
-
"eval_runtime":
|
| 97 |
-
"eval_samples_per_second":
|
| 98 |
-
"eval_steps_per_second":
|
| 99 |
"step": 6
|
| 100 |
},
|
| 101 |
{
|
| 102 |
-
"epoch":
|
| 103 |
-
"grad_norm":
|
| 104 |
-
"learning_rate":
|
| 105 |
-
"loss":
|
| 106 |
"step": 7
|
| 107 |
},
|
| 108 |
{
|
| 109 |
-
"epoch":
|
| 110 |
-
"eval_loss":
|
| 111 |
-
"eval_runtime":
|
| 112 |
-
"eval_samples_per_second":
|
| 113 |
-
"eval_steps_per_second":
|
| 114 |
"step": 7
|
| 115 |
},
|
| 116 |
{
|
| 117 |
-
"epoch":
|
| 118 |
-
"grad_norm":
|
| 119 |
-
"learning_rate":
|
| 120 |
-
"loss": 2.
|
| 121 |
"step": 8
|
| 122 |
},
|
| 123 |
{
|
| 124 |
-
"epoch":
|
| 125 |
-
"eval_loss":
|
| 126 |
-
"eval_runtime":
|
| 127 |
-
"eval_samples_per_second": 3.
|
| 128 |
-
"eval_steps_per_second":
|
| 129 |
"step": 8
|
| 130 |
},
|
| 131 |
{
|
| 132 |
-
"epoch":
|
| 133 |
-
"grad_norm":
|
| 134 |
-
"learning_rate":
|
| 135 |
-
"loss": 2.
|
| 136 |
"step": 9
|
| 137 |
},
|
| 138 |
{
|
| 139 |
-
"epoch":
|
| 140 |
-
"eval_loss":
|
| 141 |
-
"eval_runtime":
|
| 142 |
-
"eval_samples_per_second":
|
| 143 |
-
"eval_steps_per_second":
|
| 144 |
"step": 9
|
| 145 |
},
|
| 146 |
{
|
| 147 |
-
"epoch":
|
| 148 |
-
"grad_norm":
|
| 149 |
-
"learning_rate":
|
| 150 |
-
"loss":
|
| 151 |
"step": 10
|
| 152 |
},
|
| 153 |
{
|
| 154 |
-
"epoch":
|
| 155 |
-
"eval_loss":
|
| 156 |
-
"eval_runtime":
|
| 157 |
-
"eval_samples_per_second": 3.
|
| 158 |
-
"eval_steps_per_second":
|
| 159 |
"step": 10
|
| 160 |
},
|
| 161 |
{
|
| 162 |
-
"epoch":
|
| 163 |
-
"grad_norm":
|
| 164 |
-
"learning_rate":
|
| 165 |
-
"loss":
|
| 166 |
"step": 11
|
| 167 |
},
|
| 168 |
{
|
| 169 |
-
"epoch":
|
| 170 |
-
"eval_loss":
|
| 171 |
-
"eval_runtime":
|
| 172 |
-
"eval_samples_per_second":
|
| 173 |
-
"eval_steps_per_second":
|
| 174 |
"step": 11
|
| 175 |
},
|
| 176 |
{
|
| 177 |
-
"epoch":
|
| 178 |
-
"grad_norm":
|
| 179 |
-
"learning_rate":
|
| 180 |
-
"loss":
|
| 181 |
"step": 12
|
| 182 |
},
|
| 183 |
{
|
| 184 |
-
"epoch":
|
| 185 |
-
"eval_loss":
|
| 186 |
-
"eval_runtime":
|
| 187 |
-
"eval_samples_per_second":
|
| 188 |
-
"eval_steps_per_second": 0.
|
| 189 |
"step": 12
|
| 190 |
},
|
| 191 |
{
|
| 192 |
-
"epoch":
|
| 193 |
-
"grad_norm":
|
| 194 |
-
"learning_rate": 3.
|
| 195 |
-
"loss":
|
| 196 |
"step": 13
|
| 197 |
},
|
| 198 |
{
|
| 199 |
-
"epoch":
|
| 200 |
-
"eval_loss":
|
| 201 |
-
"eval_runtime":
|
| 202 |
-
"eval_samples_per_second":
|
| 203 |
-
"eval_steps_per_second":
|
| 204 |
"step": 13
|
| 205 |
},
|
| 206 |
-
{
|
| 207 |
-
"epoch": 0.4375,
|
| 208 |
-
"grad_norm": 11.233291625976562,
|
| 209 |
-
"learning_rate": 3.90625e-05,
|
| 210 |
-
"loss": 2.6233,
|
| 211 |
-
"step": 14
|
| 212 |
-
},
|
| 213 |
-
{
|
| 214 |
-
"epoch": 0.4375,
|
| 215 |
-
"eval_loss": 1.9741547107696533,
|
| 216 |
-
"eval_runtime": 14.0253,
|
| 217 |
-
"eval_samples_per_second": 3.565,
|
| 218 |
-
"eval_steps_per_second": 0.499,
|
| 219 |
-
"step": 14
|
| 220 |
-
},
|
| 221 |
-
{
|
| 222 |
-
"epoch": 0.46875,
|
| 223 |
-
"grad_norm": 16.227378845214844,
|
| 224 |
-
"learning_rate": 3.828125e-05,
|
| 225 |
-
"loss": 2.5995,
|
| 226 |
-
"step": 15
|
| 227 |
-
},
|
| 228 |
-
{
|
| 229 |
-
"epoch": 0.46875,
|
| 230 |
-
"eval_loss": 1.9386694431304932,
|
| 231 |
-
"eval_runtime": 20.5374,
|
| 232 |
-
"eval_samples_per_second": 2.435,
|
| 233 |
-
"eval_steps_per_second": 0.341,
|
| 234 |
-
"step": 15
|
| 235 |
-
},
|
| 236 |
-
{
|
| 237 |
-
"epoch": 0.5,
|
| 238 |
-
"grad_norm": 19.815183639526367,
|
| 239 |
-
"learning_rate": 3.7500000000000003e-05,
|
| 240 |
-
"loss": 2.4588,
|
| 241 |
-
"step": 16
|
| 242 |
-
},
|
| 243 |
-
{
|
| 244 |
-
"epoch": 0.5,
|
| 245 |
-
"eval_loss": 1.8799047470092773,
|
| 246 |
-
"eval_runtime": 14.1007,
|
| 247 |
-
"eval_samples_per_second": 3.546,
|
| 248 |
-
"eval_steps_per_second": 0.496,
|
| 249 |
-
"step": 16
|
| 250 |
-
},
|
| 251 |
-
{
|
| 252 |
-
"epoch": 0.53125,
|
| 253 |
-
"grad_norm": 27.6804141998291,
|
| 254 |
-
"learning_rate": 3.671875e-05,
|
| 255 |
-
"loss": 2.6622,
|
| 256 |
-
"step": 17
|
| 257 |
-
},
|
| 258 |
-
{
|
| 259 |
-
"epoch": 0.53125,
|
| 260 |
-
"eval_loss": 1.9120107889175415,
|
| 261 |
-
"eval_runtime": 19.021,
|
| 262 |
-
"eval_samples_per_second": 2.629,
|
| 263 |
-
"eval_steps_per_second": 0.368,
|
| 264 |
-
"step": 17
|
| 265 |
-
},
|
| 266 |
-
{
|
| 267 |
-
"epoch": 0.5625,
|
| 268 |
-
"grad_norm": 53.53272247314453,
|
| 269 |
-
"learning_rate": 3.59375e-05,
|
| 270 |
-
"loss": 2.6819,
|
| 271 |
-
"step": 18
|
| 272 |
-
},
|
| 273 |
-
{
|
| 274 |
-
"epoch": 0.5625,
|
| 275 |
-
"eval_loss": 1.826271414756775,
|
| 276 |
-
"eval_runtime": 14.0375,
|
| 277 |
-
"eval_samples_per_second": 3.562,
|
| 278 |
-
"eval_steps_per_second": 0.499,
|
| 279 |
-
"step": 18
|
| 280 |
-
},
|
| 281 |
-
{
|
| 282 |
-
"epoch": 0.59375,
|
| 283 |
-
"grad_norm": 31.6351318359375,
|
| 284 |
-
"learning_rate": 3.5156250000000004e-05,
|
| 285 |
-
"loss": 2.6057,
|
| 286 |
-
"step": 19
|
| 287 |
-
},
|
| 288 |
-
{
|
| 289 |
-
"epoch": 0.59375,
|
| 290 |
-
"eval_loss": 1.8063328266143799,
|
| 291 |
-
"eval_runtime": 19.3292,
|
| 292 |
-
"eval_samples_per_second": 2.587,
|
| 293 |
-
"eval_steps_per_second": 0.362,
|
| 294 |
-
"step": 19
|
| 295 |
-
},
|
| 296 |
-
{
|
| 297 |
-
"epoch": 0.625,
|
| 298 |
-
"grad_norm": 81.76189422607422,
|
| 299 |
-
"learning_rate": 3.4375e-05,
|
| 300 |
-
"loss": 2.1173,
|
| 301 |
-
"step": 20
|
| 302 |
-
},
|
| 303 |
-
{
|
| 304 |
-
"epoch": 0.625,
|
| 305 |
-
"eval_loss": 1.813103437423706,
|
| 306 |
-
"eval_runtime": 14.0502,
|
| 307 |
-
"eval_samples_per_second": 3.559,
|
| 308 |
-
"eval_steps_per_second": 0.498,
|
| 309 |
-
"step": 20
|
| 310 |
-
},
|
| 311 |
-
{
|
| 312 |
-
"epoch": 0.65625,
|
| 313 |
-
"grad_norm": 28.440521240234375,
|
| 314 |
-
"learning_rate": 3.359375e-05,
|
| 315 |
-
"loss": 2.4934,
|
| 316 |
-
"step": 21
|
| 317 |
-
},
|
| 318 |
-
{
|
| 319 |
-
"epoch": 0.65625,
|
| 320 |
-
"eval_loss": 1.8328737020492554,
|
| 321 |
-
"eval_runtime": 19.9467,
|
| 322 |
-
"eval_samples_per_second": 2.507,
|
| 323 |
-
"eval_steps_per_second": 0.351,
|
| 324 |
-
"step": 21
|
| 325 |
-
},
|
| 326 |
-
{
|
| 327 |
-
"epoch": 0.6875,
|
| 328 |
-
"grad_norm": 31.066139221191406,
|
| 329 |
-
"learning_rate": 3.2812500000000005e-05,
|
| 330 |
-
"loss": 2.4664,
|
| 331 |
-
"step": 22
|
| 332 |
-
},
|
| 333 |
-
{
|
| 334 |
-
"epoch": 0.6875,
|
| 335 |
-
"eval_loss": 1.8729884624481201,
|
| 336 |
-
"eval_runtime": 14.8515,
|
| 337 |
-
"eval_samples_per_second": 3.367,
|
| 338 |
-
"eval_steps_per_second": 0.471,
|
| 339 |
-
"step": 22
|
| 340 |
-
},
|
| 341 |
-
{
|
| 342 |
-
"epoch": 0.71875,
|
| 343 |
-
"grad_norm": 47.6856803894043,
|
| 344 |
-
"learning_rate": 3.203125e-05,
|
| 345 |
-
"loss": 2.3591,
|
| 346 |
-
"step": 23
|
| 347 |
-
},
|
| 348 |
-
{
|
| 349 |
-
"epoch": 0.71875,
|
| 350 |
-
"eval_loss": 1.9386744499206543,
|
| 351 |
-
"eval_runtime": 19.4361,
|
| 352 |
-
"eval_samples_per_second": 2.573,
|
| 353 |
-
"eval_steps_per_second": 0.36,
|
| 354 |
-
"step": 23
|
| 355 |
-
},
|
| 356 |
-
{
|
| 357 |
-
"epoch": 0.75,
|
| 358 |
-
"grad_norm": 44.08198928833008,
|
| 359 |
-
"learning_rate": 3.125e-05,
|
| 360 |
-
"loss": 2.4319,
|
| 361 |
-
"step": 24
|
| 362 |
-
},
|
| 363 |
-
{
|
| 364 |
-
"epoch": 0.75,
|
| 365 |
-
"eval_loss": 1.9019237756729126,
|
| 366 |
-
"eval_runtime": 13.9736,
|
| 367 |
-
"eval_samples_per_second": 3.578,
|
| 368 |
-
"eval_steps_per_second": 0.501,
|
| 369 |
-
"step": 24
|
| 370 |
-
},
|
| 371 |
-
{
|
| 372 |
-
"epoch": 0.78125,
|
| 373 |
-
"grad_norm": 59.23967361450195,
|
| 374 |
-
"learning_rate": 3.0468750000000002e-05,
|
| 375 |
-
"loss": 2.5579,
|
| 376 |
-
"step": 25
|
| 377 |
-
},
|
| 378 |
-
{
|
| 379 |
-
"epoch": 0.78125,
|
| 380 |
-
"eval_loss": 1.8655019998550415,
|
| 381 |
-
"eval_runtime": 19.5942,
|
| 382 |
-
"eval_samples_per_second": 2.552,
|
| 383 |
-
"eval_steps_per_second": 0.357,
|
| 384 |
-
"step": 25
|
| 385 |
-
},
|
| 386 |
-
{
|
| 387 |
-
"epoch": 0.8125,
|
| 388 |
-
"grad_norm": 44.628658294677734,
|
| 389 |
-
"learning_rate": 2.96875e-05,
|
| 390 |
-
"loss": 2.4048,
|
| 391 |
-
"step": 26
|
| 392 |
-
},
|
| 393 |
-
{
|
| 394 |
-
"epoch": 0.8125,
|
| 395 |
-
"eval_loss": 1.8706899881362915,
|
| 396 |
-
"eval_runtime": 14.1375,
|
| 397 |
-
"eval_samples_per_second": 3.537,
|
| 398 |
-
"eval_steps_per_second": 0.495,
|
| 399 |
-
"step": 26
|
| 400 |
-
},
|
| 401 |
-
{
|
| 402 |
-
"epoch": 0.84375,
|
| 403 |
-
"grad_norm": 59.57282638549805,
|
| 404 |
-
"learning_rate": 2.890625e-05,
|
| 405 |
-
"loss": 2.7706,
|
| 406 |
-
"step": 27
|
| 407 |
-
},
|
| 408 |
-
{
|
| 409 |
-
"epoch": 0.84375,
|
| 410 |
-
"eval_loss": 1.744444727897644,
|
| 411 |
-
"eval_runtime": 19.9669,
|
| 412 |
-
"eval_samples_per_second": 2.504,
|
| 413 |
-
"eval_steps_per_second": 0.351,
|
| 414 |
-
"step": 27
|
| 415 |
-
},
|
| 416 |
-
{
|
| 417 |
-
"epoch": 0.875,
|
| 418 |
-
"grad_norm": 38.48733901977539,
|
| 419 |
-
"learning_rate": 2.8125000000000003e-05,
|
| 420 |
-
"loss": 2.7279,
|
| 421 |
-
"step": 28
|
| 422 |
-
},
|
| 423 |
-
{
|
| 424 |
-
"epoch": 0.875,
|
| 425 |
-
"eval_loss": 1.6933951377868652,
|
| 426 |
-
"eval_runtime": 14.0043,
|
| 427 |
-
"eval_samples_per_second": 3.57,
|
| 428 |
-
"eval_steps_per_second": 0.5,
|
| 429 |
-
"step": 28
|
| 430 |
-
},
|
| 431 |
-
{
|
| 432 |
-
"epoch": 0.90625,
|
| 433 |
-
"grad_norm": 38.13273620605469,
|
| 434 |
-
"learning_rate": 2.734375e-05,
|
| 435 |
-
"loss": 2.5247,
|
| 436 |
-
"step": 29
|
| 437 |
-
},
|
| 438 |
-
{
|
| 439 |
-
"epoch": 0.90625,
|
| 440 |
-
"eval_loss": 1.6971945762634277,
|
| 441 |
-
"eval_runtime": 18.8375,
|
| 442 |
-
"eval_samples_per_second": 2.654,
|
| 443 |
-
"eval_steps_per_second": 0.372,
|
| 444 |
-
"step": 29
|
| 445 |
-
},
|
| 446 |
-
{
|
| 447 |
-
"epoch": 0.9375,
|
| 448 |
-
"grad_norm": 34.31296920776367,
|
| 449 |
-
"learning_rate": 2.6562500000000002e-05,
|
| 450 |
-
"loss": 2.2336,
|
| 451 |
-
"step": 30
|
| 452 |
-
},
|
| 453 |
-
{
|
| 454 |
-
"epoch": 0.9375,
|
| 455 |
-
"eval_loss": 1.7182470560073853,
|
| 456 |
-
"eval_runtime": 14.4648,
|
| 457 |
-
"eval_samples_per_second": 3.457,
|
| 458 |
-
"eval_steps_per_second": 0.484,
|
| 459 |
-
"step": 30
|
| 460 |
-
},
|
| 461 |
-
{
|
| 462 |
-
"epoch": 0.96875,
|
| 463 |
-
"grad_norm": 34.71999740600586,
|
| 464 |
-
"learning_rate": 2.578125e-05,
|
| 465 |
-
"loss": 2.2479,
|
| 466 |
-
"step": 31
|
| 467 |
-
},
|
| 468 |
-
{
|
| 469 |
-
"epoch": 0.96875,
|
| 470 |
-
"eval_loss": 1.7286863327026367,
|
| 471 |
-
"eval_runtime": 19.0105,
|
| 472 |
-
"eval_samples_per_second": 2.63,
|
| 473 |
-
"eval_steps_per_second": 0.368,
|
| 474 |
-
"step": 31
|
| 475 |
-
},
|
| 476 |
-
{
|
| 477 |
-
"epoch": 1.0,
|
| 478 |
-
"grad_norm": 54.52072525024414,
|
| 479 |
-
"learning_rate": 2.5e-05,
|
| 480 |
-
"loss": 0.9398,
|
| 481 |
-
"step": 32
|
| 482 |
-
},
|
| 483 |
-
{
|
| 484 |
-
"epoch": 1.0,
|
| 485 |
-
"eval_loss": 1.7176891565322876,
|
| 486 |
-
"eval_runtime": 15.5832,
|
| 487 |
-
"eval_samples_per_second": 3.209,
|
| 488 |
-
"eval_steps_per_second": 0.449,
|
| 489 |
-
"step": 32
|
| 490 |
-
},
|
| 491 |
-
{
|
| 492 |
-
"epoch": 1.03125,
|
| 493 |
-
"grad_norm": 37.95784378051758,
|
| 494 |
-
"learning_rate": 2.4218750000000003e-05,
|
| 495 |
-
"loss": 2.2821,
|
| 496 |
-
"step": 33
|
| 497 |
-
},
|
| 498 |
-
{
|
| 499 |
-
"epoch": 1.03125,
|
| 500 |
-
"eval_loss": 1.6573299169540405,
|
| 501 |
-
"eval_runtime": 19.2268,
|
| 502 |
-
"eval_samples_per_second": 2.601,
|
| 503 |
-
"eval_steps_per_second": 0.364,
|
| 504 |
-
"step": 33
|
| 505 |
-
},
|
| 506 |
-
{
|
| 507 |
-
"epoch": 1.0625,
|
| 508 |
-
"grad_norm": 32.960391998291016,
|
| 509 |
-
"learning_rate": 2.34375e-05,
|
| 510 |
-
"loss": 2.339,
|
| 511 |
-
"step": 34
|
| 512 |
-
},
|
| 513 |
-
{
|
| 514 |
-
"epoch": 1.0625,
|
| 515 |
-
"eval_loss": 1.623481273651123,
|
| 516 |
-
"eval_runtime": 14.5998,
|
| 517 |
-
"eval_samples_per_second": 3.425,
|
| 518 |
-
"eval_steps_per_second": 0.479,
|
| 519 |
-
"step": 34
|
| 520 |
-
},
|
| 521 |
-
{
|
| 522 |
-
"epoch": 1.09375,
|
| 523 |
-
"grad_norm": 31.484209060668945,
|
| 524 |
-
"learning_rate": 2.2656250000000002e-05,
|
| 525 |
-
"loss": 1.7187,
|
| 526 |
-
"step": 35
|
| 527 |
-
},
|
| 528 |
-
{
|
| 529 |
-
"epoch": 1.09375,
|
| 530 |
-
"eval_loss": 1.5644508600234985,
|
| 531 |
-
"eval_runtime": 20.7871,
|
| 532 |
-
"eval_samples_per_second": 2.405,
|
| 533 |
-
"eval_steps_per_second": 0.337,
|
| 534 |
-
"step": 35
|
| 535 |
-
},
|
| 536 |
-
{
|
| 537 |
-
"epoch": 1.125,
|
| 538 |
-
"grad_norm": 28.109580993652344,
|
| 539 |
-
"learning_rate": 2.1875e-05,
|
| 540 |
-
"loss": 1.585,
|
| 541 |
-
"step": 36
|
| 542 |
-
},
|
| 543 |
-
{
|
| 544 |
-
"epoch": 1.125,
|
| 545 |
-
"eval_loss": 1.4872223138809204,
|
| 546 |
-
"eval_runtime": 14.0172,
|
| 547 |
-
"eval_samples_per_second": 3.567,
|
| 548 |
-
"eval_steps_per_second": 0.499,
|
| 549 |
-
"step": 36
|
| 550 |
-
},
|
| 551 |
-
{
|
| 552 |
-
"epoch": 1.15625,
|
| 553 |
-
"grad_norm": 34.407615661621094,
|
| 554 |
-
"learning_rate": 2.109375e-05,
|
| 555 |
-
"loss": 1.5801,
|
| 556 |
-
"step": 37
|
| 557 |
-
},
|
| 558 |
-
{
|
| 559 |
-
"epoch": 1.15625,
|
| 560 |
-
"eval_loss": 1.429572582244873,
|
| 561 |
-
"eval_runtime": 19.6378,
|
| 562 |
-
"eval_samples_per_second": 2.546,
|
| 563 |
-
"eval_steps_per_second": 0.356,
|
| 564 |
-
"step": 37
|
| 565 |
-
},
|
| 566 |
-
{
|
| 567 |
-
"epoch": 1.1875,
|
| 568 |
-
"grad_norm": 34.66839599609375,
|
| 569 |
-
"learning_rate": 2.0312500000000002e-05,
|
| 570 |
-
"loss": 1.8689,
|
| 571 |
-
"step": 38
|
| 572 |
-
},
|
| 573 |
-
{
|
| 574 |
-
"epoch": 1.1875,
|
| 575 |
-
"eval_loss": 1.4055802822113037,
|
| 576 |
-
"eval_runtime": 15.0122,
|
| 577 |
-
"eval_samples_per_second": 3.331,
|
| 578 |
-
"eval_steps_per_second": 0.466,
|
| 579 |
-
"step": 38
|
| 580 |
-
},
|
| 581 |
-
{
|
| 582 |
-
"epoch": 1.21875,
|
| 583 |
-
"grad_norm": 38.05352020263672,
|
| 584 |
-
"learning_rate": 1.953125e-05,
|
| 585 |
-
"loss": 1.5591,
|
| 586 |
-
"step": 39
|
| 587 |
-
},
|
| 588 |
-
{
|
| 589 |
-
"epoch": 1.21875,
|
| 590 |
-
"eval_loss": 1.398817777633667,
|
| 591 |
-
"eval_runtime": 20.1748,
|
| 592 |
-
"eval_samples_per_second": 2.478,
|
| 593 |
-
"eval_steps_per_second": 0.347,
|
| 594 |
-
"step": 39
|
| 595 |
-
},
|
| 596 |
-
{
|
| 597 |
-
"epoch": 1.25,
|
| 598 |
-
"grad_norm": 45.93744659423828,
|
| 599 |
-
"learning_rate": 1.8750000000000002e-05,
|
| 600 |
-
"loss": 1.4769,
|
| 601 |
-
"step": 40
|
| 602 |
-
},
|
| 603 |
-
{
|
| 604 |
-
"epoch": 1.25,
|
| 605 |
-
"eval_loss": 1.3957644701004028,
|
| 606 |
-
"eval_runtime": 14.2141,
|
| 607 |
-
"eval_samples_per_second": 3.518,
|
| 608 |
-
"eval_steps_per_second": 0.492,
|
| 609 |
-
"step": 40
|
| 610 |
-
},
|
| 611 |
-
{
|
| 612 |
-
"epoch": 1.28125,
|
| 613 |
-
"grad_norm": 67.26871490478516,
|
| 614 |
-
"learning_rate": 1.796875e-05,
|
| 615 |
-
"loss": 1.2898,
|
| 616 |
-
"step": 41
|
| 617 |
-
},
|
| 618 |
-
{
|
| 619 |
-
"epoch": 1.28125,
|
| 620 |
-
"eval_loss": 1.3631155490875244,
|
| 621 |
-
"eval_runtime": 18.8012,
|
| 622 |
-
"eval_samples_per_second": 2.659,
|
| 623 |
-
"eval_steps_per_second": 0.372,
|
| 624 |
-
"step": 41
|
| 625 |
-
},
|
| 626 |
-
{
|
| 627 |
-
"epoch": 1.3125,
|
| 628 |
-
"grad_norm": 48.31477737426758,
|
| 629 |
-
"learning_rate": 1.71875e-05,
|
| 630 |
-
"loss": 1.4675,
|
| 631 |
-
"step": 42
|
| 632 |
-
},
|
| 633 |
-
{
|
| 634 |
-
"epoch": 1.3125,
|
| 635 |
-
"eval_loss": 1.3446946144104004,
|
| 636 |
-
"eval_runtime": 14.0177,
|
| 637 |
-
"eval_samples_per_second": 3.567,
|
| 638 |
-
"eval_steps_per_second": 0.499,
|
| 639 |
-
"step": 42
|
| 640 |
-
},
|
| 641 |
-
{
|
| 642 |
-
"epoch": 1.34375,
|
| 643 |
-
"grad_norm": 43.087947845458984,
|
| 644 |
-
"learning_rate": 1.6406250000000002e-05,
|
| 645 |
-
"loss": 1.0123,
|
| 646 |
-
"step": 43
|
| 647 |
-
},
|
| 648 |
-
{
|
| 649 |
-
"epoch": 1.34375,
|
| 650 |
-
"eval_loss": 1.3272167444229126,
|
| 651 |
-
"eval_runtime": 19.8127,
|
| 652 |
-
"eval_samples_per_second": 2.524,
|
| 653 |
-
"eval_steps_per_second": 0.353,
|
| 654 |
-
"step": 43
|
| 655 |
-
},
|
| 656 |
-
{
|
| 657 |
-
"epoch": 1.375,
|
| 658 |
-
"grad_norm": 50.14296340942383,
|
| 659 |
-
"learning_rate": 1.5625e-05,
|
| 660 |
-
"loss": 1.4516,
|
| 661 |
-
"step": 44
|
| 662 |
-
},
|
| 663 |
-
{
|
| 664 |
-
"epoch": 1.375,
|
| 665 |
-
"eval_loss": 1.3105698823928833,
|
| 666 |
-
"eval_runtime": 14.0327,
|
| 667 |
-
"eval_samples_per_second": 3.563,
|
| 668 |
-
"eval_steps_per_second": 0.499,
|
| 669 |
-
"step": 44
|
| 670 |
-
},
|
| 671 |
-
{
|
| 672 |
-
"epoch": 1.40625,
|
| 673 |
-
"grad_norm": 56.435394287109375,
|
| 674 |
-
"learning_rate": 1.484375e-05,
|
| 675 |
-
"loss": 1.2132,
|
| 676 |
-
"step": 45
|
| 677 |
-
},
|
| 678 |
-
{
|
| 679 |
-
"epoch": 1.40625,
|
| 680 |
-
"eval_loss": 1.2978880405426025,
|
| 681 |
-
"eval_runtime": 19.3431,
|
| 682 |
-
"eval_samples_per_second": 2.585,
|
| 683 |
-
"eval_steps_per_second": 0.362,
|
| 684 |
-
"step": 45
|
| 685 |
-
},
|
| 686 |
-
{
|
| 687 |
-
"epoch": 1.4375,
|
| 688 |
-
"grad_norm": 62.44044494628906,
|
| 689 |
-
"learning_rate": 1.4062500000000001e-05,
|
| 690 |
-
"loss": 1.1181,
|
| 691 |
-
"step": 46
|
| 692 |
-
},
|
| 693 |
-
{
|
| 694 |
-
"epoch": 1.4375,
|
| 695 |
-
"eval_loss": 1.2973986864089966,
|
| 696 |
-
"eval_runtime": 14.4572,
|
| 697 |
-
"eval_samples_per_second": 3.458,
|
| 698 |
-
"eval_steps_per_second": 0.484,
|
| 699 |
-
"step": 46
|
| 700 |
-
},
|
| 701 |
-
{
|
| 702 |
-
"epoch": 1.46875,
|
| 703 |
-
"grad_norm": 49.06483459472656,
|
| 704 |
-
"learning_rate": 1.3281250000000001e-05,
|
| 705 |
-
"loss": 0.7987,
|
| 706 |
-
"step": 47
|
| 707 |
-
},
|
| 708 |
-
{
|
| 709 |
-
"epoch": 1.46875,
|
| 710 |
-
"eval_loss": 1.3060061931610107,
|
| 711 |
-
"eval_runtime": 18.8411,
|
| 712 |
-
"eval_samples_per_second": 2.654,
|
| 713 |
-
"eval_steps_per_second": 0.372,
|
| 714 |
-
"step": 47
|
| 715 |
-
},
|
| 716 |
-
{
|
| 717 |
-
"epoch": 1.5,
|
| 718 |
-
"grad_norm": 59.63069534301758,
|
| 719 |
-
"learning_rate": 1.25e-05,
|
| 720 |
-
"loss": 0.7248,
|
| 721 |
-
"step": 48
|
| 722 |
-
},
|
| 723 |
-
{
|
| 724 |
-
"epoch": 1.5,
|
| 725 |
-
"eval_loss": 1.3088918924331665,
|
| 726 |
-
"eval_runtime": 16.3052,
|
| 727 |
-
"eval_samples_per_second": 3.067,
|
| 728 |
-
"eval_steps_per_second": 0.429,
|
| 729 |
-
"step": 48
|
| 730 |
-
},
|
| 731 |
-
{
|
| 732 |
-
"epoch": 1.53125,
|
| 733 |
-
"grad_norm": 85.45850372314453,
|
| 734 |
-
"learning_rate": 1.171875e-05,
|
| 735 |
-
"loss": 0.8957,
|
| 736 |
-
"step": 49
|
| 737 |
-
},
|
| 738 |
-
{
|
| 739 |
-
"epoch": 1.53125,
|
| 740 |
-
"eval_loss": 1.318437933921814,
|
| 741 |
-
"eval_runtime": 19.1539,
|
| 742 |
-
"eval_samples_per_second": 2.61,
|
| 743 |
-
"eval_steps_per_second": 0.365,
|
| 744 |
-
"step": 49
|
| 745 |
-
},
|
| 746 |
-
{
|
| 747 |
-
"epoch": 1.5625,
|
| 748 |
-
"grad_norm": 69.75704956054688,
|
| 749 |
-
"learning_rate": 1.09375e-05,
|
| 750 |
-
"loss": 0.8146,
|
| 751 |
-
"step": 50
|
| 752 |
-
},
|
| 753 |
-
{
|
| 754 |
-
"epoch": 1.5625,
|
| 755 |
-
"eval_loss": 1.3147114515304565,
|
| 756 |
-
"eval_runtime": 14.4752,
|
| 757 |
-
"eval_samples_per_second": 3.454,
|
| 758 |
-
"eval_steps_per_second": 0.484,
|
| 759 |
-
"step": 50
|
| 760 |
-
},
|
| 761 |
-
{
|
| 762 |
-
"epoch": 1.59375,
|
| 763 |
-
"grad_norm": 55.29493713378906,
|
| 764 |
-
"learning_rate": 1.0156250000000001e-05,
|
| 765 |
-
"loss": 0.7953,
|
| 766 |
-
"step": 51
|
| 767 |
-
},
|
| 768 |
-
{
|
| 769 |
-
"epoch": 1.59375,
|
| 770 |
-
"eval_loss": 1.2989192008972168,
|
| 771 |
-
"eval_runtime": 20.9074,
|
| 772 |
-
"eval_samples_per_second": 2.392,
|
| 773 |
-
"eval_steps_per_second": 0.335,
|
| 774 |
-
"step": 51
|
| 775 |
-
},
|
| 776 |
-
{
|
| 777 |
-
"epoch": 1.625,
|
| 778 |
-
"grad_norm": 71.8622817993164,
|
| 779 |
-
"learning_rate": 9.375000000000001e-06,
|
| 780 |
-
"loss": 1.1628,
|
| 781 |
-
"step": 52
|
| 782 |
-
},
|
| 783 |
-
{
|
| 784 |
-
"epoch": 1.625,
|
| 785 |
-
"eval_loss": 1.2810921669006348,
|
| 786 |
-
"eval_runtime": 14.0147,
|
| 787 |
-
"eval_samples_per_second": 3.568,
|
| 788 |
-
"eval_steps_per_second": 0.499,
|
| 789 |
-
"step": 52
|
| 790 |
-
},
|
| 791 |
-
{
|
| 792 |
-
"epoch": 1.65625,
|
| 793 |
-
"grad_norm": 71.86141204833984,
|
| 794 |
-
"learning_rate": 8.59375e-06,
|
| 795 |
-
"loss": 1.5174,
|
| 796 |
-
"step": 53
|
| 797 |
-
},
|
| 798 |
-
{
|
| 799 |
-
"epoch": 1.65625,
|
| 800 |
-
"eval_loss": 1.2605785131454468,
|
| 801 |
-
"eval_runtime": 18.7275,
|
| 802 |
-
"eval_samples_per_second": 2.67,
|
| 803 |
-
"eval_steps_per_second": 0.374,
|
| 804 |
-
"step": 53
|
| 805 |
-
},
|
| 806 |
-
{
|
| 807 |
-
"epoch": 1.6875,
|
| 808 |
-
"grad_norm": 62.22434997558594,
|
| 809 |
-
"learning_rate": 7.8125e-06,
|
| 810 |
-
"loss": 0.9822,
|
| 811 |
-
"step": 54
|
| 812 |
-
},
|
| 813 |
-
{
|
| 814 |
-
"epoch": 1.6875,
|
| 815 |
-
"eval_loss": 1.2227574586868286,
|
| 816 |
-
"eval_runtime": 13.9543,
|
| 817 |
-
"eval_samples_per_second": 3.583,
|
| 818 |
-
"eval_steps_per_second": 0.502,
|
| 819 |
-
"step": 54
|
| 820 |
-
},
|
| 821 |
-
{
|
| 822 |
-
"epoch": 1.71875,
|
| 823 |
-
"grad_norm": 54.53126525878906,
|
| 824 |
-
"learning_rate": 7.031250000000001e-06,
|
| 825 |
-
"loss": 1.0173,
|
| 826 |
-
"step": 55
|
| 827 |
-
},
|
| 828 |
-
{
|
| 829 |
-
"epoch": 1.71875,
|
| 830 |
-
"eval_loss": 1.1982362270355225,
|
| 831 |
-
"eval_runtime": 19.0289,
|
| 832 |
-
"eval_samples_per_second": 2.628,
|
| 833 |
-
"eval_steps_per_second": 0.368,
|
| 834 |
-
"step": 55
|
| 835 |
-
},
|
| 836 |
-
{
|
| 837 |
-
"epoch": 1.75,
|
| 838 |
-
"grad_norm": 76.07511138916016,
|
| 839 |
-
"learning_rate": 6.25e-06,
|
| 840 |
-
"loss": 1.3464,
|
| 841 |
-
"step": 56
|
| 842 |
-
},
|
| 843 |
-
{
|
| 844 |
-
"epoch": 1.75,
|
| 845 |
-
"eval_loss": 1.1850783824920654,
|
| 846 |
-
"eval_runtime": 14.0372,
|
| 847 |
-
"eval_samples_per_second": 3.562,
|
| 848 |
-
"eval_steps_per_second": 0.499,
|
| 849 |
-
"step": 56
|
| 850 |
-
},
|
| 851 |
-
{
|
| 852 |
-
"epoch": 1.78125,
|
| 853 |
-
"grad_norm": 72.6391830444336,
|
| 854 |
-
"learning_rate": 5.46875e-06,
|
| 855 |
-
"loss": 0.8926,
|
| 856 |
-
"step": 57
|
| 857 |
-
},
|
| 858 |
-
{
|
| 859 |
-
"epoch": 1.78125,
|
| 860 |
-
"eval_loss": 1.1799843311309814,
|
| 861 |
-
"eval_runtime": 22.4922,
|
| 862 |
-
"eval_samples_per_second": 2.223,
|
| 863 |
-
"eval_steps_per_second": 0.311,
|
| 864 |
-
"step": 57
|
| 865 |
-
},
|
| 866 |
-
{
|
| 867 |
-
"epoch": 1.8125,
|
| 868 |
-
"grad_norm": 56.88068771362305,
|
| 869 |
-
"learning_rate": 4.6875000000000004e-06,
|
| 870 |
-
"loss": 0.9163,
|
| 871 |
-
"step": 58
|
| 872 |
-
},
|
| 873 |
-
{
|
| 874 |
-
"epoch": 1.8125,
|
| 875 |
-
"eval_loss": 1.1798558235168457,
|
| 876 |
-
"eval_runtime": 14.0488,
|
| 877 |
-
"eval_samples_per_second": 3.559,
|
| 878 |
-
"eval_steps_per_second": 0.498,
|
| 879 |
-
"step": 58
|
| 880 |
-
},
|
| 881 |
-
{
|
| 882 |
-
"epoch": 1.84375,
|
| 883 |
-
"grad_norm": 45.068748474121094,
|
| 884 |
-
"learning_rate": 3.90625e-06,
|
| 885 |
-
"loss": 0.7737,
|
| 886 |
-
"step": 59
|
| 887 |
-
},
|
| 888 |
-
{
|
| 889 |
-
"epoch": 1.84375,
|
| 890 |
-
"eval_loss": 1.178802490234375,
|
| 891 |
-
"eval_runtime": 20.1914,
|
| 892 |
-
"eval_samples_per_second": 2.476,
|
| 893 |
-
"eval_steps_per_second": 0.347,
|
| 894 |
-
"step": 59
|
| 895 |
-
},
|
| 896 |
-
{
|
| 897 |
-
"epoch": 1.875,
|
| 898 |
-
"grad_norm": 65.94618225097656,
|
| 899 |
-
"learning_rate": 3.125e-06,
|
| 900 |
-
"loss": 0.9931,
|
| 901 |
-
"step": 60
|
| 902 |
-
},
|
| 903 |
-
{
|
| 904 |
-
"epoch": 1.875,
|
| 905 |
-
"eval_loss": 1.1730504035949707,
|
| 906 |
-
"eval_runtime": 14.2215,
|
| 907 |
-
"eval_samples_per_second": 3.516,
|
| 908 |
-
"eval_steps_per_second": 0.492,
|
| 909 |
-
"step": 60
|
| 910 |
-
},
|
| 911 |
-
{
|
| 912 |
-
"epoch": 1.90625,
|
| 913 |
-
"grad_norm": 67.28651428222656,
|
| 914 |
-
"learning_rate": 2.3437500000000002e-06,
|
| 915 |
-
"loss": 0.775,
|
| 916 |
-
"step": 61
|
| 917 |
-
},
|
| 918 |
-
{
|
| 919 |
-
"epoch": 1.90625,
|
| 920 |
-
"eval_loss": 1.1700712442398071,
|
| 921 |
-
"eval_runtime": 24.2206,
|
| 922 |
-
"eval_samples_per_second": 2.064,
|
| 923 |
-
"eval_steps_per_second": 0.289,
|
| 924 |
-
"step": 61
|
| 925 |
-
},
|
| 926 |
-
{
|
| 927 |
-
"epoch": 1.9375,
|
| 928 |
-
"grad_norm": 45.8297004699707,
|
| 929 |
-
"learning_rate": 1.5625e-06,
|
| 930 |
-
"loss": 0.655,
|
| 931 |
-
"step": 62
|
| 932 |
-
},
|
| 933 |
-
{
|
| 934 |
-
"epoch": 1.9375,
|
| 935 |
-
"eval_loss": 1.1683861017227173,
|
| 936 |
-
"eval_runtime": 13.9273,
|
| 937 |
-
"eval_samples_per_second": 3.59,
|
| 938 |
-
"eval_steps_per_second": 0.503,
|
| 939 |
-
"step": 62
|
| 940 |
-
},
|
| 941 |
-
{
|
| 942 |
-
"epoch": 1.96875,
|
| 943 |
-
"grad_norm": 64.58641052246094,
|
| 944 |
-
"learning_rate": 7.8125e-07,
|
| 945 |
-
"loss": 1.0303,
|
| 946 |
-
"step": 63
|
| 947 |
-
},
|
| 948 |
-
{
|
| 949 |
-
"epoch": 1.96875,
|
| 950 |
-
"eval_loss": 1.1670336723327637,
|
| 951 |
-
"eval_runtime": 22.0932,
|
| 952 |
-
"eval_samples_per_second": 2.263,
|
| 953 |
-
"eval_steps_per_second": 0.317,
|
| 954 |
-
"step": 63
|
| 955 |
-
},
|
| 956 |
{
|
| 957 |
"epoch": 2.0,
|
| 958 |
-
"grad_norm":
|
| 959 |
"learning_rate": 0.0,
|
| 960 |
-
"loss": 0.
|
| 961 |
-
"step":
|
| 962 |
},
|
| 963 |
{
|
| 964 |
"epoch": 2.0,
|
| 965 |
-
"eval_loss":
|
| 966 |
-
"eval_runtime":
|
| 967 |
-
"eval_samples_per_second":
|
| 968 |
-
"eval_steps_per_second": 0.
|
| 969 |
-
"step":
|
| 970 |
},
|
| 971 |
{
|
| 972 |
"epoch": 2.0,
|
| 973 |
-
"step":
|
| 974 |
-
"total_flos":
|
| 975 |
-
"train_loss": 1.
|
| 976 |
-
"train_runtime":
|
| 977 |
-
"train_samples_per_second": 0.
|
| 978 |
-
"train_steps_per_second": 0.
|
| 979 |
}
|
| 980 |
],
|
| 981 |
"logging_steps": 1.0,
|
| 982 |
-
"max_steps":
|
| 983 |
"num_input_tokens_seen": 0,
|
| 984 |
"num_train_epochs": 2,
|
| 985 |
"save_steps": 500,
|
|
@@ -995,8 +245,8 @@
|
|
| 995 |
"attributes": {}
|
| 996 |
}
|
| 997 |
},
|
| 998 |
-
"total_flos":
|
| 999 |
-
"train_batch_size":
|
| 1000 |
"trial_name": null,
|
| 1001 |
"trial_params": null
|
| 1002 |
}
|
|
|
|
| 3 |
"best_model_checkpoint": null,
|
| 4 |
"epoch": 2.0,
|
| 5 |
"eval_steps": 1.0,
|
| 6 |
+
"global_step": 14,
|
| 7 |
"is_hyper_param_search": false,
|
| 8 |
"is_local_process_zero": true,
|
| 9 |
"is_world_process_zero": true,
|
| 10 |
"log_history": [
|
| 11 |
{
|
| 12 |
+
"epoch": 0.14285714285714285,
|
| 13 |
+
"grad_norm": 203.78257751464844,
|
| 14 |
+
"learning_rate": 4.642857142857143e-05,
|
| 15 |
+
"loss": 0.8088,
|
| 16 |
"step": 1
|
| 17 |
},
|
| 18 |
{
|
| 19 |
+
"epoch": 0.14285714285714285,
|
| 20 |
+
"eval_loss": 1.0832017660140991,
|
| 21 |
+
"eval_runtime": 4.4764,
|
| 22 |
+
"eval_samples_per_second": 2.234,
|
| 23 |
+
"eval_steps_per_second": 1.117,
|
| 24 |
"step": 1
|
| 25 |
},
|
| 26 |
{
|
| 27 |
+
"epoch": 0.2857142857142857,
|
| 28 |
+
"grad_norm": 551.1321411132812,
|
| 29 |
+
"learning_rate": 4.2857142857142856e-05,
|
| 30 |
+
"loss": 1.5034,
|
| 31 |
"step": 2
|
| 32 |
},
|
| 33 |
{
|
| 34 |
+
"epoch": 0.2857142857142857,
|
| 35 |
+
"eval_loss": 0.8266311883926392,
|
| 36 |
+
"eval_runtime": 2.9969,
|
| 37 |
+
"eval_samples_per_second": 3.337,
|
| 38 |
+
"eval_steps_per_second": 1.668,
|
| 39 |
"step": 2
|
| 40 |
},
|
| 41 |
{
|
| 42 |
+
"epoch": 0.42857142857142855,
|
| 43 |
+
"grad_norm": 342.6697998046875,
|
| 44 |
+
"learning_rate": 3.928571428571429e-05,
|
| 45 |
+
"loss": 1.617,
|
| 46 |
"step": 3
|
| 47 |
},
|
| 48 |
{
|
| 49 |
+
"epoch": 0.42857142857142855,
|
| 50 |
+
"eval_loss": 0.7760497331619263,
|
| 51 |
+
"eval_runtime": 2.9732,
|
| 52 |
+
"eval_samples_per_second": 3.363,
|
| 53 |
+
"eval_steps_per_second": 1.682,
|
| 54 |
"step": 3
|
| 55 |
},
|
| 56 |
{
|
| 57 |
+
"epoch": 0.5714285714285714,
|
| 58 |
+
"grad_norm": 792.9654541015625,
|
| 59 |
+
"learning_rate": 3.571428571428572e-05,
|
| 60 |
+
"loss": 2.2375,
|
| 61 |
"step": 4
|
| 62 |
},
|
| 63 |
{
|
| 64 |
+
"epoch": 0.5714285714285714,
|
| 65 |
+
"eval_loss": 0.7384463548660278,
|
| 66 |
+
"eval_runtime": 3.1473,
|
| 67 |
+
"eval_samples_per_second": 3.177,
|
| 68 |
+
"eval_steps_per_second": 1.589,
|
| 69 |
"step": 4
|
| 70 |
},
|
| 71 |
{
|
| 72 |
+
"epoch": 0.7142857142857143,
|
| 73 |
+
"grad_norm": 210.89300537109375,
|
| 74 |
+
"learning_rate": 3.2142857142857144e-05,
|
| 75 |
+
"loss": 1.5411,
|
| 76 |
"step": 5
|
| 77 |
},
|
| 78 |
{
|
| 79 |
+
"epoch": 0.7142857142857143,
|
| 80 |
+
"eval_loss": 0.82770174741745,
|
| 81 |
+
"eval_runtime": 3.0122,
|
| 82 |
+
"eval_samples_per_second": 3.32,
|
| 83 |
+
"eval_steps_per_second": 1.66,
|
| 84 |
"step": 5
|
| 85 |
},
|
| 86 |
{
|
| 87 |
+
"epoch": 0.8571428571428571,
|
| 88 |
+
"grad_norm": 314.6759338378906,
|
| 89 |
+
"learning_rate": 2.857142857142857e-05,
|
| 90 |
+
"loss": 2.4631,
|
| 91 |
"step": 6
|
| 92 |
},
|
| 93 |
{
|
| 94 |
+
"epoch": 0.8571428571428571,
|
| 95 |
+
"eval_loss": 0.44329363107681274,
|
| 96 |
+
"eval_runtime": 4.1331,
|
| 97 |
+
"eval_samples_per_second": 2.42,
|
| 98 |
+
"eval_steps_per_second": 1.21,
|
| 99 |
"step": 6
|
| 100 |
},
|
| 101 |
{
|
| 102 |
+
"epoch": 1.0,
|
| 103 |
+
"grad_norm": 17.239919662475586,
|
| 104 |
+
"learning_rate": 2.5e-05,
|
| 105 |
+
"loss": 0.0217,
|
| 106 |
"step": 7
|
| 107 |
},
|
| 108 |
{
|
| 109 |
+
"epoch": 1.0,
|
| 110 |
+
"eval_loss": 0.9471458196640015,
|
| 111 |
+
"eval_runtime": 3.013,
|
| 112 |
+
"eval_samples_per_second": 3.319,
|
| 113 |
+
"eval_steps_per_second": 1.659,
|
| 114 |
"step": 7
|
| 115 |
},
|
| 116 |
{
|
| 117 |
+
"epoch": 1.1428571428571428,
|
| 118 |
+
"grad_norm": 213.36294555664062,
|
| 119 |
+
"learning_rate": 2.1428571428571428e-05,
|
| 120 |
+
"loss": 2.4217,
|
| 121 |
"step": 8
|
| 122 |
},
|
| 123 |
{
|
| 124 |
+
"epoch": 1.1428571428571428,
|
| 125 |
+
"eval_loss": 0.9182891845703125,
|
| 126 |
+
"eval_runtime": 2.9638,
|
| 127 |
+
"eval_samples_per_second": 3.374,
|
| 128 |
+
"eval_steps_per_second": 1.687,
|
| 129 |
"step": 8
|
| 130 |
},
|
| 131 |
{
|
| 132 |
+
"epoch": 1.2857142857142856,
|
| 133 |
+
"grad_norm": 200.6674346923828,
|
| 134 |
+
"learning_rate": 1.785714285714286e-05,
|
| 135 |
+
"loss": 2.0588,
|
| 136 |
"step": 9
|
| 137 |
},
|
| 138 |
{
|
| 139 |
+
"epoch": 1.2857142857142856,
|
| 140 |
+
"eval_loss": 0.7376034259796143,
|
| 141 |
+
"eval_runtime": 2.958,
|
| 142 |
+
"eval_samples_per_second": 3.381,
|
| 143 |
+
"eval_steps_per_second": 1.69,
|
| 144 |
"step": 9
|
| 145 |
},
|
| 146 |
{
|
| 147 |
+
"epoch": 1.4285714285714286,
|
| 148 |
+
"grad_norm": 215.76296997070312,
|
| 149 |
+
"learning_rate": 1.4285714285714285e-05,
|
| 150 |
+
"loss": 1.6484,
|
| 151 |
"step": 10
|
| 152 |
},
|
| 153 |
{
|
| 154 |
+
"epoch": 1.4285714285714286,
|
| 155 |
+
"eval_loss": 0.5826703906059265,
|
| 156 |
+
"eval_runtime": 2.9863,
|
| 157 |
+
"eval_samples_per_second": 3.349,
|
| 158 |
+
"eval_steps_per_second": 1.674,
|
| 159 |
"step": 10
|
| 160 |
},
|
| 161 |
{
|
| 162 |
+
"epoch": 1.5714285714285714,
|
| 163 |
+
"grad_norm": 122.98780059814453,
|
| 164 |
+
"learning_rate": 1.0714285714285714e-05,
|
| 165 |
+
"loss": 0.9379,
|
| 166 |
"step": 11
|
| 167 |
},
|
| 168 |
{
|
| 169 |
+
"epoch": 1.5714285714285714,
|
| 170 |
+
"eval_loss": 0.5854327082633972,
|
| 171 |
+
"eval_runtime": 2.9738,
|
| 172 |
+
"eval_samples_per_second": 3.363,
|
| 173 |
+
"eval_steps_per_second": 1.681,
|
| 174 |
"step": 11
|
| 175 |
},
|
| 176 |
{
|
| 177 |
+
"epoch": 1.7142857142857144,
|
| 178 |
+
"grad_norm": 110.82337951660156,
|
| 179 |
+
"learning_rate": 7.142857142857143e-06,
|
| 180 |
+
"loss": 0.8608,
|
| 181 |
"step": 12
|
| 182 |
},
|
| 183 |
{
|
| 184 |
+
"epoch": 1.7142857142857144,
|
| 185 |
+
"eval_loss": 0.5832847952842712,
|
| 186 |
+
"eval_runtime": 6.4862,
|
| 187 |
+
"eval_samples_per_second": 1.542,
|
| 188 |
+
"eval_steps_per_second": 0.771,
|
| 189 |
"step": 12
|
| 190 |
},
|
| 191 |
{
|
| 192 |
+
"epoch": 1.8571428571428572,
|
| 193 |
+
"grad_norm": 91.99958801269531,
|
| 194 |
+
"learning_rate": 3.5714285714285714e-06,
|
| 195 |
+
"loss": 0.958,
|
| 196 |
"step": 13
|
| 197 |
},
|
| 198 |
{
|
| 199 |
+
"epoch": 1.8571428571428572,
|
| 200 |
+
"eval_loss": 0.5970481634140015,
|
| 201 |
+
"eval_runtime": 2.9762,
|
| 202 |
+
"eval_samples_per_second": 3.36,
|
| 203 |
+
"eval_steps_per_second": 1.68,
|
| 204 |
"step": 13
|
| 205 |
},
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 206 |
{
|
| 207 |
"epoch": 2.0,
|
| 208 |
+
"grad_norm": 28.6821346282959,
|
| 209 |
"learning_rate": 0.0,
|
| 210 |
+
"loss": 0.0355,
|
| 211 |
+
"step": 14
|
| 212 |
},
|
| 213 |
{
|
| 214 |
"epoch": 2.0,
|
| 215 |
+
"eval_loss": 0.5955778956413269,
|
| 216 |
+
"eval_runtime": 5.3785,
|
| 217 |
+
"eval_samples_per_second": 1.859,
|
| 218 |
+
"eval_steps_per_second": 0.93,
|
| 219 |
+
"step": 14
|
| 220 |
},
|
| 221 |
{
|
| 222 |
"epoch": 2.0,
|
| 223 |
+
"step": 14,
|
| 224 |
+
"total_flos": 5816699796600.0,
|
| 225 |
+
"train_loss": 1.3652541448495217,
|
| 226 |
+
"train_runtime": 216.8469,
|
| 227 |
+
"train_samples_per_second": 0.461,
|
| 228 |
+
"train_steps_per_second": 0.065
|
| 229 |
}
|
| 230 |
],
|
| 231 |
"logging_steps": 1.0,
|
| 232 |
+
"max_steps": 14,
|
| 233 |
"num_input_tokens_seen": 0,
|
| 234 |
"num_train_epochs": 2,
|
| 235 |
"save_steps": 500,
|
|
|
|
| 245 |
"attributes": {}
|
| 246 |
}
|
| 247 |
},
|
| 248 |
+
"total_flos": 5816699796600.0,
|
| 249 |
+
"train_batch_size": 8,
|
| 250 |
"trial_name": null,
|
| 251 |
"trial_params": null
|
| 252 |
}
|