Kerem P
commited on
Commit
•
8f1c71f
1
Parent(s):
95a3649
End of training
Browse files- all_results.json +10 -10
- eval_results.json +6 -6
- train_results.json +5 -5
- trainer_state.json +247 -1096
all_results.json
CHANGED
|
@@ -1,14 +1,14 @@
|
|
| 1 |
{
|
| 2 |
-
"epoch":
|
| 3 |
-
"eval_f1": 0.
|
| 4 |
-
"eval_loss":
|
| 5 |
-
"eval_runtime": 20.
|
| 6 |
"eval_samples": 1916,
|
| 7 |
-
"eval_samples_per_second": 93.
|
| 8 |
-
"eval_steps_per_second": 1.
|
| 9 |
-
"train_loss":
|
| 10 |
-
"train_runtime":
|
| 11 |
"train_samples": 5743,
|
| 12 |
-
"train_samples_per_second":
|
| 13 |
-
"train_steps_per_second": 0.
|
| 14 |
}
|
|
|
|
| 1 |
{
|
| 2 |
+
"epoch": 3.0,
|
| 3 |
+
"eval_f1": 0.19553868058851448,
|
| 4 |
+
"eval_loss": 5.8615827560424805,
|
| 5 |
+
"eval_runtime": 20.4376,
|
| 6 |
"eval_samples": 1916,
|
| 7 |
+
"eval_samples_per_second": 93.749,
|
| 8 |
+
"eval_steps_per_second": 1.468,
|
| 9 |
+
"train_loss": 15.830373929303002,
|
| 10 |
+
"train_runtime": 1144.0282,
|
| 11 |
"train_samples": 5743,
|
| 12 |
+
"train_samples_per_second": 15.06,
|
| 13 |
+
"train_steps_per_second": 0.472
|
| 14 |
}
|
eval_results.json
CHANGED
|
@@ -1,9 +1,9 @@
|
|
| 1 |
{
|
| 2 |
-
"epoch":
|
| 3 |
-
"eval_f1": 0.
|
| 4 |
-
"eval_loss":
|
| 5 |
-
"eval_runtime": 20.
|
| 6 |
"eval_samples": 1916,
|
| 7 |
-
"eval_samples_per_second": 93.
|
| 8 |
-
"eval_steps_per_second": 1.
|
| 9 |
}
|
|
|
|
| 1 |
{
|
| 2 |
+
"epoch": 3.0,
|
| 3 |
+
"eval_f1": 0.19553868058851448,
|
| 4 |
+
"eval_loss": 5.8615827560424805,
|
| 5 |
+
"eval_runtime": 20.4376,
|
| 6 |
"eval_samples": 1916,
|
| 7 |
+
"eval_samples_per_second": 93.749,
|
| 8 |
+
"eval_steps_per_second": 1.468
|
| 9 |
}
|
train_results.json
CHANGED
|
@@ -1,8 +1,8 @@
|
|
| 1 |
{
|
| 2 |
-
"epoch":
|
| 3 |
-
"train_loss":
|
| 4 |
-
"train_runtime":
|
| 5 |
"train_samples": 5743,
|
| 6 |
-
"train_samples_per_second":
|
| 7 |
-
"train_steps_per_second": 0.
|
| 8 |
}
|
|
|
|
| 1 |
{
|
| 2 |
+
"epoch": 3.0,
|
| 3 |
+
"train_loss": 15.830373929303002,
|
| 4 |
+
"train_runtime": 1144.0282,
|
| 5 |
"train_samples": 5743,
|
| 6 |
+
"train_samples_per_second": 15.06,
|
| 7 |
+
"train_steps_per_second": 0.472
|
| 8 |
}
|
trainer_state.json
CHANGED
|
@@ -1,1230 +1,381 @@
|
|
| 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": 0.11,
|
| 13 |
-
"learning_rate":
|
| 14 |
-
"loss":
|
| 15 |
"step": 20
|
| 16 |
},
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 17 |
{
|
| 18 |
"epoch": 0.22,
|
| 19 |
-
"learning_rate":
|
| 20 |
-
"loss":
|
| 21 |
"step": 40
|
| 22 |
},
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 23 |
{
|
| 24 |
"epoch": 0.33,
|
| 25 |
-
"learning_rate":
|
| 26 |
-
"loss":
|
| 27 |
"step": 60
|
| 28 |
},
|
| 29 |
{
|
| 30 |
-
"epoch": 0.
|
| 31 |
-
"learning_rate":
|
| 32 |
-
"loss":
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 33 |
"step": 80
|
| 34 |
},
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 35 |
{
|
| 36 |
"epoch": 0.56,
|
| 37 |
-
"learning_rate":
|
| 38 |
-
"loss":
|
| 39 |
"step": 100
|
| 40 |
},
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 41 |
{
|
| 42 |
"epoch": 0.67,
|
| 43 |
-
"learning_rate":
|
| 44 |
-
"loss":
|
| 45 |
"step": 120
|
| 46 |
},
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 47 |
{
|
| 48 |
"epoch": 0.78,
|
| 49 |
-
"learning_rate":
|
| 50 |
-
"loss":
|
| 51 |
"step": 140
|
| 52 |
},
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 53 |
{
|
| 54 |
"epoch": 0.89,
|
| 55 |
-
"learning_rate":
|
| 56 |
-
"loss":
|
| 57 |
"step": 160
|
| 58 |
},
|
| 59 |
{
|
| 60 |
-
"epoch":
|
| 61 |
-
"
|
| 62 |
-
"
|
| 63 |
-
"
|
| 64 |
-
"eval_samples_per_second": 82.614,
|
| 65 |
-
"eval_steps_per_second": 1.294,
|
| 66 |
-
"step": 179
|
| 67 |
},
|
| 68 |
{
|
| 69 |
"epoch": 1.0,
|
| 70 |
-
"learning_rate":
|
| 71 |
-
"loss":
|
| 72 |
"step": 180
|
| 73 |
},
|
| 74 |
{
|
| 75 |
-
"epoch": 1.
|
| 76 |
-
"learning_rate": 2e-05,
|
| 77 |
-
"loss": 0.3674,
|
| 78 |
-
"step": 200
|
| 79 |
-
},
|
| 80 |
-
{
|
| 81 |
-
"epoch": 1.23,
|
| 82 |
-
"learning_rate": 2e-05,
|
| 83 |
-
"loss": 0.5531,
|
| 84 |
-
"step": 220
|
| 85 |
-
},
|
| 86 |
-
{
|
| 87 |
-
"epoch": 1.34,
|
| 88 |
-
"learning_rate": 2e-05,
|
| 89 |
-
"loss": 0.2448,
|
| 90 |
-
"step": 240
|
| 91 |
-
},
|
| 92 |
-
{
|
| 93 |
-
"epoch": 1.45,
|
| 94 |
-
"learning_rate": 2e-05,
|
| 95 |
-
"loss": 0.1894,
|
| 96 |
-
"step": 260
|
| 97 |
-
},
|
| 98 |
-
{
|
| 99 |
-
"epoch": 1.56,
|
| 100 |
-
"learning_rate": 2e-05,
|
| 101 |
-
"loss": 0.2385,
|
| 102 |
-
"step": 280
|
| 103 |
-
},
|
| 104 |
-
{
|
| 105 |
-
"epoch": 1.67,
|
| 106 |
-
"learning_rate": 2e-05,
|
| 107 |
-
"loss": 0.2773,
|
| 108 |
-
"step": 300
|
| 109 |
-
},
|
| 110 |
-
{
|
| 111 |
-
"epoch": 1.78,
|
| 112 |
-
"learning_rate": 2e-05,
|
| 113 |
-
"loss": 0.1879,
|
| 114 |
-
"step": 320
|
| 115 |
-
},
|
| 116 |
-
{
|
| 117 |
-
"epoch": 1.89,
|
| 118 |
-
"learning_rate": 2e-05,
|
| 119 |
-
"loss": 0.1541,
|
| 120 |
-
"step": 340
|
| 121 |
-
},
|
| 122 |
-
{
|
| 123 |
-
"epoch": 2.0,
|
| 124 |
-
"eval_f1": 0.1941564561734213,
|
| 125 |
-
"eval_loss": 8.468379020690918,
|
| 126 |
-
"eval_runtime": 22.9961,
|
| 127 |
-
"eval_samples_per_second": 83.318,
|
| 128 |
-
"eval_steps_per_second": 1.305,
|
| 129 |
-
"step": 359
|
| 130 |
-
},
|
| 131 |
-
{
|
| 132 |
-
"epoch": 2.01,
|
| 133 |
-
"learning_rate": 2e-05,
|
| 134 |
-
"loss": 0.1381,
|
| 135 |
-
"step": 360
|
| 136 |
-
},
|
| 137 |
-
{
|
| 138 |
-
"epoch": 2.12,
|
| 139 |
-
"learning_rate": 2e-05,
|
| 140 |
-
"loss": 0.1464,
|
| 141 |
-
"step": 380
|
| 142 |
-
},
|
| 143 |
-
{
|
| 144 |
-
"epoch": 2.23,
|
| 145 |
-
"learning_rate": 2e-05,
|
| 146 |
-
"loss": 0.2856,
|
| 147 |
-
"step": 400
|
| 148 |
-
},
|
| 149 |
-
{
|
| 150 |
-
"epoch": 2.34,
|
| 151 |
-
"learning_rate": 2e-05,
|
| 152 |
-
"loss": 0.3727,
|
| 153 |
-
"step": 420
|
| 154 |
-
},
|
| 155 |
-
{
|
| 156 |
-
"epoch": 2.45,
|
| 157 |
-
"learning_rate": 2e-05,
|
| 158 |
-
"loss": 0.2145,
|
| 159 |
-
"step": 440
|
| 160 |
-
},
|
| 161 |
-
{
|
| 162 |
-
"epoch": 2.56,
|
| 163 |
-
"learning_rate": 2e-05,
|
| 164 |
-
"loss": 0.0867,
|
| 165 |
-
"step": 460
|
| 166 |
-
},
|
| 167 |
-
{
|
| 168 |
-
"epoch": 2.67,
|
| 169 |
-
"learning_rate": 2e-05,
|
| 170 |
-
"loss": 0.232,
|
| 171 |
-
"step": 480
|
| 172 |
-
},
|
| 173 |
-
{
|
| 174 |
-
"epoch": 2.79,
|
| 175 |
-
"learning_rate": 2e-05,
|
| 176 |
-
"loss": 0.1407,
|
| 177 |
-
"step": 500
|
| 178 |
-
},
|
| 179 |
-
{
|
| 180 |
-
"epoch": 2.9,
|
| 181 |
-
"learning_rate": 2e-05,
|
| 182 |
-
"loss": 0.1257,
|
| 183 |
-
"step": 520
|
| 184 |
-
},
|
| 185 |
-
{
|
| 186 |
-
"epoch": 3.0,
|
| 187 |
"eval_f1": 0.1941564561734213,
|
| 188 |
-
"eval_loss":
|
| 189 |
-
"eval_runtime":
|
| 190 |
-
"eval_samples_per_second":
|
| 191 |
-
"eval_steps_per_second": 1.
|
| 192 |
-
"step":
|
| 193 |
-
},
|
| 194 |
-
{
|
| 195 |
-
"epoch": 3.01,
|
| 196 |
-
"learning_rate": 2e-05,
|
| 197 |
-
"loss": 0.1963,
|
| 198 |
-
"step": 540
|
| 199 |
},
|
| 200 |
{
|
| 201 |
-
"epoch":
|
| 202 |
-
"learning_rate":
|
| 203 |
-
"loss":
|
| 204 |
-
"step":
|
| 205 |
},
|
| 206 |
{
|
| 207 |
-
"epoch":
|
| 208 |
-
"learning_rate":
|
| 209 |
-
"loss":
|
| 210 |
-
"step":
|
| 211 |
},
|
| 212 |
{
|
| 213 |
-
"epoch":
|
| 214 |
-
"learning_rate":
|
| 215 |
-
"loss": 0.
|
| 216 |
-
"step":
|
| 217 |
},
|
| 218 |
{
|
| 219 |
-
"epoch":
|
| 220 |
-
"learning_rate":
|
| 221 |
-
"loss": 0.
|
| 222 |
-
"step":
|
| 223 |
},
|
| 224 |
{
|
| 225 |
-
"epoch":
|
| 226 |
-
"learning_rate":
|
| 227 |
-
"loss":
|
| 228 |
-
"step":
|
| 229 |
},
|
| 230 |
{
|
| 231 |
-
"epoch":
|
| 232 |
-
"learning_rate":
|
| 233 |
-
"loss": 0.
|
| 234 |
-
"step":
|
| 235 |
},
|
| 236 |
{
|
| 237 |
-
"epoch":
|
| 238 |
-
"learning_rate":
|
| 239 |
-
"loss": 0.
|
| 240 |
-
"step":
|
| 241 |
},
|
| 242 |
{
|
| 243 |
-
"epoch":
|
| 244 |
-
"learning_rate":
|
| 245 |
-
"loss": 0.
|
| 246 |
-
"step":
|
| 247 |
},
|
| 248 |
{
|
| 249 |
-
"epoch":
|
| 250 |
-
"
|
| 251 |
-
"
|
| 252 |
-
"
|
| 253 |
-
"eval_samples_per_second": 83.42,
|
| 254 |
-
"eval_steps_per_second": 1.306,
|
| 255 |
-
"step": 718
|
| 256 |
},
|
| 257 |
{
|
| 258 |
-
"epoch":
|
| 259 |
-
"learning_rate":
|
| 260 |
-
"loss": 0.
|
| 261 |
-
"step":
|
| 262 |
},
|
| 263 |
{
|
| 264 |
-
"epoch":
|
| 265 |
-
"learning_rate":
|
| 266 |
-
"loss": 0.
|
| 267 |
-
"step":
|
| 268 |
},
|
| 269 |
{
|
| 270 |
-
"epoch":
|
| 271 |
-
"learning_rate":
|
| 272 |
-
"loss":
|
| 273 |
-
"step":
|
| 274 |
},
|
| 275 |
{
|
| 276 |
-
"epoch":
|
| 277 |
-
"learning_rate":
|
| 278 |
-
"loss": 0.
|
| 279 |
-
"step":
|
| 280 |
},
|
| 281 |
{
|
| 282 |
-
"epoch":
|
| 283 |
-
"learning_rate":
|
| 284 |
-
"loss":
|
| 285 |
-
"step":
|
| 286 |
},
|
| 287 |
{
|
| 288 |
-
"epoch":
|
| 289 |
-
"learning_rate":
|
| 290 |
-
"loss": 0.
|
| 291 |
-
"step":
|
| 292 |
},
|
| 293 |
{
|
| 294 |
-
"epoch":
|
| 295 |
-
"learning_rate":
|
| 296 |
-
"loss": 0.
|
| 297 |
-
"step":
|
| 298 |
},
|
| 299 |
{
|
| 300 |
-
"epoch":
|
| 301 |
-
"learning_rate":
|
| 302 |
-
"loss": 0.
|
| 303 |
-
"step":
|
| 304 |
},
|
| 305 |
{
|
| 306 |
-
"epoch":
|
| 307 |
-
"learning_rate":
|
| 308 |
-
"loss": 0.
|
| 309 |
-
"step":
|
| 310 |
},
|
| 311 |
{
|
| 312 |
-
"epoch":
|
| 313 |
"eval_f1": 0.1941564561734213,
|
| 314 |
-
"eval_loss":
|
| 315 |
-
"eval_runtime":
|
| 316 |
-
"eval_samples_per_second":
|
| 317 |
-
"eval_steps_per_second": 1.
|
| 318 |
-
"step":
|
| 319 |
-
},
|
| 320 |
-
{
|
| 321 |
-
"epoch": 5.01,
|
| 322 |
-
"learning_rate": 2e-05,
|
| 323 |
-
"loss": 0.0993,
|
| 324 |
-
"step": 900
|
| 325 |
-
},
|
| 326 |
-
{
|
| 327 |
-
"epoch": 5.13,
|
| 328 |
-
"learning_rate": 2e-05,
|
| 329 |
-
"loss": 0.0972,
|
| 330 |
-
"step": 920
|
| 331 |
-
},
|
| 332 |
-
{
|
| 333 |
-
"epoch": 5.24,
|
| 334 |
-
"learning_rate": 2e-05,
|
| 335 |
-
"loss": 0.0389,
|
| 336 |
-
"step": 940
|
| 337 |
-
},
|
| 338 |
-
{
|
| 339 |
-
"epoch": 5.35,
|
| 340 |
-
"learning_rate": 2e-05,
|
| 341 |
-
"loss": 0.1366,
|
| 342 |
-
"step": 960
|
| 343 |
-
},
|
| 344 |
-
{
|
| 345 |
-
"epoch": 5.46,
|
| 346 |
-
"learning_rate": 2e-05,
|
| 347 |
-
"loss": 0.0833,
|
| 348 |
-
"step": 980
|
| 349 |
-
},
|
| 350 |
-
{
|
| 351 |
-
"epoch": 5.57,
|
| 352 |
-
"learning_rate": 2e-05,
|
| 353 |
-
"loss": 0.1634,
|
| 354 |
-
"step": 1000
|
| 355 |
-
},
|
| 356 |
-
{
|
| 357 |
-
"epoch": 5.68,
|
| 358 |
-
"learning_rate": 2e-05,
|
| 359 |
-
"loss": 0.0691,
|
| 360 |
-
"step": 1020
|
| 361 |
-
},
|
| 362 |
-
{
|
| 363 |
-
"epoch": 5.79,
|
| 364 |
-
"learning_rate": 2e-05,
|
| 365 |
-
"loss": 0.1487,
|
| 366 |
-
"step": 1040
|
| 367 |
-
},
|
| 368 |
-
{
|
| 369 |
-
"epoch": 5.91,
|
| 370 |
-
"learning_rate": 2e-05,
|
| 371 |
-
"loss": 0.145,
|
| 372 |
-
"step": 1060
|
| 373 |
-
},
|
| 374 |
-
{
|
| 375 |
-
"epoch": 6.0,
|
| 376 |
-
"eval_f1": 0.7984031936127745,
|
| 377 |
-
"eval_loss": 0.2693595290184021,
|
| 378 |
-
"eval_runtime": 22.9118,
|
| 379 |
-
"eval_samples_per_second": 83.625,
|
| 380 |
-
"eval_steps_per_second": 1.309,
|
| 381 |
-
"step": 1077
|
| 382 |
-
},
|
| 383 |
-
{
|
| 384 |
-
"epoch": 6.02,
|
| 385 |
-
"learning_rate": 2e-05,
|
| 386 |
-
"loss": 0.0373,
|
| 387 |
-
"step": 1080
|
| 388 |
-
},
|
| 389 |
-
{
|
| 390 |
-
"epoch": 6.13,
|
| 391 |
-
"learning_rate": 2e-05,
|
| 392 |
-
"loss": 0.0409,
|
| 393 |
-
"step": 1100
|
| 394 |
-
},
|
| 395 |
-
{
|
| 396 |
-
"epoch": 6.24,
|
| 397 |
-
"learning_rate": 2e-05,
|
| 398 |
-
"loss": 0.0714,
|
| 399 |
-
"step": 1120
|
| 400 |
-
},
|
| 401 |
-
{
|
| 402 |
-
"epoch": 6.35,
|
| 403 |
-
"learning_rate": 2e-05,
|
| 404 |
-
"loss": 0.0915,
|
| 405 |
-
"step": 1140
|
| 406 |
-
},
|
| 407 |
-
{
|
| 408 |
-
"epoch": 6.46,
|
| 409 |
-
"learning_rate": 2e-05,
|
| 410 |
-
"loss": 0.1359,
|
| 411 |
-
"step": 1160
|
| 412 |
-
},
|
| 413 |
-
{
|
| 414 |
-
"epoch": 6.57,
|
| 415 |
-
"learning_rate": 2e-05,
|
| 416 |
-
"loss": 0.1016,
|
| 417 |
-
"step": 1180
|
| 418 |
-
},
|
| 419 |
-
{
|
| 420 |
-
"epoch": 6.69,
|
| 421 |
-
"learning_rate": 2e-05,
|
| 422 |
-
"loss": 0.0346,
|
| 423 |
-
"step": 1200
|
| 424 |
-
},
|
| 425 |
-
{
|
| 426 |
-
"epoch": 6.8,
|
| 427 |
-
"learning_rate": 2e-05,
|
| 428 |
-
"loss": 0.0437,
|
| 429 |
-
"step": 1220
|
| 430 |
-
},
|
| 431 |
-
{
|
| 432 |
-
"epoch": 6.91,
|
| 433 |
-
"learning_rate": 2e-05,
|
| 434 |
-
"loss": 0.1191,
|
| 435 |
-
"step": 1240
|
| 436 |
-
},
|
| 437 |
-
{
|
| 438 |
-
"epoch": 7.0,
|
| 439 |
-
"eval_f1": 0.20265617314313825,
|
| 440 |
-
"eval_loss": 2.941455602645874,
|
| 441 |
-
"eval_runtime": 22.6391,
|
| 442 |
-
"eval_samples_per_second": 84.632,
|
| 443 |
-
"eval_steps_per_second": 1.325,
|
| 444 |
-
"step": 1256
|
| 445 |
-
},
|
| 446 |
-
{
|
| 447 |
-
"epoch": 7.02,
|
| 448 |
-
"learning_rate": 2e-05,
|
| 449 |
-
"loss": 0.069,
|
| 450 |
-
"step": 1260
|
| 451 |
-
},
|
| 452 |
-
{
|
| 453 |
-
"epoch": 7.13,
|
| 454 |
-
"learning_rate": 2e-05,
|
| 455 |
-
"loss": 0.009,
|
| 456 |
-
"step": 1280
|
| 457 |
-
},
|
| 458 |
-
{
|
| 459 |
-
"epoch": 7.24,
|
| 460 |
-
"learning_rate": 2e-05,
|
| 461 |
-
"loss": 0.0485,
|
| 462 |
-
"step": 1300
|
| 463 |
-
},
|
| 464 |
-
{
|
| 465 |
-
"epoch": 7.35,
|
| 466 |
-
"learning_rate": 2e-05,
|
| 467 |
-
"loss": 0.0105,
|
| 468 |
-
"step": 1320
|
| 469 |
-
},
|
| 470 |
-
{
|
| 471 |
-
"epoch": 7.47,
|
| 472 |
-
"learning_rate": 2e-05,
|
| 473 |
-
"loss": 0.0835,
|
| 474 |
-
"step": 1340
|
| 475 |
-
},
|
| 476 |
-
{
|
| 477 |
-
"epoch": 7.58,
|
| 478 |
-
"learning_rate": 2e-05,
|
| 479 |
-
"loss": 0.1458,
|
| 480 |
-
"step": 1360
|
| 481 |
-
},
|
| 482 |
-
{
|
| 483 |
-
"epoch": 7.69,
|
| 484 |
-
"learning_rate": 2e-05,
|
| 485 |
-
"loss": 0.0553,
|
| 486 |
-
"step": 1380
|
| 487 |
-
},
|
| 488 |
-
{
|
| 489 |
-
"epoch": 7.8,
|
| 490 |
-
"learning_rate": 2e-05,
|
| 491 |
-
"loss": 0.005,
|
| 492 |
-
"step": 1400
|
| 493 |
-
},
|
| 494 |
-
{
|
| 495 |
-
"epoch": 7.91,
|
| 496 |
-
"learning_rate": 2e-05,
|
| 497 |
-
"loss": 0.1008,
|
| 498 |
-
"step": 1420
|
| 499 |
-
},
|
| 500 |
-
{
|
| 501 |
-
"epoch": 8.0,
|
| 502 |
-
"eval_f1": 0.9023255813953488,
|
| 503 |
-
"eval_loss": 0.17851048707962036,
|
| 504 |
-
"eval_runtime": 22.8798,
|
| 505 |
-
"eval_samples_per_second": 83.742,
|
| 506 |
-
"eval_steps_per_second": 1.311,
|
| 507 |
-
"step": 1436
|
| 508 |
-
},
|
| 509 |
-
{
|
| 510 |
-
"epoch": 8.02,
|
| 511 |
-
"learning_rate": 2e-05,
|
| 512 |
-
"loss": 0.0414,
|
| 513 |
-
"step": 1440
|
| 514 |
-
},
|
| 515 |
-
{
|
| 516 |
-
"epoch": 8.13,
|
| 517 |
-
"learning_rate": 2e-05,
|
| 518 |
-
"loss": 0.0351,
|
| 519 |
-
"step": 1460
|
| 520 |
-
},
|
| 521 |
-
{
|
| 522 |
-
"epoch": 8.25,
|
| 523 |
-
"learning_rate": 2e-05,
|
| 524 |
-
"loss": 0.115,
|
| 525 |
-
"step": 1480
|
| 526 |
-
},
|
| 527 |
-
{
|
| 528 |
-
"epoch": 8.36,
|
| 529 |
-
"learning_rate": 2e-05,
|
| 530 |
-
"loss": 0.0453,
|
| 531 |
-
"step": 1500
|
| 532 |
-
},
|
| 533 |
-
{
|
| 534 |
-
"epoch": 4.23,
|
| 535 |
-
"learning_rate": 0.001,
|
| 536 |
-
"loss": 18.7223,
|
| 537 |
-
"step": 1520
|
| 538 |
-
},
|
| 539 |
-
{
|
| 540 |
-
"epoch": 4.29,
|
| 541 |
-
"learning_rate": 0.001,
|
| 542 |
-
"loss": 5.5701,
|
| 543 |
-
"step": 1540
|
| 544 |
-
},
|
| 545 |
-
{
|
| 546 |
-
"epoch": 4.35,
|
| 547 |
-
"learning_rate": 0.001,
|
| 548 |
-
"loss": 1.2935,
|
| 549 |
-
"step": 1560
|
| 550 |
-
},
|
| 551 |
-
{
|
| 552 |
-
"epoch": 4.4,
|
| 553 |
-
"learning_rate": 0.001,
|
| 554 |
-
"loss": 0.6161,
|
| 555 |
-
"step": 1580
|
| 556 |
-
},
|
| 557 |
-
{
|
| 558 |
-
"epoch": 4.46,
|
| 559 |
-
"learning_rate": 0.001,
|
| 560 |
-
"loss": 0.757,
|
| 561 |
-
"step": 1600
|
| 562 |
-
},
|
| 563 |
-
{
|
| 564 |
-
"epoch": 4.51,
|
| 565 |
-
"learning_rate": 0.001,
|
| 566 |
-
"loss": 0.6241,
|
| 567 |
-
"step": 1620
|
| 568 |
-
},
|
| 569 |
-
{
|
| 570 |
-
"epoch": 4.57,
|
| 571 |
-
"learning_rate": 0.001,
|
| 572 |
-
"loss": 0.5211,
|
| 573 |
-
"step": 1640
|
| 574 |
-
},
|
| 575 |
-
{
|
| 576 |
-
"epoch": 4.62,
|
| 577 |
-
"learning_rate": 0.001,
|
| 578 |
-
"loss": 0.4467,
|
| 579 |
-
"step": 1660
|
| 580 |
-
},
|
| 581 |
-
{
|
| 582 |
-
"epoch": 4.68,
|
| 583 |
-
"learning_rate": 0.001,
|
| 584 |
-
"loss": 0.424,
|
| 585 |
-
"step": 1680
|
| 586 |
-
},
|
| 587 |
-
{
|
| 588 |
-
"epoch": 4.74,
|
| 589 |
-
"learning_rate": 0.001,
|
| 590 |
-
"loss": 0.3741,
|
| 591 |
-
"step": 1700
|
| 592 |
-
},
|
| 593 |
-
{
|
| 594 |
-
"epoch": 4.79,
|
| 595 |
-
"learning_rate": 0.001,
|
| 596 |
-
"loss": 0.3276,
|
| 597 |
-
"step": 1720
|
| 598 |
},
|
| 599 |
{
|
| 600 |
-
"epoch":
|
| 601 |
-
"learning_rate":
|
| 602 |
-
"loss": 0.
|
| 603 |
-
"step":
|
| 604 |
},
|
| 605 |
{
|
| 606 |
-
"epoch":
|
| 607 |
-
"learning_rate":
|
| 608 |
"loss": 0.3626,
|
| 609 |
-
"step":
|
| 610 |
-
},
|
| 611 |
-
{
|
| 612 |
-
"epoch": 4.96,
|
| 613 |
-
"learning_rate": 0.001,
|
| 614 |
-
"loss": 0.3698,
|
| 615 |
-
"step": 1780
|
| 616 |
-
},
|
| 617 |
-
{
|
| 618 |
-
"epoch": 5.0,
|
| 619 |
-
"eval_f1": 0.0,
|
| 620 |
-
"eval_loss": 0.3513816297054291,
|
| 621 |
-
"eval_runtime": 20.8512,
|
| 622 |
-
"eval_samples_per_second": 91.889,
|
| 623 |
-
"eval_steps_per_second": 1.439,
|
| 624 |
-
"step": 1795
|
| 625 |
-
},
|
| 626 |
-
{
|
| 627 |
-
"epoch": 5.01,
|
| 628 |
-
"learning_rate": 0.001,
|
| 629 |
-
"loss": 0.3672,
|
| 630 |
-
"step": 1800
|
| 631 |
-
},
|
| 632 |
-
{
|
| 633 |
-
"epoch": 5.07,
|
| 634 |
-
"learning_rate": 0.001,
|
| 635 |
-
"loss": 0.3879,
|
| 636 |
-
"step": 1820
|
| 637 |
-
},
|
| 638 |
-
{
|
| 639 |
-
"epoch": 5.13,
|
| 640 |
-
"learning_rate": 0.001,
|
| 641 |
-
"loss": 0.458,
|
| 642 |
-
"step": 1840
|
| 643 |
-
},
|
| 644 |
-
{
|
| 645 |
-
"epoch": 5.18,
|
| 646 |
-
"learning_rate": 0.001,
|
| 647 |
-
"loss": 0.3949,
|
| 648 |
-
"step": 1860
|
| 649 |
-
},
|
| 650 |
-
{
|
| 651 |
-
"epoch": 5.24,
|
| 652 |
-
"learning_rate": 0.001,
|
| 653 |
-
"loss": 0.372,
|
| 654 |
-
"step": 1880
|
| 655 |
-
},
|
| 656 |
-
{
|
| 657 |
-
"epoch": 5.29,
|
| 658 |
-
"learning_rate": 0.001,
|
| 659 |
-
"loss": 0.3578,
|
| 660 |
-
"step": 1900
|
| 661 |
-
},
|
| 662 |
-
{
|
| 663 |
-
"epoch": 5.35,
|
| 664 |
-
"learning_rate": 0.001,
|
| 665 |
-
"loss": 0.3906,
|
| 666 |
-
"step": 1920
|
| 667 |
-
},
|
| 668 |
-
{
|
| 669 |
-
"epoch": 5.4,
|
| 670 |
-
"learning_rate": 0.001,
|
| 671 |
-
"loss": 0.3888,
|
| 672 |
-
"step": 1940
|
| 673 |
-
},
|
| 674 |
-
{
|
| 675 |
-
"epoch": 5.46,
|
| 676 |
-
"learning_rate": 0.001,
|
| 677 |
-
"loss": 0.4049,
|
| 678 |
-
"step": 1960
|
| 679 |
-
},
|
| 680 |
-
{
|
| 681 |
-
"epoch": 5.52,
|
| 682 |
-
"learning_rate": 0.001,
|
| 683 |
-
"loss": 0.3692,
|
| 684 |
-
"step": 1980
|
| 685 |
-
},
|
| 686 |
-
{
|
| 687 |
-
"epoch": 5.57,
|
| 688 |
-
"learning_rate": 0.001,
|
| 689 |
-
"loss": 0.3299,
|
| 690 |
-
"step": 2000
|
| 691 |
-
},
|
| 692 |
-
{
|
| 693 |
-
"epoch": 5.63,
|
| 694 |
-
"learning_rate": 0.001,
|
| 695 |
-
"loss": 0.3714,
|
| 696 |
-
"step": 2020
|
| 697 |
-
},
|
| 698 |
-
{
|
| 699 |
-
"epoch": 5.68,
|
| 700 |
-
"learning_rate": 0.001,
|
| 701 |
-
"loss": 0.3423,
|
| 702 |
-
"step": 2040
|
| 703 |
-
},
|
| 704 |
-
{
|
| 705 |
-
"epoch": 5.74,
|
| 706 |
-
"learning_rate": 0.001,
|
| 707 |
-
"loss": 0.3534,
|
| 708 |
-
"step": 2060
|
| 709 |
-
},
|
| 710 |
-
{
|
| 711 |
-
"epoch": 5.79,
|
| 712 |
-
"learning_rate": 0.001,
|
| 713 |
-
"loss": 0.3426,
|
| 714 |
-
"step": 2080
|
| 715 |
-
},
|
| 716 |
-
{
|
| 717 |
-
"epoch": 5.85,
|
| 718 |
-
"learning_rate": 0.001,
|
| 719 |
-
"loss": 0.3684,
|
| 720 |
-
"step": 2100
|
| 721 |
-
},
|
| 722 |
-
{
|
| 723 |
-
"epoch": 5.91,
|
| 724 |
-
"learning_rate": 0.001,
|
| 725 |
-
"loss": 0.3472,
|
| 726 |
-
"step": 2120
|
| 727 |
-
},
|
| 728 |
-
{
|
| 729 |
-
"epoch": 5.96,
|
| 730 |
-
"learning_rate": 0.001,
|
| 731 |
-
"loss": 0.299,
|
| 732 |
-
"step": 2140
|
| 733 |
-
},
|
| 734 |
-
{
|
| 735 |
-
"epoch": 6.0,
|
| 736 |
-
"eval_f1": 0.0,
|
| 737 |
-
"eval_loss": 0.3469391465187073,
|
| 738 |
-
"eval_runtime": 20.5335,
|
| 739 |
-
"eval_samples_per_second": 93.311,
|
| 740 |
-
"eval_steps_per_second": 1.461,
|
| 741 |
-
"step": 2154
|
| 742 |
-
},
|
| 743 |
-
{
|
| 744 |
-
"epoch": 6.02,
|
| 745 |
-
"learning_rate": 0.001,
|
| 746 |
-
"loss": 0.3336,
|
| 747 |
-
"step": 2160
|
| 748 |
-
},
|
| 749 |
-
{
|
| 750 |
-
"epoch": 6.07,
|
| 751 |
-
"learning_rate": 0.001,
|
| 752 |
-
"loss": 0.4366,
|
| 753 |
-
"step": 2180
|
| 754 |
-
},
|
| 755 |
-
{
|
| 756 |
-
"epoch": 6.13,
|
| 757 |
-
"learning_rate": 0.001,
|
| 758 |
-
"loss": 0.3709,
|
| 759 |
-
"step": 2200
|
| 760 |
-
},
|
| 761 |
-
{
|
| 762 |
-
"epoch": 6.18,
|
| 763 |
-
"learning_rate": 0.001,
|
| 764 |
-
"loss": 0.3357,
|
| 765 |
-
"step": 2220
|
| 766 |
-
},
|
| 767 |
-
{
|
| 768 |
-
"epoch": 6.24,
|
| 769 |
-
"learning_rate": 0.001,
|
| 770 |
-
"loss": 0.4034,
|
| 771 |
-
"step": 2240
|
| 772 |
-
},
|
| 773 |
-
{
|
| 774 |
-
"epoch": 6.3,
|
| 775 |
-
"learning_rate": 0.001,
|
| 776 |
-
"loss": 0.3868,
|
| 777 |
-
"step": 2260
|
| 778 |
-
},
|
| 779 |
-
{
|
| 780 |
-
"epoch": 6.35,
|
| 781 |
-
"learning_rate": 0.001,
|
| 782 |
-
"loss": 0.3328,
|
| 783 |
-
"step": 2280
|
| 784 |
-
},
|
| 785 |
-
{
|
| 786 |
-
"epoch": 6.41,
|
| 787 |
-
"learning_rate": 0.001,
|
| 788 |
-
"loss": 0.3974,
|
| 789 |
-
"step": 2300
|
| 790 |
-
},
|
| 791 |
-
{
|
| 792 |
-
"epoch": 6.46,
|
| 793 |
-
"learning_rate": 0.001,
|
| 794 |
-
"loss": 0.3707,
|
| 795 |
-
"step": 2320
|
| 796 |
-
},
|
| 797 |
-
{
|
| 798 |
-
"epoch": 6.52,
|
| 799 |
-
"learning_rate": 0.001,
|
| 800 |
-
"loss": 0.3753,
|
| 801 |
-
"step": 2340
|
| 802 |
-
},
|
| 803 |
-
{
|
| 804 |
-
"epoch": 6.57,
|
| 805 |
-
"learning_rate": 0.001,
|
| 806 |
-
"loss": 0.3255,
|
| 807 |
-
"step": 2360
|
| 808 |
-
},
|
| 809 |
-
{
|
| 810 |
-
"epoch": 6.63,
|
| 811 |
-
"learning_rate": 0.001,
|
| 812 |
-
"loss": 0.4284,
|
| 813 |
-
"step": 2380
|
| 814 |
-
},
|
| 815 |
-
{
|
| 816 |
-
"epoch": 6.69,
|
| 817 |
-
"learning_rate": 0.001,
|
| 818 |
-
"loss": 0.3699,
|
| 819 |
-
"step": 2400
|
| 820 |
-
},
|
| 821 |
-
{
|
| 822 |
-
"epoch": 6.74,
|
| 823 |
-
"learning_rate": 0.001,
|
| 824 |
-
"loss": 0.3705,
|
| 825 |
-
"step": 2420
|
| 826 |
-
},
|
| 827 |
-
{
|
| 828 |
-
"epoch": 6.8,
|
| 829 |
-
"learning_rate": 0.001,
|
| 830 |
-
"loss": 0.2841,
|
| 831 |
-
"step": 2440
|
| 832 |
-
},
|
| 833 |
-
{
|
| 834 |
-
"epoch": 6.85,
|
| 835 |
-
"learning_rate": 0.001,
|
| 836 |
-
"loss": 0.2687,
|
| 837 |
-
"step": 2460
|
| 838 |
-
},
|
| 839 |
-
{
|
| 840 |
-
"epoch": 6.91,
|
| 841 |
-
"learning_rate": 0.001,
|
| 842 |
-
"loss": 0.3294,
|
| 843 |
-
"step": 2480
|
| 844 |
-
},
|
| 845 |
-
{
|
| 846 |
-
"epoch": 6.96,
|
| 847 |
-
"learning_rate": 0.001,
|
| 848 |
-
"loss": 0.3531,
|
| 849 |
-
"step": 2500
|
| 850 |
-
},
|
| 851 |
-
{
|
| 852 |
-
"epoch": 7.0,
|
| 853 |
-
"eval_f1": 0.0,
|
| 854 |
-
"eval_loss": 0.3420043885707855,
|
| 855 |
-
"eval_runtime": 20.5195,
|
| 856 |
-
"eval_samples_per_second": 93.374,
|
| 857 |
-
"eval_steps_per_second": 1.462,
|
| 858 |
-
"step": 2513
|
| 859 |
-
},
|
| 860 |
-
{
|
| 861 |
-
"epoch": 7.02,
|
| 862 |
-
"learning_rate": 0.001,
|
| 863 |
-
"loss": 0.3396,
|
| 864 |
-
"step": 2520
|
| 865 |
-
},
|
| 866 |
-
{
|
| 867 |
-
"epoch": 7.08,
|
| 868 |
-
"learning_rate": 0.001,
|
| 869 |
-
"loss": 0.3824,
|
| 870 |
-
"step": 2540
|
| 871 |
-
},
|
| 872 |
-
{
|
| 873 |
-
"epoch": 7.13,
|
| 874 |
-
"learning_rate": 0.001,
|
| 875 |
-
"loss": 0.2518,
|
| 876 |
-
"step": 2560
|
| 877 |
-
},
|
| 878 |
-
{
|
| 879 |
-
"epoch": 7.19,
|
| 880 |
-
"learning_rate": 0.001,
|
| 881 |
-
"loss": 0.3822,
|
| 882 |
-
"step": 2580
|
| 883 |
-
},
|
| 884 |
-
{
|
| 885 |
-
"epoch": 7.24,
|
| 886 |
-
"learning_rate": 0.001,
|
| 887 |
-
"loss": 0.3969,
|
| 888 |
-
"step": 2600
|
| 889 |
-
},
|
| 890 |
-
{
|
| 891 |
-
"epoch": 7.3,
|
| 892 |
-
"learning_rate": 0.001,
|
| 893 |
-
"loss": 0.2551,
|
| 894 |
-
"step": 2620
|
| 895 |
-
},
|
| 896 |
-
{
|
| 897 |
-
"epoch": 7.35,
|
| 898 |
-
"learning_rate": 0.001,
|
| 899 |
-
"loss": 0.3387,
|
| 900 |
-
"step": 2640
|
| 901 |
-
},
|
| 902 |
-
{
|
| 903 |
-
"epoch": 7.41,
|
| 904 |
-
"learning_rate": 0.001,
|
| 905 |
-
"loss": 0.3761,
|
| 906 |
-
"step": 2660
|
| 907 |
-
},
|
| 908 |
-
{
|
| 909 |
-
"epoch": 7.47,
|
| 910 |
-
"learning_rate": 0.001,
|
| 911 |
-
"loss": 0.3899,
|
| 912 |
-
"step": 2680
|
| 913 |
-
},
|
| 914 |
-
{
|
| 915 |
-
"epoch": 7.52,
|
| 916 |
-
"learning_rate": 0.001,
|
| 917 |
-
"loss": 0.3691,
|
| 918 |
-
"step": 2700
|
| 919 |
-
},
|
| 920 |
-
{
|
| 921 |
-
"epoch": 7.58,
|
| 922 |
-
"learning_rate": 0.001,
|
| 923 |
-
"loss": 0.3172,
|
| 924 |
-
"step": 2720
|
| 925 |
-
},
|
| 926 |
-
{
|
| 927 |
-
"epoch": 7.63,
|
| 928 |
-
"learning_rate": 0.001,
|
| 929 |
-
"loss": 0.3358,
|
| 930 |
-
"step": 2740
|
| 931 |
-
},
|
| 932 |
-
{
|
| 933 |
-
"epoch": 7.69,
|
| 934 |
-
"learning_rate": 0.001,
|
| 935 |
-
"loss": 0.3459,
|
| 936 |
-
"step": 2760
|
| 937 |
-
},
|
| 938 |
-
{
|
| 939 |
-
"epoch": 7.74,
|
| 940 |
-
"learning_rate": 0.001,
|
| 941 |
-
"loss": 0.3347,
|
| 942 |
-
"step": 2780
|
| 943 |
-
},
|
| 944 |
-
{
|
| 945 |
-
"epoch": 7.8,
|
| 946 |
-
"learning_rate": 0.001,
|
| 947 |
-
"loss": 0.3459,
|
| 948 |
-
"step": 2800
|
| 949 |
-
},
|
| 950 |
-
{
|
| 951 |
-
"epoch": 7.86,
|
| 952 |
-
"learning_rate": 0.001,
|
| 953 |
-
"loss": 0.3797,
|
| 954 |
-
"step": 2820
|
| 955 |
-
},
|
| 956 |
-
{
|
| 957 |
-
"epoch": 7.91,
|
| 958 |
-
"learning_rate": 0.001,
|
| 959 |
-
"loss": 0.3721,
|
| 960 |
-
"step": 2840
|
| 961 |
-
},
|
| 962 |
-
{
|
| 963 |
-
"epoch": 7.97,
|
| 964 |
-
"learning_rate": 0.001,
|
| 965 |
-
"loss": 0.3892,
|
| 966 |
-
"step": 2860
|
| 967 |
-
},
|
| 968 |
-
{
|
| 969 |
-
"epoch": 8.0,
|
| 970 |
-
"eval_f1": 0.0,
|
| 971 |
-
"eval_loss": 0.34283891320228577,
|
| 972 |
-
"eval_runtime": 20.5477,
|
| 973 |
-
"eval_samples_per_second": 93.247,
|
| 974 |
-
"eval_steps_per_second": 1.46,
|
| 975 |
-
"step": 2872
|
| 976 |
-
},
|
| 977 |
-
{
|
| 978 |
-
"epoch": 8.02,
|
| 979 |
-
"learning_rate": 0.001,
|
| 980 |
-
"loss": 0.3234,
|
| 981 |
-
"step": 2880
|
| 982 |
-
},
|
| 983 |
-
{
|
| 984 |
-
"epoch": 8.08,
|
| 985 |
-
"learning_rate": 0.001,
|
| 986 |
-
"loss": 0.3979,
|
| 987 |
-
"step": 2900
|
| 988 |
-
},
|
| 989 |
-
{
|
| 990 |
-
"epoch": 8.13,
|
| 991 |
-
"learning_rate": 0.001,
|
| 992 |
-
"loss": 0.4032,
|
| 993 |
-
"step": 2920
|
| 994 |
-
},
|
| 995 |
-
{
|
| 996 |
-
"epoch": 8.19,
|
| 997 |
-
"learning_rate": 0.001,
|
| 998 |
-
"loss": 0.3787,
|
| 999 |
-
"step": 2940
|
| 1000 |
-
},
|
| 1001 |
-
{
|
| 1002 |
-
"epoch": 8.25,
|
| 1003 |
-
"learning_rate": 0.001,
|
| 1004 |
-
"loss": 0.3144,
|
| 1005 |
-
"step": 2960
|
| 1006 |
-
},
|
| 1007 |
-
{
|
| 1008 |
-
"epoch": 8.3,
|
| 1009 |
-
"learning_rate": 0.001,
|
| 1010 |
-
"loss": 0.4071,
|
| 1011 |
-
"step": 2980
|
| 1012 |
-
},
|
| 1013 |
-
{
|
| 1014 |
-
"epoch": 8.36,
|
| 1015 |
-
"learning_rate": 0.001,
|
| 1016 |
-
"loss": 0.3192,
|
| 1017 |
-
"step": 3000
|
| 1018 |
-
},
|
| 1019 |
-
{
|
| 1020 |
-
"epoch": 8.41,
|
| 1021 |
-
"learning_rate": 0.001,
|
| 1022 |
-
"loss": 0.3194,
|
| 1023 |
-
"step": 3020
|
| 1024 |
-
},
|
| 1025 |
-
{
|
| 1026 |
-
"epoch": 8.47,
|
| 1027 |
-
"learning_rate": 0.001,
|
| 1028 |
-
"loss": 0.3468,
|
| 1029 |
-
"step": 3040
|
| 1030 |
-
},
|
| 1031 |
-
{
|
| 1032 |
-
"epoch": 8.52,
|
| 1033 |
-
"learning_rate": 0.001,
|
| 1034 |
-
"loss": 0.325,
|
| 1035 |
-
"step": 3060
|
| 1036 |
-
},
|
| 1037 |
-
{
|
| 1038 |
-
"epoch": 8.58,
|
| 1039 |
-
"learning_rate": 0.001,
|
| 1040 |
-
"loss": 0.3631,
|
| 1041 |
-
"step": 3080
|
| 1042 |
-
},
|
| 1043 |
-
{
|
| 1044 |
-
"epoch": 8.64,
|
| 1045 |
-
"learning_rate": 0.001,
|
| 1046 |
-
"loss": 0.3464,
|
| 1047 |
-
"step": 3100
|
| 1048 |
-
},
|
| 1049 |
-
{
|
| 1050 |
-
"epoch": 8.69,
|
| 1051 |
-
"learning_rate": 0.001,
|
| 1052 |
-
"loss": 0.3378,
|
| 1053 |
-
"step": 3120
|
| 1054 |
-
},
|
| 1055 |
-
{
|
| 1056 |
-
"epoch": 8.75,
|
| 1057 |
-
"learning_rate": 0.001,
|
| 1058 |
-
"loss": 0.3808,
|
| 1059 |
-
"step": 3140
|
| 1060 |
-
},
|
| 1061 |
-
{
|
| 1062 |
-
"epoch": 8.8,
|
| 1063 |
-
"learning_rate": 0.001,
|
| 1064 |
-
"loss": 0.3668,
|
| 1065 |
-
"step": 3160
|
| 1066 |
-
},
|
| 1067 |
-
{
|
| 1068 |
-
"epoch": 8.86,
|
| 1069 |
-
"learning_rate": 0.001,
|
| 1070 |
-
"loss": 0.3045,
|
| 1071 |
-
"step": 3180
|
| 1072 |
-
},
|
| 1073 |
-
{
|
| 1074 |
-
"epoch": 8.91,
|
| 1075 |
-
"learning_rate": 0.001,
|
| 1076 |
-
"loss": 0.2805,
|
| 1077 |
-
"step": 3200
|
| 1078 |
-
},
|
| 1079 |
-
{
|
| 1080 |
-
"epoch": 8.97,
|
| 1081 |
-
"learning_rate": 0.001,
|
| 1082 |
-
"loss": 0.3706,
|
| 1083 |
-
"step": 3220
|
| 1084 |
-
},
|
| 1085 |
-
{
|
| 1086 |
-
"epoch": 9.0,
|
| 1087 |
-
"eval_f1": 0.0,
|
| 1088 |
-
"eval_loss": 0.3420598804950714,
|
| 1089 |
-
"eval_runtime": 20.5266,
|
| 1090 |
-
"eval_samples_per_second": 93.342,
|
| 1091 |
-
"eval_steps_per_second": 1.462,
|
| 1092 |
-
"step": 3231
|
| 1093 |
-
},
|
| 1094 |
-
{
|
| 1095 |
-
"epoch": 9.03,
|
| 1096 |
-
"learning_rate": 0.001,
|
| 1097 |
-
"loss": 0.3502,
|
| 1098 |
-
"step": 3240
|
| 1099 |
-
},
|
| 1100 |
-
{
|
| 1101 |
-
"epoch": 9.08,
|
| 1102 |
-
"learning_rate": 0.001,
|
| 1103 |
-
"loss": 0.3414,
|
| 1104 |
-
"step": 3260
|
| 1105 |
},
|
| 1106 |
{
|
| 1107 |
-
"epoch":
|
| 1108 |
-
"learning_rate":
|
| 1109 |
-
"loss": 0.
|
| 1110 |
-
"step":
|
| 1111 |
},
|
| 1112 |
{
|
| 1113 |
-
"epoch":
|
| 1114 |
-
"learning_rate":
|
| 1115 |
-
"loss": 0.
|
| 1116 |
-
"step":
|
| 1117 |
},
|
| 1118 |
{
|
| 1119 |
-
"epoch":
|
| 1120 |
-
"learning_rate":
|
| 1121 |
-
"loss": 0.
|
| 1122 |
-
"step":
|
| 1123 |
},
|
| 1124 |
{
|
| 1125 |
-
"epoch":
|
| 1126 |
-
"learning_rate":
|
| 1127 |
-
"loss": 0.
|
| 1128 |
-
"step":
|
| 1129 |
},
|
| 1130 |
{
|
| 1131 |
-
"epoch":
|
| 1132 |
-
"learning_rate":
|
| 1133 |
-
"loss": 0.
|
| 1134 |
-
"step":
|
| 1135 |
},
|
| 1136 |
{
|
| 1137 |
-
"epoch":
|
| 1138 |
-
"learning_rate":
|
| 1139 |
-
"loss": 0.
|
| 1140 |
-
"step":
|
| 1141 |
},
|
| 1142 |
{
|
| 1143 |
-
"epoch":
|
| 1144 |
-
"learning_rate":
|
| 1145 |
-
"loss": 0.
|
| 1146 |
-
"step":
|
| 1147 |
},
|
| 1148 |
{
|
| 1149 |
-
"epoch":
|
| 1150 |
-
"learning_rate":
|
| 1151 |
-
"loss": 0.
|
| 1152 |
-
"step":
|
| 1153 |
},
|
| 1154 |
{
|
| 1155 |
-
"epoch":
|
| 1156 |
-
"learning_rate":
|
| 1157 |
-
"loss": 0.
|
| 1158 |
-
"step":
|
| 1159 |
},
|
| 1160 |
{
|
| 1161 |
-
"epoch":
|
| 1162 |
-
"learning_rate":
|
| 1163 |
-
"loss": 0.
|
| 1164 |
-
"step":
|
| 1165 |
},
|
| 1166 |
{
|
| 1167 |
-
"epoch":
|
| 1168 |
-
"learning_rate":
|
| 1169 |
-
"loss": 0.
|
| 1170 |
-
"step":
|
| 1171 |
},
|
| 1172 |
{
|
| 1173 |
-
"epoch":
|
| 1174 |
-
"learning_rate":
|
| 1175 |
-
"loss": 0.
|
| 1176 |
-
"step":
|
| 1177 |
},
|
| 1178 |
{
|
| 1179 |
-
"epoch":
|
| 1180 |
-
"learning_rate":
|
| 1181 |
-
"loss": 0.
|
| 1182 |
-
"step":
|
| 1183 |
},
|
| 1184 |
{
|
| 1185 |
-
"epoch":
|
| 1186 |
-
"learning_rate":
|
| 1187 |
-
"loss": 0.
|
| 1188 |
-
"step":
|
| 1189 |
},
|
| 1190 |
{
|
| 1191 |
-
"epoch":
|
| 1192 |
-
"learning_rate":
|
| 1193 |
-
"loss": 0.
|
| 1194 |
-
"step":
|
| 1195 |
},
|
| 1196 |
{
|
| 1197 |
-
"epoch":
|
| 1198 |
-
"learning_rate": 0.
|
| 1199 |
-
"loss": 0.
|
| 1200 |
-
"step":
|
| 1201 |
},
|
| 1202 |
{
|
| 1203 |
-
"epoch":
|
| 1204 |
-
"eval_f1": 0.
|
| 1205 |
-
"eval_loss":
|
| 1206 |
-
"eval_runtime": 20.
|
| 1207 |
-
"eval_samples_per_second": 93.
|
| 1208 |
-
"eval_steps_per_second": 1.
|
| 1209 |
-
"step":
|
| 1210 |
},
|
| 1211 |
{
|
| 1212 |
-
"epoch":
|
| 1213 |
-
"step":
|
| 1214 |
-
"total_flos":
|
| 1215 |
-
"train_loss":
|
| 1216 |
-
"train_runtime":
|
| 1217 |
-
"train_samples_per_second":
|
| 1218 |
-
"train_steps_per_second": 0.
|
| 1219 |
}
|
| 1220 |
],
|
| 1221 |
-
"logging_steps":
|
| 1222 |
-
"max_steps":
|
| 1223 |
"num_input_tokens_seen": 0,
|
| 1224 |
-
"num_train_epochs":
|
| 1225 |
"save_steps": 500,
|
| 1226 |
-
"total_flos":
|
| 1227 |
-
"train_batch_size":
|
| 1228 |
"trial_name": null,
|
| 1229 |
"trial_params": null
|
| 1230 |
}
|
|
|
|
| 1 |
{
|
| 2 |
"best_metric": null,
|
| 3 |
"best_model_checkpoint": null,
|
| 4 |
+
"epoch": 3.0,
|
| 5 |
"eval_steps": 500,
|
| 6 |
+
"global_step": 540,
|
| 7 |
"is_hyper_param_search": false,
|
| 8 |
"is_local_process_zero": true,
|
| 9 |
"is_world_process_zero": true,
|
| 10 |
"log_history": [
|
| 11 |
+
{
|
| 12 |
+
"epoch": 0.06,
|
| 13 |
+
"learning_rate": 1.4814814814814812e-06,
|
| 14 |
+
"loss": 118.6129,
|
| 15 |
+
"step": 10
|
| 16 |
+
},
|
| 17 |
{
|
| 18 |
"epoch": 0.11,
|
| 19 |
+
"learning_rate": 2.9629629629629625e-06,
|
| 20 |
+
"loss": 96.0751,
|
| 21 |
"step": 20
|
| 22 |
},
|
| 23 |
+
{
|
| 24 |
+
"epoch": 0.17,
|
| 25 |
+
"learning_rate": 4.444444444444444e-06,
|
| 26 |
+
"loss": 96.9119,
|
| 27 |
+
"step": 30
|
| 28 |
+
},
|
| 29 |
{
|
| 30 |
"epoch": 0.22,
|
| 31 |
+
"learning_rate": 5.925925925925925e-06,
|
| 32 |
+
"loss": 85.8261,
|
| 33 |
"step": 40
|
| 34 |
},
|
| 35 |
+
{
|
| 36 |
+
"epoch": 0.28,
|
| 37 |
+
"learning_rate": 7.4074074074074075e-06,
|
| 38 |
+
"loss": 76.0406,
|
| 39 |
+
"step": 50
|
| 40 |
+
},
|
| 41 |
{
|
| 42 |
"epoch": 0.33,
|
| 43 |
+
"learning_rate": 7.99699181001692e-06,
|
| 44 |
+
"loss": 62.1596,
|
| 45 |
"step": 60
|
| 46 |
},
|
| 47 |
{
|
| 48 |
+
"epoch": 0.39,
|
| 49 |
+
"learning_rate": 7.978624809626406e-06,
|
| 50 |
+
"loss": 60.1637,
|
| 51 |
+
"step": 70
|
| 52 |
+
},
|
| 53 |
+
{
|
| 54 |
+
"epoch": 0.44,
|
| 55 |
+
"learning_rate": 7.943638653379184e-06,
|
| 56 |
+
"loss": 49.0233,
|
| 57 |
"step": 80
|
| 58 |
},
|
| 59 |
+
{
|
| 60 |
+
"epoch": 0.5,
|
| 61 |
+
"learning_rate": 7.892179482319294e-06,
|
| 62 |
+
"loss": 61.9877,
|
| 63 |
+
"step": 90
|
| 64 |
+
},
|
| 65 |
{
|
| 66 |
"epoch": 0.56,
|
| 67 |
+
"learning_rate": 7.824462247095518e-06,
|
| 68 |
+
"loss": 41.0057,
|
| 69 |
"step": 100
|
| 70 |
},
|
| 71 |
+
{
|
| 72 |
+
"epoch": 0.61,
|
| 73 |
+
"learning_rate": 7.740769810088759e-06,
|
| 74 |
+
"loss": 28.3225,
|
| 75 |
+
"step": 110
|
| 76 |
+
},
|
| 77 |
{
|
| 78 |
"epoch": 0.67,
|
| 79 |
+
"learning_rate": 7.641451763864587e-06,
|
| 80 |
+
"loss": 21.0979,
|
| 81 |
"step": 120
|
| 82 |
},
|
| 83 |
+
{
|
| 84 |
+
"epoch": 0.72,
|
| 85 |
+
"learning_rate": 7.526922970886431e-06,
|
| 86 |
+
"loss": 18.2794,
|
| 87 |
+
"step": 130
|
| 88 |
+
},
|
| 89 |
{
|
| 90 |
"epoch": 0.78,
|
| 91 |
+
"learning_rate": 7.3976618305891895e-06,
|
| 92 |
+
"loss": 9.044,
|
| 93 |
"step": 140
|
| 94 |
},
|
| 95 |
+
{
|
| 96 |
+
"epoch": 0.83,
|
| 97 |
+
"learning_rate": 7.2542082810518696e-06,
|
| 98 |
+
"loss": 6.2416,
|
| 99 |
+
"step": 150
|
| 100 |
+
},
|
| 101 |
{
|
| 102 |
"epoch": 0.89,
|
| 103 |
+
"learning_rate": 7.097161543616529e-06,
|
| 104 |
+
"loss": 2.6869,
|
| 105 |
"step": 160
|
| 106 |
},
|
| 107 |
{
|
| 108 |
+
"epoch": 0.94,
|
| 109 |
+
"learning_rate": 6.927177619874449e-06,
|
| 110 |
+
"loss": 1.5778,
|
| 111 |
+
"step": 170
|
|
|
|
|
|
|
|
|
|
| 112 |
},
|
| 113 |
{
|
| 114 |
"epoch": 1.0,
|
| 115 |
+
"learning_rate": 6.744966551474935e-06,
|
| 116 |
+
"loss": 1.5514,
|
| 117 |
"step": 180
|
| 118 |
},
|
| 119 |
{
|
| 120 |
+
"epoch": 1.0,
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 121 |
"eval_f1": 0.1941564561734213,
|
| 122 |
+
"eval_loss": 15.080679893493652,
|
| 123 |
+
"eval_runtime": 20.7366,
|
| 124 |
+
"eval_samples_per_second": 92.397,
|
| 125 |
+
"eval_steps_per_second": 1.447,
|
| 126 |
+
"step": 180
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 127 |
},
|
| 128 |
{
|
| 129 |
+
"epoch": 1.06,
|
| 130 |
+
"learning_rate": 6.551289454202823e-06,
|
| 131 |
+
"loss": 1.0316,
|
| 132 |
+
"step": 190
|
| 133 |
},
|
| 134 |
{
|
| 135 |
+
"epoch": 1.11,
|
| 136 |
+
"learning_rate": 6.346955338713672e-06,
|
| 137 |
+
"loss": 1.3113,
|
| 138 |
+
"step": 200
|
| 139 |
},
|
| 140 |
{
|
| 141 |
+
"epoch": 1.17,
|
| 142 |
+
"learning_rate": 6.132817731206765e-06,
|
| 143 |
+
"loss": 0.9913,
|
| 144 |
+
"step": 210
|
| 145 |
},
|
| 146 |
{
|
| 147 |
+
"epoch": 1.22,
|
| 148 |
+
"learning_rate": 5.9097711081517955e-06,
|
| 149 |
+
"loss": 0.9418,
|
| 150 |
+
"step": 220
|
| 151 |
},
|
| 152 |
{
|
| 153 |
+
"epoch": 1.28,
|
| 154 |
+
"learning_rate": 5.678747159961716e-06,
|
| 155 |
+
"loss": 1.3579,
|
| 156 |
+
"step": 230
|
| 157 |
},
|
| 158 |
{
|
| 159 |
+
"epoch": 1.33,
|
| 160 |
+
"learning_rate": 5.440710899218841e-06,
|
| 161 |
+
"loss": 0.7793,
|
| 162 |
+
"step": 240
|
| 163 |
},
|
| 164 |
{
|
| 165 |
+
"epoch": 1.39,
|
| 166 |
+
"learning_rate": 5.196656629710567e-06,
|
| 167 |
+
"loss": 0.8691,
|
| 168 |
+
"step": 250
|
| 169 |
},
|
| 170 |
{
|
| 171 |
+
"epoch": 1.44,
|
| 172 |
+
"learning_rate": 4.947603793112476e-06,
|
| 173 |
+
"loss": 0.7268,
|
| 174 |
+
"step": 260
|
| 175 |
},
|
| 176 |
{
|
| 177 |
+
"epoch": 1.5,
|
| 178 |
+
"learning_rate": 4.694592710667722e-06,
|
| 179 |
+
"loss": 1.3357,
|
| 180 |
+
"step": 270
|
|
|
|
|
|
|
|
|
|
| 181 |
},
|
| 182 |
{
|
| 183 |
+
"epoch": 1.56,
|
| 184 |
+
"learning_rate": 4.438680237650126e-06,
|
| 185 |
+
"loss": 0.7497,
|
| 186 |
+
"step": 280
|
| 187 |
},
|
| 188 |
{
|
| 189 |
+
"epoch": 1.61,
|
| 190 |
+
"learning_rate": 4.180935348762825e-06,
|
| 191 |
+
"loss": 0.5343,
|
| 192 |
+
"step": 290
|
| 193 |
},
|
| 194 |
{
|
| 195 |
+
"epoch": 1.67,
|
| 196 |
+
"learning_rate": 3.922434672912703e-06,
|
| 197 |
+
"loss": 1.0287,
|
| 198 |
+
"step": 300
|
| 199 |
},
|
| 200 |
{
|
| 201 |
+
"epoch": 1.72,
|
| 202 |
+
"learning_rate": 3.664257996012372e-06,
|
| 203 |
+
"loss": 0.45,
|
| 204 |
+
"step": 310
|
| 205 |
},
|
| 206 |
{
|
| 207 |
+
"epoch": 1.78,
|
| 208 |
+
"learning_rate": 3.4074837505950056e-06,
|
| 209 |
+
"loss": 1.4606,
|
| 210 |
+
"step": 320
|
| 211 |
},
|
| 212 |
{
|
| 213 |
+
"epoch": 1.83,
|
| 214 |
+
"learning_rate": 3.1531845110823583e-06,
|
| 215 |
+
"loss": 0.4987,
|
| 216 |
+
"step": 330
|
| 217 |
},
|
| 218 |
{
|
| 219 |
+
"epoch": 1.89,
|
| 220 |
+
"learning_rate": 2.9024225135227944e-06,
|
| 221 |
+
"loss": 0.4837,
|
| 222 |
+
"step": 340
|
| 223 |
},
|
| 224 |
{
|
| 225 |
+
"epoch": 1.94,
|
| 226 |
+
"learning_rate": 2.6562452185138318e-06,
|
| 227 |
+
"loss": 0.4317,
|
| 228 |
+
"step": 350
|
| 229 |
},
|
| 230 |
{
|
| 231 |
+
"epoch": 2.0,
|
| 232 |
+
"learning_rate": 2.4156809358433725e-06,
|
| 233 |
+
"loss": 0.309,
|
| 234 |
+
"step": 360
|
| 235 |
},
|
| 236 |
{
|
| 237 |
+
"epoch": 2.0,
|
| 238 |
"eval_f1": 0.1941564561734213,
|
| 239 |
+
"eval_loss": 13.09052562713623,
|
| 240 |
+
"eval_runtime": 20.4809,
|
| 241 |
+
"eval_samples_per_second": 93.55,
|
| 242 |
+
"eval_steps_per_second": 1.465,
|
| 243 |
+
"step": 360
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 244 |
},
|
| 245 |
{
|
| 246 |
+
"epoch": 2.06,
|
| 247 |
+
"learning_rate": 2.181734529125998e-06,
|
| 248 |
+
"loss": 0.4364,
|
| 249 |
+
"step": 370
|
| 250 |
},
|
| 251 |
{
|
| 252 |
+
"epoch": 2.11,
|
| 253 |
+
"learning_rate": 1.9553832183765073e-06,
|
| 254 |
"loss": 0.3626,
|
| 255 |
+
"step": 380
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 256 |
},
|
| 257 |
{
|
| 258 |
+
"epoch": 2.17,
|
| 259 |
+
"learning_rate": 1.7375724980538462e-06,
|
| 260 |
+
"loss": 0.1876,
|
| 261 |
+
"step": 390
|
| 262 |
},
|
| 263 |
{
|
| 264 |
+
"epoch": 2.22,
|
| 265 |
+
"learning_rate": 1.529212187626172e-06,
|
| 266 |
+
"loss": 0.1443,
|
| 267 |
+
"step": 400
|
| 268 |
},
|
| 269 |
{
|
| 270 |
+
"epoch": 2.28,
|
| 271 |
+
"learning_rate": 1.3311726311543211e-06,
|
| 272 |
+
"loss": 0.1844,
|
| 273 |
+
"step": 410
|
| 274 |
},
|
| 275 |
{
|
| 276 |
+
"epoch": 2.33,
|
| 277 |
+
"learning_rate": 1.1442810617684044e-06,
|
| 278 |
+
"loss": 0.1862,
|
| 279 |
+
"step": 420
|
| 280 |
},
|
| 281 |
{
|
| 282 |
+
"epoch": 2.39,
|
| 283 |
+
"learning_rate": 9.693181462235283e-07,
|
| 284 |
+
"loss": 0.0915,
|
| 285 |
+
"step": 430
|
| 286 |
},
|
| 287 |
{
|
| 288 |
+
"epoch": 2.44,
|
| 289 |
+
"learning_rate": 8.070147239684279e-07,
|
| 290 |
+
"loss": 0.0219,
|
| 291 |
+
"step": 440
|
| 292 |
},
|
| 293 |
{
|
| 294 |
+
"epoch": 2.5,
|
| 295 |
+
"learning_rate": 6.580487543482549e-07,
|
| 296 |
+
"loss": 0.1046,
|
| 297 |
+
"step": 450
|
| 298 |
},
|
| 299 |
{
|
| 300 |
+
"epoch": 2.56,
|
| 301 |
+
"learning_rate": 5.230424846934088e-07,
|
| 302 |
+
"loss": 0.0806,
|
| 303 |
+
"step": 460
|
| 304 |
},
|
| 305 |
{
|
| 306 |
+
"epoch": 2.61,
|
| 307 |
+
"learning_rate": 4.025598511236281e-07,
|
| 308 |
+
"loss": 0.1472,
|
| 309 |
+
"step": 470
|
| 310 |
},
|
| 311 |
{
|
| 312 |
+
"epoch": 2.67,
|
| 313 |
+
"learning_rate": 2.9710412292443863e-07,
|
| 314 |
+
"loss": 0.1118,
|
| 315 |
+
"step": 480
|
| 316 |
},
|
| 317 |
{
|
| 318 |
+
"epoch": 2.72,
|
| 319 |
+
"learning_rate": 2.071158003356941e-07,
|
| 320 |
+
"loss": 0.0869,
|
| 321 |
+
"step": 490
|
| 322 |
},
|
| 323 |
{
|
| 324 |
+
"epoch": 2.78,
|
| 325 |
+
"learning_rate": 1.3297077453335193e-07,
|
| 326 |
+
"loss": 0.1594,
|
| 327 |
+
"step": 500
|
| 328 |
},
|
| 329 |
{
|
| 330 |
+
"epoch": 2.83,
|
| 331 |
+
"learning_rate": 7.497875749046122e-08,
|
| 332 |
+
"loss": 0.1333,
|
| 333 |
+
"step": 510
|
| 334 |
},
|
| 335 |
{
|
| 336 |
+
"epoch": 2.89,
|
| 337 |
+
"learning_rate": 3.3381988275995585e-08,
|
| 338 |
+
"loss": 0.1171,
|
| 339 |
+
"step": 520
|
| 340 |
},
|
| 341 |
{
|
| 342 |
+
"epoch": 2.94,
|
| 343 |
+
"learning_rate": 8.354221195471912e-09,
|
| 344 |
+
"loss": 0.2449,
|
| 345 |
+
"step": 530
|
| 346 |
},
|
| 347 |
{
|
| 348 |
+
"epoch": 3.0,
|
| 349 |
+
"learning_rate": 0.0,
|
| 350 |
+
"loss": 0.1401,
|
| 351 |
+
"step": 540
|
| 352 |
},
|
| 353 |
{
|
| 354 |
+
"epoch": 3.0,
|
| 355 |
+
"eval_f1": 0.19553868058851448,
|
| 356 |
+
"eval_loss": 5.8615827560424805,
|
| 357 |
+
"eval_runtime": 20.4644,
|
| 358 |
+
"eval_samples_per_second": 93.626,
|
| 359 |
+
"eval_steps_per_second": 1.466,
|
| 360 |
+
"step": 540
|
| 361 |
},
|
| 362 |
{
|
| 363 |
+
"epoch": 3.0,
|
| 364 |
+
"step": 540,
|
| 365 |
+
"total_flos": 17340365537280.0,
|
| 366 |
+
"train_loss": 15.830373929303002,
|
| 367 |
+
"train_runtime": 1144.0282,
|
| 368 |
+
"train_samples_per_second": 15.06,
|
| 369 |
+
"train_steps_per_second": 0.472
|
| 370 |
}
|
| 371 |
],
|
| 372 |
+
"logging_steps": 10,
|
| 373 |
+
"max_steps": 540,
|
| 374 |
"num_input_tokens_seen": 0,
|
| 375 |
+
"num_train_epochs": 3,
|
| 376 |
"save_steps": 500,
|
| 377 |
+
"total_flos": 17340365537280.0,
|
| 378 |
+
"train_batch_size": 4,
|
| 379 |
"trial_name": null,
|
| 380 |
"trial_params": null
|
| 381 |
}
|