End of training
Browse files
all_results.json
CHANGED
|
@@ -1,8 +1,8 @@
|
|
| 1 |
{
|
| 2 |
-
"epoch":
|
| 3 |
-
"eval_accuracy": 0.
|
| 4 |
-
"eval_loss": 2.
|
| 5 |
-
"eval_runtime":
|
| 6 |
-
"eval_samples_per_second": 5.
|
| 7 |
-
"eval_steps_per_second":
|
| 8 |
}
|
|
|
|
| 1 |
{
|
| 2 |
+
"epoch": 7.12,
|
| 3 |
+
"eval_accuracy": 0.225,
|
| 4 |
+
"eval_loss": 2.501384973526001,
|
| 5 |
+
"eval_runtime": 945.2716,
|
| 6 |
+
"eval_samples_per_second": 5.247,
|
| 7 |
+
"eval_steps_per_second": 0.656
|
| 8 |
}
|
runs/Mar17_19-35-48_500f3adbbe43/events.out.tfevents.1710729400.500f3adbbe43.26.1
CHANGED
|
@@ -1,3 +1,3 @@
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
-
oid sha256:
|
| 3 |
-
size
|
|
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:9a8ae09f173638e3a9b5cf378ee3b2bb4856529d7bcaabbb980f10e6a01cbe73
|
| 3 |
+
size 734
|
test_results.json
CHANGED
|
@@ -1,8 +1,8 @@
|
|
| 1 |
{
|
| 2 |
-
"epoch":
|
| 3 |
-
"eval_accuracy": 0.
|
| 4 |
-
"eval_loss": 2.
|
| 5 |
-
"eval_runtime":
|
| 6 |
-
"eval_samples_per_second": 5.
|
| 7 |
-
"eval_steps_per_second":
|
| 8 |
}
|
|
|
|
| 1 |
{
|
| 2 |
+
"epoch": 7.12,
|
| 3 |
+
"eval_accuracy": 0.225,
|
| 4 |
+
"eval_loss": 2.501384973526001,
|
| 5 |
+
"eval_runtime": 945.2716,
|
| 6 |
+
"eval_samples_per_second": 5.247,
|
| 7 |
+
"eval_steps_per_second": 0.656
|
| 8 |
}
|
trainer_state.json
CHANGED
|
@@ -1,7 +1,7 @@
|
|
| 1 |
{
|
| 2 |
-
"best_metric": 0.
|
| 3 |
-
"best_model_checkpoint": "videomae-base-finetuned-ucfcrime-full2/checkpoint-
|
| 4 |
-
"epoch":
|
| 5 |
"eval_steps": 500,
|
| 6 |
"global_step": 700,
|
| 7 |
"is_hyper_param_search": false,
|
|
@@ -10,1055 +10,601 @@
|
|
| 10 |
"log_history": [
|
| 11 |
{
|
| 12 |
"epoch": 0.01,
|
| 13 |
-
"grad_norm":
|
| 14 |
-
"learning_rate": 3.5714285714285714e-06,
|
| 15 |
-
"loss": 2.6878,
|
| 16 |
-
"step": 5
|
| 17 |
-
},
|
| 18 |
-
{
|
| 19 |
-
"epoch": 0.01,
|
| 20 |
-
"grad_norm": 21.39861297607422,
|
| 21 |
"learning_rate": 7.142857142857143e-06,
|
| 22 |
-
"loss": 2.
|
| 23 |
"step": 10
|
| 24 |
},
|
| 25 |
-
{
|
| 26 |
-
"epoch": 0.02,
|
| 27 |
-
"grad_norm": 17.860671997070312,
|
| 28 |
-
"learning_rate": 1.0714285714285714e-05,
|
| 29 |
-
"loss": 2.7214,
|
| 30 |
-
"step": 15
|
| 31 |
-
},
|
| 32 |
{
|
| 33 |
"epoch": 0.03,
|
| 34 |
-
"grad_norm":
|
| 35 |
"learning_rate": 1.4285714285714285e-05,
|
| 36 |
-
"loss": 2.
|
| 37 |
"step": 20
|
| 38 |
},
|
| 39 |
{
|
| 40 |
"epoch": 0.04,
|
| 41 |
-
"grad_norm":
|
| 42 |
-
"learning_rate": 1.785714285714286e-05,
|
| 43 |
-
"loss": 2.689,
|
| 44 |
-
"step": 25
|
| 45 |
-
},
|
| 46 |
-
{
|
| 47 |
-
"epoch": 0.04,
|
| 48 |
-
"grad_norm": 12.049698829650879,
|
| 49 |
"learning_rate": 2.1428571428571428e-05,
|
| 50 |
-
"loss": 2.
|
| 51 |
"step": 30
|
| 52 |
},
|
| 53 |
-
{
|
| 54 |
-
"epoch": 0.05,
|
| 55 |
-
"grad_norm": 13.192450523376465,
|
| 56 |
-
"learning_rate": 2.5e-05,
|
| 57 |
-
"loss": 2.5365,
|
| 58 |
-
"step": 35
|
| 59 |
-
},
|
| 60 |
{
|
| 61 |
"epoch": 0.06,
|
| 62 |
-
"grad_norm":
|
| 63 |
"learning_rate": 2.857142857142857e-05,
|
| 64 |
-
"loss": 2.
|
| 65 |
"step": 40
|
| 66 |
},
|
| 67 |
-
{
|
| 68 |
-
"epoch": 0.06,
|
| 69 |
-
"grad_norm": 10.616167068481445,
|
| 70 |
-
"learning_rate": 3.2142857142857144e-05,
|
| 71 |
-
"loss": 2.7132,
|
| 72 |
-
"step": 45
|
| 73 |
-
},
|
| 74 |
{
|
| 75 |
"epoch": 0.07,
|
| 76 |
-
"grad_norm":
|
| 77 |
"learning_rate": 3.571428571428572e-05,
|
| 78 |
-
"loss": 2.
|
| 79 |
"step": 50
|
| 80 |
},
|
| 81 |
-
{
|
| 82 |
-
"epoch": 0.08,
|
| 83 |
-
"grad_norm": 8.298579216003418,
|
| 84 |
-
"learning_rate": 3.928571428571429e-05,
|
| 85 |
-
"loss": 2.4331,
|
| 86 |
-
"step": 55
|
| 87 |
-
},
|
| 88 |
{
|
| 89 |
"epoch": 0.09,
|
| 90 |
-
"grad_norm":
|
| 91 |
"learning_rate": 4.2857142857142856e-05,
|
| 92 |
-
"loss": 2.
|
| 93 |
"step": 60
|
| 94 |
},
|
| 95 |
-
{
|
| 96 |
-
"epoch": 0.09,
|
| 97 |
-
"grad_norm": 10.443252563476562,
|
| 98 |
-
"learning_rate": 4.642857142857143e-05,
|
| 99 |
-
"loss": 2.7184,
|
| 100 |
-
"step": 65
|
| 101 |
-
},
|
| 102 |
{
|
| 103 |
"epoch": 0.1,
|
| 104 |
-
"grad_norm":
|
| 105 |
"learning_rate": 5e-05,
|
| 106 |
-
"loss": 2.
|
| 107 |
"step": 70
|
| 108 |
},
|
| 109 |
{
|
| 110 |
"epoch": 0.11,
|
| 111 |
-
"grad_norm": 8.
|
| 112 |
-
"learning_rate": 4.960317460317461e-05,
|
| 113 |
-
"loss": 2.3317,
|
| 114 |
-
"step": 75
|
| 115 |
-
},
|
| 116 |
-
{
|
| 117 |
-
"epoch": 0.11,
|
| 118 |
-
"grad_norm": 9.528456687927246,
|
| 119 |
"learning_rate": 4.9206349206349204e-05,
|
| 120 |
-
"loss": 2.
|
| 121 |
"step": 80
|
| 122 |
},
|
| 123 |
{
|
| 124 |
-
"epoch": 0.
|
| 125 |
-
"
|
| 126 |
-
"
|
| 127 |
-
"
|
| 128 |
-
"
|
|
|
|
|
|
|
| 129 |
},
|
| 130 |
{
|
| 131 |
-
"epoch": 0
|
| 132 |
-
"grad_norm":
|
| 133 |
"learning_rate": 4.841269841269841e-05,
|
| 134 |
-
"loss": 2.
|
| 135 |
"step": 90
|
| 136 |
},
|
| 137 |
{
|
| 138 |
-
"epoch":
|
| 139 |
-
"grad_norm":
|
| 140 |
-
"learning_rate": 4.801587301587302e-05,
|
| 141 |
-
"loss": 2.4065,
|
| 142 |
-
"step": 95
|
| 143 |
-
},
|
| 144 |
-
{
|
| 145 |
-
"epoch": 0.14,
|
| 146 |
-
"grad_norm": 10.122940063476562,
|
| 147 |
"learning_rate": 4.761904761904762e-05,
|
| 148 |
-
"loss": 2.
|
| 149 |
"step": 100
|
| 150 |
},
|
| 151 |
{
|
| 152 |
-
"epoch":
|
| 153 |
-
"grad_norm":
|
| 154 |
-
"learning_rate": 4.722222222222222e-05,
|
| 155 |
-
"loss": 2.5572,
|
| 156 |
-
"step": 105
|
| 157 |
-
},
|
| 158 |
-
{
|
| 159 |
-
"epoch": 0.16,
|
| 160 |
-
"grad_norm": 10.649138450622559,
|
| 161 |
"learning_rate": 4.682539682539683e-05,
|
| 162 |
-
"loss": 2.
|
| 163 |
"step": 110
|
| 164 |
},
|
| 165 |
{
|
| 166 |
-
"epoch":
|
| 167 |
-
"grad_norm":
|
| 168 |
-
"learning_rate": 4.642857142857143e-05,
|
| 169 |
-
"loss": 2.6114,
|
| 170 |
-
"step": 115
|
| 171 |
-
},
|
| 172 |
-
{
|
| 173 |
-
"epoch": 0.17,
|
| 174 |
-
"grad_norm": 7.953385829925537,
|
| 175 |
"learning_rate": 4.603174603174603e-05,
|
| 176 |
-
"loss": 2.
|
| 177 |
"step": 120
|
| 178 |
},
|
| 179 |
{
|
| 180 |
-
"epoch":
|
| 181 |
-
"grad_norm":
|
| 182 |
-
"learning_rate": 4.563492063492064e-05,
|
| 183 |
-
"loss": 2.6787,
|
| 184 |
-
"step": 125
|
| 185 |
-
},
|
| 186 |
-
{
|
| 187 |
-
"epoch": 0.19,
|
| 188 |
-
"grad_norm": 8.697818756103516,
|
| 189 |
"learning_rate": 4.523809523809524e-05,
|
| 190 |
-
"loss": 2.
|
| 191 |
"step": 130
|
| 192 |
},
|
| 193 |
{
|
| 194 |
-
"epoch":
|
| 195 |
-
"grad_norm":
|
| 196 |
-
"learning_rate": 4.4841269841269846e-05,
|
| 197 |
-
"loss": 2.6221,
|
| 198 |
-
"step": 135
|
| 199 |
-
},
|
| 200 |
-
{
|
| 201 |
-
"epoch": 0.2,
|
| 202 |
-
"grad_norm": 9.103715896606445,
|
| 203 |
"learning_rate": 4.4444444444444447e-05,
|
| 204 |
-
"loss": 2.
|
| 205 |
"step": 140
|
| 206 |
},
|
| 207 |
{
|
| 208 |
-
"epoch":
|
| 209 |
-
"grad_norm":
|
| 210 |
-
"learning_rate": 4.404761904761905e-05,
|
| 211 |
-
"loss": 2.6136,
|
| 212 |
-
"step": 145
|
| 213 |
-
},
|
| 214 |
-
{
|
| 215 |
-
"epoch": 0.21,
|
| 216 |
-
"grad_norm": 10.060530662536621,
|
| 217 |
"learning_rate": 4.3650793650793655e-05,
|
| 218 |
-
"loss": 2.
|
| 219 |
"step": 150
|
| 220 |
},
|
| 221 |
{
|
| 222 |
-
"epoch":
|
| 223 |
-
"grad_norm":
|
| 224 |
-
"learning_rate": 4.3253968253968256e-05,
|
| 225 |
-
"loss": 2.571,
|
| 226 |
-
"step": 155
|
| 227 |
-
},
|
| 228 |
-
{
|
| 229 |
-
"epoch": 0.23,
|
| 230 |
-
"grad_norm": 11.69568157196045,
|
| 231 |
"learning_rate": 4.2857142857142856e-05,
|
| 232 |
-
"loss": 2.
|
| 233 |
"step": 160
|
| 234 |
},
|
| 235 |
{
|
| 236 |
-
"epoch":
|
| 237 |
-
"grad_norm":
|
| 238 |
-
"learning_rate": 4.2460317460317464e-05,
|
| 239 |
-
"loss": 2.4975,
|
| 240 |
-
"step": 165
|
| 241 |
-
},
|
| 242 |
-
{
|
| 243 |
-
"epoch": 0.24,
|
| 244 |
-
"grad_norm": 8.202585220336914,
|
| 245 |
"learning_rate": 4.2063492063492065e-05,
|
| 246 |
-
"loss": 2.
|
| 247 |
"step": 170
|
| 248 |
},
|
| 249 |
{
|
| 250 |
-
"epoch":
|
| 251 |
-
"
|
| 252 |
-
"
|
| 253 |
-
"
|
| 254 |
-
"
|
| 255 |
-
|
| 256 |
-
|
| 257 |
-
"epoch": 0.25,
|
| 258 |
-
"eval_accuracy": 0.17190058044351839,
|
| 259 |
-
"eval_loss": 2.5464329719543457,
|
| 260 |
-
"eval_runtime": 2790.7409,
|
| 261 |
-
"eval_samples_per_second": 4.815,
|
| 262 |
-
"eval_steps_per_second": 1.204,
|
| 263 |
-
"step": 175
|
| 264 |
},
|
| 265 |
{
|
| 266 |
-
"epoch":
|
| 267 |
-
"grad_norm": 8.
|
| 268 |
"learning_rate": 4.126984126984127e-05,
|
| 269 |
-
"loss": 2.
|
| 270 |
"step": 180
|
| 271 |
},
|
| 272 |
{
|
| 273 |
-
"epoch":
|
| 274 |
-
"grad_norm": 7.
|
| 275 |
-
"learning_rate": 4.0873015873015874e-05,
|
| 276 |
-
"loss": 2.6691,
|
| 277 |
-
"step": 185
|
| 278 |
-
},
|
| 279 |
-
{
|
| 280 |
-
"epoch": 1.02,
|
| 281 |
-
"grad_norm": 7.451665878295898,
|
| 282 |
"learning_rate": 4.047619047619048e-05,
|
| 283 |
-
"loss": 2.
|
| 284 |
"step": 190
|
| 285 |
},
|
| 286 |
{
|
| 287 |
-
"epoch":
|
| 288 |
-
"grad_norm":
|
| 289 |
-
"learning_rate": 4.007936507936508e-05,
|
| 290 |
-
"loss": 2.7581,
|
| 291 |
-
"step": 195
|
| 292 |
-
},
|
| 293 |
-
{
|
| 294 |
-
"epoch": 1.04,
|
| 295 |
-
"grad_norm": 8.437348365783691,
|
| 296 |
"learning_rate": 3.968253968253968e-05,
|
| 297 |
-
"loss": 2.
|
| 298 |
"step": 200
|
| 299 |
},
|
| 300 |
{
|
| 301 |
-
"epoch":
|
| 302 |
-
"grad_norm":
|
| 303 |
-
"learning_rate": 3.928571428571429e-05,
|
| 304 |
-
"loss": 2.7035,
|
| 305 |
-
"step": 205
|
| 306 |
-
},
|
| 307 |
-
{
|
| 308 |
-
"epoch": 1.05,
|
| 309 |
-
"grad_norm": 7.769816875457764,
|
| 310 |
"learning_rate": 3.888888888888889e-05,
|
| 311 |
-
"loss": 2.
|
| 312 |
"step": 210
|
| 313 |
},
|
| 314 |
{
|
| 315 |
-
"epoch":
|
| 316 |
-
"grad_norm":
|
| 317 |
-
"learning_rate": 3.84920634920635e-05,
|
| 318 |
-
"loss": 2.5088,
|
| 319 |
-
"step": 215
|
| 320 |
-
},
|
| 321 |
-
{
|
| 322 |
-
"epoch": 1.06,
|
| 323 |
-
"grad_norm": 7.872964859008789,
|
| 324 |
"learning_rate": 3.809523809523809e-05,
|
| 325 |
-
"loss": 2.
|
| 326 |
"step": 220
|
| 327 |
},
|
| 328 |
{
|
| 329 |
-
"epoch":
|
| 330 |
-
"grad_norm":
|
| 331 |
-
"learning_rate": 3.76984126984127e-05,
|
| 332 |
-
"loss": 2.6319,
|
| 333 |
-
"step": 225
|
| 334 |
-
},
|
| 335 |
-
{
|
| 336 |
-
"epoch": 1.08,
|
| 337 |
-
"grad_norm": 7.4448394775390625,
|
| 338 |
"learning_rate": 3.730158730158731e-05,
|
| 339 |
-
"loss": 2.
|
| 340 |
"step": 230
|
| 341 |
},
|
| 342 |
{
|
| 343 |
-
"epoch":
|
| 344 |
-
"grad_norm":
|
| 345 |
-
"learning_rate": 3.690476190476191e-05,
|
| 346 |
-
"loss": 2.4276,
|
| 347 |
-
"step": 235
|
| 348 |
-
},
|
| 349 |
-
{
|
| 350 |
-
"epoch": 1.09,
|
| 351 |
-
"grad_norm": 7.45771598815918,
|
| 352 |
"learning_rate": 3.650793650793651e-05,
|
| 353 |
-
"loss": 2.
|
| 354 |
"step": 240
|
| 355 |
},
|
| 356 |
{
|
| 357 |
-
"epoch":
|
| 358 |
-
"grad_norm":
|
| 359 |
-
"learning_rate": 3.611111111111111e-05,
|
| 360 |
-
"loss": 2.4295,
|
| 361 |
-
"step": 245
|
| 362 |
-
},
|
| 363 |
-
{
|
| 364 |
-
"epoch": 1.11,
|
| 365 |
-
"grad_norm": 7.304623603820801,
|
| 366 |
"learning_rate": 3.571428571428572e-05,
|
| 367 |
-
"loss": 2.
|
| 368 |
"step": 250
|
| 369 |
},
|
| 370 |
{
|
| 371 |
-
"epoch":
|
| 372 |
-
"grad_norm": 8.
|
| 373 |
-
"learning_rate": 3.5317460317460324e-05,
|
| 374 |
-
"loss": 2.5164,
|
| 375 |
-
"step": 255
|
| 376 |
-
},
|
| 377 |
-
{
|
| 378 |
-
"epoch": 1.12,
|
| 379 |
-
"grad_norm": 6.037725448608398,
|
| 380 |
"learning_rate": 3.492063492063492e-05,
|
| 381 |
-
"loss": 2.
|
| 382 |
"step": 260
|
| 383 |
},
|
| 384 |
{
|
| 385 |
-
"epoch":
|
| 386 |
-
"
|
| 387 |
-
"
|
| 388 |
-
"
|
| 389 |
-
"
|
|
|
|
|
|
|
| 390 |
},
|
| 391 |
{
|
| 392 |
-
"epoch":
|
| 393 |
-
"grad_norm":
|
| 394 |
"learning_rate": 3.412698412698413e-05,
|
| 395 |
-
"loss": 2.
|
| 396 |
"step": 270
|
| 397 |
},
|
| 398 |
{
|
| 399 |
-
"epoch":
|
| 400 |
-
"grad_norm":
|
| 401 |
-
"learning_rate": 3.3730158730158734e-05,
|
| 402 |
-
"loss": 2.5689,
|
| 403 |
-
"step": 275
|
| 404 |
-
},
|
| 405 |
-
{
|
| 406 |
-
"epoch": 1.15,
|
| 407 |
-
"grad_norm": 8.133126258850098,
|
| 408 |
"learning_rate": 3.3333333333333335e-05,
|
| 409 |
-
"loss": 2.
|
| 410 |
"step": 280
|
| 411 |
},
|
| 412 |
{
|
| 413 |
-
"epoch":
|
| 414 |
-
"grad_norm":
|
| 415 |
-
"learning_rate": 3.2936507936507936e-05,
|
| 416 |
-
"loss": 2.4184,
|
| 417 |
-
"step": 285
|
| 418 |
-
},
|
| 419 |
-
{
|
| 420 |
-
"epoch": 1.16,
|
| 421 |
-
"grad_norm": 7.738976001739502,
|
| 422 |
"learning_rate": 3.253968253968254e-05,
|
| 423 |
-
"loss": 2.
|
| 424 |
"step": 290
|
| 425 |
},
|
| 426 |
{
|
| 427 |
-
"epoch":
|
| 428 |
-
"grad_norm":
|
| 429 |
-
"learning_rate": 3.2142857142857144e-05,
|
| 430 |
-
"loss": 2.3066,
|
| 431 |
-
"step": 295
|
| 432 |
-
},
|
| 433 |
-
{
|
| 434 |
-
"epoch": 1.18,
|
| 435 |
-
"grad_norm": 7.5370588302612305,
|
| 436 |
"learning_rate": 3.1746031746031745e-05,
|
| 437 |
-
"loss": 2.
|
| 438 |
"step": 300
|
| 439 |
},
|
| 440 |
{
|
| 441 |
-
"epoch":
|
| 442 |
-
"grad_norm": 7.
|
| 443 |
-
"learning_rate": 3.134920634920635e-05,
|
| 444 |
-
"loss": 2.4807,
|
| 445 |
-
"step": 305
|
| 446 |
-
},
|
| 447 |
-
{
|
| 448 |
-
"epoch": 1.19,
|
| 449 |
-
"grad_norm": 8.314027786254883,
|
| 450 |
"learning_rate": 3.095238095238095e-05,
|
| 451 |
-
"loss": 2.
|
| 452 |
"step": 310
|
| 453 |
},
|
| 454 |
{
|
| 455 |
-
"epoch":
|
| 456 |
-
"grad_norm":
|
| 457 |
-
"learning_rate": 3.055555555555556e-05,
|
| 458 |
-
"loss": 2.5207,
|
| 459 |
-
"step": 315
|
| 460 |
-
},
|
| 461 |
-
{
|
| 462 |
-
"epoch": 1.21,
|
| 463 |
-
"grad_norm": 8.472153663635254,
|
| 464 |
"learning_rate": 3.0158730158730158e-05,
|
| 465 |
-
"loss": 2.
|
| 466 |
"step": 320
|
| 467 |
},
|
| 468 |
{
|
| 469 |
-
"epoch":
|
| 470 |
-
"grad_norm": 8.
|
| 471 |
-
"learning_rate": 2.9761904761904762e-05,
|
| 472 |
-
"loss": 2.5238,
|
| 473 |
-
"step": 325
|
| 474 |
-
},
|
| 475 |
-
{
|
| 476 |
-
"epoch": 1.22,
|
| 477 |
-
"grad_norm": 8.0416898727417,
|
| 478 |
"learning_rate": 2.9365079365079366e-05,
|
| 479 |
-
"loss": 2.
|
| 480 |
"step": 330
|
| 481 |
},
|
| 482 |
{
|
| 483 |
-
"epoch":
|
| 484 |
-
"grad_norm":
|
| 485 |
-
"learning_rate": 2.8968253968253974e-05,
|
| 486 |
-
"loss": 2.3977,
|
| 487 |
-
"step": 335
|
| 488 |
-
},
|
| 489 |
-
{
|
| 490 |
-
"epoch": 1.24,
|
| 491 |
-
"grad_norm": 8.505749702453613,
|
| 492 |
"learning_rate": 2.857142857142857e-05,
|
| 493 |
-
"loss":
|
| 494 |
"step": 340
|
| 495 |
},
|
| 496 |
{
|
| 497 |
-
"epoch":
|
| 498 |
-
"grad_norm": 7.
|
| 499 |
-
"learning_rate": 2.8174603174603175e-05,
|
| 500 |
-
"loss": 1.993,
|
| 501 |
-
"step": 345
|
| 502 |
-
},
|
| 503 |
-
{
|
| 504 |
-
"epoch": 1.25,
|
| 505 |
-
"grad_norm": 8.76427936553955,
|
| 506 |
"learning_rate": 2.777777777777778e-05,
|
| 507 |
-
"loss":
|
| 508 |
"step": 350
|
| 509 |
},
|
| 510 |
{
|
| 511 |
-
"epoch":
|
| 512 |
-
"eval_accuracy": 0.
|
| 513 |
-
"eval_loss": 2.
|
| 514 |
-
"eval_runtime":
|
| 515 |
-
"eval_samples_per_second":
|
| 516 |
-
"eval_steps_per_second":
|
| 517 |
-
"step":
|
| 518 |
-
},
|
| 519 |
-
{
|
| 520 |
-
"epoch": 2.01,
|
| 521 |
-
"grad_norm": 9.777434349060059,
|
| 522 |
-
"learning_rate": 2.7380952380952383e-05,
|
| 523 |
-
"loss": 2.3064,
|
| 524 |
-
"step": 355
|
| 525 |
},
|
| 526 |
{
|
| 527 |
-
"epoch":
|
| 528 |
-
"grad_norm":
|
| 529 |
"learning_rate": 2.6984126984126984e-05,
|
| 530 |
-
"loss":
|
| 531 |
"step": 360
|
| 532 |
},
|
| 533 |
{
|
| 534 |
-
"epoch":
|
| 535 |
-
"grad_norm":
|
| 536 |
-
"learning_rate": 2.6587301587301588e-05,
|
| 537 |
-
"loss": 2.3623,
|
| 538 |
-
"step": 365
|
| 539 |
-
},
|
| 540 |
-
{
|
| 541 |
-
"epoch": 2.03,
|
| 542 |
-
"grad_norm": 9.471118927001953,
|
| 543 |
"learning_rate": 2.6190476190476192e-05,
|
| 544 |
-
"loss":
|
| 545 |
"step": 370
|
| 546 |
},
|
| 547 |
{
|
| 548 |
-
"epoch":
|
| 549 |
-
"grad_norm":
|
| 550 |
-
"learning_rate": 2.5793650793650796e-05,
|
| 551 |
-
"loss": 2.6313,
|
| 552 |
-
"step": 375
|
| 553 |
-
},
|
| 554 |
-
{
|
| 555 |
-
"epoch": 2.04,
|
| 556 |
-
"grad_norm": 9.770784378051758,
|
| 557 |
"learning_rate": 2.5396825396825397e-05,
|
| 558 |
-
"loss": 2.
|
| 559 |
"step": 380
|
| 560 |
},
|
| 561 |
{
|
| 562 |
-
"epoch":
|
| 563 |
-
"grad_norm":
|
| 564 |
-
"learning_rate": 2.5e-05,
|
| 565 |
-
"loss": 2.4206,
|
| 566 |
-
"step": 385
|
| 567 |
-
},
|
| 568 |
-
{
|
| 569 |
-
"epoch": 2.06,
|
| 570 |
-
"grad_norm": 10.88526725769043,
|
| 571 |
"learning_rate": 2.4603174603174602e-05,
|
| 572 |
-
"loss": 2.
|
| 573 |
"step": 390
|
| 574 |
},
|
| 575 |
{
|
| 576 |
-
"epoch":
|
| 577 |
-
"grad_norm":
|
| 578 |
-
"learning_rate": 2.4206349206349206e-05,
|
| 579 |
-
"loss": 2.3234,
|
| 580 |
-
"step": 395
|
| 581 |
-
},
|
| 582 |
-
{
|
| 583 |
-
"epoch": 2.07,
|
| 584 |
-
"grad_norm": 10.911561012268066,
|
| 585 |
"learning_rate": 2.380952380952381e-05,
|
| 586 |
-
"loss": 2.
|
| 587 |
"step": 400
|
| 588 |
},
|
| 589 |
{
|
| 590 |
-
"epoch":
|
| 591 |
-
"grad_norm":
|
| 592 |
-
"learning_rate": 2.3412698412698414e-05,
|
| 593 |
-
"loss": 2.209,
|
| 594 |
-
"step": 405
|
| 595 |
-
},
|
| 596 |
-
{
|
| 597 |
-
"epoch": 2.09,
|
| 598 |
-
"grad_norm": 10.131900787353516,
|
| 599 |
"learning_rate": 2.3015873015873015e-05,
|
| 600 |
-
"loss": 2.
|
| 601 |
"step": 410
|
| 602 |
},
|
| 603 |
{
|
| 604 |
-
"epoch":
|
| 605 |
-
"grad_norm":
|
| 606 |
-
"learning_rate": 2.261904761904762e-05,
|
| 607 |
-
"loss": 2.2815,
|
| 608 |
-
"step": 415
|
| 609 |
-
},
|
| 610 |
-
{
|
| 611 |
-
"epoch": 2.1,
|
| 612 |
-
"grad_norm": 8.873805046081543,
|
| 613 |
"learning_rate": 2.2222222222222223e-05,
|
| 614 |
-
"loss": 2.
|
| 615 |
"step": 420
|
| 616 |
},
|
| 617 |
{
|
| 618 |
-
"epoch":
|
| 619 |
-
"grad_norm":
|
| 620 |
-
"learning_rate": 2.1825396825396827e-05,
|
| 621 |
-
"loss": 2.4577,
|
| 622 |
-
"step": 425
|
| 623 |
-
},
|
| 624 |
-
{
|
| 625 |
-
"epoch": 2.11,
|
| 626 |
-
"grad_norm": 9.494528770446777,
|
| 627 |
"learning_rate": 2.1428571428571428e-05,
|
| 628 |
-
"loss": 2.
|
| 629 |
"step": 430
|
| 630 |
},
|
| 631 |
{
|
| 632 |
-
"epoch":
|
| 633 |
-
"grad_norm":
|
| 634 |
-
"learning_rate": 2.1031746031746032e-05,
|
| 635 |
-
"loss": 2.1842,
|
| 636 |
-
"step": 435
|
| 637 |
-
},
|
| 638 |
-
{
|
| 639 |
-
"epoch": 2.13,
|
| 640 |
-
"grad_norm": 11.144043922424316,
|
| 641 |
"learning_rate": 2.0634920634920636e-05,
|
| 642 |
-
"loss": 2.
|
| 643 |
"step": 440
|
| 644 |
},
|
| 645 |
{
|
| 646 |
-
"epoch":
|
| 647 |
-
"
|
| 648 |
-
"
|
| 649 |
-
"
|
| 650 |
-
"
|
|
|
|
|
|
|
| 651 |
},
|
| 652 |
{
|
| 653 |
-
"epoch":
|
| 654 |
-
"grad_norm":
|
| 655 |
"learning_rate": 1.984126984126984e-05,
|
| 656 |
-
"loss":
|
| 657 |
"step": 450
|
| 658 |
},
|
| 659 |
{
|
| 660 |
-
"epoch":
|
| 661 |
-
"grad_norm":
|
| 662 |
-
"learning_rate": 1.9444444444444445e-05,
|
| 663 |
-
"loss": 2.385,
|
| 664 |
-
"step": 455
|
| 665 |
-
},
|
| 666 |
-
{
|
| 667 |
-
"epoch": 2.16,
|
| 668 |
-
"grad_norm": 15.662388801574707,
|
| 669 |
"learning_rate": 1.9047619047619046e-05,
|
| 670 |
-
"loss":
|
| 671 |
"step": 460
|
| 672 |
},
|
| 673 |
{
|
| 674 |
-
"epoch":
|
| 675 |
-
"grad_norm":
|
| 676 |
-
"learning_rate": 1.8650793650793654e-05,
|
| 677 |
-
"loss": 2.3551,
|
| 678 |
-
"step": 465
|
| 679 |
-
},
|
| 680 |
-
{
|
| 681 |
-
"epoch": 2.17,
|
| 682 |
-
"grad_norm": 8.680765151977539,
|
| 683 |
"learning_rate": 1.8253968253968254e-05,
|
| 684 |
-
"loss":
|
| 685 |
"step": 470
|
| 686 |
},
|
| 687 |
{
|
| 688 |
-
"epoch":
|
| 689 |
-
"grad_norm":
|
| 690 |
-
"learning_rate": 1.785714285714286e-05,
|
| 691 |
-
"loss": 2.2117,
|
| 692 |
-
"step": 475
|
| 693 |
-
},
|
| 694 |
-
{
|
| 695 |
-
"epoch": 2.19,
|
| 696 |
-
"grad_norm": 8.637816429138184,
|
| 697 |
"learning_rate": 1.746031746031746e-05,
|
| 698 |
-
"loss":
|
| 699 |
"step": 480
|
| 700 |
},
|
| 701 |
{
|
| 702 |
-
"epoch":
|
| 703 |
-
"grad_norm":
|
| 704 |
-
"learning_rate": 1.7063492063492063e-05,
|
| 705 |
-
"loss": 2.5038,
|
| 706 |
-
"step": 485
|
| 707 |
-
},
|
| 708 |
-
{
|
| 709 |
-
"epoch": 2.2,
|
| 710 |
-
"grad_norm": 9.811333656311035,
|
| 711 |
"learning_rate": 1.6666666666666667e-05,
|
| 712 |
-
"loss": 2.
|
| 713 |
"step": 490
|
| 714 |
},
|
| 715 |
{
|
| 716 |
-
"epoch":
|
| 717 |
-
"grad_norm": 9.
|
| 718 |
-
"learning_rate": 1.626984126984127e-05,
|
| 719 |
-
"loss": 2.7302,
|
| 720 |
-
"step": 495
|
| 721 |
-
},
|
| 722 |
-
{
|
| 723 |
-
"epoch": 2.21,
|
| 724 |
-
"grad_norm": 8.803714752197266,
|
| 725 |
"learning_rate": 1.5873015873015872e-05,
|
| 726 |
-
"loss": 2.
|
| 727 |
"step": 500
|
| 728 |
},
|
| 729 |
{
|
| 730 |
-
"epoch":
|
| 731 |
-
"grad_norm":
|
| 732 |
-
"learning_rate": 1.5476190476190476e-05,
|
| 733 |
-
"loss": 2.3112,
|
| 734 |
-
"step": 505
|
| 735 |
-
},
|
| 736 |
-
{
|
| 737 |
-
"epoch": 2.23,
|
| 738 |
-
"grad_norm": 8.800261497497559,
|
| 739 |
"learning_rate": 1.5079365079365079e-05,
|
| 740 |
-
"loss":
|
| 741 |
"step": 510
|
| 742 |
},
|
| 743 |
{
|
| 744 |
-
"epoch":
|
| 745 |
-
"grad_norm":
|
| 746 |
-
"learning_rate": 1.4682539682539683e-05,
|
| 747 |
-
"loss": 2.4612,
|
| 748 |
-
"step": 515
|
| 749 |
-
},
|
| 750 |
-
{
|
| 751 |
-
"epoch": 2.24,
|
| 752 |
-
"grad_norm": 11.743760108947754,
|
| 753 |
"learning_rate": 1.4285714285714285e-05,
|
| 754 |
-
"loss":
|
| 755 |
"step": 520
|
| 756 |
},
|
| 757 |
{
|
| 758 |
-
"epoch":
|
| 759 |
-
"
|
| 760 |
-
"
|
| 761 |
-
"
|
| 762 |
-
"
|
| 763 |
-
|
| 764 |
-
|
| 765 |
-
"epoch": 2.25,
|
| 766 |
-
"eval_accuracy": 0.13528798928411967,
|
| 767 |
-
"eval_loss": 2.775129556655884,
|
| 768 |
-
"eval_runtime": 2705.5287,
|
| 769 |
-
"eval_samples_per_second": 4.967,
|
| 770 |
-
"eval_steps_per_second": 1.242,
|
| 771 |
-
"step": 525
|
| 772 |
},
|
| 773 |
{
|
| 774 |
-
"epoch":
|
| 775 |
-
"grad_norm":
|
| 776 |
"learning_rate": 1.3492063492063492e-05,
|
| 777 |
-
"loss":
|
| 778 |
"step": 530
|
| 779 |
},
|
| 780 |
{
|
| 781 |
-
"epoch":
|
| 782 |
-
"grad_norm":
|
| 783 |
-
"learning_rate": 1.3095238095238096e-05,
|
| 784 |
-
"loss": 2.7047,
|
| 785 |
-
"step": 535
|
| 786 |
-
},
|
| 787 |
-
{
|
| 788 |
-
"epoch": 3.02,
|
| 789 |
-
"grad_norm": 9.126996040344238,
|
| 790 |
"learning_rate": 1.2698412698412699e-05,
|
| 791 |
-
"loss":
|
| 792 |
"step": 540
|
| 793 |
},
|
| 794 |
{
|
| 795 |
-
"epoch":
|
| 796 |
-
"grad_norm": 8.
|
| 797 |
-
"learning_rate": 1.2301587301587301e-05,
|
| 798 |
-
"loss": 2.292,
|
| 799 |
-
"step": 545
|
| 800 |
-
},
|
| 801 |
-
{
|
| 802 |
-
"epoch": 3.04,
|
| 803 |
-
"grad_norm": 9.275003433227539,
|
| 804 |
"learning_rate": 1.1904761904761905e-05,
|
| 805 |
-
"loss":
|
| 806 |
"step": 550
|
| 807 |
},
|
| 808 |
{
|
| 809 |
-
"epoch":
|
| 810 |
-
"grad_norm":
|
| 811 |
-
"learning_rate": 1.1507936507936508e-05,
|
| 812 |
-
"loss": 2.3276,
|
| 813 |
-
"step": 555
|
| 814 |
-
},
|
| 815 |
-
{
|
| 816 |
-
"epoch": 3.05,
|
| 817 |
-
"grad_norm": 9.327167510986328,
|
| 818 |
"learning_rate": 1.1111111111111112e-05,
|
| 819 |
-
"loss":
|
| 820 |
"step": 560
|
| 821 |
},
|
| 822 |
{
|
| 823 |
-
"epoch":
|
| 824 |
-
"grad_norm":
|
| 825 |
-
"learning_rate": 1.0714285714285714e-05,
|
| 826 |
-
"loss": 2.1437,
|
| 827 |
-
"step": 565
|
| 828 |
-
},
|
| 829 |
-
{
|
| 830 |
-
"epoch": 3.06,
|
| 831 |
-
"grad_norm": 10.702942848205566,
|
| 832 |
"learning_rate": 1.0317460317460318e-05,
|
| 833 |
-
"loss":
|
| 834 |
"step": 570
|
| 835 |
},
|
| 836 |
{
|
| 837 |
-
"epoch":
|
| 838 |
-
"grad_norm": 13.
|
| 839 |
-
"learning_rate": 9.92063492063492e-06,
|
| 840 |
-
"loss": 2.5248,
|
| 841 |
-
"step": 575
|
| 842 |
-
},
|
| 843 |
-
{
|
| 844 |
-
"epoch": 3.08,
|
| 845 |
-
"grad_norm": 7.934788227081299,
|
| 846 |
"learning_rate": 9.523809523809523e-06,
|
| 847 |
-
"loss":
|
| 848 |
"step": 580
|
| 849 |
},
|
| 850 |
{
|
| 851 |
-
"epoch":
|
| 852 |
-
"grad_norm":
|
| 853 |
-
"learning_rate": 9.126984126984127e-06,
|
| 854 |
-
"loss": 2.1392,
|
| 855 |
-
"step": 585
|
| 856 |
-
},
|
| 857 |
-
{
|
| 858 |
-
"epoch": 3.09,
|
| 859 |
-
"grad_norm": 11.382800102233887,
|
| 860 |
"learning_rate": 8.73015873015873e-06,
|
| 861 |
-
"loss":
|
| 862 |
"step": 590
|
| 863 |
},
|
| 864 |
{
|
| 865 |
-
"epoch":
|
| 866 |
-
"grad_norm":
|
| 867 |
-
"learning_rate": 8.333333333333334e-06,
|
| 868 |
-
"loss": 2.1905,
|
| 869 |
-
"step": 595
|
| 870 |
-
},
|
| 871 |
-
{
|
| 872 |
-
"epoch": 3.11,
|
| 873 |
-
"grad_norm": 9.530142784118652,
|
| 874 |
"learning_rate": 7.936507936507936e-06,
|
| 875 |
-
"loss":
|
| 876 |
"step": 600
|
| 877 |
},
|
| 878 |
{
|
| 879 |
-
"epoch":
|
| 880 |
-
"grad_norm":
|
| 881 |
-
"learning_rate": 7.5396825396825394e-06,
|
| 882 |
-
"loss": 2.4007,
|
| 883 |
-
"step": 605
|
| 884 |
-
},
|
| 885 |
-
{
|
| 886 |
-
"epoch": 3.12,
|
| 887 |
-
"grad_norm": 10.5969820022583,
|
| 888 |
"learning_rate": 7.142857142857143e-06,
|
| 889 |
-
"loss":
|
| 890 |
"step": 610
|
| 891 |
},
|
| 892 |
{
|
| 893 |
-
"epoch":
|
| 894 |
-
"
|
| 895 |
-
"
|
| 896 |
-
"
|
| 897 |
-
"
|
|
|
|
|
|
|
| 898 |
},
|
| 899 |
{
|
| 900 |
-
"epoch":
|
| 901 |
-
"grad_norm":
|
| 902 |
"learning_rate": 6.349206349206349e-06,
|
| 903 |
-
"loss":
|
| 904 |
"step": 620
|
| 905 |
},
|
| 906 |
{
|
| 907 |
-
"epoch":
|
| 908 |
-
"grad_norm":
|
| 909 |
-
"learning_rate": 5.9523809523809525e-06,
|
| 910 |
-
"loss": 2.2321,
|
| 911 |
-
"step": 625
|
| 912 |
-
},
|
| 913 |
-
{
|
| 914 |
-
"epoch": 3.15,
|
| 915 |
-
"grad_norm": 8.392168045043945,
|
| 916 |
"learning_rate": 5.555555555555556e-06,
|
| 917 |
-
"loss":
|
| 918 |
"step": 630
|
| 919 |
},
|
| 920 |
{
|
| 921 |
-
"epoch":
|
| 922 |
-
"grad_norm":
|
| 923 |
-
"learning_rate": 5.158730158730159e-06,
|
| 924 |
-
"loss": 2.2595,
|
| 925 |
-
"step": 635
|
| 926 |
-
},
|
| 927 |
-
{
|
| 928 |
-
"epoch": 3.16,
|
| 929 |
-
"grad_norm": 8.25489616394043,
|
| 930 |
"learning_rate": 4.7619047619047615e-06,
|
| 931 |
-
"loss":
|
| 932 |
"step": 640
|
| 933 |
},
|
| 934 |
{
|
| 935 |
-
"epoch":
|
| 936 |
-
"grad_norm":
|
| 937 |
-
"learning_rate": 4.365079365079365e-06,
|
| 938 |
-
"loss": 2.4893,
|
| 939 |
-
"step": 645
|
| 940 |
-
},
|
| 941 |
-
{
|
| 942 |
-
"epoch": 3.18,
|
| 943 |
-
"grad_norm": 10.919437408447266,
|
| 944 |
"learning_rate": 3.968253968253968e-06,
|
| 945 |
-
"loss":
|
| 946 |
"step": 650
|
| 947 |
},
|
| 948 |
{
|
| 949 |
-
"epoch":
|
| 950 |
-
"grad_norm":
|
| 951 |
-
"learning_rate": 3.5714285714285714e-06,
|
| 952 |
-
"loss": 2.1834,
|
| 953 |
-
"step": 655
|
| 954 |
-
},
|
| 955 |
-
{
|
| 956 |
-
"epoch": 3.19,
|
| 957 |
-
"grad_norm": 11.596048355102539,
|
| 958 |
"learning_rate": 3.1746031746031746e-06,
|
| 959 |
-
"loss":
|
| 960 |
"step": 660
|
| 961 |
},
|
| 962 |
{
|
| 963 |
-
"epoch":
|
| 964 |
-
"grad_norm": 11.
|
| 965 |
-
"learning_rate": 2.777777777777778e-06,
|
| 966 |
-
"loss": 2.3287,
|
| 967 |
-
"step": 665
|
| 968 |
-
},
|
| 969 |
-
{
|
| 970 |
-
"epoch": 3.21,
|
| 971 |
-
"grad_norm": 9.233076095581055,
|
| 972 |
"learning_rate": 2.3809523809523808e-06,
|
| 973 |
-
"loss":
|
| 974 |
"step": 670
|
| 975 |
},
|
| 976 |
{
|
| 977 |
-
"epoch":
|
| 978 |
-
"grad_norm":
|
| 979 |
-
"learning_rate": 1.984126984126984e-06,
|
| 980 |
-
"loss": 2.2577,
|
| 981 |
-
"step": 675
|
| 982 |
-
},
|
| 983 |
-
{
|
| 984 |
-
"epoch": 3.22,
|
| 985 |
-
"grad_norm": 8.783037185668945,
|
| 986 |
"learning_rate": 1.5873015873015873e-06,
|
| 987 |
-
"loss":
|
| 988 |
"step": 680
|
| 989 |
},
|
| 990 |
{
|
| 991 |
-
"epoch":
|
| 992 |
-
"grad_norm":
|
| 993 |
-
"learning_rate": 1.1904761904761904e-06,
|
| 994 |
-
"loss": 2.3779,
|
| 995 |
-
"step": 685
|
| 996 |
-
},
|
| 997 |
-
{
|
| 998 |
-
"epoch": 3.24,
|
| 999 |
-
"grad_norm": 9.461167335510254,
|
| 1000 |
"learning_rate": 7.936507936507937e-07,
|
| 1001 |
-
"loss": 1.
|
| 1002 |
"step": 690
|
| 1003 |
},
|
| 1004 |
{
|
| 1005 |
-
"epoch":
|
| 1006 |
-
"grad_norm":
|
| 1007 |
-
"learning_rate": 3.9682539682539683e-07,
|
| 1008 |
-
"loss": 2.0514,
|
| 1009 |
-
"step": 695
|
| 1010 |
-
},
|
| 1011 |
-
{
|
| 1012 |
-
"epoch": 3.25,
|
| 1013 |
-
"grad_norm": 11.628158569335938,
|
| 1014 |
"learning_rate": 0.0,
|
| 1015 |
-
"loss": 1.
|
| 1016 |
"step": 700
|
| 1017 |
},
|
| 1018 |
{
|
| 1019 |
-
"epoch":
|
| 1020 |
-
"eval_accuracy": 0.
|
| 1021 |
-
"eval_loss": 2.
|
| 1022 |
-
"eval_runtime":
|
| 1023 |
-
"eval_samples_per_second":
|
| 1024 |
-
"eval_steps_per_second":
|
| 1025 |
"step": 700
|
| 1026 |
},
|
| 1027 |
{
|
| 1028 |
-
"epoch":
|
| 1029 |
"step": 700,
|
| 1030 |
-
"total_flos":
|
| 1031 |
-
"train_loss": 2.
|
| 1032 |
-
"train_runtime":
|
| 1033 |
-
"train_samples_per_second": 0.
|
| 1034 |
-
"train_steps_per_second": 0.
|
| 1035 |
-
},
|
| 1036 |
-
{
|
| 1037 |
-
"epoch":
|
| 1038 |
-
"eval_accuracy": 0.
|
| 1039 |
-
"eval_loss": 2.
|
| 1040 |
-
"eval_runtime":
|
| 1041 |
-
"eval_samples_per_second": 5.
|
| 1042 |
-
"eval_steps_per_second":
|
| 1043 |
"step": 700
|
| 1044 |
},
|
| 1045 |
{
|
| 1046 |
-
"epoch":
|
| 1047 |
-
"eval_accuracy": 0.
|
| 1048 |
-
"eval_loss": 2.
|
| 1049 |
-
"eval_runtime":
|
| 1050 |
-
"eval_samples_per_second": 5.
|
| 1051 |
-
"eval_steps_per_second":
|
| 1052 |
"step": 700
|
| 1053 |
}
|
| 1054 |
],
|
| 1055 |
-
"logging_steps":
|
| 1056 |
"max_steps": 700,
|
| 1057 |
"num_input_tokens_seen": 0,
|
| 1058 |
"num_train_epochs": 9223372036854775807,
|
| 1059 |
"save_steps": 500,
|
| 1060 |
-
"total_flos":
|
| 1061 |
-
"train_batch_size":
|
| 1062 |
"trial_name": null,
|
| 1063 |
"trial_params": null
|
| 1064 |
}
|
|
|
|
| 1 |
{
|
| 2 |
+
"best_metric": 0.23411221908022026,
|
| 3 |
+
"best_model_checkpoint": "videomae-base-finetuned-ucfcrime-full2/checkpoint-616",
|
| 4 |
+
"epoch": 7.12,
|
| 5 |
"eval_steps": 500,
|
| 6 |
"global_step": 700,
|
| 7 |
"is_hyper_param_search": false,
|
|
|
|
| 10 |
"log_history": [
|
| 11 |
{
|
| 12 |
"epoch": 0.01,
|
| 13 |
+
"grad_norm": 9.479973793029785,
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 14 |
"learning_rate": 7.142857142857143e-06,
|
| 15 |
+
"loss": 2.6626,
|
| 16 |
"step": 10
|
| 17 |
},
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 18 |
{
|
| 19 |
"epoch": 0.03,
|
| 20 |
+
"grad_norm": 9.114124298095703,
|
| 21 |
"learning_rate": 1.4285714285714285e-05,
|
| 22 |
+
"loss": 2.6526,
|
| 23 |
"step": 20
|
| 24 |
},
|
| 25 |
{
|
| 26 |
"epoch": 0.04,
|
| 27 |
+
"grad_norm": 8.465902328491211,
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 28 |
"learning_rate": 2.1428571428571428e-05,
|
| 29 |
+
"loss": 2.6297,
|
| 30 |
"step": 30
|
| 31 |
},
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 32 |
{
|
| 33 |
"epoch": 0.06,
|
| 34 |
+
"grad_norm": 8.733708381652832,
|
| 35 |
"learning_rate": 2.857142857142857e-05,
|
| 36 |
+
"loss": 2.5219,
|
| 37 |
"step": 40
|
| 38 |
},
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 39 |
{
|
| 40 |
"epoch": 0.07,
|
| 41 |
+
"grad_norm": 7.121254920959473,
|
| 42 |
"learning_rate": 3.571428571428572e-05,
|
| 43 |
+
"loss": 2.5271,
|
| 44 |
"step": 50
|
| 45 |
},
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 46 |
{
|
| 47 |
"epoch": 0.09,
|
| 48 |
+
"grad_norm": 7.228466510772705,
|
| 49 |
"learning_rate": 4.2857142857142856e-05,
|
| 50 |
+
"loss": 2.5257,
|
| 51 |
"step": 60
|
| 52 |
},
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 53 |
{
|
| 54 |
"epoch": 0.1,
|
| 55 |
+
"grad_norm": 7.514686584472656,
|
| 56 |
"learning_rate": 5e-05,
|
| 57 |
+
"loss": 2.5942,
|
| 58 |
"step": 70
|
| 59 |
},
|
| 60 |
{
|
| 61 |
"epoch": 0.11,
|
| 62 |
+
"grad_norm": 8.028372764587402,
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 63 |
"learning_rate": 4.9206349206349204e-05,
|
| 64 |
+
"loss": 2.5836,
|
| 65 |
"step": 80
|
| 66 |
},
|
| 67 |
{
|
| 68 |
+
"epoch": 0.13,
|
| 69 |
+
"eval_accuracy": 0.20799226075308827,
|
| 70 |
+
"eval_loss": 2.494396924972534,
|
| 71 |
+
"eval_runtime": 2684.2234,
|
| 72 |
+
"eval_samples_per_second": 5.006,
|
| 73 |
+
"eval_steps_per_second": 0.626,
|
| 74 |
+
"step": 88
|
| 75 |
},
|
| 76 |
{
|
| 77 |
+
"epoch": 1.0,
|
| 78 |
+
"grad_norm": 9.068702697753906,
|
| 79 |
"learning_rate": 4.841269841269841e-05,
|
| 80 |
+
"loss": 2.5532,
|
| 81 |
"step": 90
|
| 82 |
},
|
| 83 |
{
|
| 84 |
+
"epoch": 1.02,
|
| 85 |
+
"grad_norm": 9.19058609008789,
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 86 |
"learning_rate": 4.761904761904762e-05,
|
| 87 |
+
"loss": 2.4752,
|
| 88 |
"step": 100
|
| 89 |
},
|
| 90 |
{
|
| 91 |
+
"epoch": 1.03,
|
| 92 |
+
"grad_norm": 6.300441741943359,
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 93 |
"learning_rate": 4.682539682539683e-05,
|
| 94 |
+
"loss": 2.512,
|
| 95 |
"step": 110
|
| 96 |
},
|
| 97 |
{
|
| 98 |
+
"epoch": 1.05,
|
| 99 |
+
"grad_norm": 6.24762487411499,
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 100 |
"learning_rate": 4.603174603174603e-05,
|
| 101 |
+
"loss": 2.6188,
|
| 102 |
"step": 120
|
| 103 |
},
|
| 104 |
{
|
| 105 |
+
"epoch": 1.06,
|
| 106 |
+
"grad_norm": 6.59229850769043,
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 107 |
"learning_rate": 4.523809523809524e-05,
|
| 108 |
+
"loss": 2.3887,
|
| 109 |
"step": 130
|
| 110 |
},
|
| 111 |
{
|
| 112 |
+
"epoch": 1.07,
|
| 113 |
+
"grad_norm": 7.241441249847412,
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 114 |
"learning_rate": 4.4444444444444447e-05,
|
| 115 |
+
"loss": 2.4597,
|
| 116 |
"step": 140
|
| 117 |
},
|
| 118 |
{
|
| 119 |
+
"epoch": 1.09,
|
| 120 |
+
"grad_norm": 6.4515275955200195,
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 121 |
"learning_rate": 4.3650793650793655e-05,
|
| 122 |
+
"loss": 2.4792,
|
| 123 |
"step": 150
|
| 124 |
},
|
| 125 |
{
|
| 126 |
+
"epoch": 1.1,
|
| 127 |
+
"grad_norm": 6.551385879516602,
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 128 |
"learning_rate": 4.2857142857142856e-05,
|
| 129 |
+
"loss": 2.4636,
|
| 130 |
"step": 160
|
| 131 |
},
|
| 132 |
{
|
| 133 |
+
"epoch": 1.12,
|
| 134 |
+
"grad_norm": 5.237429618835449,
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 135 |
"learning_rate": 4.2063492063492065e-05,
|
| 136 |
+
"loss": 2.3212,
|
| 137 |
"step": 170
|
| 138 |
},
|
| 139 |
{
|
| 140 |
+
"epoch": 1.13,
|
| 141 |
+
"eval_accuracy": 0.17725852061318648,
|
| 142 |
+
"eval_loss": 2.5854506492614746,
|
| 143 |
+
"eval_runtime": 2671.3135,
|
| 144 |
+
"eval_samples_per_second": 5.03,
|
| 145 |
+
"eval_steps_per_second": 0.629,
|
| 146 |
+
"step": 176
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 147 |
},
|
| 148 |
{
|
| 149 |
+
"epoch": 2.01,
|
| 150 |
+
"grad_norm": 8.875786781311035,
|
| 151 |
"learning_rate": 4.126984126984127e-05,
|
| 152 |
+
"loss": 2.3138,
|
| 153 |
"step": 180
|
| 154 |
},
|
| 155 |
{
|
| 156 |
+
"epoch": 2.02,
|
| 157 |
+
"grad_norm": 7.928117275238037,
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 158 |
"learning_rate": 4.047619047619048e-05,
|
| 159 |
+
"loss": 2.535,
|
| 160 |
"step": 190
|
| 161 |
},
|
| 162 |
{
|
| 163 |
+
"epoch": 2.03,
|
| 164 |
+
"grad_norm": 7.844710350036621,
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 165 |
"learning_rate": 3.968253968253968e-05,
|
| 166 |
+
"loss": 2.4373,
|
| 167 |
"step": 200
|
| 168 |
},
|
| 169 |
{
|
| 170 |
+
"epoch": 2.05,
|
| 171 |
+
"grad_norm": 7.789624214172363,
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 172 |
"learning_rate": 3.888888888888889e-05,
|
| 173 |
+
"loss": 2.3984,
|
| 174 |
"step": 210
|
| 175 |
},
|
| 176 |
{
|
| 177 |
+
"epoch": 2.06,
|
| 178 |
+
"grad_norm": 7.342363357543945,
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 179 |
"learning_rate": 3.809523809523809e-05,
|
| 180 |
+
"loss": 2.1948,
|
| 181 |
"step": 220
|
| 182 |
},
|
| 183 |
{
|
| 184 |
+
"epoch": 2.08,
|
| 185 |
+
"grad_norm": 6.116743564605713,
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 186 |
"learning_rate": 3.730158730158731e-05,
|
| 187 |
+
"loss": 2.3234,
|
| 188 |
"step": 230
|
| 189 |
},
|
| 190 |
{
|
| 191 |
+
"epoch": 2.09,
|
| 192 |
+
"grad_norm": 8.138971328735352,
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 193 |
"learning_rate": 3.650793650793651e-05,
|
| 194 |
+
"loss": 2.3012,
|
| 195 |
"step": 240
|
| 196 |
},
|
| 197 |
{
|
| 198 |
+
"epoch": 2.11,
|
| 199 |
+
"grad_norm": 7.563113689422607,
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 200 |
"learning_rate": 3.571428571428572e-05,
|
| 201 |
+
"loss": 2.4211,
|
| 202 |
"step": 250
|
| 203 |
},
|
| 204 |
{
|
| 205 |
+
"epoch": 2.12,
|
| 206 |
+
"grad_norm": 8.180752754211426,
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 207 |
"learning_rate": 3.492063492063492e-05,
|
| 208 |
+
"loss": 2.2333,
|
| 209 |
"step": 260
|
| 210 |
},
|
| 211 |
{
|
| 212 |
+
"epoch": 2.13,
|
| 213 |
+
"eval_accuracy": 0.10455424914421789,
|
| 214 |
+
"eval_loss": 2.627028226852417,
|
| 215 |
+
"eval_runtime": 2669.9747,
|
| 216 |
+
"eval_samples_per_second": 5.033,
|
| 217 |
+
"eval_steps_per_second": 0.629,
|
| 218 |
+
"step": 264
|
| 219 |
},
|
| 220 |
{
|
| 221 |
+
"epoch": 3.01,
|
| 222 |
+
"grad_norm": 8.733146667480469,
|
| 223 |
"learning_rate": 3.412698412698413e-05,
|
| 224 |
+
"loss": 2.3931,
|
| 225 |
"step": 270
|
| 226 |
},
|
| 227 |
{
|
| 228 |
+
"epoch": 3.02,
|
| 229 |
+
"grad_norm": 8.307793617248535,
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 230 |
"learning_rate": 3.3333333333333335e-05,
|
| 231 |
+
"loss": 2.1948,
|
| 232 |
"step": 280
|
| 233 |
},
|
| 234 |
{
|
| 235 |
+
"epoch": 3.04,
|
| 236 |
+
"grad_norm": 9.008221626281738,
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 237 |
"learning_rate": 3.253968253968254e-05,
|
| 238 |
+
"loss": 2.3545,
|
| 239 |
"step": 290
|
| 240 |
},
|
| 241 |
{
|
| 242 |
+
"epoch": 3.05,
|
| 243 |
+
"grad_norm": 8.743054389953613,
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 244 |
"learning_rate": 3.1746031746031745e-05,
|
| 245 |
+
"loss": 2.1182,
|
| 246 |
"step": 300
|
| 247 |
},
|
| 248 |
{
|
| 249 |
+
"epoch": 3.07,
|
| 250 |
+
"grad_norm": 7.782846450805664,
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 251 |
"learning_rate": 3.095238095238095e-05,
|
| 252 |
+
"loss": 2.2446,
|
| 253 |
"step": 310
|
| 254 |
},
|
| 255 |
{
|
| 256 |
+
"epoch": 3.08,
|
| 257 |
+
"grad_norm": 9.437503814697266,
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 258 |
"learning_rate": 3.0158730158730158e-05,
|
| 259 |
+
"loss": 2.3095,
|
| 260 |
"step": 320
|
| 261 |
},
|
| 262 |
{
|
| 263 |
+
"epoch": 3.09,
|
| 264 |
+
"grad_norm": 8.776910781860352,
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 265 |
"learning_rate": 2.9365079365079366e-05,
|
| 266 |
+
"loss": 2.2165,
|
| 267 |
"step": 330
|
| 268 |
},
|
| 269 |
{
|
| 270 |
+
"epoch": 3.11,
|
| 271 |
+
"grad_norm": 7.262299060821533,
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 272 |
"learning_rate": 2.857142857142857e-05,
|
| 273 |
+
"loss": 1.9958,
|
| 274 |
"step": 340
|
| 275 |
},
|
| 276 |
{
|
| 277 |
+
"epoch": 3.12,
|
| 278 |
+
"grad_norm": 7.0002288818359375,
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 279 |
"learning_rate": 2.777777777777778e-05,
|
| 280 |
+
"loss": 1.985,
|
| 281 |
"step": 350
|
| 282 |
},
|
| 283 |
{
|
| 284 |
+
"epoch": 3.13,
|
| 285 |
+
"eval_accuracy": 0.2108944783449918,
|
| 286 |
+
"eval_loss": 2.4058477878570557,
|
| 287 |
+
"eval_runtime": 2676.3678,
|
| 288 |
+
"eval_samples_per_second": 5.021,
|
| 289 |
+
"eval_steps_per_second": 0.628,
|
| 290 |
+
"step": 352
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 291 |
},
|
| 292 |
{
|
| 293 |
+
"epoch": 4.01,
|
| 294 |
+
"grad_norm": 13.075250625610352,
|
| 295 |
"learning_rate": 2.6984126984126984e-05,
|
| 296 |
+
"loss": 1.9445,
|
| 297 |
"step": 360
|
| 298 |
},
|
| 299 |
{
|
| 300 |
+
"epoch": 4.03,
|
| 301 |
+
"grad_norm": 11.625497817993164,
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 302 |
"learning_rate": 2.6190476190476192e-05,
|
| 303 |
+
"loss": 1.9884,
|
| 304 |
"step": 370
|
| 305 |
},
|
| 306 |
{
|
| 307 |
+
"epoch": 4.04,
|
| 308 |
+
"grad_norm": 9.822185516357422,
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 309 |
"learning_rate": 2.5396825396825397e-05,
|
| 310 |
+
"loss": 2.1677,
|
| 311 |
"step": 380
|
| 312 |
},
|
| 313 |
{
|
| 314 |
+
"epoch": 4.05,
|
| 315 |
+
"grad_norm": 9.551963806152344,
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 316 |
"learning_rate": 2.4603174603174602e-05,
|
| 317 |
+
"loss": 2.1942,
|
| 318 |
"step": 390
|
| 319 |
},
|
| 320 |
{
|
| 321 |
+
"epoch": 4.07,
|
| 322 |
+
"grad_norm": 10.00942325592041,
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 323 |
"learning_rate": 2.380952380952381e-05,
|
| 324 |
+
"loss": 2.1495,
|
| 325 |
"step": 400
|
| 326 |
},
|
| 327 |
{
|
| 328 |
+
"epoch": 4.08,
|
| 329 |
+
"grad_norm": 7.805147171020508,
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 330 |
"learning_rate": 2.3015873015873015e-05,
|
| 331 |
+
"loss": 2.1318,
|
| 332 |
"step": 410
|
| 333 |
},
|
| 334 |
{
|
| 335 |
+
"epoch": 4.1,
|
| 336 |
+
"grad_norm": 10.390979766845703,
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 337 |
"learning_rate": 2.2222222222222223e-05,
|
| 338 |
+
"loss": 2.0618,
|
| 339 |
"step": 420
|
| 340 |
},
|
| 341 |
{
|
| 342 |
+
"epoch": 4.11,
|
| 343 |
+
"grad_norm": 8.040055274963379,
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 344 |
"learning_rate": 2.1428571428571428e-05,
|
| 345 |
+
"loss": 2.1917,
|
| 346 |
"step": 430
|
| 347 |
},
|
| 348 |
{
|
| 349 |
+
"epoch": 4.13,
|
| 350 |
+
"grad_norm": 19.272586822509766,
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 351 |
"learning_rate": 2.0634920634920636e-05,
|
| 352 |
+
"loss": 2.194,
|
| 353 |
"step": 440
|
| 354 |
},
|
| 355 |
{
|
| 356 |
+
"epoch": 4.13,
|
| 357 |
+
"eval_accuracy": 0.22354517041226374,
|
| 358 |
+
"eval_loss": 2.3653502464294434,
|
| 359 |
+
"eval_runtime": 2678.1886,
|
| 360 |
+
"eval_samples_per_second": 5.018,
|
| 361 |
+
"eval_steps_per_second": 0.627,
|
| 362 |
+
"step": 440
|
| 363 |
},
|
| 364 |
{
|
| 365 |
+
"epoch": 5.01,
|
| 366 |
+
"grad_norm": 8.98901081085205,
|
| 367 |
"learning_rate": 1.984126984126984e-05,
|
| 368 |
+
"loss": 1.9553,
|
| 369 |
"step": 450
|
| 370 |
},
|
| 371 |
{
|
| 372 |
+
"epoch": 5.03,
|
| 373 |
+
"grad_norm": 9.513757705688477,
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 374 |
"learning_rate": 1.9047619047619046e-05,
|
| 375 |
+
"loss": 1.9827,
|
| 376 |
"step": 460
|
| 377 |
},
|
| 378 |
{
|
| 379 |
+
"epoch": 5.04,
|
| 380 |
+
"grad_norm": 8.948848724365234,
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 381 |
"learning_rate": 1.8253968253968254e-05,
|
| 382 |
+
"loss": 1.8577,
|
| 383 |
"step": 470
|
| 384 |
},
|
| 385 |
{
|
| 386 |
+
"epoch": 5.06,
|
| 387 |
+
"grad_norm": 10.43582534790039,
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 388 |
"learning_rate": 1.746031746031746e-05,
|
| 389 |
+
"loss": 1.8543,
|
| 390 |
"step": 480
|
| 391 |
},
|
| 392 |
{
|
| 393 |
+
"epoch": 5.07,
|
| 394 |
+
"grad_norm": 10.061915397644043,
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 395 |
"learning_rate": 1.6666666666666667e-05,
|
| 396 |
+
"loss": 2.0255,
|
| 397 |
"step": 490
|
| 398 |
},
|
| 399 |
{
|
| 400 |
+
"epoch": 5.09,
|
| 401 |
+
"grad_norm": 9.82913875579834,
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 402 |
"learning_rate": 1.5873015873015872e-05,
|
| 403 |
+
"loss": 2.0622,
|
| 404 |
"step": 500
|
| 405 |
},
|
| 406 |
{
|
| 407 |
+
"epoch": 5.1,
|
| 408 |
+
"grad_norm": 7.94363260269165,
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 409 |
"learning_rate": 1.5079365079365079e-05,
|
| 410 |
+
"loss": 1.9135,
|
| 411 |
"step": 510
|
| 412 |
},
|
| 413 |
{
|
| 414 |
+
"epoch": 5.11,
|
| 415 |
+
"grad_norm": 8.4102144241333,
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 416 |
"learning_rate": 1.4285714285714285e-05,
|
| 417 |
+
"loss": 1.9796,
|
| 418 |
"step": 520
|
| 419 |
},
|
| 420 |
{
|
| 421 |
+
"epoch": 5.13,
|
| 422 |
+
"eval_accuracy": 0.22354517041226374,
|
| 423 |
+
"eval_loss": 2.2609341144561768,
|
| 424 |
+
"eval_runtime": 2669.641,
|
| 425 |
+
"eval_samples_per_second": 5.034,
|
| 426 |
+
"eval_steps_per_second": 0.629,
|
| 427 |
+
"step": 528
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 428 |
},
|
| 429 |
{
|
| 430 |
+
"epoch": 6.0,
|
| 431 |
+
"grad_norm": 9.409048080444336,
|
| 432 |
"learning_rate": 1.3492063492063492e-05,
|
| 433 |
+
"loss": 1.7939,
|
| 434 |
"step": 530
|
| 435 |
},
|
| 436 |
{
|
| 437 |
+
"epoch": 6.02,
|
| 438 |
+
"grad_norm": 19.078195571899414,
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 439 |
"learning_rate": 1.2698412698412699e-05,
|
| 440 |
+
"loss": 1.7682,
|
| 441 |
"step": 540
|
| 442 |
},
|
| 443 |
{
|
| 444 |
+
"epoch": 6.03,
|
| 445 |
+
"grad_norm": 8.306180000305176,
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 446 |
"learning_rate": 1.1904761904761905e-05,
|
| 447 |
+
"loss": 1.8795,
|
| 448 |
"step": 550
|
| 449 |
},
|
| 450 |
{
|
| 451 |
+
"epoch": 6.05,
|
| 452 |
+
"grad_norm": 6.871161460876465,
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 453 |
"learning_rate": 1.1111111111111112e-05,
|
| 454 |
+
"loss": 1.8303,
|
| 455 |
"step": 560
|
| 456 |
},
|
| 457 |
{
|
| 458 |
+
"epoch": 6.06,
|
| 459 |
+
"grad_norm": 11.705153465270996,
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 460 |
"learning_rate": 1.0317460317460318e-05,
|
| 461 |
+
"loss": 1.7997,
|
| 462 |
"step": 570
|
| 463 |
},
|
| 464 |
{
|
| 465 |
+
"epoch": 6.07,
|
| 466 |
+
"grad_norm": 13.744942665100098,
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 467 |
"learning_rate": 9.523809523809523e-06,
|
| 468 |
+
"loss": 1.7532,
|
| 469 |
"step": 580
|
| 470 |
},
|
| 471 |
{
|
| 472 |
+
"epoch": 6.09,
|
| 473 |
+
"grad_norm": 11.410266876220703,
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 474 |
"learning_rate": 8.73015873015873e-06,
|
| 475 |
+
"loss": 1.7416,
|
| 476 |
"step": 590
|
| 477 |
},
|
| 478 |
{
|
| 479 |
+
"epoch": 6.1,
|
| 480 |
+
"grad_norm": 15.524273872375488,
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 481 |
"learning_rate": 7.936507936507936e-06,
|
| 482 |
+
"loss": 1.9096,
|
| 483 |
"step": 600
|
| 484 |
},
|
| 485 |
{
|
| 486 |
+
"epoch": 6.12,
|
| 487 |
+
"grad_norm": 9.335684776306152,
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 488 |
"learning_rate": 7.142857142857143e-06,
|
| 489 |
+
"loss": 1.8786,
|
| 490 |
"step": 610
|
| 491 |
},
|
| 492 |
{
|
| 493 |
+
"epoch": 6.13,
|
| 494 |
+
"eval_accuracy": 0.23411221908022026,
|
| 495 |
+
"eval_loss": 2.2724666595458984,
|
| 496 |
+
"eval_runtime": 2674.1892,
|
| 497 |
+
"eval_samples_per_second": 5.025,
|
| 498 |
+
"eval_steps_per_second": 0.628,
|
| 499 |
+
"step": 616
|
| 500 |
},
|
| 501 |
{
|
| 502 |
+
"epoch": 7.01,
|
| 503 |
+
"grad_norm": 12.772497177124023,
|
| 504 |
"learning_rate": 6.349206349206349e-06,
|
| 505 |
+
"loss": 1.6914,
|
| 506 |
"step": 620
|
| 507 |
},
|
| 508 |
{
|
| 509 |
+
"epoch": 7.02,
|
| 510 |
+
"grad_norm": 11.079833984375,
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 511 |
"learning_rate": 5.555555555555556e-06,
|
| 512 |
+
"loss": 1.6596,
|
| 513 |
"step": 630
|
| 514 |
},
|
| 515 |
{
|
| 516 |
+
"epoch": 7.03,
|
| 517 |
+
"grad_norm": 11.924140930175781,
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 518 |
"learning_rate": 4.7619047619047615e-06,
|
| 519 |
+
"loss": 1.7996,
|
| 520 |
"step": 640
|
| 521 |
},
|
| 522 |
{
|
| 523 |
+
"epoch": 7.05,
|
| 524 |
+
"grad_norm": 12.28601360321045,
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 525 |
"learning_rate": 3.968253968253968e-06,
|
| 526 |
+
"loss": 1.5564,
|
| 527 |
"step": 650
|
| 528 |
},
|
| 529 |
{
|
| 530 |
+
"epoch": 7.06,
|
| 531 |
+
"grad_norm": 12.644978523254395,
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 532 |
"learning_rate": 3.1746031746031746e-06,
|
| 533 |
+
"loss": 1.8793,
|
| 534 |
"step": 660
|
| 535 |
},
|
| 536 |
{
|
| 537 |
+
"epoch": 7.08,
|
| 538 |
+
"grad_norm": 11.849274635314941,
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 539 |
"learning_rate": 2.3809523809523808e-06,
|
| 540 |
+
"loss": 1.6106,
|
| 541 |
"step": 670
|
| 542 |
},
|
| 543 |
{
|
| 544 |
+
"epoch": 7.09,
|
| 545 |
+
"grad_norm": 12.6458101272583,
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 546 |
"learning_rate": 1.5873015873015873e-06,
|
| 547 |
+
"loss": 1.9571,
|
| 548 |
"step": 680
|
| 549 |
},
|
| 550 |
{
|
| 551 |
+
"epoch": 7.11,
|
| 552 |
+
"grad_norm": 12.309945106506348,
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 553 |
"learning_rate": 7.936507936507937e-07,
|
| 554 |
+
"loss": 1.6146,
|
| 555 |
"step": 690
|
| 556 |
},
|
| 557 |
{
|
| 558 |
+
"epoch": 7.12,
|
| 559 |
+
"grad_norm": 11.306781768798828,
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 560 |
"learning_rate": 0.0,
|
| 561 |
+
"loss": 1.71,
|
| 562 |
"step": 700
|
| 563 |
},
|
| 564 |
{
|
| 565 |
+
"epoch": 7.12,
|
| 566 |
+
"eval_accuracy": 0.22257776454829586,
|
| 567 |
+
"eval_loss": 2.2227871417999268,
|
| 568 |
+
"eval_runtime": 2677.8894,
|
| 569 |
+
"eval_samples_per_second": 5.018,
|
| 570 |
+
"eval_steps_per_second": 0.627,
|
| 571 |
"step": 700
|
| 572 |
},
|
| 573 |
{
|
| 574 |
+
"epoch": 7.12,
|
| 575 |
"step": 700,
|
| 576 |
+
"total_flos": 6.943814924023628e+18,
|
| 577 |
+
"train_loss": 2.1489518587929863,
|
| 578 |
+
"train_runtime": 24272.1999,
|
| 579 |
+
"train_samples_per_second": 0.231,
|
| 580 |
+
"train_steps_per_second": 0.029
|
| 581 |
+
},
|
| 582 |
+
{
|
| 583 |
+
"epoch": 7.12,
|
| 584 |
+
"eval_accuracy": 0.225,
|
| 585 |
+
"eval_loss": 2.50138521194458,
|
| 586 |
+
"eval_runtime": 955.2551,
|
| 587 |
+
"eval_samples_per_second": 5.192,
|
| 588 |
+
"eval_steps_per_second": 0.649,
|
| 589 |
"step": 700
|
| 590 |
},
|
| 591 |
{
|
| 592 |
+
"epoch": 7.12,
|
| 593 |
+
"eval_accuracy": 0.225,
|
| 594 |
+
"eval_loss": 2.501384973526001,
|
| 595 |
+
"eval_runtime": 945.2716,
|
| 596 |
+
"eval_samples_per_second": 5.247,
|
| 597 |
+
"eval_steps_per_second": 0.656,
|
| 598 |
"step": 700
|
| 599 |
}
|
| 600 |
],
|
| 601 |
+
"logging_steps": 10,
|
| 602 |
"max_steps": 700,
|
| 603 |
"num_input_tokens_seen": 0,
|
| 604 |
"num_train_epochs": 9223372036854775807,
|
| 605 |
"save_steps": 500,
|
| 606 |
+
"total_flos": 6.943814924023628e+18,
|
| 607 |
+
"train_batch_size": 8,
|
| 608 |
"trial_name": null,
|
| 609 |
"trial_params": null
|
| 610 |
}
|