Commit
•
92976fa
1
Parent(s):
d59fdc5
End of training
Browse files- all_results.json +9 -9
- eval_results.json +5 -5
- runs/Jul19_10-25-19_amy-2-gpu/events.out.tfevents.1689762522.amy-2-gpu.77175.1 +3 -0
- train_results.json +5 -5
- trainer_state.json +84 -404
all_results.json
CHANGED
|
@@ -1,11 +1,11 @@
|
|
| 1 |
{
|
| 2 |
-
"epoch":
|
| 3 |
-
"eval_loss":
|
| 4 |
-
"eval_runtime":
|
| 5 |
-
"eval_samples_per_second":
|
| 6 |
-
"eval_steps_per_second": 1.
|
| 7 |
-
"train_loss":
|
| 8 |
-
"train_runtime":
|
| 9 |
-
"train_samples_per_second":
|
| 10 |
-
"train_steps_per_second": 0.
|
| 11 |
}
|
|
|
|
| 1 |
{
|
| 2 |
+
"epoch": 10.0,
|
| 3 |
+
"eval_loss": 6395.85986328125,
|
| 4 |
+
"eval_runtime": 0.6634,
|
| 5 |
+
"eval_samples_per_second": 12.059,
|
| 6 |
+
"eval_steps_per_second": 1.507,
|
| 7 |
+
"train_loss": 6197.140625,
|
| 8 |
+
"train_runtime": 176.1588,
|
| 9 |
+
"train_samples_per_second": 1.817,
|
| 10 |
+
"train_steps_per_second": 0.227
|
| 11 |
}
|
eval_results.json
CHANGED
|
@@ -1,7 +1,7 @@
|
|
| 1 |
{
|
| 2 |
-
"epoch":
|
| 3 |
-
"eval_loss":
|
| 4 |
-
"eval_runtime":
|
| 5 |
-
"eval_samples_per_second":
|
| 6 |
-
"eval_steps_per_second": 1.
|
| 7 |
}
|
|
|
|
| 1 |
{
|
| 2 |
+
"epoch": 10.0,
|
| 3 |
+
"eval_loss": 6395.85986328125,
|
| 4 |
+
"eval_runtime": 0.6634,
|
| 5 |
+
"eval_samples_per_second": 12.059,
|
| 6 |
+
"eval_steps_per_second": 1.507
|
| 7 |
}
|
runs/Jul19_10-25-19_amy-2-gpu/events.out.tfevents.1689762522.amy-2-gpu.77175.1
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:86c2ec01e027a24a3cefc096b6ea2c3abea038efcb737c42ceeef61dc245e2e4
|
| 3 |
+
size 354
|
train_results.json
CHANGED
|
@@ -1,7 +1,7 @@
|
|
| 1 |
{
|
| 2 |
-
"epoch":
|
| 3 |
-
"train_loss":
|
| 4 |
-
"train_runtime":
|
| 5 |
-
"train_samples_per_second":
|
| 6 |
-
"train_steps_per_second": 0.
|
| 7 |
}
|
|
|
|
| 1 |
{
|
| 2 |
+
"epoch": 10.0,
|
| 3 |
+
"train_loss": 6197.140625,
|
| 4 |
+
"train_runtime": 176.1588,
|
| 5 |
+
"train_samples_per_second": 1.817,
|
| 6 |
+
"train_steps_per_second": 0.227
|
| 7 |
}
|
trainer_state.json
CHANGED
|
@@ -1,449 +1,129 @@
|
|
| 1 |
{
|
| 2 |
-
"best_metric":
|
| 3 |
-
"best_model_checkpoint": "./coco_outputs/checkpoint-
|
| 4 |
-
"epoch":
|
| 5 |
-
"global_step":
|
| 6 |
"is_hyper_param_search": false,
|
| 7 |
"is_local_process_zero": true,
|
| 8 |
"is_world_process_zero": true,
|
| 9 |
"log_history": [
|
| 10 |
-
{
|
| 11 |
-
"epoch": 0.08,
|
| 12 |
-
"learning_rate": 1.96875e-07,
|
| 13 |
-
"loss": 6346.4363,
|
| 14 |
-
"step": 10
|
| 15 |
-
},
|
| 16 |
-
{
|
| 17 |
-
"epoch": 0.16,
|
| 18 |
-
"learning_rate": 1.9375e-07,
|
| 19 |
-
"loss": 6401.3863,
|
| 20 |
-
"step": 20
|
| 21 |
-
},
|
| 22 |
-
{
|
| 23 |
-
"epoch": 0.23,
|
| 24 |
-
"learning_rate": 1.90625e-07,
|
| 25 |
-
"loss": 6297.2812,
|
| 26 |
-
"step": 30
|
| 27 |
-
},
|
| 28 |
-
{
|
| 29 |
-
"epoch": 0.31,
|
| 30 |
-
"learning_rate": 1.875e-07,
|
| 31 |
-
"loss": 6283.0867,
|
| 32 |
-
"step": 40
|
| 33 |
-
},
|
| 34 |
-
{
|
| 35 |
-
"epoch": 0.39,
|
| 36 |
-
"learning_rate": 1.8437499999999998e-07,
|
| 37 |
-
"loss": 6141.998,
|
| 38 |
-
"step": 50
|
| 39 |
-
},
|
| 40 |
-
{
|
| 41 |
-
"epoch": 0.47,
|
| 42 |
-
"learning_rate": 1.8124999999999999e-07,
|
| 43 |
-
"loss": 6092.3895,
|
| 44 |
-
"step": 60
|
| 45 |
-
},
|
| 46 |
-
{
|
| 47 |
-
"epoch": 0.55,
|
| 48 |
-
"learning_rate": 1.78125e-07,
|
| 49 |
-
"loss": 6228.5742,
|
| 50 |
-
"step": 70
|
| 51 |
-
},
|
| 52 |
-
{
|
| 53 |
-
"epoch": 0.62,
|
| 54 |
-
"learning_rate": 1.75e-07,
|
| 55 |
-
"loss": 6152.1148,
|
| 56 |
-
"step": 80
|
| 57 |
-
},
|
| 58 |
-
{
|
| 59 |
-
"epoch": 0.7,
|
| 60 |
-
"learning_rate": 1.71875e-07,
|
| 61 |
-
"loss": 6431.5375,
|
| 62 |
-
"step": 90
|
| 63 |
-
},
|
| 64 |
-
{
|
| 65 |
-
"epoch": 0.78,
|
| 66 |
-
"learning_rate": 1.6875e-07,
|
| 67 |
-
"loss": 6201.1875,
|
| 68 |
-
"step": 100
|
| 69 |
-
},
|
| 70 |
-
{
|
| 71 |
-
"epoch": 0.86,
|
| 72 |
-
"learning_rate": 1.65625e-07,
|
| 73 |
-
"loss": 6414.7582,
|
| 74 |
-
"step": 110
|
| 75 |
-
},
|
| 76 |
-
{
|
| 77 |
-
"epoch": 0.94,
|
| 78 |
-
"learning_rate": 1.6249999999999998e-07,
|
| 79 |
-
"loss": 6463.6664,
|
| 80 |
-
"step": 120
|
| 81 |
-
},
|
| 82 |
{
|
| 83 |
"epoch": 1.0,
|
| 84 |
-
"eval_loss":
|
| 85 |
-
"eval_runtime":
|
| 86 |
-
"eval_samples_per_second": 14.
|
| 87 |
-
"eval_steps_per_second": 1.
|
| 88 |
-
"step":
|
| 89 |
-
},
|
| 90 |
-
{
|
| 91 |
-
"epoch": 1.02,
|
| 92 |
-
"learning_rate": 1.5937499999999998e-07,
|
| 93 |
-
"loss": 6203.2496,
|
| 94 |
-
"step": 130
|
| 95 |
-
},
|
| 96 |
-
{
|
| 97 |
-
"epoch": 1.09,
|
| 98 |
-
"learning_rate": 1.5624999999999999e-07,
|
| 99 |
-
"loss": 6195.1863,
|
| 100 |
-
"step": 140
|
| 101 |
-
},
|
| 102 |
-
{
|
| 103 |
-
"epoch": 1.17,
|
| 104 |
-
"learning_rate": 1.53125e-07,
|
| 105 |
-
"loss": 6365.1023,
|
| 106 |
-
"step": 150
|
| 107 |
-
},
|
| 108 |
-
{
|
| 109 |
-
"epoch": 1.25,
|
| 110 |
-
"learning_rate": 1.5e-07,
|
| 111 |
-
"loss": 6115.4832,
|
| 112 |
-
"step": 160
|
| 113 |
-
},
|
| 114 |
-
{
|
| 115 |
-
"epoch": 1.33,
|
| 116 |
-
"learning_rate": 1.46875e-07,
|
| 117 |
-
"loss": 6507.116,
|
| 118 |
-
"step": 170
|
| 119 |
-
},
|
| 120 |
-
{
|
| 121 |
-
"epoch": 1.41,
|
| 122 |
-
"learning_rate": 1.4375e-07,
|
| 123 |
-
"loss": 6235.0008,
|
| 124 |
-
"step": 180
|
| 125 |
-
},
|
| 126 |
-
{
|
| 127 |
-
"epoch": 1.48,
|
| 128 |
-
"learning_rate": 1.40625e-07,
|
| 129 |
-
"loss": 6424.7523,
|
| 130 |
-
"step": 190
|
| 131 |
-
},
|
| 132 |
-
{
|
| 133 |
-
"epoch": 1.56,
|
| 134 |
-
"learning_rate": 1.375e-07,
|
| 135 |
-
"loss": 6032.757,
|
| 136 |
-
"step": 200
|
| 137 |
-
},
|
| 138 |
-
{
|
| 139 |
-
"epoch": 1.64,
|
| 140 |
-
"learning_rate": 1.3437499999999998e-07,
|
| 141 |
-
"loss": 5972.9164,
|
| 142 |
-
"step": 210
|
| 143 |
-
},
|
| 144 |
-
{
|
| 145 |
-
"epoch": 1.72,
|
| 146 |
-
"learning_rate": 1.3125e-07,
|
| 147 |
-
"loss": 5884.6973,
|
| 148 |
-
"step": 220
|
| 149 |
-
},
|
| 150 |
-
{
|
| 151 |
-
"epoch": 1.8,
|
| 152 |
-
"learning_rate": 1.28125e-07,
|
| 153 |
-
"loss": 6157.3324,
|
| 154 |
-
"step": 230
|
| 155 |
-
},
|
| 156 |
-
{
|
| 157 |
-
"epoch": 1.88,
|
| 158 |
-
"learning_rate": 1.25e-07,
|
| 159 |
-
"loss": 6338.2691,
|
| 160 |
-
"step": 240
|
| 161 |
-
},
|
| 162 |
-
{
|
| 163 |
-
"epoch": 1.95,
|
| 164 |
-
"learning_rate": 1.21875e-07,
|
| 165 |
-
"loss": 6242.4801,
|
| 166 |
-
"step": 250
|
| 167 |
},
|
| 168 |
{
|
| 169 |
"epoch": 2.0,
|
| 170 |
-
"eval_loss":
|
| 171 |
-
"eval_runtime":
|
| 172 |
-
"eval_samples_per_second": 14.
|
| 173 |
-
"eval_steps_per_second": 1.
|
| 174 |
-
"step":
|
| 175 |
-
},
|
| 176 |
-
{
|
| 177 |
-
"epoch": 2.03,
|
| 178 |
-
"learning_rate": 1.1875e-07,
|
| 179 |
-
"loss": 6287.9883,
|
| 180 |
-
"step": 260
|
| 181 |
-
},
|
| 182 |
-
{
|
| 183 |
-
"epoch": 2.11,
|
| 184 |
-
"learning_rate": 1.1562499999999999e-07,
|
| 185 |
-
"loss": 6261.4707,
|
| 186 |
-
"step": 270
|
| 187 |
-
},
|
| 188 |
-
{
|
| 189 |
-
"epoch": 2.19,
|
| 190 |
-
"learning_rate": 1.125e-07,
|
| 191 |
-
"loss": 6337.7199,
|
| 192 |
-
"step": 280
|
| 193 |
-
},
|
| 194 |
-
{
|
| 195 |
-
"epoch": 2.27,
|
| 196 |
-
"learning_rate": 1.09375e-07,
|
| 197 |
-
"loss": 6193.377,
|
| 198 |
-
"step": 290
|
| 199 |
-
},
|
| 200 |
-
{
|
| 201 |
-
"epoch": 2.34,
|
| 202 |
-
"learning_rate": 1.0624999999999999e-07,
|
| 203 |
-
"loss": 5926.5789,
|
| 204 |
-
"step": 300
|
| 205 |
-
},
|
| 206 |
-
{
|
| 207 |
-
"epoch": 2.42,
|
| 208 |
-
"learning_rate": 1.0312499999999999e-07,
|
| 209 |
-
"loss": 6122.5051,
|
| 210 |
-
"step": 310
|
| 211 |
},
|
| 212 |
{
|
| 213 |
"epoch": 2.5,
|
| 214 |
-
"learning_rate":
|
| 215 |
-
"loss":
|
| 216 |
-
"step":
|
| 217 |
-
},
|
| 218 |
-
{
|
| 219 |
-
"epoch": 2.58,
|
| 220 |
-
"learning_rate": 9.6875e-08,
|
| 221 |
-
"loss": 6162.8938,
|
| 222 |
-
"step": 330
|
| 223 |
-
},
|
| 224 |
-
{
|
| 225 |
-
"epoch": 2.66,
|
| 226 |
-
"learning_rate": 9.375e-08,
|
| 227 |
-
"loss": 6154.5023,
|
| 228 |
-
"step": 340
|
| 229 |
-
},
|
| 230 |
-
{
|
| 231 |
-
"epoch": 2.73,
|
| 232 |
-
"learning_rate": 9.062499999999999e-08,
|
| 233 |
-
"loss": 6107.1387,
|
| 234 |
-
"step": 350
|
| 235 |
-
},
|
| 236 |
-
{
|
| 237 |
-
"epoch": 2.81,
|
| 238 |
-
"learning_rate": 8.75e-08,
|
| 239 |
-
"loss": 6092.2199,
|
| 240 |
-
"step": 360
|
| 241 |
-
},
|
| 242 |
-
{
|
| 243 |
-
"epoch": 2.89,
|
| 244 |
-
"learning_rate": 8.4375e-08,
|
| 245 |
-
"loss": 6291.2375,
|
| 246 |
-
"step": 370
|
| 247 |
-
},
|
| 248 |
-
{
|
| 249 |
-
"epoch": 2.97,
|
| 250 |
-
"learning_rate": 8.124999999999999e-08,
|
| 251 |
-
"loss": 6344.1949,
|
| 252 |
-
"step": 380
|
| 253 |
},
|
| 254 |
{
|
| 255 |
"epoch": 3.0,
|
| 256 |
-
"eval_loss":
|
| 257 |
-
"eval_runtime":
|
| 258 |
-
"eval_samples_per_second": 14.
|
| 259 |
-
"eval_steps_per_second": 1.
|
| 260 |
-
"step":
|
| 261 |
-
},
|
| 262 |
-
{
|
| 263 |
-
"epoch": 3.05,
|
| 264 |
-
"learning_rate": 7.812499999999999e-08,
|
| 265 |
-
"loss": 6475.0023,
|
| 266 |
-
"step": 390
|
| 267 |
-
},
|
| 268 |
-
{
|
| 269 |
-
"epoch": 3.12,
|
| 270 |
-
"learning_rate": 7.5e-08,
|
| 271 |
-
"loss": 6335.7672,
|
| 272 |
-
"step": 400
|
| 273 |
-
},
|
| 274 |
-
{
|
| 275 |
-
"epoch": 3.2,
|
| 276 |
-
"learning_rate": 7.1875e-08,
|
| 277 |
-
"loss": 6222.3957,
|
| 278 |
-
"step": 410
|
| 279 |
-
},
|
| 280 |
-
{
|
| 281 |
-
"epoch": 3.28,
|
| 282 |
-
"learning_rate": 6.875e-08,
|
| 283 |
-
"loss": 6346.1539,
|
| 284 |
-
"step": 420
|
| 285 |
-
},
|
| 286 |
-
{
|
| 287 |
-
"epoch": 3.36,
|
| 288 |
-
"learning_rate": 6.5625e-08,
|
| 289 |
-
"loss": 6196.7348,
|
| 290 |
-
"step": 430
|
| 291 |
-
},
|
| 292 |
-
{
|
| 293 |
-
"epoch": 3.44,
|
| 294 |
-
"learning_rate": 6.25e-08,
|
| 295 |
-
"loss": 6440.0867,
|
| 296 |
-
"step": 440
|
| 297 |
-
},
|
| 298 |
-
{
|
| 299 |
-
"epoch": 3.52,
|
| 300 |
-
"learning_rate": 5.9375e-08,
|
| 301 |
-
"loss": 6186.3539,
|
| 302 |
-
"step": 450
|
| 303 |
-
},
|
| 304 |
-
{
|
| 305 |
-
"epoch": 3.59,
|
| 306 |
-
"learning_rate": 5.625e-08,
|
| 307 |
-
"loss": 6364.8961,
|
| 308 |
-
"step": 460
|
| 309 |
-
},
|
| 310 |
-
{
|
| 311 |
-
"epoch": 3.67,
|
| 312 |
-
"learning_rate": 5.3124999999999994e-08,
|
| 313 |
-
"loss": 6580.9484,
|
| 314 |
-
"step": 470
|
| 315 |
-
},
|
| 316 |
-
{
|
| 317 |
-
"epoch": 3.75,
|
| 318 |
-
"learning_rate": 5e-08,
|
| 319 |
-
"loss": 6384.6832,
|
| 320 |
-
"step": 480
|
| 321 |
-
},
|
| 322 |
-
{
|
| 323 |
-
"epoch": 3.83,
|
| 324 |
-
"learning_rate": 4.6875e-08,
|
| 325 |
-
"loss": 6195.5973,
|
| 326 |
-
"step": 490
|
| 327 |
-
},
|
| 328 |
-
{
|
| 329 |
-
"epoch": 3.91,
|
| 330 |
-
"learning_rate": 4.375e-08,
|
| 331 |
-
"loss": 6409.9172,
|
| 332 |
-
"step": 500
|
| 333 |
-
},
|
| 334 |
-
{
|
| 335 |
-
"epoch": 3.98,
|
| 336 |
-
"learning_rate": 4.0624999999999995e-08,
|
| 337 |
-
"loss": 6528.6664,
|
| 338 |
-
"step": 510
|
| 339 |
},
|
| 340 |
{
|
| 341 |
"epoch": 4.0,
|
| 342 |
-
"eval_loss":
|
| 343 |
-
"eval_runtime":
|
| 344 |
-
"eval_samples_per_second": 14.
|
| 345 |
-
"eval_steps_per_second": 1.
|
| 346 |
-
"step":
|
| 347 |
-
},
|
| 348 |
-
{
|
| 349 |
-
"epoch": 4.06,
|
| 350 |
-
"learning_rate": 3.75e-08,
|
| 351 |
-
"loss": 6024.7227,
|
| 352 |
-
"step": 520
|
| 353 |
-
},
|
| 354 |
-
{
|
| 355 |
-
"epoch": 4.14,
|
| 356 |
-
"learning_rate": 3.4375e-08,
|
| 357 |
-
"loss": 5942.5199,
|
| 358 |
-
"step": 530
|
| 359 |
},
|
| 360 |
{
|
| 361 |
-
"epoch":
|
| 362 |
-
"learning_rate":
|
| 363 |
-
"loss":
|
| 364 |
-
"step":
|
| 365 |
-
},
|
| 366 |
-
{
|
| 367 |
-
"epoch": 4.3,
|
| 368 |
-
"learning_rate": 2.8125e-08,
|
| 369 |
-
"loss": 6593.8445,
|
| 370 |
-
"step": 550
|
| 371 |
-
},
|
| 372 |
-
{
|
| 373 |
-
"epoch": 4.38,
|
| 374 |
-
"learning_rate": 2.5e-08,
|
| 375 |
-
"loss": 6542.1891,
|
| 376 |
-
"step": 560
|
| 377 |
-
},
|
| 378 |
-
{
|
| 379 |
-
"epoch": 4.45,
|
| 380 |
-
"learning_rate": 2.1875e-08,
|
| 381 |
-
"loss": 6388.3813,
|
| 382 |
-
"step": 570
|
| 383 |
},
|
| 384 |
{
|
| 385 |
-
"epoch":
|
| 386 |
-
"
|
| 387 |
-
"
|
| 388 |
-
"
|
|
|
|
|
|
|
| 389 |
},
|
| 390 |
{
|
| 391 |
-
"epoch":
|
| 392 |
-
"
|
| 393 |
-
"
|
| 394 |
-
"
|
|
|
|
|
|
|
| 395 |
},
|
| 396 |
{
|
| 397 |
-
"epoch":
|
| 398 |
-
"
|
| 399 |
-
"
|
| 400 |
-
"
|
|
|
|
|
|
|
| 401 |
},
|
| 402 |
{
|
| 403 |
-
"epoch":
|
| 404 |
-
"learning_rate":
|
| 405 |
-
"loss":
|
| 406 |
-
"step":
|
| 407 |
},
|
| 408 |
{
|
| 409 |
-
"epoch":
|
| 410 |
-
"
|
| 411 |
-
"
|
| 412 |
-
"
|
|
|
|
|
|
|
| 413 |
},
|
| 414 |
{
|
| 415 |
-
"epoch":
|
| 416 |
-
"
|
| 417 |
-
"
|
| 418 |
-
"
|
|
|
|
|
|
|
| 419 |
},
|
| 420 |
{
|
| 421 |
-
"epoch":
|
| 422 |
"learning_rate": 0.0,
|
| 423 |
-
"loss":
|
| 424 |
-
"step":
|
| 425 |
},
|
| 426 |
{
|
| 427 |
-
"epoch":
|
| 428 |
-
"eval_loss":
|
| 429 |
-
"eval_runtime":
|
| 430 |
-
"eval_samples_per_second": 14.
|
| 431 |
-
"eval_steps_per_second": 1.
|
| 432 |
-
"step":
|
| 433 |
},
|
| 434 |
{
|
| 435 |
-
"epoch":
|
| 436 |
-
"step":
|
| 437 |
-
"total_flos":
|
| 438 |
-
"train_loss":
|
| 439 |
-
"train_runtime":
|
| 440 |
-
"train_samples_per_second":
|
| 441 |
-
"train_steps_per_second": 0.
|
| 442 |
}
|
| 443 |
],
|
| 444 |
-
"max_steps":
|
| 445 |
-
"num_train_epochs":
|
| 446 |
-
"total_flos":
|
| 447 |
"trial_name": null,
|
| 448 |
"trial_params": null
|
| 449 |
}
|
|
|
|
| 1 |
{
|
| 2 |
+
"best_metric": 6395.85986328125,
|
| 3 |
+
"best_model_checkpoint": "./coco_outputs/checkpoint-40",
|
| 4 |
+
"epoch": 10.0,
|
| 5 |
+
"global_step": 40,
|
| 6 |
"is_hyper_param_search": false,
|
| 7 |
"is_local_process_zero": true,
|
| 8 |
"is_world_process_zero": true,
|
| 9 |
"log_history": [
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 10 |
{
|
| 11 |
"epoch": 1.0,
|
| 12 |
+
"eval_loss": 6398.54345703125,
|
| 13 |
+
"eval_runtime": 0.5491,
|
| 14 |
+
"eval_samples_per_second": 14.569,
|
| 15 |
+
"eval_steps_per_second": 1.821,
|
| 16 |
+
"step": 4
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 17 |
},
|
| 18 |
{
|
| 19 |
"epoch": 2.0,
|
| 20 |
+
"eval_loss": 6397.9765625,
|
| 21 |
+
"eval_runtime": 0.5423,
|
| 22 |
+
"eval_samples_per_second": 14.751,
|
| 23 |
+
"eval_steps_per_second": 1.844,
|
| 24 |
+
"step": 8
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 25 |
},
|
| 26 |
{
|
| 27 |
"epoch": 2.5,
|
| 28 |
+
"learning_rate": 1.5000000000000002e-05,
|
| 29 |
+
"loss": 6175.8828,
|
| 30 |
+
"step": 10
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 31 |
},
|
| 32 |
{
|
| 33 |
"epoch": 3.0,
|
| 34 |
+
"eval_loss": 6397.42529296875,
|
| 35 |
+
"eval_runtime": 0.5517,
|
| 36 |
+
"eval_samples_per_second": 14.501,
|
| 37 |
+
"eval_steps_per_second": 1.813,
|
| 38 |
+
"step": 12
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 39 |
},
|
| 40 |
{
|
| 41 |
"epoch": 4.0,
|
| 42 |
+
"eval_loss": 6396.962890625,
|
| 43 |
+
"eval_runtime": 0.555,
|
| 44 |
+
"eval_samples_per_second": 14.415,
|
| 45 |
+
"eval_steps_per_second": 1.802,
|
| 46 |
+
"step": 16
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 47 |
},
|
| 48 |
{
|
| 49 |
+
"epoch": 5.0,
|
| 50 |
+
"learning_rate": 1e-05,
|
| 51 |
+
"loss": 6292.3258,
|
| 52 |
+
"step": 20
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 53 |
},
|
| 54 |
{
|
| 55 |
+
"epoch": 5.0,
|
| 56 |
+
"eval_loss": 6396.57177734375,
|
| 57 |
+
"eval_runtime": 0.5448,
|
| 58 |
+
"eval_samples_per_second": 14.685,
|
| 59 |
+
"eval_steps_per_second": 1.836,
|
| 60 |
+
"step": 20
|
| 61 |
},
|
| 62 |
{
|
| 63 |
+
"epoch": 6.0,
|
| 64 |
+
"eval_loss": 6396.27880859375,
|
| 65 |
+
"eval_runtime": 0.5466,
|
| 66 |
+
"eval_samples_per_second": 14.636,
|
| 67 |
+
"eval_steps_per_second": 1.829,
|
| 68 |
+
"step": 24
|
| 69 |
},
|
| 70 |
{
|
| 71 |
+
"epoch": 7.0,
|
| 72 |
+
"eval_loss": 6396.07666015625,
|
| 73 |
+
"eval_runtime": 0.5439,
|
| 74 |
+
"eval_samples_per_second": 14.708,
|
| 75 |
+
"eval_steps_per_second": 1.839,
|
| 76 |
+
"step": 28
|
| 77 |
},
|
| 78 |
{
|
| 79 |
+
"epoch": 7.5,
|
| 80 |
+
"learning_rate": 5e-06,
|
| 81 |
+
"loss": 6230.4844,
|
| 82 |
+
"step": 30
|
| 83 |
},
|
| 84 |
{
|
| 85 |
+
"epoch": 8.0,
|
| 86 |
+
"eval_loss": 6395.9599609375,
|
| 87 |
+
"eval_runtime": 0.5536,
|
| 88 |
+
"eval_samples_per_second": 14.451,
|
| 89 |
+
"eval_steps_per_second": 1.806,
|
| 90 |
+
"step": 32
|
| 91 |
},
|
| 92 |
{
|
| 93 |
+
"epoch": 9.0,
|
| 94 |
+
"eval_loss": 6395.8671875,
|
| 95 |
+
"eval_runtime": 0.5708,
|
| 96 |
+
"eval_samples_per_second": 14.015,
|
| 97 |
+
"eval_steps_per_second": 1.752,
|
| 98 |
+
"step": 36
|
| 99 |
},
|
| 100 |
{
|
| 101 |
+
"epoch": 10.0,
|
| 102 |
"learning_rate": 0.0,
|
| 103 |
+
"loss": 6089.8695,
|
| 104 |
+
"step": 40
|
| 105 |
},
|
| 106 |
{
|
| 107 |
+
"epoch": 10.0,
|
| 108 |
+
"eval_loss": 6395.85986328125,
|
| 109 |
+
"eval_runtime": 0.5496,
|
| 110 |
+
"eval_samples_per_second": 14.557,
|
| 111 |
+
"eval_steps_per_second": 1.82,
|
| 112 |
+
"step": 40
|
| 113 |
},
|
| 114 |
{
|
| 115 |
+
"epoch": 10.0,
|
| 116 |
+
"step": 40,
|
| 117 |
+
"total_flos": 1.530004230144e+17,
|
| 118 |
+
"train_loss": 6197.140625,
|
| 119 |
+
"train_runtime": 176.1588,
|
| 120 |
+
"train_samples_per_second": 1.817,
|
| 121 |
+
"train_steps_per_second": 0.227
|
| 122 |
}
|
| 123 |
],
|
| 124 |
+
"max_steps": 40,
|
| 125 |
+
"num_train_epochs": 10,
|
| 126 |
+
"total_flos": 1.530004230144e+17,
|
| 127 |
"trial_name": null,
|
| 128 |
"trial_params": null
|
| 129 |
}
|