Commit
•
cf0ef27
1
Parent(s):
f6b2ead
End of training
Browse files- all_results.json +9 -9
- eval_nbest_predictions.json +2 -2
- eval_predictions.json +0 -0
- eval_results.json +4 -4
- runs/Aug11_08-26-39_1be7fe59a580/events.out.tfevents.1660225523.1be7fe59a580.532.2 +3 -0
- train_results.json +6 -6
- trainer_state.json +107 -263
all_results.json
CHANGED
|
@@ -1,11 +1,11 @@
|
|
| 1 |
{
|
| 2 |
-
"epoch":
|
| 3 |
-
"eval_exact_match":
|
| 4 |
-
"eval_f1":
|
| 5 |
-
"eval_samples":
|
| 6 |
-
"train_loss":
|
| 7 |
-
"train_runtime":
|
| 8 |
-
"train_samples":
|
| 9 |
-
"train_samples_per_second":
|
| 10 |
-
"train_steps_per_second":
|
| 11 |
}
|
|
|
|
| 1 |
{
|
| 2 |
+
"epoch": 1.0,
|
| 3 |
+
"eval_exact_match": 83.75591296121098,
|
| 4 |
+
"eval_f1": 90.51169124706182,
|
| 5 |
+
"eval_samples": 12097,
|
| 6 |
+
"train_loss": 1.0714931627842768,
|
| 7 |
+
"train_runtime": 16914.7753,
|
| 8 |
+
"train_samples": 97408,
|
| 9 |
+
"train_samples_per_second": 5.759,
|
| 10 |
+
"train_steps_per_second": 0.96
|
| 11 |
}
|
eval_nbest_predictions.json
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 sha256:1e6a5c147e8344efe83d2f97e842013d36cf4e74e6b41928b3d814982f0ad31e
|
| 3 |
+
size 49605151
|
eval_predictions.json
CHANGED
|
The diff for this file is too large to render.
See raw diff
|
|
|
eval_results.json
CHANGED
|
@@ -1,6 +1,6 @@
|
|
| 1 |
{
|
| 2 |
-
"epoch":
|
| 3 |
-
"eval_exact_match":
|
| 4 |
-
"eval_f1":
|
| 5 |
-
"eval_samples":
|
| 6 |
}
|
|
|
|
| 1 |
{
|
| 2 |
+
"epoch": 1.0,
|
| 3 |
+
"eval_exact_match": 83.75591296121098,
|
| 4 |
+
"eval_f1": 90.51169124706182,
|
| 5 |
+
"eval_samples": 12097
|
| 6 |
}
|
runs/Aug11_08-26-39_1be7fe59a580/events.out.tfevents.1660225523.1be7fe59a580.532.2
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:738f7e50bbbc36c527f40ba4ea3acba0fd5d2f3c068dde75c6c202ca2326a021
|
| 3 |
+
size 191
|
train_results.json
CHANGED
|
@@ -1,8 +1,8 @@
|
|
| 1 |
{
|
| 2 |
-
"epoch":
|
| 3 |
-
"train_loss":
|
| 4 |
-
"train_runtime":
|
| 5 |
-
"train_samples":
|
| 6 |
-
"train_samples_per_second":
|
| 7 |
-
"train_steps_per_second":
|
| 8 |
}
|
|
|
|
| 1 |
{
|
| 2 |
+
"epoch": 1.0,
|
| 3 |
+
"train_loss": 1.0714931627842768,
|
| 4 |
+
"train_runtime": 16914.7753,
|
| 5 |
+
"train_samples": 97408,
|
| 6 |
+
"train_samples_per_second": 5.759,
|
| 7 |
+
"train_steps_per_second": 0.96
|
| 8 |
}
|
trainer_state.json
CHANGED
|
@@ -1,373 +1,217 @@
|
|
| 1 |
{
|
| 2 |
"best_metric": null,
|
| 3 |
"best_model_checkpoint": null,
|
| 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": 0.03,
|
| 12 |
-
"learning_rate": 2.
|
| 13 |
-
"loss": 2.
|
| 14 |
"step": 500
|
| 15 |
},
|
| 16 |
{
|
| 17 |
-
"epoch": 0.
|
| 18 |
-
"learning_rate": 2.
|
| 19 |
-
"loss": 1.
|
| 20 |
"step": 1000
|
| 21 |
},
|
| 22 |
{
|
| 23 |
-
"epoch": 0.
|
| 24 |
-
"learning_rate": 2.
|
| 25 |
-
"loss": 1.
|
| 26 |
"step": 1500
|
| 27 |
},
|
| 28 |
{
|
| 29 |
-
"epoch": 0.
|
| 30 |
-
"learning_rate": 2.
|
| 31 |
-
"loss": 1.
|
| 32 |
"step": 2000
|
| 33 |
},
|
| 34 |
{
|
| 35 |
-
"epoch": 0.
|
| 36 |
-
"learning_rate": 2.
|
| 37 |
-
"loss": 1.
|
| 38 |
"step": 2500
|
| 39 |
},
|
| 40 |
{
|
| 41 |
-
"epoch": 0.
|
| 42 |
-
"learning_rate": 2.
|
| 43 |
-
"loss": 1.
|
| 44 |
"step": 3000
|
| 45 |
},
|
| 46 |
{
|
| 47 |
-
"epoch": 0.
|
| 48 |
-
"learning_rate": 2.
|
| 49 |
-
"loss": 1.
|
| 50 |
"step": 3500
|
| 51 |
},
|
| 52 |
{
|
| 53 |
-
"epoch": 0.
|
| 54 |
-
"learning_rate": 2.
|
| 55 |
-
"loss": 1.
|
| 56 |
"step": 4000
|
| 57 |
},
|
| 58 |
{
|
| 59 |
-
"epoch": 0.
|
| 60 |
-
"learning_rate": 2.
|
| 61 |
-
"loss": 1.
|
| 62 |
"step": 4500
|
| 63 |
},
|
| 64 |
{
|
| 65 |
-
"epoch": 0.
|
| 66 |
-
"learning_rate": 2.
|
| 67 |
-
"loss": 1.
|
| 68 |
"step": 5000
|
| 69 |
},
|
| 70 |
{
|
| 71 |
-
"epoch": 0.
|
| 72 |
-
"learning_rate":
|
| 73 |
-
"loss": 1.
|
| 74 |
"step": 5500
|
| 75 |
},
|
| 76 |
{
|
| 77 |
-
"epoch": 0.
|
| 78 |
-
"learning_rate":
|
| 79 |
-
"loss": 1.
|
| 80 |
"step": 6000
|
| 81 |
},
|
| 82 |
{
|
| 83 |
-
"epoch": 0.
|
| 84 |
-
"learning_rate":
|
| 85 |
-
"loss": 1.
|
| 86 |
"step": 6500
|
| 87 |
},
|
| 88 |
{
|
| 89 |
-
"epoch": 0.
|
| 90 |
-
"learning_rate":
|
| 91 |
-
"loss": 1.
|
| 92 |
"step": 7000
|
| 93 |
},
|
| 94 |
{
|
| 95 |
-
"epoch": 0.
|
| 96 |
-
"learning_rate":
|
| 97 |
-
"loss":
|
| 98 |
"step": 7500
|
| 99 |
},
|
| 100 |
{
|
| 101 |
-
"epoch": 0.
|
| 102 |
-
"learning_rate":
|
| 103 |
-
"loss":
|
| 104 |
"step": 8000
|
| 105 |
},
|
| 106 |
{
|
| 107 |
-
"epoch": 0.
|
| 108 |
-
"learning_rate":
|
| 109 |
-
"loss": 1.
|
| 110 |
"step": 8500
|
| 111 |
},
|
| 112 |
{
|
| 113 |
-
"epoch": 0.
|
| 114 |
-
"learning_rate":
|
| 115 |
-
"loss":
|
| 116 |
"step": 9000
|
| 117 |
},
|
| 118 |
{
|
| 119 |
-
"epoch": 0.
|
| 120 |
-
"learning_rate":
|
| 121 |
-
"loss":
|
| 122 |
"step": 9500
|
| 123 |
},
|
| 124 |
{
|
| 125 |
-
"epoch": 0.
|
| 126 |
-
"learning_rate": 1.
|
| 127 |
-
"loss":
|
| 128 |
"step": 10000
|
| 129 |
},
|
| 130 |
{
|
| 131 |
-
"epoch": 0.
|
| 132 |
-
"learning_rate": 1.
|
| 133 |
-
"loss":
|
| 134 |
"step": 10500
|
| 135 |
},
|
| 136 |
{
|
| 137 |
-
"epoch": 0.
|
| 138 |
-
"learning_rate":
|
| 139 |
-
"loss":
|
| 140 |
"step": 11000
|
| 141 |
},
|
| 142 |
{
|
| 143 |
-
"epoch": 0.
|
| 144 |
-
"learning_rate":
|
| 145 |
-
"loss":
|
| 146 |
"step": 11500
|
| 147 |
},
|
| 148 |
{
|
| 149 |
-
"epoch": 0.
|
| 150 |
-
"learning_rate":
|
| 151 |
-
"loss":
|
| 152 |
"step": 12000
|
| 153 |
},
|
| 154 |
{
|
| 155 |
-
"epoch": 0.
|
| 156 |
-
"learning_rate":
|
| 157 |
-
"loss":
|
| 158 |
"step": 12500
|
| 159 |
},
|
| 160 |
{
|
| 161 |
-
"epoch": 0.
|
| 162 |
-
"learning_rate":
|
| 163 |
-
"loss":
|
| 164 |
"step": 13000
|
| 165 |
},
|
| 166 |
{
|
| 167 |
-
"epoch": 0.
|
| 168 |
-
"learning_rate":
|
| 169 |
-
"loss":
|
| 170 |
"step": 13500
|
| 171 |
},
|
| 172 |
{
|
| 173 |
-
"epoch": 0.
|
| 174 |
-
"learning_rate":
|
| 175 |
-
"loss":
|
| 176 |
"step": 14000
|
| 177 |
},
|
| 178 |
{
|
| 179 |
-
"epoch": 0.
|
| 180 |
-
"learning_rate":
|
| 181 |
-
"loss":
|
| 182 |
"step": 14500
|
| 183 |
},
|
| 184 |
{
|
| 185 |
-
"epoch":
|
| 186 |
-
"learning_rate":
|
| 187 |
-
"loss": 0.
|
| 188 |
"step": 15000
|
| 189 |
},
|
| 190 |
{
|
| 191 |
-
"epoch":
|
| 192 |
-
"learning_rate": 1.
|
| 193 |
-
"loss": 0.
|
| 194 |
"step": 15500
|
| 195 |
},
|
| 196 |
{
|
| 197 |
-
"epoch":
|
| 198 |
-
"learning_rate":
|
| 199 |
-
"loss": 0.
|
| 200 |
"step": 16000
|
| 201 |
},
|
| 202 |
{
|
| 203 |
-
"epoch": 1.
|
| 204 |
-
"
|
| 205 |
-
"
|
| 206 |
-
"
|
| 207 |
-
|
| 208 |
-
|
| 209 |
-
"
|
| 210 |
-
"learning_rate": 1.255694643956495e-05,
|
| 211 |
-
"loss": 0.6894,
|
| 212 |
-
"step": 17000
|
| 213 |
-
},
|
| 214 |
-
{
|
| 215 |
-
"epoch": 1.2,
|
| 216 |
-
"learning_rate": 1.2043915452493331e-05,
|
| 217 |
-
"loss": 0.6848,
|
| 218 |
-
"step": 17500
|
| 219 |
-
},
|
| 220 |
-
{
|
| 221 |
-
"epoch": 1.23,
|
| 222 |
-
"learning_rate": 1.1530884465421712e-05,
|
| 223 |
-
"loss": 0.7217,
|
| 224 |
-
"step": 18000
|
| 225 |
-
},
|
| 226 |
-
{
|
| 227 |
-
"epoch": 1.27,
|
| 228 |
-
"learning_rate": 1.1017853478350093e-05,
|
| 229 |
-
"loss": 0.7294,
|
| 230 |
-
"step": 18500
|
| 231 |
-
},
|
| 232 |
-
{
|
| 233 |
-
"epoch": 1.3,
|
| 234 |
-
"learning_rate": 1.0504822491278474e-05,
|
| 235 |
-
"loss": 0.7065,
|
| 236 |
-
"step": 19000
|
| 237 |
-
},
|
| 238 |
-
{
|
| 239 |
-
"epoch": 1.33,
|
| 240 |
-
"learning_rate": 9.991791504206853e-06,
|
| 241 |
-
"loss": 0.704,
|
| 242 |
-
"step": 19500
|
| 243 |
-
},
|
| 244 |
-
{
|
| 245 |
-
"epoch": 1.37,
|
| 246 |
-
"learning_rate": 9.478760517135234e-06,
|
| 247 |
-
"loss": 0.7225,
|
| 248 |
-
"step": 20000
|
| 249 |
-
},
|
| 250 |
-
{
|
| 251 |
-
"epoch": 1.4,
|
| 252 |
-
"learning_rate": 8.965729530063615e-06,
|
| 253 |
-
"loss": 0.6821,
|
| 254 |
-
"step": 20500
|
| 255 |
-
},
|
| 256 |
-
{
|
| 257 |
-
"epoch": 1.44,
|
| 258 |
-
"learning_rate": 8.452698542991998e-06,
|
| 259 |
-
"loss": 0.6722,
|
| 260 |
-
"step": 21000
|
| 261 |
-
},
|
| 262 |
-
{
|
| 263 |
-
"epoch": 1.47,
|
| 264 |
-
"learning_rate": 7.939667555920379e-06,
|
| 265 |
-
"loss": 0.7089,
|
| 266 |
-
"step": 21500
|
| 267 |
-
},
|
| 268 |
-
{
|
| 269 |
-
"epoch": 1.5,
|
| 270 |
-
"learning_rate": 7.426636568848759e-06,
|
| 271 |
-
"loss": 0.71,
|
| 272 |
-
"step": 22000
|
| 273 |
-
},
|
| 274 |
-
{
|
| 275 |
-
"epoch": 1.54,
|
| 276 |
-
"learning_rate": 6.913605581777139e-06,
|
| 277 |
-
"loss": 0.7121,
|
| 278 |
-
"step": 22500
|
| 279 |
-
},
|
| 280 |
-
{
|
| 281 |
-
"epoch": 1.57,
|
| 282 |
-
"learning_rate": 6.40057459470552e-06,
|
| 283 |
-
"loss": 0.7063,
|
| 284 |
-
"step": 23000
|
| 285 |
-
},
|
| 286 |
-
{
|
| 287 |
-
"epoch": 1.61,
|
| 288 |
-
"learning_rate": 5.887543607633901e-06,
|
| 289 |
-
"loss": 0.7064,
|
| 290 |
-
"step": 23500
|
| 291 |
-
},
|
| 292 |
-
{
|
| 293 |
-
"epoch": 1.64,
|
| 294 |
-
"learning_rate": 5.374512620562282e-06,
|
| 295 |
-
"loss": 0.7128,
|
| 296 |
-
"step": 24000
|
| 297 |
-
},
|
| 298 |
-
{
|
| 299 |
-
"epoch": 1.68,
|
| 300 |
-
"learning_rate": 4.861481633490662e-06,
|
| 301 |
-
"loss": 0.6573,
|
| 302 |
-
"step": 24500
|
| 303 |
-
},
|
| 304 |
-
{
|
| 305 |
-
"epoch": 1.71,
|
| 306 |
-
"learning_rate": 4.348450646419044e-06,
|
| 307 |
-
"loss": 0.6695,
|
| 308 |
-
"step": 25000
|
| 309 |
-
},
|
| 310 |
-
{
|
| 311 |
-
"epoch": 1.74,
|
| 312 |
-
"learning_rate": 3.835419659347425e-06,
|
| 313 |
-
"loss": 0.6928,
|
| 314 |
-
"step": 25500
|
| 315 |
-
},
|
| 316 |
-
{
|
| 317 |
-
"epoch": 1.78,
|
| 318 |
-
"learning_rate": 3.3223886722758055e-06,
|
| 319 |
-
"loss": 0.6625,
|
| 320 |
-
"step": 26000
|
| 321 |
-
},
|
| 322 |
-
{
|
| 323 |
-
"epoch": 1.81,
|
| 324 |
-
"learning_rate": 2.809357685204186e-06,
|
| 325 |
-
"loss": 0.6704,
|
| 326 |
-
"step": 26500
|
| 327 |
-
},
|
| 328 |
-
{
|
| 329 |
-
"epoch": 1.85,
|
| 330 |
-
"learning_rate": 2.2963266981325674e-06,
|
| 331 |
-
"loss": 0.6487,
|
| 332 |
-
"step": 27000
|
| 333 |
-
},
|
| 334 |
-
{
|
| 335 |
-
"epoch": 1.88,
|
| 336 |
-
"learning_rate": 1.783295711060948e-06,
|
| 337 |
-
"loss": 0.6707,
|
| 338 |
-
"step": 27500
|
| 339 |
-
},
|
| 340 |
-
{
|
| 341 |
-
"epoch": 1.92,
|
| 342 |
-
"learning_rate": 1.270264723989329e-06,
|
| 343 |
-
"loss": 0.6782,
|
| 344 |
-
"step": 28000
|
| 345 |
-
},
|
| 346 |
-
{
|
| 347 |
-
"epoch": 1.95,
|
| 348 |
-
"learning_rate": 7.572337369177098e-07,
|
| 349 |
-
"loss": 0.6796,
|
| 350 |
-
"step": 28500
|
| 351 |
-
},
|
| 352 |
-
{
|
| 353 |
-
"epoch": 1.98,
|
| 354 |
-
"learning_rate": 2.442027498460907e-07,
|
| 355 |
-
"loss": 0.6222,
|
| 356 |
-
"step": 29000
|
| 357 |
-
},
|
| 358 |
-
{
|
| 359 |
-
"epoch": 2.0,
|
| 360 |
-
"step": 29238,
|
| 361 |
-
"total_flos": 4.583876744068301e+16,
|
| 362 |
-
"train_loss": 0.9686727332664389,
|
| 363 |
-
"train_runtime": 18778.1209,
|
| 364 |
-
"train_samples_per_second": 9.342,
|
| 365 |
-
"train_steps_per_second": 1.557
|
| 366 |
}
|
| 367 |
],
|
| 368 |
-
"max_steps":
|
| 369 |
-
"num_train_epochs":
|
| 370 |
-
"total_flos": 4.
|
| 371 |
"trial_name": null,
|
| 372 |
"trial_params": null
|
| 373 |
}
|
|
|
|
| 1 |
{
|
| 2 |
"best_metric": null,
|
| 3 |
"best_model_checkpoint": null,
|
| 4 |
+
"epoch": 1.0,
|
| 5 |
+
"global_step": 16235,
|
| 6 |
"is_hyper_param_search": false,
|
| 7 |
"is_local_process_zero": true,
|
| 8 |
"is_world_process_zero": true,
|
| 9 |
"log_history": [
|
| 10 |
{
|
| 11 |
"epoch": 0.03,
|
| 12 |
+
"learning_rate": 2.9076070218663384e-05,
|
| 13 |
+
"loss": 2.2393,
|
| 14 |
"step": 500
|
| 15 |
},
|
| 16 |
{
|
| 17 |
+
"epoch": 0.06,
|
| 18 |
+
"learning_rate": 2.8152140437326767e-05,
|
| 19 |
+
"loss": 1.4831,
|
| 20 |
"step": 1000
|
| 21 |
},
|
| 22 |
{
|
| 23 |
+
"epoch": 0.09,
|
| 24 |
+
"learning_rate": 2.7228210655990143e-05,
|
| 25 |
+
"loss": 1.3501,
|
| 26 |
"step": 1500
|
| 27 |
},
|
| 28 |
{
|
| 29 |
+
"epoch": 0.12,
|
| 30 |
+
"learning_rate": 2.6304280874653526e-05,
|
| 31 |
+
"loss": 1.3197,
|
| 32 |
"step": 2000
|
| 33 |
},
|
| 34 |
{
|
| 35 |
+
"epoch": 0.15,
|
| 36 |
+
"learning_rate": 2.538035109331691e-05,
|
| 37 |
+
"loss": 1.2581,
|
| 38 |
"step": 2500
|
| 39 |
},
|
| 40 |
{
|
| 41 |
+
"epoch": 0.18,
|
| 42 |
+
"learning_rate": 2.4456421311980292e-05,
|
| 43 |
+
"loss": 1.2174,
|
| 44 |
"step": 3000
|
| 45 |
},
|
| 46 |
{
|
| 47 |
+
"epoch": 0.22,
|
| 48 |
+
"learning_rate": 2.3532491530643672e-05,
|
| 49 |
+
"loss": 1.1655,
|
| 50 |
"step": 3500
|
| 51 |
},
|
| 52 |
{
|
| 53 |
+
"epoch": 0.25,
|
| 54 |
+
"learning_rate": 2.2608561749307052e-05,
|
| 55 |
+
"loss": 1.1402,
|
| 56 |
"step": 4000
|
| 57 |
},
|
| 58 |
{
|
| 59 |
+
"epoch": 0.28,
|
| 60 |
+
"learning_rate": 2.1684631967970435e-05,
|
| 61 |
+
"loss": 1.0944,
|
| 62 |
"step": 4500
|
| 63 |
},
|
| 64 |
{
|
| 65 |
+
"epoch": 0.31,
|
| 66 |
+
"learning_rate": 2.0760702186633818e-05,
|
| 67 |
+
"loss": 1.152,
|
| 68 |
"step": 5000
|
| 69 |
},
|
| 70 |
{
|
| 71 |
+
"epoch": 0.34,
|
| 72 |
+
"learning_rate": 1.9836772405297198e-05,
|
| 73 |
+
"loss": 1.1138,
|
| 74 |
"step": 5500
|
| 75 |
},
|
| 76 |
{
|
| 77 |
+
"epoch": 0.37,
|
| 78 |
+
"learning_rate": 1.8912842623960577e-05,
|
| 79 |
+
"loss": 1.0516,
|
| 80 |
"step": 6000
|
| 81 |
},
|
| 82 |
{
|
| 83 |
+
"epoch": 0.4,
|
| 84 |
+
"learning_rate": 1.798891284262396e-05,
|
| 85 |
+
"loss": 1.0912,
|
| 86 |
"step": 6500
|
| 87 |
},
|
| 88 |
{
|
| 89 |
+
"epoch": 0.43,
|
| 90 |
+
"learning_rate": 1.7064983061287344e-05,
|
| 91 |
+
"loss": 1.0322,
|
| 92 |
"step": 7000
|
| 93 |
},
|
| 94 |
{
|
| 95 |
+
"epoch": 0.46,
|
| 96 |
+
"learning_rate": 1.6141053279950723e-05,
|
| 97 |
+
"loss": 0.9863,
|
| 98 |
"step": 7500
|
| 99 |
},
|
| 100 |
{
|
| 101 |
+
"epoch": 0.49,
|
| 102 |
+
"learning_rate": 1.5217123498614105e-05,
|
| 103 |
+
"loss": 0.9888,
|
| 104 |
"step": 8000
|
| 105 |
},
|
| 106 |
{
|
| 107 |
+
"epoch": 0.52,
|
| 108 |
+
"learning_rate": 1.4293193717277488e-05,
|
| 109 |
+
"loss": 1.0046,
|
| 110 |
"step": 8500
|
| 111 |
},
|
| 112 |
{
|
| 113 |
+
"epoch": 0.55,
|
| 114 |
+
"learning_rate": 1.3369263935940867e-05,
|
| 115 |
+
"loss": 0.9813,
|
| 116 |
"step": 9000
|
| 117 |
},
|
| 118 |
{
|
| 119 |
+
"epoch": 0.59,
|
| 120 |
+
"learning_rate": 1.244533415460425e-05,
|
| 121 |
+
"loss": 0.9854,
|
| 122 |
"step": 9500
|
| 123 |
},
|
| 124 |
{
|
| 125 |
+
"epoch": 0.62,
|
| 126 |
+
"learning_rate": 1.1521404373267632e-05,
|
| 127 |
+
"loss": 0.9457,
|
| 128 |
"step": 10000
|
| 129 |
},
|
| 130 |
{
|
| 131 |
+
"epoch": 0.65,
|
| 132 |
+
"learning_rate": 1.0597474591931013e-05,
|
| 133 |
+
"loss": 0.931,
|
| 134 |
"step": 10500
|
| 135 |
},
|
| 136 |
{
|
| 137 |
+
"epoch": 0.68,
|
| 138 |
+
"learning_rate": 9.673544810594396e-06,
|
| 139 |
+
"loss": 0.8925,
|
| 140 |
"step": 11000
|
| 141 |
},
|
| 142 |
{
|
| 143 |
+
"epoch": 0.71,
|
| 144 |
+
"learning_rate": 8.749615029257776e-06,
|
| 145 |
+
"loss": 0.9469,
|
| 146 |
"step": 11500
|
| 147 |
},
|
| 148 |
{
|
| 149 |
+
"epoch": 0.74,
|
| 150 |
+
"learning_rate": 7.82568524792116e-06,
|
| 151 |
+
"loss": 0.9721,
|
| 152 |
"step": 12000
|
| 153 |
},
|
| 154 |
{
|
| 155 |
+
"epoch": 0.77,
|
| 156 |
+
"learning_rate": 6.90175546658454e-06,
|
| 157 |
+
"loss": 0.9377,
|
| 158 |
"step": 12500
|
| 159 |
},
|
| 160 |
{
|
| 161 |
+
"epoch": 0.8,
|
| 162 |
+
"learning_rate": 5.977825685247921e-06,
|
| 163 |
+
"loss": 0.8359,
|
| 164 |
"step": 13000
|
| 165 |
},
|
| 166 |
{
|
| 167 |
+
"epoch": 0.83,
|
| 168 |
+
"learning_rate": 5.0538959039113025e-06,
|
| 169 |
+
"loss": 0.9046,
|
| 170 |
"step": 13500
|
| 171 |
},
|
| 172 |
{
|
| 173 |
+
"epoch": 0.86,
|
| 174 |
+
"learning_rate": 4.129966122574684e-06,
|
| 175 |
+
"loss": 0.8857,
|
| 176 |
"step": 14000
|
| 177 |
},
|
| 178 |
{
|
| 179 |
+
"epoch": 0.89,
|
| 180 |
+
"learning_rate": 3.206036341238066e-06,
|
| 181 |
+
"loss": 0.8674,
|
| 182 |
"step": 14500
|
| 183 |
},
|
| 184 |
{
|
| 185 |
+
"epoch": 0.92,
|
| 186 |
+
"learning_rate": 2.2821065599014475e-06,
|
| 187 |
+
"loss": 0.8429,
|
| 188 |
"step": 15000
|
| 189 |
},
|
| 190 |
{
|
| 191 |
+
"epoch": 0.95,
|
| 192 |
+
"learning_rate": 1.3581767785648291e-06,
|
| 193 |
+
"loss": 0.9049,
|
| 194 |
"step": 15500
|
| 195 |
},
|
| 196 |
{
|
| 197 |
+
"epoch": 0.99,
|
| 198 |
+
"learning_rate": 4.3424699722821066e-07,
|
| 199 |
+
"loss": 0.8825,
|
| 200 |
"step": 16000
|
| 201 |
},
|
| 202 |
{
|
| 203 |
+
"epoch": 1.0,
|
| 204 |
+
"step": 16235,
|
| 205 |
+
"total_flos": 4.523174534460211e+16,
|
| 206 |
+
"train_loss": 1.0714931627842768,
|
| 207 |
+
"train_runtime": 16914.7753,
|
| 208 |
+
"train_samples_per_second": 5.759,
|
| 209 |
+
"train_steps_per_second": 0.96
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 210 |
}
|
| 211 |
],
|
| 212 |
+
"max_steps": 16235,
|
| 213 |
+
"num_train_epochs": 1,
|
| 214 |
+
"total_flos": 4.523174534460211e+16,
|
| 215 |
"trial_name": null,
|
| 216 |
"trial_params": null
|
| 217 |
}
|