Training in progress, step 99, checkpoint
Browse files
last-checkpoint/adapter_config.json
CHANGED
|
@@ -20,9 +20,9 @@
|
|
| 20 |
"rank_pattern": {},
|
| 21 |
"revision": null,
|
| 22 |
"target_modules": [
|
|
|
|
| 23 |
"Wqkv",
|
| 24 |
-
"layer"
|
| 25 |
-
"out_proj"
|
| 26 |
],
|
| 27 |
"task_type": "CAUSAL_LM",
|
| 28 |
"use_dora": false,
|
|
|
|
| 20 |
"rank_pattern": {},
|
| 21 |
"revision": null,
|
| 22 |
"target_modules": [
|
| 23 |
+
"out_proj",
|
| 24 |
"Wqkv",
|
| 25 |
+
"layer"
|
|
|
|
| 26 |
],
|
| 27 |
"task_type": "CAUSAL_LM",
|
| 28 |
"use_dora": false,
|
last-checkpoint/adapter_model.safetensors
CHANGED
|
@@ -1,3 +1,3 @@
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
-
oid sha256:
|
| 3 |
size 5752
|
|
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:fe3a9f1e13fc722192c8866e2604075b038e5ff2515d3b8dc7ccb353997e493d
|
| 3 |
size 5752
|
last-checkpoint/optimizer.pt
CHANGED
|
@@ -1,3 +1,3 @@
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
-
oid sha256:
|
| 3 |
size 15814
|
|
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:e3e09bf1bdfbf403552733db55f2cce3f7c6b52579e7d1bab8f62474f3a3a59d
|
| 3 |
size 15814
|
last-checkpoint/rng_state.pth
CHANGED
|
@@ -1,3 +1,3 @@
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
-
oid sha256:
|
| 3 |
size 14244
|
|
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:fd7949088d7233b9cf27831b62c9b3897ec0e564f2b4de82b056a51e812411f4
|
| 3 |
size 14244
|
last-checkpoint/trainer_state.json
CHANGED
|
@@ -1,7 +1,7 @@
|
|
| 1 |
{
|
| 2 |
"best_metric": null,
|
| 3 |
"best_model_checkpoint": null,
|
| 4 |
-
"epoch": 0.
|
| 5 |
"eval_steps": 50,
|
| 6 |
"global_step": 99,
|
| 7 |
"is_hyper_param_search": false,
|
|
@@ -9,716 +9,51 @@
|
|
| 9 |
"is_world_process_zero": true,
|
| 10 |
"log_history": [
|
| 11 |
{
|
| 12 |
-
"epoch": 0.
|
| 13 |
-
"grad_norm": 2.7270019927527755e-05,
|
| 14 |
-
"learning_rate": 5e-05,
|
| 15 |
-
"loss": 46.0,
|
| 16 |
-
"step": 1
|
| 17 |
-
},
|
| 18 |
-
{
|
| 19 |
-
"epoch": 0.0015552099533437014,
|
| 20 |
"eval_loss": 11.5,
|
| 21 |
-
"eval_runtime":
|
| 22 |
-
"eval_samples_per_second":
|
| 23 |
-
"eval_steps_per_second":
|
| 24 |
"step": 1
|
| 25 |
},
|
| 26 |
{
|
| 27 |
-
"epoch": 0.
|
| 28 |
-
"grad_norm":
|
| 29 |
-
"learning_rate": 0.0001,
|
| 30 |
-
"loss": 46.0,
|
| 31 |
-
"step": 2
|
| 32 |
-
},
|
| 33 |
-
{
|
| 34 |
-
"epoch": 0.004665629860031105,
|
| 35 |
-
"grad_norm": 3.66069762094412e-05,
|
| 36 |
-
"learning_rate": 0.00015,
|
| 37 |
-
"loss": 46.0,
|
| 38 |
-
"step": 3
|
| 39 |
-
},
|
| 40 |
-
{
|
| 41 |
-
"epoch": 0.006220839813374806,
|
| 42 |
-
"grad_norm": 1.8183880456490442e-05,
|
| 43 |
-
"learning_rate": 0.0002,
|
| 44 |
-
"loss": 46.0,
|
| 45 |
-
"step": 4
|
| 46 |
-
},
|
| 47 |
-
{
|
| 48 |
-
"epoch": 0.007776049766718507,
|
| 49 |
-
"grad_norm": 3.452602322795428e-05,
|
| 50 |
-
"learning_rate": 0.00025,
|
| 51 |
-
"loss": 46.0,
|
| 52 |
-
"step": 5
|
| 53 |
-
},
|
| 54 |
-
{
|
| 55 |
-
"epoch": 0.00933125972006221,
|
| 56 |
-
"grad_norm": 4.146944411331788e-05,
|
| 57 |
-
"learning_rate": 0.0003,
|
| 58 |
-
"loss": 46.0,
|
| 59 |
-
"step": 6
|
| 60 |
-
},
|
| 61 |
-
{
|
| 62 |
-
"epoch": 0.01088646967340591,
|
| 63 |
-
"grad_norm": 2.2120035282569006e-05,
|
| 64 |
-
"learning_rate": 0.00035,
|
| 65 |
-
"loss": 46.0,
|
| 66 |
-
"step": 7
|
| 67 |
-
},
|
| 68 |
-
{
|
| 69 |
-
"epoch": 0.012441679626749611,
|
| 70 |
-
"grad_norm": 3.082965122302994e-05,
|
| 71 |
-
"learning_rate": 0.0004,
|
| 72 |
-
"loss": 46.0,
|
| 73 |
-
"step": 8
|
| 74 |
-
},
|
| 75 |
-
{
|
| 76 |
-
"epoch": 0.013996889580093312,
|
| 77 |
-
"grad_norm": 2.718676842050627e-05,
|
| 78 |
-
"learning_rate": 0.00045000000000000004,
|
| 79 |
-
"loss": 46.0,
|
| 80 |
-
"step": 9
|
| 81 |
-
},
|
| 82 |
-
{
|
| 83 |
-
"epoch": 0.015552099533437015,
|
| 84 |
-
"grad_norm": 5.129418786964379e-05,
|
| 85 |
"learning_rate": 0.0005,
|
| 86 |
-
"loss":
|
| 87 |
-
"step": 10
|
| 88 |
-
},
|
| 89 |
-
{
|
| 90 |
-
"epoch": 0.017107309486780714,
|
| 91 |
-
"grad_norm": 3.0237812097766437e-05,
|
| 92 |
-
"learning_rate": 0.0004998442655654946,
|
| 93 |
-
"loss": 46.0,
|
| 94 |
-
"step": 11
|
| 95 |
-
},
|
| 96 |
-
{
|
| 97 |
-
"epoch": 0.01866251944012442,
|
| 98 |
-
"grad_norm": 3.804243533522822e-05,
|
| 99 |
-
"learning_rate": 0.0004993772562876909,
|
| 100 |
-
"loss": 46.0,
|
| 101 |
-
"step": 12
|
| 102 |
-
},
|
| 103 |
-
{
|
| 104 |
-
"epoch": 0.02021772939346812,
|
| 105 |
-
"grad_norm": 4.027284376206808e-05,
|
| 106 |
-
"learning_rate": 0.0004985995540019955,
|
| 107 |
-
"loss": 46.0,
|
| 108 |
-
"step": 13
|
| 109 |
-
},
|
| 110 |
-
{
|
| 111 |
-
"epoch": 0.02177293934681182,
|
| 112 |
-
"grad_norm": 3.257339994888753e-05,
|
| 113 |
-
"learning_rate": 0.0004975121276286136,
|
| 114 |
-
"loss": 46.0,
|
| 115 |
-
"step": 14
|
| 116 |
-
},
|
| 117 |
-
{
|
| 118 |
-
"epoch": 0.02332814930015552,
|
| 119 |
-
"grad_norm": 2.8166376068838872e-05,
|
| 120 |
-
"learning_rate": 0.0004961163319653958,
|
| 121 |
-
"loss": 46.0,
|
| 122 |
-
"step": 15
|
| 123 |
-
},
|
| 124 |
-
{
|
| 125 |
-
"epoch": 0.024883359253499222,
|
| 126 |
-
"grad_norm": 3.2243387977359816e-05,
|
| 127 |
-
"learning_rate": 0.0004944139059999286,
|
| 128 |
-
"loss": 46.0,
|
| 129 |
-
"step": 16
|
| 130 |
-
},
|
| 131 |
-
{
|
| 132 |
-
"epoch": 0.026438569206842923,
|
| 133 |
-
"grad_norm": 2.9551172701758333e-05,
|
| 134 |
-
"learning_rate": 0.000492406970742972,
|
| 135 |
-
"loss": 46.0,
|
| 136 |
-
"step": 17
|
| 137 |
-
},
|
| 138 |
-
{
|
| 139 |
-
"epoch": 0.027993779160186624,
|
| 140 |
-
"grad_norm": 3.1649913580622524e-05,
|
| 141 |
-
"learning_rate": 0.0004900980265859448,
|
| 142 |
-
"loss": 46.0,
|
| 143 |
-
"step": 18
|
| 144 |
-
},
|
| 145 |
-
{
|
| 146 |
-
"epoch": 0.029548989113530325,
|
| 147 |
-
"grad_norm": 3.053872205782682e-05,
|
| 148 |
-
"learning_rate": 0.0004874899501857477,
|
| 149 |
-
"loss": 46.0,
|
| 150 |
-
"step": 19
|
| 151 |
-
},
|
| 152 |
-
{
|
| 153 |
-
"epoch": 0.03110419906687403,
|
| 154 |
-
"grad_norm": 3.0835763027425855e-05,
|
| 155 |
-
"learning_rate": 0.00048458599088080736,
|
| 156 |
-
"loss": 46.0,
|
| 157 |
"step": 20
|
| 158 |
},
|
| 159 |
{
|
| 160 |
-
"epoch": 0.
|
| 161 |
-
"grad_norm":
|
| 162 |
-
"learning_rate": 0.
|
| 163 |
-
"loss":
|
| 164 |
-
"step": 21
|
| 165 |
-
},
|
| 166 |
-
{
|
| 167 |
-
"epoch": 0.03421461897356143,
|
| 168 |
-
"grad_norm": 3.770717012230307e-05,
|
| 169 |
-
"learning_rate": 0.00047790525956913543,
|
| 170 |
-
"loss": 46.0,
|
| 171 |
-
"step": 22
|
| 172 |
-
},
|
| 173 |
-
{
|
| 174 |
-
"epoch": 0.03576982892690513,
|
| 175 |
-
"grad_norm": 3.4990694985026494e-05,
|
| 176 |
-
"learning_rate": 0.0004741368109217071,
|
| 177 |
-
"loss": 46.0,
|
| 178 |
-
"step": 23
|
| 179 |
-
},
|
| 180 |
-
{
|
| 181 |
-
"epoch": 0.03732503888024884,
|
| 182 |
-
"grad_norm": 2.7294441679259762e-05,
|
| 183 |
-
"learning_rate": 0.00047008911571827283,
|
| 184 |
-
"loss": 46.0,
|
| 185 |
-
"step": 24
|
| 186 |
-
},
|
| 187 |
-
{
|
| 188 |
-
"epoch": 0.038880248833592534,
|
| 189 |
-
"grad_norm": 2.366018816246651e-05,
|
| 190 |
-
"learning_rate": 0.00046576721688302105,
|
| 191 |
-
"loss": 46.0,
|
| 192 |
-
"step": 25
|
| 193 |
-
},
|
| 194 |
-
{
|
| 195 |
-
"epoch": 0.04043545878693624,
|
| 196 |
-
"grad_norm": 3.290159293101169e-05,
|
| 197 |
-
"learning_rate": 0.0004611764989637205,
|
| 198 |
-
"loss": 46.0,
|
| 199 |
-
"step": 26
|
| 200 |
-
},
|
| 201 |
-
{
|
| 202 |
-
"epoch": 0.041990668740279936,
|
| 203 |
-
"grad_norm": 2.414073605905287e-05,
|
| 204 |
-
"learning_rate": 0.0004563226814232444,
|
| 205 |
-
"loss": 46.0,
|
| 206 |
-
"step": 27
|
| 207 |
-
},
|
| 208 |
-
{
|
| 209 |
-
"epoch": 0.04354587869362364,
|
| 210 |
-
"grad_norm": 5.248881279840134e-05,
|
| 211 |
-
"learning_rate": 0.0004512118115138315,
|
| 212 |
-
"loss": 46.0,
|
| 213 |
-
"step": 28
|
| 214 |
-
},
|
| 215 |
-
{
|
| 216 |
-
"epoch": 0.04510108864696734,
|
| 217 |
-
"grad_norm": 3.5208511690143496e-05,
|
| 218 |
-
"learning_rate": 0.0004458502567429631,
|
| 219 |
-
"loss": 46.0,
|
| 220 |
-
"step": 29
|
| 221 |
-
},
|
| 222 |
-
{
|
| 223 |
-
"epoch": 0.04665629860031104,
|
| 224 |
-
"grad_norm": 2.0407787815202028e-05,
|
| 225 |
-
"learning_rate": 0.00044024469694024196,
|
| 226 |
-
"loss": 46.0,
|
| 227 |
-
"step": 30
|
| 228 |
-
},
|
| 229 |
-
{
|
| 230 |
-
"epoch": 0.04821150855365474,
|
| 231 |
-
"grad_norm": 3.576717062969692e-05,
|
| 232 |
-
"learning_rate": 0.00043440211593515554,
|
| 233 |
-
"loss": 46.0,
|
| 234 |
-
"step": 31
|
| 235 |
-
},
|
| 236 |
-
{
|
| 237 |
-
"epoch": 0.049766718506998445,
|
| 238 |
-
"grad_norm": 3.6498990084510297e-05,
|
| 239 |
-
"learning_rate": 0.0004283297928560951,
|
| 240 |
-
"loss": 46.0,
|
| 241 |
-
"step": 32
|
| 242 |
-
},
|
| 243 |
-
{
|
| 244 |
-
"epoch": 0.05132192846034215,
|
| 245 |
-
"grad_norm": 4.3066760554211214e-05,
|
| 246 |
-
"learning_rate": 0.0004220352930614672,
|
| 247 |
-
"loss": 46.0,
|
| 248 |
-
"step": 33
|
| 249 |
-
},
|
| 250 |
-
{
|
| 251 |
-
"epoch": 0.05287713841368585,
|
| 252 |
-
"grad_norm": 3.5059842048212886e-05,
|
| 253 |
-
"learning_rate": 0.00041552645871420013,
|
| 254 |
-
"loss": 46.0,
|
| 255 |
-
"step": 34
|
| 256 |
-
},
|
| 257 |
-
{
|
| 258 |
-
"epoch": 0.05443234836702955,
|
| 259 |
-
"grad_norm": 4.324290421209298e-05,
|
| 260 |
-
"learning_rate": 0.00040881139901138467,
|
| 261 |
-
"loss": 46.0,
|
| 262 |
-
"step": 35
|
| 263 |
-
},
|
| 264 |
-
{
|
| 265 |
-
"epoch": 0.05598755832037325,
|
| 266 |
-
"grad_norm": 3.7152800359763205e-05,
|
| 267 |
-
"learning_rate": 0.00040189848008122475,
|
| 268 |
-
"loss": 46.0,
|
| 269 |
-
"step": 36
|
| 270 |
-
},
|
| 271 |
-
{
|
| 272 |
-
"epoch": 0.05754276827371695,
|
| 273 |
-
"grad_norm": 5.383112875279039e-05,
|
| 274 |
-
"learning_rate": 0.00039479631455988334,
|
| 275 |
-
"loss": 46.0,
|
| 276 |
-
"step": 37
|
| 277 |
-
},
|
| 278 |
-
{
|
| 279 |
-
"epoch": 0.05909797822706065,
|
| 280 |
-
"grad_norm": 4.21604527218733e-05,
|
| 281 |
-
"learning_rate": 0.0003875137508612103,
|
| 282 |
-
"loss": 46.0,
|
| 283 |
-
"step": 38
|
| 284 |
-
},
|
| 285 |
-
{
|
| 286 |
-
"epoch": 0.060653188180404355,
|
| 287 |
-
"grad_norm": 5.1183244067942724e-05,
|
| 288 |
-
"learning_rate": 0.00038005986215272055,
|
| 289 |
-
"loss": 46.0,
|
| 290 |
-
"step": 39
|
| 291 |
-
},
|
| 292 |
-
{
|
| 293 |
-
"epoch": 0.06220839813374806,
|
| 294 |
-
"grad_norm": 5.168653660803102e-05,
|
| 295 |
-
"learning_rate": 0.0003724439350515571,
|
| 296 |
-
"loss": 46.0,
|
| 297 |
"step": 40
|
| 298 |
},
|
| 299 |
{
|
| 300 |
-
"epoch": 0.
|
| 301 |
-
"grad_norm": 4.977727439836599e-05,
|
| 302 |
-
"learning_rate": 0.0003646754580545226,
|
| 303 |
-
"loss": 46.0,
|
| 304 |
-
"step": 41
|
| 305 |
-
},
|
| 306 |
-
{
|
| 307 |
-
"epoch": 0.06531881804043546,
|
| 308 |
-
"grad_norm": 3.4292854252271354e-05,
|
| 309 |
-
"learning_rate": 0.000356764109716594,
|
| 310 |
-
"loss": 46.0,
|
| 311 |
-
"step": 42
|
| 312 |
-
},
|
| 313 |
-
{
|
| 314 |
-
"epoch": 0.06687402799377916,
|
| 315 |
-
"grad_norm": 5.1678060117410496e-05,
|
| 316 |
-
"learning_rate": 0.00034871974659264783,
|
| 317 |
-
"loss": 46.0,
|
| 318 |
-
"step": 43
|
| 319 |
-
},
|
| 320 |
-
{
|
| 321 |
-
"epoch": 0.06842923794712286,
|
| 322 |
-
"grad_norm": 4.2354509787401184e-05,
|
| 323 |
-
"learning_rate": 0.0003405523909574206,
|
| 324 |
-
"loss": 46.0,
|
| 325 |
-
"step": 44
|
| 326 |
-
},
|
| 327 |
-
{
|
| 328 |
-
"epoch": 0.06998444790046657,
|
| 329 |
-
"grad_norm": 4.2241175833623856e-05,
|
| 330 |
-
"learning_rate": 0.0003322722183190025,
|
| 331 |
-
"loss": 46.0,
|
| 332 |
-
"step": 45
|
| 333 |
-
},
|
| 334 |
-
{
|
| 335 |
-
"epoch": 0.07153965785381027,
|
| 336 |
-
"grad_norm": 4.768397775478661e-05,
|
| 337 |
-
"learning_rate": 0.0003238895447414211,
|
| 338 |
-
"loss": 46.0,
|
| 339 |
-
"step": 46
|
| 340 |
-
},
|
| 341 |
-
{
|
| 342 |
-
"epoch": 0.07309486780715396,
|
| 343 |
-
"grad_norm": 3.0143904950818978e-05,
|
| 344 |
-
"learning_rate": 0.0003154148139921102,
|
| 345 |
-
"loss": 46.0,
|
| 346 |
-
"step": 47
|
| 347 |
-
},
|
| 348 |
-
{
|
| 349 |
-
"epoch": 0.07465007776049767,
|
| 350 |
-
"grad_norm": 4.8561680159764364e-05,
|
| 351 |
-
"learning_rate": 0.00030685858453027663,
|
| 352 |
-
"loss": 46.0,
|
| 353 |
-
"step": 48
|
| 354 |
-
},
|
| 355 |
-
{
|
| 356 |
-
"epoch": 0.07620528771384137,
|
| 357 |
-
"grad_norm": 2.9335944418562576e-05,
|
| 358 |
-
"learning_rate": 0.0002982315163523742,
|
| 359 |
-
"loss": 46.0,
|
| 360 |
-
"step": 49
|
| 361 |
-
},
|
| 362 |
-
{
|
| 363 |
-
"epoch": 0.07776049766718507,
|
| 364 |
-
"grad_norm": 3.3190059184562415e-05,
|
| 365 |
-
"learning_rate": 0.000289544357711076,
|
| 366 |
-
"loss": 46.0,
|
| 367 |
-
"step": 50
|
| 368 |
-
},
|
| 369 |
-
{
|
| 370 |
-
"epoch": 0.07776049766718507,
|
| 371 |
"eval_loss": 11.5,
|
| 372 |
-
"eval_runtime":
|
| 373 |
-
"eval_samples_per_second":
|
| 374 |
-
"eval_steps_per_second":
|
| 375 |
"step": 50
|
| 376 |
},
|
| 377 |
{
|
| 378 |
-
"epoch": 0.
|
| 379 |
-
"grad_norm":
|
| 380 |
-
"learning_rate": 0.
|
| 381 |
-
"loss":
|
| 382 |
-
"step": 51
|
| 383 |
-
},
|
| 384 |
-
{
|
| 385 |
-
"epoch": 0.08087091757387248,
|
| 386 |
-
"grad_norm": 4.568342774291523e-05,
|
| 387 |
-
"learning_rate": 0.0002720331228909005,
|
| 388 |
-
"loss": 46.0,
|
| 389 |
-
"step": 52
|
| 390 |
-
},
|
| 391 |
-
{
|
| 392 |
-
"epoch": 0.08242612752721618,
|
| 393 |
-
"grad_norm": 3.215171454939991e-05,
|
| 394 |
-
"learning_rate": 0.00026323086353004075,
|
| 395 |
-
"loss": 46.0,
|
| 396 |
-
"step": 53
|
| 397 |
-
},
|
| 398 |
-
{
|
| 399 |
-
"epoch": 0.08398133748055987,
|
| 400 |
-
"grad_norm": 2.91961150651332e-05,
|
| 401 |
-
"learning_rate": 0.0002544121201607822,
|
| 402 |
-
"loss": 46.0,
|
| 403 |
-
"step": 54
|
| 404 |
-
},
|
| 405 |
-
{
|
| 406 |
-
"epoch": 0.08553654743390357,
|
| 407 |
-
"grad_norm": 7.603203994221985e-05,
|
| 408 |
-
"learning_rate": 0.00024558787983921783,
|
| 409 |
-
"loss": 46.0,
|
| 410 |
-
"step": 55
|
| 411 |
-
},
|
| 412 |
-
{
|
| 413 |
-
"epoch": 0.08709175738724728,
|
| 414 |
-
"grad_norm": 5.6477994803572074e-05,
|
| 415 |
-
"learning_rate": 0.0002367691364699592,
|
| 416 |
-
"loss": 46.0,
|
| 417 |
-
"step": 56
|
| 418 |
-
},
|
| 419 |
-
{
|
| 420 |
-
"epoch": 0.08864696734059098,
|
| 421 |
-
"grad_norm": 3.0733051971765235e-05,
|
| 422 |
-
"learning_rate": 0.00022796687710909964,
|
| 423 |
-
"loss": 46.0,
|
| 424 |
-
"step": 57
|
| 425 |
-
},
|
| 426 |
-
{
|
| 427 |
-
"epoch": 0.09020217729393468,
|
| 428 |
-
"grad_norm": 3.2454528991365805e-05,
|
| 429 |
-
"learning_rate": 0.00021919206827571036,
|
| 430 |
-
"loss": 46.0,
|
| 431 |
-
"step": 58
|
| 432 |
-
},
|
| 433 |
-
{
|
| 434 |
-
"epoch": 0.09175738724727839,
|
| 435 |
-
"grad_norm": 7.772906974423677e-05,
|
| 436 |
-
"learning_rate": 0.00021045564228892402,
|
| 437 |
-
"loss": 46.0,
|
| 438 |
-
"step": 59
|
| 439 |
-
},
|
| 440 |
-
{
|
| 441 |
-
"epoch": 0.09331259720062209,
|
| 442 |
-
"grad_norm": 4.852826532442123e-05,
|
| 443 |
-
"learning_rate": 0.00020176848364762578,
|
| 444 |
-
"loss": 46.0,
|
| 445 |
"step": 60
|
| 446 |
},
|
| 447 |
{
|
| 448 |
-
"epoch": 0.
|
| 449 |
-
"grad_norm":
|
| 450 |
-
"learning_rate":
|
| 451 |
-
"loss":
|
| 452 |
-
"step": 61
|
| 453 |
-
},
|
| 454 |
-
{
|
| 455 |
-
"epoch": 0.09642301710730948,
|
| 456 |
-
"grad_norm": 4.146797800785862e-05,
|
| 457 |
-
"learning_rate": 0.00018458518600788986,
|
| 458 |
-
"loss": 46.0,
|
| 459 |
-
"step": 62
|
| 460 |
-
},
|
| 461 |
-
{
|
| 462 |
-
"epoch": 0.09797822706065319,
|
| 463 |
-
"grad_norm": 5.218683509156108e-05,
|
| 464 |
-
"learning_rate": 0.00017611045525857898,
|
| 465 |
-
"loss": 46.0,
|
| 466 |
-
"step": 63
|
| 467 |
-
},
|
| 468 |
-
{
|
| 469 |
-
"epoch": 0.09953343701399689,
|
| 470 |
-
"grad_norm": 6.322271656244993e-05,
|
| 471 |
-
"learning_rate": 0.0001677277816809975,
|
| 472 |
-
"loss": 46.0,
|
| 473 |
-
"step": 64
|
| 474 |
-
},
|
| 475 |
-
{
|
| 476 |
-
"epoch": 0.10108864696734059,
|
| 477 |
-
"grad_norm": 3.459403887973167e-05,
|
| 478 |
-
"learning_rate": 0.00015944760904257942,
|
| 479 |
-
"loss": 46.0,
|
| 480 |
-
"step": 65
|
| 481 |
-
},
|
| 482 |
-
{
|
| 483 |
-
"epoch": 0.1026438569206843,
|
| 484 |
-
"grad_norm": 5.7651352108223364e-05,
|
| 485 |
-
"learning_rate": 0.0001512802534073522,
|
| 486 |
-
"loss": 46.0,
|
| 487 |
-
"step": 66
|
| 488 |
-
},
|
| 489 |
-
{
|
| 490 |
-
"epoch": 0.104199066874028,
|
| 491 |
-
"grad_norm": 5.704832437913865e-05,
|
| 492 |
-
"learning_rate": 0.00014323589028340596,
|
| 493 |
-
"loss": 46.0,
|
| 494 |
-
"step": 67
|
| 495 |
-
},
|
| 496 |
-
{
|
| 497 |
-
"epoch": 0.1057542768273717,
|
| 498 |
-
"grad_norm": 3.9827515138313174e-05,
|
| 499 |
-
"learning_rate": 0.00013532454194547733,
|
| 500 |
-
"loss": 46.0,
|
| 501 |
-
"step": 68
|
| 502 |
-
},
|
| 503 |
-
{
|
| 504 |
-
"epoch": 0.10730948678071539,
|
| 505 |
-
"grad_norm": 7.096585613908246e-05,
|
| 506 |
-
"learning_rate": 0.00012755606494844294,
|
| 507 |
-
"loss": 46.0,
|
| 508 |
-
"step": 69
|
| 509 |
-
},
|
| 510 |
-
{
|
| 511 |
-
"epoch": 0.1088646967340591,
|
| 512 |
-
"grad_norm": 5.004853301215917e-05,
|
| 513 |
-
"learning_rate": 0.00011994013784727947,
|
| 514 |
-
"loss": 46.0,
|
| 515 |
-
"step": 70
|
| 516 |
-
},
|
| 517 |
-
{
|
| 518 |
-
"epoch": 0.1104199066874028,
|
| 519 |
-
"grad_norm": 3.065542477997951e-05,
|
| 520 |
-
"learning_rate": 0.00011248624913878966,
|
| 521 |
-
"loss": 46.0,
|
| 522 |
-
"step": 71
|
| 523 |
-
},
|
| 524 |
-
{
|
| 525 |
-
"epoch": 0.1119751166407465,
|
| 526 |
-
"grad_norm": 3.73735892935656e-05,
|
| 527 |
-
"learning_rate": 0.0001052036854401166,
|
| 528 |
-
"loss": 46.0,
|
| 529 |
-
"step": 72
|
| 530 |
-
},
|
| 531 |
-
{
|
| 532 |
-
"epoch": 0.11353032659409021,
|
| 533 |
-
"grad_norm": 8.166713814716786e-05,
|
| 534 |
-
"learning_rate": 9.810151991877531e-05,
|
| 535 |
-
"loss": 46.0,
|
| 536 |
-
"step": 73
|
| 537 |
-
},
|
| 538 |
-
{
|
| 539 |
-
"epoch": 0.1150855365474339,
|
| 540 |
-
"grad_norm": 5.165397305972874e-05,
|
| 541 |
-
"learning_rate": 9.118860098861537e-05,
|
| 542 |
-
"loss": 46.0,
|
| 543 |
-
"step": 74
|
| 544 |
-
},
|
| 545 |
-
{
|
| 546 |
-
"epoch": 0.1166407465007776,
|
| 547 |
-
"grad_norm": 4.3509502575034276e-05,
|
| 548 |
-
"learning_rate": 8.44735412857999e-05,
|
| 549 |
-
"loss": 46.0,
|
| 550 |
-
"step": 75
|
| 551 |
-
},
|
| 552 |
-
{
|
| 553 |
-
"epoch": 0.1181959564541213,
|
| 554 |
-
"grad_norm": 3.0016582968528382e-05,
|
| 555 |
-
"learning_rate": 7.79647069385328e-05,
|
| 556 |
-
"loss": 46.0,
|
| 557 |
-
"step": 76
|
| 558 |
-
},
|
| 559 |
-
{
|
| 560 |
-
"epoch": 0.11975116640746501,
|
| 561 |
-
"grad_norm": 4.617219383362681e-05,
|
| 562 |
-
"learning_rate": 7.167020714390501e-05,
|
| 563 |
-
"loss": 46.0,
|
| 564 |
-
"step": 77
|
| 565 |
-
},
|
| 566 |
-
{
|
| 567 |
-
"epoch": 0.12130637636080871,
|
| 568 |
-
"grad_norm": 6.536076398333535e-05,
|
| 569 |
-
"learning_rate": 6.559788406484446e-05,
|
| 570 |
-
"loss": 46.0,
|
| 571 |
-
"step": 78
|
| 572 |
-
},
|
| 573 |
-
{
|
| 574 |
-
"epoch": 0.12286158631415241,
|
| 575 |
-
"grad_norm": 5.4571206419495866e-05,
|
| 576 |
-
"learning_rate": 5.975530305975807e-05,
|
| 577 |
-
"loss": 46.0,
|
| 578 |
-
"step": 79
|
| 579 |
-
},
|
| 580 |
-
{
|
| 581 |
-
"epoch": 0.12441679626749612,
|
| 582 |
-
"grad_norm": 4.393588096718304e-05,
|
| 583 |
-
"learning_rate": 5.414974325703686e-05,
|
| 584 |
-
"loss": 46.0,
|
| 585 |
"step": 80
|
| 586 |
-
},
|
| 587 |
-
{
|
| 588 |
-
"epoch": 0.12597200622083982,
|
| 589 |
-
"grad_norm": 0.00010264776210533455,
|
| 590 |
-
"learning_rate": 4.8788188486168616e-05,
|
| 591 |
-
"loss": 46.0,
|
| 592 |
-
"step": 81
|
| 593 |
-
},
|
| 594 |
-
{
|
| 595 |
-
"epoch": 0.12752721617418353,
|
| 596 |
-
"grad_norm": 8.71722077135928e-05,
|
| 597 |
-
"learning_rate": 4.367731857675569e-05,
|
| 598 |
-
"loss": 46.0,
|
| 599 |
-
"step": 82
|
| 600 |
-
},
|
| 601 |
-
{
|
| 602 |
-
"epoch": 0.1290824261275272,
|
| 603 |
-
"grad_norm": 8.793560118647292e-05,
|
| 604 |
-
"learning_rate": 3.882350103627952e-05,
|
| 605 |
-
"loss": 46.0,
|
| 606 |
-
"step": 83
|
| 607 |
-
},
|
| 608 |
-
{
|
| 609 |
-
"epoch": 0.13063763608087092,
|
| 610 |
-
"grad_norm": 6.0609807405853644e-05,
|
| 611 |
-
"learning_rate": 3.423278311697897e-05,
|
| 612 |
-
"loss": 46.0,
|
| 613 |
-
"step": 84
|
| 614 |
-
},
|
| 615 |
-
{
|
| 616 |
-
"epoch": 0.1321928460342146,
|
| 617 |
-
"grad_norm": 4.071524745086208e-05,
|
| 618 |
-
"learning_rate": 2.9910884281727225e-05,
|
| 619 |
-
"loss": 46.0,
|
| 620 |
-
"step": 85
|
| 621 |
-
},
|
| 622 |
-
{
|
| 623 |
-
"epoch": 0.13374805598755832,
|
| 624 |
-
"grad_norm": 4.141516546951607e-05,
|
| 625 |
-
"learning_rate": 2.586318907829291e-05,
|
| 626 |
-
"loss": 46.0,
|
| 627 |
-
"step": 86
|
| 628 |
-
},
|
| 629 |
-
{
|
| 630 |
-
"epoch": 0.13530326594090203,
|
| 631 |
-
"grad_norm": 5.893910929444246e-05,
|
| 632 |
-
"learning_rate": 2.209474043086457e-05,
|
| 633 |
-
"loss": 46.0,
|
| 634 |
-
"step": 87
|
| 635 |
-
},
|
| 636 |
-
{
|
| 637 |
-
"epoch": 0.1368584758942457,
|
| 638 |
-
"grad_norm": 3.4431377571308985e-05,
|
| 639 |
-
"learning_rate": 1.861023335719475e-05,
|
| 640 |
-
"loss": 46.0,
|
| 641 |
-
"step": 88
|
| 642 |
-
},
|
| 643 |
-
{
|
| 644 |
-
"epoch": 0.13841368584758942,
|
| 645 |
-
"grad_norm": 8.926719601731747e-05,
|
| 646 |
-
"learning_rate": 1.5414009119192633e-05,
|
| 647 |
-
"loss": 46.0,
|
| 648 |
-
"step": 89
|
| 649 |
-
},
|
| 650 |
-
{
|
| 651 |
-
"epoch": 0.13996889580093314,
|
| 652 |
-
"grad_norm": 5.9101457736687735e-05,
|
| 653 |
-
"learning_rate": 1.25100498142523e-05,
|
| 654 |
-
"loss": 46.0,
|
| 655 |
-
"step": 90
|
| 656 |
-
},
|
| 657 |
-
{
|
| 658 |
-
"epoch": 0.14152410575427682,
|
| 659 |
-
"grad_norm": 5.563394006458111e-05,
|
| 660 |
-
"learning_rate": 9.901973414055187e-06,
|
| 661 |
-
"loss": 46.0,
|
| 662 |
-
"step": 91
|
| 663 |
-
},
|
| 664 |
-
{
|
| 665 |
-
"epoch": 0.14307931570762053,
|
| 666 |
-
"grad_norm": 3.268073487561196e-05,
|
| 667 |
-
"learning_rate": 7.593029257027956e-06,
|
| 668 |
-
"loss": 46.0,
|
| 669 |
-
"step": 92
|
| 670 |
-
},
|
| 671 |
-
{
|
| 672 |
-
"epoch": 0.14463452566096424,
|
| 673 |
-
"grad_norm": 4.3672669562511146e-05,
|
| 674 |
-
"learning_rate": 5.5860940000714015e-06,
|
| 675 |
-
"loss": 46.0,
|
| 676 |
-
"step": 93
|
| 677 |
-
},
|
| 678 |
-
{
|
| 679 |
-
"epoch": 0.14618973561430793,
|
| 680 |
-
"grad_norm": 6.923845648998395e-05,
|
| 681 |
-
"learning_rate": 3.8836680346041594e-06,
|
| 682 |
-
"loss": 46.0,
|
| 683 |
-
"step": 94
|
| 684 |
-
},
|
| 685 |
-
{
|
| 686 |
-
"epoch": 0.14774494556765164,
|
| 687 |
-
"grad_norm": 8.794792665867135e-05,
|
| 688 |
-
"learning_rate": 2.487872371386424e-06,
|
| 689 |
-
"loss": 46.0,
|
| 690 |
-
"step": 95
|
| 691 |
-
},
|
| 692 |
-
{
|
| 693 |
-
"epoch": 0.14930015552099535,
|
| 694 |
-
"grad_norm": 5.365633114706725e-05,
|
| 695 |
-
"learning_rate": 1.4004459980045125e-06,
|
| 696 |
-
"loss": 46.0,
|
| 697 |
-
"step": 96
|
| 698 |
-
},
|
| 699 |
-
{
|
| 700 |
-
"epoch": 0.15085536547433903,
|
| 701 |
-
"grad_norm": 4.416866795509122e-05,
|
| 702 |
-
"learning_rate": 6.22743712309054e-07,
|
| 703 |
-
"loss": 46.0,
|
| 704 |
-
"step": 97
|
| 705 |
-
},
|
| 706 |
-
{
|
| 707 |
-
"epoch": 0.15241057542768274,
|
| 708 |
-
"grad_norm": 9.724321716930717e-05,
|
| 709 |
-
"learning_rate": 1.557344345054501e-07,
|
| 710 |
-
"loss": 46.0,
|
| 711 |
-
"step": 98
|
| 712 |
-
},
|
| 713 |
-
{
|
| 714 |
-
"epoch": 0.15396578538102643,
|
| 715 |
-
"grad_norm": 4.769267616211437e-05,
|
| 716 |
-
"learning_rate": 0.0,
|
| 717 |
-
"loss": 46.0,
|
| 718 |
-
"step": 99
|
| 719 |
}
|
| 720 |
],
|
| 721 |
-
"logging_steps":
|
| 722 |
"max_steps": 99,
|
| 723 |
"num_input_tokens_seen": 0,
|
| 724 |
"num_train_epochs": 1,
|
|
@@ -735,8 +70,8 @@
|
|
| 735 |
"attributes": {}
|
| 736 |
}
|
| 737 |
},
|
| 738 |
-
"total_flos":
|
| 739 |
-
"train_batch_size":
|
| 740 |
"trial_name": null,
|
| 741 |
"trial_params": null
|
| 742 |
}
|
|
|
|
| 1 |
{
|
| 2 |
"best_metric": null,
|
| 3 |
"best_model_checkpoint": null,
|
| 4 |
+
"epoch": 0.30745341614906835,
|
| 5 |
"eval_steps": 50,
|
| 6 |
"global_step": 99,
|
| 7 |
"is_hyper_param_search": false,
|
|
|
|
| 9 |
"is_world_process_zero": true,
|
| 10 |
"log_history": [
|
| 11 |
{
|
| 12 |
+
"epoch": 0.003105590062111801,
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 13 |
"eval_loss": 11.5,
|
| 14 |
+
"eval_runtime": 0.766,
|
| 15 |
+
"eval_samples_per_second": 353.798,
|
| 16 |
+
"eval_steps_per_second": 22.194,
|
| 17 |
"step": 1
|
| 18 |
},
|
| 19 |
{
|
| 20 |
+
"epoch": 0.062111801242236024,
|
| 21 |
+
"grad_norm": 8.48894524096977e-06,
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 22 |
"learning_rate": 0.0005,
|
| 23 |
+
"loss": 11.5,
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 24 |
"step": 20
|
| 25 |
},
|
| 26 |
{
|
| 27 |
+
"epoch": 0.12422360248447205,
|
| 28 |
+
"grad_norm": 4.7956536946003325e-06,
|
| 29 |
+
"learning_rate": 0.00042501051864235636,
|
| 30 |
+
"loss": 11.5,
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 31 |
"step": 40
|
| 32 |
},
|
| 33 |
{
|
| 34 |
+
"epoch": 0.15527950310559005,
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 35 |
"eval_loss": 11.5,
|
| 36 |
+
"eval_runtime": 0.6897,
|
| 37 |
+
"eval_samples_per_second": 392.927,
|
| 38 |
+
"eval_steps_per_second": 24.649,
|
| 39 |
"step": 50
|
| 40 |
},
|
| 41 |
{
|
| 42 |
+
"epoch": 0.18633540372670807,
|
| 43 |
+
"grad_norm": 8.120113307086285e-06,
|
| 44 |
+
"learning_rate": 0.00024502945308373244,
|
| 45 |
+
"loss": 11.5,
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 46 |
"step": 60
|
| 47 |
},
|
| 48 |
{
|
| 49 |
+
"epoch": 0.2484472049689441,
|
| 50 |
+
"grad_norm": 8.012266334844753e-06,
|
| 51 |
+
"learning_rate": 6.803029740762648e-05,
|
| 52 |
+
"loss": 11.5,
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 53 |
"step": 80
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 54 |
}
|
| 55 |
],
|
| 56 |
+
"logging_steps": 20,
|
| 57 |
"max_steps": 99,
|
| 58 |
"num_input_tokens_seen": 0,
|
| 59 |
"num_train_epochs": 1,
|
|
|
|
| 70 |
"attributes": {}
|
| 71 |
}
|
| 72 |
},
|
| 73 |
+
"total_flos": 3944729935872.0,
|
| 74 |
+
"train_batch_size": 16,
|
| 75 |
"trial_name": null,
|
| 76 |
"trial_params": null
|
| 77 |
}
|
last-checkpoint/training_args.bin
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 sha256:8c894c75a1cba5d37858d46fa46d51225ce080fd0ac9d5db564f04c7de591383
|
| 3 |
+
size 6776
|