Delete trainer_state.json
Browse files- trainer_state.json +0 -1409
trainer_state.json
DELETED
|
@@ -1,1409 +0,0 @@
|
|
| 1 |
-
{
|
| 2 |
-
"best_metric": null,
|
| 3 |
-
"best_model_checkpoint": null,
|
| 4 |
-
"epoch": 0.9882571793977445,
|
| 5 |
-
"eval_steps": 500,
|
| 6 |
-
"global_step": 8500,
|
| 7 |
-
"is_hyper_param_search": false,
|
| 8 |
-
"is_local_process_zero": true,
|
| 9 |
-
"is_world_process_zero": true,
|
| 10 |
-
"log_history": [
|
| 11 |
-
{
|
| 12 |
-
"clip_ratio": 0.0,
|
| 13 |
-
"completion_length": 9.5546875,
|
| 14 |
-
"epoch": 0.0001162655505173817,
|
| 15 |
-
"grad_norm": 4.281470286839314,
|
| 16 |
-
"kl": 0.0,
|
| 17 |
-
"learning_rate": 1.1494252873563218e-08,
|
| 18 |
-
"loss": 0.024247150868177414,
|
| 19 |
-
"memory(GiB)": 38.7,
|
| 20 |
-
"response_clip_ratio": 0.0,
|
| 21 |
-
"reward": 1.191904902458191,
|
| 22 |
-
"reward_std": 0.2820184826850891,
|
| 23 |
-
"rewards/MultiModalMultiActionNoThinkNewAction": 1.191904902458191,
|
| 24 |
-
"step": 1,
|
| 25 |
-
"train_speed(iter/s)": 0.026594
|
| 26 |
-
},
|
| 27 |
-
{
|
| 28 |
-
"clip_ratio": 0.0,
|
| 29 |
-
"completion_length": 9.550307765151516,
|
| 30 |
-
"epoch": 0.01162655505173817,
|
| 31 |
-
"grad_norm": 3.349521431788205,
|
| 32 |
-
"kl": 0.042687281213625514,
|
| 33 |
-
"learning_rate": 9.99994247477391e-07,
|
| 34 |
-
"loss": 0.006692861065720067,
|
| 35 |
-
"memory(GiB)": 49.8,
|
| 36 |
-
"response_clip_ratio": 0.0,
|
| 37 |
-
"reward": 1.103371890506359,
|
| 38 |
-
"reward_std": 0.4002408231749679,
|
| 39 |
-
"rewards/MultiModalMultiActionNoThinkNewAction": 1.103371890506359,
|
| 40 |
-
"step": 100,
|
| 41 |
-
"train_speed(iter/s)": 0.093985
|
| 42 |
-
},
|
| 43 |
-
{
|
| 44 |
-
"clip_ratio": 0.0,
|
| 45 |
-
"completion_length": 9.4905078125,
|
| 46 |
-
"epoch": 0.02325311010347634,
|
| 47 |
-
"grad_norm": 6.366169338883484,
|
| 48 |
-
"kl": 0.09955322265625,
|
| 49 |
-
"learning_rate": 9.995654233098478e-07,
|
| 50 |
-
"loss": 0.0052225708961486815,
|
| 51 |
-
"memory(GiB)": 49.84,
|
| 52 |
-
"response_clip_ratio": 0.0,
|
| 53 |
-
"reward": 1.2144100672006608,
|
| 54 |
-
"reward_std": 0.26068811796605584,
|
| 55 |
-
"rewards/MultiModalMultiActionNoThinkNewAction": 1.2144100672006608,
|
| 56 |
-
"step": 200,
|
| 57 |
-
"train_speed(iter/s)": 0.09568
|
| 58 |
-
},
|
| 59 |
-
{
|
| 60 |
-
"clip_ratio": 0.0,
|
| 61 |
-
"completion_length": 9.5164453125,
|
| 62 |
-
"epoch": 0.03487966515521451,
|
| 63 |
-
"grad_norm": 8.965565147378136,
|
| 64 |
-
"kl": 0.123232421875,
|
| 65 |
-
"learning_rate": 9.98456494798275e-07,
|
| 66 |
-
"loss": 0.005987527966499329,
|
| 67 |
-
"memory(GiB)": 49.84,
|
| 68 |
-
"response_clip_ratio": 0.0,
|
| 69 |
-
"reward": 1.268085294365883,
|
| 70 |
-
"reward_std": 0.2155047995969653,
|
| 71 |
-
"rewards/MultiModalMultiActionNoThinkNewAction": 1.268085294365883,
|
| 72 |
-
"step": 300,
|
| 73 |
-
"train_speed(iter/s)": 0.095899
|
| 74 |
-
},
|
| 75 |
-
{
|
| 76 |
-
"clip_ratio": 0.0,
|
| 77 |
-
"completion_length": 9.3789453125,
|
| 78 |
-
"epoch": 0.04650622020695268,
|
| 79 |
-
"grad_norm": 5.398711016783826,
|
| 80 |
-
"kl": 0.1314990234375,
|
| 81 |
-
"learning_rate": 9.966689716290176e-07,
|
| 82 |
-
"loss": 0.006581841707229614,
|
| 83 |
-
"memory(GiB)": 49.84,
|
| 84 |
-
"response_clip_ratio": 0.0,
|
| 85 |
-
"reward": 1.3075979512929916,
|
| 86 |
-
"reward_std": 0.1733351560495794,
|
| 87 |
-
"rewards/MultiModalMultiActionNoThinkNewAction": 1.3075979512929916,
|
| 88 |
-
"step": 400,
|
| 89 |
-
"train_speed(iter/s)": 0.095915
|
| 90 |
-
},
|
| 91 |
-
{
|
| 92 |
-
"clip_ratio": 0.0,
|
| 93 |
-
"completion_length": 9.464140625,
|
| 94 |
-
"epoch": 0.05813277525869085,
|
| 95 |
-
"grad_norm": 11.929118986441576,
|
| 96 |
-
"kl": 0.13150390625,
|
| 97 |
-
"learning_rate": 9.942052873217221e-07,
|
| 98 |
-
"loss": 0.004319159388542175,
|
| 99 |
-
"memory(GiB)": 49.84,
|
| 100 |
-
"response_clip_ratio": 0.0,
|
| 101 |
-
"reward": 1.3076169914007187,
|
| 102 |
-
"reward_std": 0.16386293478310107,
|
| 103 |
-
"rewards/MultiModalMultiActionNoThinkNewAction": 1.3076169914007187,
|
| 104 |
-
"step": 500,
|
| 105 |
-
"train_speed(iter/s)": 0.095747
|
| 106 |
-
},
|
| 107 |
-
{
|
| 108 |
-
"clip_ratio": 0.0,
|
| 109 |
-
"completion_length": 9.3632421875,
|
| 110 |
-
"epoch": 0.06975933031042902,
|
| 111 |
-
"grad_norm": 9.252344124235806,
|
| 112 |
-
"kl": 0.1381982421875,
|
| 113 |
-
"learning_rate": 9.910687959163634e-07,
|
| 114 |
-
"loss": 0.0042449763417243954,
|
| 115 |
-
"memory(GiB)": 49.84,
|
| 116 |
-
"response_clip_ratio": 0.0,
|
| 117 |
-
"reward": 1.304671415090561,
|
| 118 |
-
"reward_std": 0.1764876712858677,
|
| 119 |
-
"rewards/MultiModalMultiActionNoThinkNewAction": 1.304671415090561,
|
| 120 |
-
"step": 600,
|
| 121 |
-
"train_speed(iter/s)": 0.094624
|
| 122 |
-
},
|
| 123 |
-
{
|
| 124 |
-
"clip_ratio": 0.0,
|
| 125 |
-
"completion_length": 9.49859375,
|
| 126 |
-
"epoch": 0.08138588536216719,
|
| 127 |
-
"grad_norm": 10.5649786127299,
|
| 128 |
-
"kl": 0.14923828125,
|
| 129 |
-
"learning_rate": 9.872637674070805e-07,
|
| 130 |
-
"loss": 0.005914233922958374,
|
| 131 |
-
"memory(GiB)": 49.84,
|
| 132 |
-
"response_clip_ratio": 0.0,
|
| 133 |
-
"reward": 1.34539220392704,
|
| 134 |
-
"reward_std": 0.14272778324782848,
|
| 135 |
-
"rewards/MultiModalMultiActionNoThinkNewAction": 1.34539220392704,
|
| 136 |
-
"step": 700,
|
| 137 |
-
"train_speed(iter/s)": 0.094626
|
| 138 |
-
},
|
| 139 |
-
{
|
| 140 |
-
"clip_ratio": 0.0,
|
| 141 |
-
"completion_length": 9.431015625,
|
| 142 |
-
"epoch": 0.09301244041390536,
|
| 143 |
-
"grad_norm": 13.958988545352236,
|
| 144 |
-
"kl": 0.1456884765625,
|
| 145 |
-
"learning_rate": 9.827953819290425e-07,
|
| 146 |
-
"loss": 0.00618122935295105,
|
| 147 |
-
"memory(GiB)": 49.86,
|
| 148 |
-
"response_clip_ratio": 0.0,
|
| 149 |
-
"reward": 1.3057015722990035,
|
| 150 |
-
"reward_std": 0.15137971622869373,
|
| 151 |
-
"rewards/MultiModalMultiActionNoThinkNewAction": 1.3057015722990035,
|
| 152 |
-
"step": 800,
|
| 153 |
-
"train_speed(iter/s)": 0.09467
|
| 154 |
-
},
|
| 155 |
-
{
|
| 156 |
-
"clip_ratio": 0.0,
|
| 157 |
-
"completion_length": 9.5357421875,
|
| 158 |
-
"epoch": 0.10463899546564354,
|
| 159 |
-
"grad_norm": 8.423788144030903,
|
| 160 |
-
"kl": 0.14693359375,
|
| 161 |
-
"learning_rate": 9.776697227062527e-07,
|
| 162 |
-
"loss": 0.004831492304801941,
|
| 163 |
-
"memory(GiB)": 49.86,
|
| 164 |
-
"response_clip_ratio": 0.0,
|
| 165 |
-
"reward": 1.347001107931137,
|
| 166 |
-
"reward_std": 0.14807809382677078,
|
| 167 |
-
"rewards/MultiModalMultiActionNoThinkNewAction": 1.347001107931137,
|
| 168 |
-
"step": 900,
|
| 169 |
-
"train_speed(iter/s)": 0.094728
|
| 170 |
-
},
|
| 171 |
-
{
|
| 172 |
-
"clip_ratio": 0.0,
|
| 173 |
-
"completion_length": 9.3253515625,
|
| 174 |
-
"epoch": 0.1162655505173817,
|
| 175 |
-
"grad_norm": 6.695960097747582,
|
| 176 |
-
"kl": 0.1556298828125,
|
| 177 |
-
"learning_rate": 9.718937677698976e-07,
|
| 178 |
-
"loss": 0.006615055799484253,
|
| 179 |
-
"memory(GiB)": 49.86,
|
| 180 |
-
"response_clip_ratio": 0.0,
|
| 181 |
-
"reward": 1.3540274119377136,
|
| 182 |
-
"reward_std": 0.12457055719569325,
|
| 183 |
-
"rewards/MultiModalMultiActionNoThinkNewAction": 1.3540274119377136,
|
| 184 |
-
"step": 1000,
|
| 185 |
-
"train_speed(iter/s)": 0.094744
|
| 186 |
-
},
|
| 187 |
-
{
|
| 188 |
-
"clip_ratio": 0.0,
|
| 189 |
-
"completion_length": 9.465390625,
|
| 190 |
-
"epoch": 0.12789210556911987,
|
| 191 |
-
"grad_norm": 10.285617962993532,
|
| 192 |
-
"kl": 0.16072265625,
|
| 193 |
-
"learning_rate": 9.654753804585103e-07,
|
| 194 |
-
"loss": 0.005601688623428345,
|
| 195 |
-
"memory(GiB)": 49.86,
|
| 196 |
-
"response_clip_ratio": 0.0,
|
| 197 |
-
"reward": 1.2828907597064971,
|
| 198 |
-
"reward_std": 0.14970409277826546,
|
| 199 |
-
"rewards/MultiModalMultiActionNoThinkNewAction": 1.2828907597064971,
|
| 200 |
-
"step": 1100,
|
| 201 |
-
"train_speed(iter/s)": 0.094415
|
| 202 |
-
},
|
| 203 |
-
{
|
| 204 |
-
"clip_ratio": 0.0,
|
| 205 |
-
"completion_length": 9.42546875,
|
| 206 |
-
"epoch": 0.13951866062085805,
|
| 207 |
-
"grad_norm": 10.836784655966541,
|
| 208 |
-
"kl": 0.158349609375,
|
| 209 |
-
"learning_rate": 9.584232987128862e-07,
|
| 210 |
-
"loss": 0.005478205680847168,
|
| 211 |
-
"memory(GiB)": 49.86,
|
| 212 |
-
"response_clip_ratio": 0.0,
|
| 213 |
-
"reward": 1.3187702250480653,
|
| 214 |
-
"reward_std": 0.1322044050693512,
|
| 215 |
-
"rewards/MultiModalMultiActionNoThinkNewAction": 1.3187702250480653,
|
| 216 |
-
"step": 1200,
|
| 217 |
-
"train_speed(iter/s)": 0.094452
|
| 218 |
-
},
|
| 219 |
-
{
|
| 220 |
-
"clip_ratio": 0.0,
|
| 221 |
-
"completion_length": 9.5623828125,
|
| 222 |
-
"epoch": 0.1511452156725962,
|
| 223 |
-
"grad_norm": 11.103606735921945,
|
| 224 |
-
"kl": 0.1616162109375,
|
| 225 |
-
"learning_rate": 9.507471231803197e-07,
|
| 226 |
-
"loss": 0.00721156120300293,
|
| 227 |
-
"memory(GiB)": 49.86,
|
| 228 |
-
"response_clip_ratio": 0.0,
|
| 229 |
-
"reward": 1.3138669067621231,
|
| 230 |
-
"reward_std": 0.14569927806034685,
|
| 231 |
-
"rewards/MultiModalMultiActionNoThinkNewAction": 1.3138669067621231,
|
| 232 |
-
"step": 1300,
|
| 233 |
-
"train_speed(iter/s)": 0.094427
|
| 234 |
-
},
|
| 235 |
-
{
|
| 236 |
-
"clip_ratio": 0.0,
|
| 237 |
-
"completion_length": 9.3422265625,
|
| 238 |
-
"epoch": 0.16277177072433438,
|
| 239 |
-
"grad_norm": 13.981767782488324,
|
| 240 |
-
"kl": 0.154521484375,
|
| 241 |
-
"learning_rate": 9.424573041443602e-07,
|
| 242 |
-
"loss": 0.0046518009901046755,
|
| 243 |
-
"memory(GiB)": 49.88,
|
| 244 |
-
"response_clip_ratio": 0.0,
|
| 245 |
-
"reward": 1.355497771501541,
|
| 246 |
-
"reward_std": 0.1318685195595026,
|
| 247 |
-
"rewards/MultiModalMultiActionNoThinkNewAction": 1.355497771501541,
|
| 248 |
-
"step": 1400,
|
| 249 |
-
"train_speed(iter/s)": 0.094456
|
| 250 |
-
},
|
| 251 |
-
{
|
| 252 |
-
"clip_ratio": 0.0,
|
| 253 |
-
"completion_length": 9.40296875,
|
| 254 |
-
"epoch": 0.17439832577607256,
|
| 255 |
-
"grad_norm": 15.113419365506667,
|
| 256 |
-
"kl": 0.1561474609375,
|
| 257 |
-
"learning_rate": 9.335651272978812e-07,
|
| 258 |
-
"loss": 0.007406370639801025,
|
| 259 |
-
"memory(GiB)": 49.88,
|
| 260 |
-
"response_clip_ratio": 0.0,
|
| 261 |
-
"reward": 1.3447466862201691,
|
| 262 |
-
"reward_std": 0.11326077262870968,
|
| 263 |
-
"rewards/MultiModalMultiActionNoThinkNewAction": 1.3447466862201691,
|
| 264 |
-
"step": 1500,
|
| 265 |
-
"train_speed(iter/s)": 0.094463
|
| 266 |
-
},
|
| 267 |
-
{
|
| 268 |
-
"clip_ratio": 0.0,
|
| 269 |
-
"completion_length": 9.2856640625,
|
| 270 |
-
"epoch": 0.1860248808278107,
|
| 271 |
-
"grad_norm": 9.743718453316509,
|
| 272 |
-
"kl": 0.1603076171875,
|
| 273 |
-
"learning_rate": 9.240826983788282e-07,
|
| 274 |
-
"loss": 0.005493613481521606,
|
| 275 |
-
"memory(GiB)": 49.88,
|
| 276 |
-
"response_clip_ratio": 0.0,
|
| 277 |
-
"reward": 1.3119167065620423,
|
| 278 |
-
"reward_std": 0.12479189267382025,
|
| 279 |
-
"rewards/MultiModalMultiActionNoThinkNewAction": 1.3119167065620423,
|
| 280 |
-
"step": 1600,
|
| 281 |
-
"train_speed(iter/s)": 0.094167
|
| 282 |
-
},
|
| 283 |
-
{
|
| 284 |
-
"clip_ratio": 0.0,
|
| 285 |
-
"completion_length": 9.5353515625,
|
| 286 |
-
"epoch": 0.1976514358795489,
|
| 287 |
-
"grad_norm": 13.264394110416175,
|
| 288 |
-
"kl": 0.1698193359375,
|
| 289 |
-
"learning_rate": 9.140229266895642e-07,
|
| 290 |
-
"loss": 0.006227902770042419,
|
| 291 |
-
"memory(GiB)": 49.88,
|
| 292 |
-
"response_clip_ratio": 0.0,
|
| 293 |
-
"reward": 1.3038281148672104,
|
| 294 |
-
"reward_std": 0.11284614092670381,
|
| 295 |
-
"rewards/MultiModalMultiActionNoThinkNewAction": 1.3038281148672104,
|
| 296 |
-
"step": 1700,
|
| 297 |
-
"train_speed(iter/s)": 0.094188
|
| 298 |
-
},
|
| 299 |
-
{
|
| 300 |
-
"clip_ratio": 0.0,
|
| 301 |
-
"completion_length": 9.4226171875,
|
| 302 |
-
"epoch": 0.20927799093128707,
|
| 303 |
-
"grad_norm": 9.544011725041532,
|
| 304 |
-
"kl": 0.16232421875,
|
| 305 |
-
"learning_rate": 9.033995075222507e-07,
|
| 306 |
-
"loss": 0.005826195478439331,
|
| 307 |
-
"memory(GiB)": 49.88,
|
| 308 |
-
"response_clip_ratio": 0.0,
|
| 309 |
-
"reward": 1.3212931084632873,
|
| 310 |
-
"reward_std": 0.11722485709935426,
|
| 311 |
-
"rewards/MultiModalMultiActionNoThinkNewAction": 1.3212931084632873,
|
| 312 |
-
"step": 1800,
|
| 313 |
-
"train_speed(iter/s)": 0.094171
|
| 314 |
-
},
|
| 315 |
-
{
|
| 316 |
-
"clip_ratio": 0.0,
|
| 317 |
-
"completion_length": 9.4878125,
|
| 318 |
-
"epoch": 0.22090454598302522,
|
| 319 |
-
"grad_norm": 10.343651509670643,
|
| 320 |
-
"kl": 0.17587890625,
|
| 321 |
-
"learning_rate": 8.922269035141858e-07,
|
| 322 |
-
"loss": 0.006521174311637878,
|
| 323 |
-
"memory(GiB)": 49.88,
|
| 324 |
-
"response_clip_ratio": 0.0,
|
| 325 |
-
"reward": 1.3666004729270935,
|
| 326 |
-
"reward_std": 0.11584680547006428,
|
| 327 |
-
"rewards/MultiModalMultiActionNoThinkNewAction": 1.3666004729270935,
|
| 328 |
-
"step": 1900,
|
| 329 |
-
"train_speed(iter/s)": 0.094172
|
| 330 |
-
},
|
| 331 |
-
{
|
| 332 |
-
"clip_ratio": 0.0,
|
| 333 |
-
"completion_length": 9.4062109375,
|
| 334 |
-
"epoch": 0.2325311010347634,
|
| 335 |
-
"grad_norm": 14.015908728784089,
|
| 336 |
-
"kl": 0.1614306640625,
|
| 337 |
-
"learning_rate": 8.805203249584874e-07,
|
| 338 |
-
"loss": 0.005633658170700074,
|
| 339 |
-
"memory(GiB)": 49.88,
|
| 340 |
-
"response_clip_ratio": 0.0,
|
| 341 |
-
"reward": 1.3424965512752534,
|
| 342 |
-
"reward_std": 0.11278555382974446,
|
| 343 |
-
"rewards/MultiModalMultiActionNoThinkNewAction": 1.3424965512752534,
|
| 344 |
-
"step": 2000,
|
| 345 |
-
"train_speed(iter/s)": 0.094185
|
| 346 |
-
},
|
| 347 |
-
{
|
| 348 |
-
"clip_ratio": 0.0,
|
| 349 |
-
"completion_length": 9.5082421875,
|
| 350 |
-
"epoch": 0.24415765608650156,
|
| 351 |
-
"grad_norm": 18.652060056142428,
|
| 352 |
-
"kl": 0.16919921875,
|
| 353 |
-
"learning_rate": 8.682957090969219e-07,
|
| 354 |
-
"loss": 0.005501749515533447,
|
| 355 |
-
"memory(GiB)": 49.96,
|
| 356 |
-
"response_clip_ratio": 0.0,
|
| 357 |
-
"reward": 1.3379686850309371,
|
| 358 |
-
"reward_std": 0.11816087782382965,
|
| 359 |
-
"rewards/MultiModalMultiActionNoThinkNewAction": 1.3379686850309371,
|
| 360 |
-
"step": 2100,
|
| 361 |
-
"train_speed(iter/s)": 0.094031
|
| 362 |
-
},
|
| 363 |
-
{
|
| 364 |
-
"clip_ratio": 0.0,
|
| 365 |
-
"completion_length": 9.449453125,
|
| 366 |
-
"epoch": 0.25578421113823974,
|
| 367 |
-
"grad_norm": 13.30826804477268,
|
| 368 |
-
"kl": 0.1685107421875,
|
| 369 |
-
"learning_rate": 8.555696984230717e-07,
|
| 370 |
-
"loss": 0.006635627746582031,
|
| 371 |
-
"memory(GiB)": 49.96,
|
| 372 |
-
"response_clip_ratio": 0.0,
|
| 373 |
-
"reward": 1.3248258876800536,
|
| 374 |
-
"reward_std": 0.12071031459607184,
|
| 375 |
-
"rewards/MultiModalMultiActionNoThinkNewAction": 1.3248258876800536,
|
| 376 |
-
"step": 2200,
|
| 377 |
-
"train_speed(iter/s)": 0.094031
|
| 378 |
-
},
|
| 379 |
-
{
|
| 380 |
-
"clip_ratio": 0.0,
|
| 381 |
-
"completion_length": 9.582734375,
|
| 382 |
-
"epoch": 0.2674107661899779,
|
| 383 |
-
"grad_norm": 14.474933819131337,
|
| 384 |
-
"kl": 0.1704052734375,
|
| 385 |
-
"learning_rate": 8.423596180253791e-07,
|
| 386 |
-
"loss": 0.005957164168357849,
|
| 387 |
-
"memory(GiB)": 49.96,
|
| 388 |
-
"response_clip_ratio": 0.0,
|
| 389 |
-
"reward": 1.3814373064041137,
|
| 390 |
-
"reward_std": 0.1127387316338718,
|
| 391 |
-
"rewards/MultiModalMultiActionNoThinkNewAction": 1.3814373064041137,
|
| 392 |
-
"step": 2300,
|
| 393 |
-
"train_speed(iter/s)": 0.094051
|
| 394 |
-
},
|
| 395 |
-
{
|
| 396 |
-
"clip_ratio": 0.0,
|
| 397 |
-
"completion_length": 9.45703125,
|
| 398 |
-
"epoch": 0.2790373212417161,
|
| 399 |
-
"grad_norm": 9.422596375579234,
|
| 400 |
-
"kl": 0.167822265625,
|
| 401 |
-
"learning_rate": 8.286834520009099e-07,
|
| 402 |
-
"loss": 0.0051918733119964595,
|
| 403 |
-
"memory(GiB)": 49.96,
|
| 404 |
-
"response_clip_ratio": 0.0,
|
| 405 |
-
"reward": 1.3118427366018295,
|
| 406 |
-
"reward_std": 0.12126734969206154,
|
| 407 |
-
"rewards/MultiModalMultiActionNoThinkNewAction": 1.3118427366018295,
|
| 408 |
-
"step": 2400,
|
| 409 |
-
"train_speed(iter/s)": 0.09409
|
| 410 |
-
},
|
| 411 |
-
{
|
| 412 |
-
"clip_ratio": 0.0,
|
| 413 |
-
"completion_length": 9.4315234375,
|
| 414 |
-
"epoch": 0.29066387629345425,
|
| 415 |
-
"grad_norm": 9.38274122340621,
|
| 416 |
-
"kl": 0.1697900390625,
|
| 417 |
-
"learning_rate": 8.145598189719482e-07,
|
| 418 |
-
"loss": 0.005171371102333069,
|
| 419 |
-
"memory(GiB)": 49.96,
|
| 420 |
-
"response_clip_ratio": 0.0,
|
| 421 |
-
"reward": 1.3867295610904693,
|
| 422 |
-
"reward_std": 0.11929248780943454,
|
| 423 |
-
"rewards/MultiModalMultiActionNoThinkNewAction": 1.3867295610904693,
|
| 424 |
-
"step": 2500,
|
| 425 |
-
"train_speed(iter/s)": 0.094103
|
| 426 |
-
},
|
| 427 |
-
{
|
| 428 |
-
"clip_ratio": 0.0,
|
| 429 |
-
"completion_length": 9.479296875,
|
| 430 |
-
"epoch": 0.3022904313451924,
|
| 431 |
-
"grad_norm": 13.737979351496811,
|
| 432 |
-
"kl": 0.176787109375,
|
| 433 |
-
"learning_rate": 8.000079467387547e-07,
|
| 434 |
-
"loss": 0.007171725034713745,
|
| 435 |
-
"memory(GiB)": 49.97,
|
| 436 |
-
"response_clip_ratio": 0.0,
|
| 437 |
-
"reward": 1.3833684372901915,
|
| 438 |
-
"reward_std": 0.1071634407620877,
|
| 439 |
-
"rewards/MultiModalMultiActionNoThinkNewAction": 1.3833684372901915,
|
| 440 |
-
"step": 2600,
|
| 441 |
-
"train_speed(iter/s)": 0.093982
|
| 442 |
-
},
|
| 443 |
-
{
|
| 444 |
-
"clip_ratio": 0.0,
|
| 445 |
-
"completion_length": 9.4278515625,
|
| 446 |
-
"epoch": 0.3139169863969306,
|
| 447 |
-
"grad_norm": 15.580608222516146,
|
| 448 |
-
"kl": 0.171650390625,
|
| 449 |
-
"learning_rate": 7.850476461029935e-07,
|
| 450 |
-
"loss": 0.004804742336273193,
|
| 451 |
-
"memory(GiB)": 49.97,
|
| 452 |
-
"response_clip_ratio": 0.0,
|
| 453 |
-
"reward": 1.3642382991313935,
|
| 454 |
-
"reward_std": 0.10643348384648561,
|
| 455 |
-
"rewards/MultiModalMultiActionNoThinkNewAction": 1.3642382991313935,
|
| 456 |
-
"step": 2700,
|
| 457 |
-
"train_speed(iter/s)": 0.094
|
| 458 |
-
},
|
| 459 |
-
{
|
| 460 |
-
"clip_ratio": 0.0,
|
| 461 |
-
"completion_length": 9.5507421875,
|
| 462 |
-
"epoch": 0.32554354144866876,
|
| 463 |
-
"grad_norm": 11.114845313442226,
|
| 464 |
-
"kl": 0.1594287109375,
|
| 465 |
-
"learning_rate": 7.696992838974672e-07,
|
| 466 |
-
"loss": 0.0048704689741134646,
|
| 467 |
-
"memory(GiB)": 49.97,
|
| 468 |
-
"response_clip_ratio": 0.0,
|
| 469 |
-
"reward": 1.3298124277591705,
|
| 470 |
-
"reward_std": 0.11607776273973286,
|
| 471 |
-
"rewards/MultiModalMultiActionNoThinkNewAction": 1.3298124277591705,
|
| 472 |
-
"step": 2800,
|
| 473 |
-
"train_speed(iter/s)": 0.093983
|
| 474 |
-
},
|
| 475 |
-
{
|
| 476 |
-
"clip_ratio": 0.0,
|
| 477 |
-
"completion_length": 9.53078125,
|
| 478 |
-
"epoch": 0.3371700965004069,
|
| 479 |
-
"grad_norm": 12.360373764757135,
|
| 480 |
-
"kl": 0.174365234375,
|
| 481 |
-
"learning_rate": 7.53983755258874e-07,
|
| 482 |
-
"loss": 0.005847660303115845,
|
| 483 |
-
"memory(GiB)": 49.97,
|
| 484 |
-
"response_clip_ratio": 0.0,
|
| 485 |
-
"reward": 1.3589847725629807,
|
| 486 |
-
"reward_std": 0.10737374008633196,
|
| 487 |
-
"rewards/MultiModalMultiActionNoThinkNewAction": 1.3589847725629807,
|
| 488 |
-
"step": 2900,
|
| 489 |
-
"train_speed(iter/s)": 0.093989
|
| 490 |
-
},
|
| 491 |
-
{
|
| 492 |
-
"clip_ratio": 0.0,
|
| 493 |
-
"completion_length": 9.5336328125,
|
| 494 |
-
"epoch": 0.3487966515521451,
|
| 495 |
-
"grad_norm": 10.064735245824536,
|
| 496 |
-
"kl": 0.1720654296875,
|
| 497 |
-
"learning_rate": 7.379224551813364e-07,
|
| 498 |
-
"loss": 0.005320903062820435,
|
| 499 |
-
"memory(GiB)": 49.97,
|
| 500 |
-
"response_clip_ratio": 0.0,
|
| 501 |
-
"reward": 1.3229967230558395,
|
| 502 |
-
"reward_std": 0.11362701586447656,
|
| 503 |
-
"rewards/MultiModalMultiActionNoThinkNewAction": 1.3229967230558395,
|
| 504 |
-
"step": 3000,
|
| 505 |
-
"train_speed(iter/s)": 0.094004
|
| 506 |
-
},
|
| 507 |
-
{
|
| 508 |
-
"clip_ratio": 0.0,
|
| 509 |
-
"completion_length": 9.531796875,
|
| 510 |
-
"epoch": 0.3604232066038833,
|
| 511 |
-
"grad_norm": 9.746163845034996,
|
| 512 |
-
"kl": 0.1750537109375,
|
| 513 |
-
"learning_rate": 7.215372493894305e-07,
|
| 514 |
-
"loss": 0.005580630302429199,
|
| 515 |
-
"memory(GiB)": 49.97,
|
| 516 |
-
"response_clip_ratio": 0.0,
|
| 517 |
-
"reward": 1.394041805267334,
|
| 518 |
-
"reward_std": 0.10719707342796028,
|
| 519 |
-
"rewards/MultiModalMultiActionNoThinkNewAction": 1.394041805267334,
|
| 520 |
-
"step": 3100,
|
| 521 |
-
"train_speed(iter/s)": 0.093906
|
| 522 |
-
},
|
| 523 |
-
{
|
| 524 |
-
"clip_ratio": 0.0,
|
| 525 |
-
"completion_length": 9.569140625,
|
| 526 |
-
"epoch": 0.3720497616556214,
|
| 527 |
-
"grad_norm": 10.075852679167767,
|
| 528 |
-
"kl": 0.164130859375,
|
| 529 |
-
"learning_rate": 7.048504445703623e-07,
|
| 530 |
-
"loss": 0.00647194504737854,
|
| 531 |
-
"memory(GiB)": 49.97,
|
| 532 |
-
"response_clip_ratio": 0.0,
|
| 533 |
-
"reward": 1.3294458091259003,
|
| 534 |
-
"reward_std": 0.10014143475331366,
|
| 535 |
-
"rewards/MultiModalMultiActionNoThinkNewAction": 1.3294458091259003,
|
| 536 |
-
"step": 3200,
|
| 537 |
-
"train_speed(iter/s)": 0.093919
|
| 538 |
-
},
|
| 539 |
-
{
|
| 540 |
-
"clip_ratio": 0.0,
|
| 541 |
-
"completion_length": 9.46984375,
|
| 542 |
-
"epoch": 0.38367631670735963,
|
| 543 |
-
"grad_norm": 9.861222956771822,
|
| 544 |
-
"kl": 0.16787109375,
|
| 545 |
-
"learning_rate": 6.87884758005825e-07,
|
| 546 |
-
"loss": 0.005488141775131226,
|
| 547 |
-
"memory(GiB)": 49.97,
|
| 548 |
-
"response_clip_ratio": 0.0,
|
| 549 |
-
"reward": 1.3671020436286927,
|
| 550 |
-
"reward_std": 0.10204395545646548,
|
| 551 |
-
"rewards/MultiModalMultiActionNoThinkNewAction": 1.3671020436286927,
|
| 552 |
-
"step": 3300,
|
| 553 |
-
"train_speed(iter/s)": 0.093924
|
| 554 |
-
},
|
| 555 |
-
{
|
| 556 |
-
"clip_ratio": 0.0,
|
| 557 |
-
"completion_length": 9.4744921875,
|
| 558 |
-
"epoch": 0.3953028717590978,
|
| 559 |
-
"grad_norm": 10.703294199835678,
|
| 560 |
-
"kl": 0.1692626953125,
|
| 561 |
-
"learning_rate": 6.706632866448739e-07,
|
| 562 |
-
"loss": 0.005845343470573425,
|
| 563 |
-
"memory(GiB)": 49.97,
|
| 564 |
-
"response_clip_ratio": 0.0,
|
| 565 |
-
"reward": 1.3646071863174438,
|
| 566 |
-
"reward_std": 0.10554393734782934,
|
| 567 |
-
"rewards/MultiModalMultiActionNoThinkNewAction": 1.3646071863174438,
|
| 568 |
-
"step": 3400,
|
| 569 |
-
"train_speed(iter/s)": 0.093949
|
| 570 |
-
},
|
| 571 |
-
{
|
| 572 |
-
"clip_ratio": 0.0,
|
| 573 |
-
"completion_length": 9.5876953125,
|
| 574 |
-
"epoch": 0.40692942681083594,
|
| 575 |
-
"grad_norm": 12.156014006973646,
|
| 576 |
-
"kl": 0.164482421875,
|
| 577 |
-
"learning_rate": 6.53209475659926e-07,
|
| 578 |
-
"loss": 0.005560991764068604,
|
| 579 |
-
"memory(GiB)": 49.97,
|
| 580 |
-
"response_clip_ratio": 0.0,
|
| 581 |
-
"reward": 1.3740809667110443,
|
| 582 |
-
"reward_std": 0.10303839593194425,
|
| 583 |
-
"rewards/MultiModalMultiActionNoThinkNewAction": 1.3740809667110443,
|
| 584 |
-
"step": 3500,
|
| 585 |
-
"train_speed(iter/s)": 0.093941
|
| 586 |
-
},
|
| 587 |
-
{
|
| 588 |
-
"clip_ratio": 0.0,
|
| 589 |
-
"completion_length": 9.4395703125,
|
| 590 |
-
"epoch": 0.41855598186257414,
|
| 591 |
-
"grad_norm": 10.579136209736488,
|
| 592 |
-
"kl": 0.167001953125,
|
| 593 |
-
"learning_rate": 6.355470865286917e-07,
|
| 594 |
-
"loss": 0.0047145700454711914,
|
| 595 |
-
"memory(GiB)": 49.97,
|
| 596 |
-
"response_clip_ratio": 0.0,
|
| 597 |
-
"reward": 1.3503324526548386,
|
| 598 |
-
"reward_std": 0.09954373368062079,
|
| 599 |
-
"rewards/MultiModalMultiActionNoThinkNewAction": 1.3503324526548386,
|
| 600 |
-
"step": 3600,
|
| 601 |
-
"train_speed(iter/s)": 0.09383
|
| 602 |
-
},
|
| 603 |
-
{
|
| 604 |
-
"clip_ratio": 0.0,
|
| 605 |
-
"completion_length": 9.5458984375,
|
| 606 |
-
"epoch": 0.4301825369143123,
|
| 607 |
-
"grad_norm": 12.621743422766063,
|
| 608 |
-
"kl": 0.1657080078125,
|
| 609 |
-
"learning_rate": 6.177001646854896e-07,
|
| 610 |
-
"loss": 0.006506719589233398,
|
| 611 |
-
"memory(GiB)": 49.97,
|
| 612 |
-
"response_clip_ratio": 3.90625e-05,
|
| 613 |
-
"reward": 1.3525580525398255,
|
| 614 |
-
"reward_std": 0.09826488124206662,
|
| 615 |
-
"rewards/MultiModalMultiActionNoThinkNewAction": 1.3525580525398255,
|
| 616 |
-
"step": 3700,
|
| 617 |
-
"train_speed(iter/s)": 0.093795
|
| 618 |
-
},
|
| 619 |
-
{
|
| 620 |
-
"clip_ratio": 0.0,
|
| 621 |
-
"completion_length": 9.4311328125,
|
| 622 |
-
"epoch": 0.44180909196605045,
|
| 623 |
-
"grad_norm": 9.354617068193154,
|
| 624 |
-
"kl": 0.173662109375,
|
| 625 |
-
"learning_rate": 5.996930067859863e-07,
|
| 626 |
-
"loss": 0.005135659575462342,
|
| 627 |
-
"memory(GiB)": 49.97,
|
| 628 |
-
"response_clip_ratio": 0.0,
|
| 629 |
-
"reward": 1.3989291822910308,
|
| 630 |
-
"reward_std": 0.09648296672850848,
|
| 631 |
-
"rewards/MultiModalMultiActionNoThinkNewAction": 1.3989291822910308,
|
| 632 |
-
"step": 3800,
|
| 633 |
-
"train_speed(iter/s)": 0.093833
|
| 634 |
-
},
|
| 635 |
-
{
|
| 636 |
-
"clip_ratio": 0.0,
|
| 637 |
-
"completion_length": 9.49171875,
|
| 638 |
-
"epoch": 0.45343564701778866,
|
| 639 |
-
"grad_norm": 14.90852834265166,
|
| 640 |
-
"kl": 0.16689453125,
|
| 641 |
-
"learning_rate": 5.815501276299251e-07,
|
| 642 |
-
"loss": 0.0054639244079589845,
|
| 643 |
-
"memory(GiB)": 49.97,
|
| 644 |
-
"response_clip_ratio": 0.0,
|
| 645 |
-
"reward": 1.3303378784656525,
|
| 646 |
-
"reward_std": 0.1025526038557291,
|
| 647 |
-
"rewards/MultiModalMultiActionNoThinkNewAction": 1.3303378784656525,
|
| 648 |
-
"step": 3900,
|
| 649 |
-
"train_speed(iter/s)": 0.093865
|
| 650 |
-
},
|
| 651 |
-
{
|
| 652 |
-
"clip_ratio": 0.0,
|
| 653 |
-
"completion_length": 9.45296875,
|
| 654 |
-
"epoch": 0.4650622020695268,
|
| 655 |
-
"grad_norm": 9.790046480640859,
|
| 656 |
-
"kl": 0.171201171875,
|
| 657 |
-
"learning_rate": 5.632962267868746e-07,
|
| 658 |
-
"loss": 0.005732476711273193,
|
| 659 |
-
"memory(GiB)": 49.97,
|
| 660 |
-
"response_clip_ratio": 0.0,
|
| 661 |
-
"reward": 1.3618503904342651,
|
| 662 |
-
"reward_std": 0.09345575381070376,
|
| 663 |
-
"rewards/MultiModalMultiActionNoThinkNewAction": 1.3618503904342651,
|
| 664 |
-
"step": 4000,
|
| 665 |
-
"train_speed(iter/s)": 0.093884
|
| 666 |
-
},
|
| 667 |
-
{
|
| 668 |
-
"clip_ratio": 0.0,
|
| 669 |
-
"completion_length": 9.4981640625,
|
| 670 |
-
"epoch": 0.47668875712126496,
|
| 671 |
-
"grad_norm": 16.23171809625287,
|
| 672 |
-
"kl": 0.1683154296875,
|
| 673 |
-
"learning_rate": 5.449561549704347e-07,
|
| 674 |
-
"loss": 0.006350870132446289,
|
| 675 |
-
"memory(GiB)": 49.97,
|
| 676 |
-
"response_clip_ratio": 0.0,
|
| 677 |
-
"reward": 1.3815095353126525,
|
| 678 |
-
"reward_std": 0.10499328017234802,
|
| 679 |
-
"rewards/MultiModalMultiActionNoThinkNewAction": 1.3815095353126525,
|
| 680 |
-
"step": 4100,
|
| 681 |
-
"train_speed(iter/s)": 0.093808
|
| 682 |
-
},
|
| 683 |
-
{
|
| 684 |
-
"clip_ratio": 0.0,
|
| 685 |
-
"completion_length": 9.4899609375,
|
| 686 |
-
"epoch": 0.4883153121730031,
|
| 687 |
-
"grad_norm": 10.028625827275622,
|
| 688 |
-
"kl": 0.172294921875,
|
| 689 |
-
"learning_rate": 5.265548802066736e-07,
|
| 690 |
-
"loss": 0.005829288959503174,
|
| 691 |
-
"memory(GiB)": 50.01,
|
| 692 |
-
"response_clip_ratio": 0.0,
|
| 693 |
-
"reward": 1.3365002036094666,
|
| 694 |
-
"reward_std": 0.09621842056512833,
|
| 695 |
-
"rewards/MultiModalMultiActionNoThinkNewAction": 1.3365002036094666,
|
| 696 |
-
"step": 4200,
|
| 697 |
-
"train_speed(iter/s)": 0.093821
|
| 698 |
-
},
|
| 699 |
-
{
|
| 700 |
-
"clip_ratio": 0.0,
|
| 701 |
-
"completion_length": 9.44015625,
|
| 702 |
-
"epoch": 0.4999418672247413,
|
| 703 |
-
"grad_norm": 10.282442330807516,
|
| 704 |
-
"kl": 0.1719873046875,
|
| 705 |
-
"learning_rate": 5.081174538428596e-07,
|
| 706 |
-
"loss": 0.006229421496391297,
|
| 707 |
-
"memory(GiB)": 50.01,
|
| 708 |
-
"response_clip_ratio": 0.0,
|
| 709 |
-
"reward": 1.3299858927726746,
|
| 710 |
-
"reward_std": 0.09780080110765993,
|
| 711 |
-
"rewards/MultiModalMultiActionNoThinkNewAction": 1.3299858927726746,
|
| 712 |
-
"step": 4300,
|
| 713 |
-
"train_speed(iter/s)": 0.093842
|
| 714 |
-
},
|
| 715 |
-
{
|
| 716 |
-
"clip_ratio": 0.0,
|
| 717 |
-
"completion_length": 9.4265234375,
|
| 718 |
-
"epoch": 0.5115684222764795,
|
| 719 |
-
"grad_norm": 10.479527710442367,
|
| 720 |
-
"kl": 0.171787109375,
|
| 721 |
-
"learning_rate": 4.89668976442758e-07,
|
| 722 |
-
"loss": 0.004759013652801514,
|
| 723 |
-
"memory(GiB)": 50.01,
|
| 724 |
-
"response_clip_ratio": 0.0,
|
| 725 |
-
"reward": 1.4068960046768189,
|
| 726 |
-
"reward_std": 0.09772759121842682,
|
| 727 |
-
"rewards/MultiModalMultiActionNoThinkNewAction": 1.4068960046768189,
|
| 728 |
-
"step": 4400,
|
| 729 |
-
"train_speed(iter/s)": 0.093846
|
| 730 |
-
},
|
| 731 |
-
{
|
| 732 |
-
"clip_ratio": 0.0,
|
| 733 |
-
"completion_length": 9.43875,
|
| 734 |
-
"epoch": 0.5231949773282176,
|
| 735 |
-
"grad_norm": 17.529468974754778,
|
| 736 |
-
"kl": 0.1623486328125,
|
| 737 |
-
"learning_rate": 4.7123456361492607e-07,
|
| 738 |
-
"loss": 0.004565647840499878,
|
| 739 |
-
"memory(GiB)": 50.01,
|
| 740 |
-
"response_clip_ratio": 0.0,
|
| 741 |
-
"reward": 1.3555949538946153,
|
| 742 |
-
"reward_std": 0.09924329521134495,
|
| 743 |
-
"rewards/MultiModalMultiActionNoThinkNewAction": 1.3555949538946153,
|
| 744 |
-
"step": 4500,
|
| 745 |
-
"train_speed(iter/s)": 0.093871
|
| 746 |
-
},
|
| 747 |
-
{
|
| 748 |
-
"clip_ratio": 0.0,
|
| 749 |
-
"completion_length": 9.5710546875,
|
| 750 |
-
"epoch": 0.5348215323799558,
|
| 751 |
-
"grad_norm": 13.65773402603584,
|
| 752 |
-
"kl": 0.16982421875,
|
| 753 |
-
"learning_rate": 4.528393118205257e-07,
|
| 754 |
-
"loss": 0.005337421298027039,
|
| 755 |
-
"memory(GiB)": 50.01,
|
| 756 |
-
"response_clip_ratio": 0.0,
|
| 757 |
-
"reward": 1.3599874556064606,
|
| 758 |
-
"reward_std": 0.10061670660972595,
|
| 759 |
-
"rewards/MultiModalMultiActionNoThinkNewAction": 1.3599874556064606,
|
| 760 |
-
"step": 4600,
|
| 761 |
-
"train_speed(iter/s)": 0.093805
|
| 762 |
-
},
|
| 763 |
-
{
|
| 764 |
-
"clip_ratio": 0.0,
|
| 765 |
-
"completion_length": 9.53828125,
|
| 766 |
-
"epoch": 0.546448087431694,
|
| 767 |
-
"grad_norm": 8.826143181240491,
|
| 768 |
-
"kl": 0.177373046875,
|
| 769 |
-
"learning_rate": 4.3450826420720283e-07,
|
| 770 |
-
"loss": 0.0059269857406616215,
|
| 771 |
-
"memory(GiB)": 50.01,
|
| 772 |
-
"response_clip_ratio": 3.90625e-05,
|
| 773 |
-
"reward": 1.3456268101930617,
|
| 774 |
-
"reward_std": 0.0843733003223315,
|
| 775 |
-
"rewards/MultiModalMultiActionNoThinkNewAction": 1.3456268101930617,
|
| 776 |
-
"step": 4700,
|
| 777 |
-
"train_speed(iter/s)": 0.093771
|
| 778 |
-
},
|
| 779 |
-
{
|
| 780 |
-
"clip_ratio": 0.0,
|
| 781 |
-
"completion_length": 9.7103125,
|
| 782 |
-
"epoch": 0.5580746424834322,
|
| 783 |
-
"grad_norm": 10.962918399547704,
|
| 784 |
-
"kl": 0.178984375,
|
| 785 |
-
"learning_rate": 4.1626637651554855e-07,
|
| 786 |
-
"loss": 0.005514188408851623,
|
| 787 |
-
"memory(GiB)": 50.01,
|
| 788 |
-
"response_clip_ratio": 0.0,
|
| 789 |
-
"reward": 1.3955884772539138,
|
| 790 |
-
"reward_std": 0.0897440404444933,
|
| 791 |
-
"rewards/MultiModalMultiActionNoThinkNewAction": 1.3955884772539138,
|
| 792 |
-
"step": 4800,
|
| 793 |
-
"train_speed(iter/s)": 0.093778
|
| 794 |
-
},
|
| 795 |
-
{
|
| 796 |
-
"clip_ratio": 0.0,
|
| 797 |
-
"completion_length": 9.421328125,
|
| 798 |
-
"epoch": 0.5697011975351703,
|
| 799 |
-
"grad_norm": 13.288533082686738,
|
| 800 |
-
"kl": 0.1740966796875,
|
| 801 |
-
"learning_rate": 3.9813848310455493e-07,
|
| 802 |
-
"loss": 0.004496717453002929,
|
| 803 |
-
"memory(GiB)": 50.01,
|
| 804 |
-
"response_clip_ratio": 0.0,
|
| 805 |
-
"reward": 1.3486594069004059,
|
| 806 |
-
"reward_std": 0.09252091265749186,
|
| 807 |
-
"rewards/MultiModalMultiActionNoThinkNewAction": 1.3486594069004059,
|
| 808 |
-
"step": 4900,
|
| 809 |
-
"train_speed(iter/s)": 0.093792
|
| 810 |
-
},
|
| 811 |
-
{
|
| 812 |
-
"clip_ratio": 0.0,
|
| 813 |
-
"completion_length": 9.6050390625,
|
| 814 |
-
"epoch": 0.5813277525869085,
|
| 815 |
-
"grad_norm": 13.576182979053868,
|
| 816 |
-
"kl": 0.16875,
|
| 817 |
-
"learning_rate": 3.801492631423183e-07,
|
| 818 |
-
"loss": 0.006311448216438293,
|
| 819 |
-
"memory(GiB)": 50.01,
|
| 820 |
-
"response_clip_ratio": 0.0,
|
| 821 |
-
"reward": 1.3385675930976868,
|
| 822 |
-
"reward_std": 0.08789137057960034,
|
| 823 |
-
"rewards/MultiModalMultiActionNoThinkNewAction": 1.3385675930976868,
|
| 824 |
-
"step": 5000,
|
| 825 |
-
"train_speed(iter/s)": 0.093819
|
| 826 |
-
},
|
| 827 |
-
{
|
| 828 |
-
"clip_ratio": 0.0,
|
| 829 |
-
"completion_length": 9.4701953125,
|
| 830 |
-
"epoch": 0.5929543076386466,
|
| 831 |
-
"grad_norm": 10.883793005621234,
|
| 832 |
-
"kl": 0.1636376953125,
|
| 833 |
-
"learning_rate": 3.6232320700801824e-07,
|
| 834 |
-
"loss": 0.005545246005058288,
|
| 835 |
-
"memory(GiB)": 50.01,
|
| 836 |
-
"response_clip_ratio": 0.0,
|
| 837 |
-
"reward": 1.3905673331022264,
|
| 838 |
-
"reward_std": 0.08755960265174507,
|
| 839 |
-
"rewards/MultiModalMultiActionNoThinkNewAction": 1.3905673331022264,
|
| 840 |
-
"step": 5100,
|
| 841 |
-
"train_speed(iter/s)": 0.093749
|
| 842 |
-
},
|
| 843 |
-
{
|
| 844 |
-
"clip_ratio": 0.0,
|
| 845 |
-
"completion_length": 9.4654296875,
|
| 846 |
-
"epoch": 0.6045808626903848,
|
| 847 |
-
"grad_norm": 10.540876662145202,
|
| 848 |
-
"kl": 0.1647607421875,
|
| 849 |
-
"learning_rate": 3.446845829509133e-07,
|
| 850 |
-
"loss": 0.004752608835697174,
|
| 851 |
-
"memory(GiB)": 50.01,
|
| 852 |
-
"response_clip_ratio": 0.0,
|
| 853 |
-
"reward": 1.428147051334381,
|
| 854 |
-
"reward_std": 0.08269149933941662,
|
| 855 |
-
"rewards/MultiModalMultiActionNoThinkNewAction": 1.428147051334381,
|
| 856 |
-
"step": 5200,
|
| 857 |
-
"train_speed(iter/s)": 0.093767
|
| 858 |
-
},
|
| 859 |
-
{
|
| 860 |
-
"clip_ratio": 0.0,
|
| 861 |
-
"completion_length": 9.4448046875,
|
| 862 |
-
"epoch": 0.616207417742123,
|
| 863 |
-
"grad_norm": 13.475346015876513,
|
| 864 |
-
"kl": 0.16580078125,
|
| 865 |
-
"learning_rate": 3.2725740405174216e-07,
|
| 866 |
-
"loss": 0.0050477349758148195,
|
| 867 |
-
"memory(GiB)": 50.01,
|
| 868 |
-
"response_clip_ratio": 0.0,
|
| 869 |
-
"reward": 1.404239798784256,
|
| 870 |
-
"reward_std": 0.0858971696998924,
|
| 871 |
-
"rewards/MultiModalMultiActionNoThinkNewAction": 1.404239798784256,
|
| 872 |
-
"step": 5300,
|
| 873 |
-
"train_speed(iter/s)": 0.093778
|
| 874 |
-
},
|
| 875 |
-
{
|
| 876 |
-
"clip_ratio": 0.0,
|
| 877 |
-
"completion_length": 9.4316796875,
|
| 878 |
-
"epoch": 0.6278339727938612,
|
| 879 |
-
"grad_norm": 12.403978855236371,
|
| 880 |
-
"kl": 0.1601806640625,
|
| 881 |
-
"learning_rate": 3.1006539553150724e-07,
|
| 882 |
-
"loss": 0.005034840106964112,
|
| 883 |
-
"memory(GiB)": 50.01,
|
| 884 |
-
"response_clip_ratio": 0.0,
|
| 885 |
-
"reward": 1.3801133859157562,
|
| 886 |
-
"reward_std": 0.0971566170360893,
|
| 887 |
-
"rewards/MultiModalMultiActionNoThinkNewAction": 1.3801133859157562,
|
| 888 |
-
"step": 5400,
|
| 889 |
-
"train_speed(iter/s)": 0.093788
|
| 890 |
-
},
|
| 891 |
-
{
|
| 892 |
-
"clip_ratio": 0.0,
|
| 893 |
-
"completion_length": 9.47546875,
|
| 894 |
-
"epoch": 0.6394605278455994,
|
| 895 |
-
"grad_norm": 13.196174592922217,
|
| 896 |
-
"kl": 0.1671484375,
|
| 897 |
-
"learning_rate": 2.931319624521521e-07,
|
| 898 |
-
"loss": 0.00633289098739624,
|
| 899 |
-
"memory(GiB)": 50.01,
|
| 900 |
-
"response_clip_ratio": 0.0,
|
| 901 |
-
"reward": 1.3824928963184357,
|
| 902 |
-
"reward_std": 0.09388638647273183,
|
| 903 |
-
"rewards/MultiModalMultiActionNoThinkNewAction": 1.3824928963184357,
|
| 904 |
-
"step": 5500,
|
| 905 |
-
"train_speed(iter/s)": 0.093796
|
| 906 |
-
},
|
| 907 |
-
{
|
| 908 |
-
"clip_ratio": 0.0,
|
| 909 |
-
"completion_length": 9.5144921875,
|
| 910 |
-
"epoch": 0.6510870828973375,
|
| 911 |
-
"grad_norm": 10.532572334095438,
|
| 912 |
-
"kl": 0.17203125,
|
| 913 |
-
"learning_rate": 2.7648015785309695e-07,
|
| 914 |
-
"loss": 0.004889653325080872,
|
| 915 |
-
"memory(GiB)": 50.01,
|
| 916 |
-
"response_clip_ratio": 0.0,
|
| 917 |
-
"reward": 1.4016336119174957,
|
| 918 |
-
"reward_std": 0.09526587199419737,
|
| 919 |
-
"rewards/MultiModalMultiActionNoThinkNewAction": 1.4016336119174957,
|
| 920 |
-
"step": 5600,
|
| 921 |
-
"train_speed(iter/s)": 0.09374
|
| 922 |
-
},
|
| 923 |
-
{
|
| 924 |
-
"clip_ratio": 0.0,
|
| 925 |
-
"completion_length": 9.5614453125,
|
| 926 |
-
"epoch": 0.6627136379490757,
|
| 927 |
-
"grad_norm": 12.720205008880969,
|
| 928 |
-
"kl": 0.16912109375,
|
| 929 |
-
"learning_rate": 2.60132651367017e-07,
|
| 930 |
-
"loss": 0.00633256196975708,
|
| 931 |
-
"memory(GiB)": 50.01,
|
| 932 |
-
"response_clip_ratio": 0.0,
|
| 933 |
-
"reward": 1.3703772777318954,
|
| 934 |
-
"reward_std": 0.09573625258170068,
|
| 935 |
-
"rewards/MultiModalMultiActionNoThinkNewAction": 1.3703772777318954,
|
| 936 |
-
"step": 5700,
|
| 937 |
-
"train_speed(iter/s)": 0.093756
|
| 938 |
-
},
|
| 939 |
-
{
|
| 940 |
-
"clip_ratio": 0.0,
|
| 941 |
-
"completion_length": 9.571796875,
|
| 942 |
-
"epoch": 0.6743401930008138,
|
| 943 |
-
"grad_norm": 6.872908071660086,
|
| 944 |
-
"kl": 0.177060546875,
|
| 945 |
-
"learning_rate": 2.441116983575876e-07,
|
| 946 |
-
"loss": 0.004491334557533264,
|
| 947 |
-
"memory(GiB)": 50.01,
|
| 948 |
-
"response_clip_ratio": 0.0,
|
| 949 |
-
"reward": 1.3781077253818512,
|
| 950 |
-
"reward_std": 0.0949486710689962,
|
| 951 |
-
"rewards/MultiModalMultiActionNoThinkNewAction": 1.3781077253818512,
|
| 952 |
-
"step": 5800,
|
| 953 |
-
"train_speed(iter/s)": 0.093772
|
| 954 |
-
},
|
| 955 |
-
{
|
| 956 |
-
"clip_ratio": 0.0,
|
| 957 |
-
"completion_length": 9.511953125,
|
| 958 |
-
"epoch": 0.685966748052552,
|
| 959 |
-
"grad_norm": 10.828437072129354,
|
| 960 |
-
"kl": 0.1688671875,
|
| 961 |
-
"learning_rate": 2.2843910962121028e-07,
|
| 962 |
-
"loss": 0.004997452199459076,
|
| 963 |
-
"memory(GiB)": 50.01,
|
| 964 |
-
"response_clip_ratio": 0.0,
|
| 965 |
-
"reward": 1.3701044476032258,
|
| 966 |
-
"reward_std": 0.08872982957400381,
|
| 967 |
-
"rewards/MultiModalMultiActionNoThinkNewAction": 1.3701044476032258,
|
| 968 |
-
"step": 5900,
|
| 969 |
-
"train_speed(iter/s)": 0.0938
|
| 970 |
-
},
|
| 971 |
-
{
|
| 972 |
-
"clip_ratio": 0.0,
|
| 973 |
-
"completion_length": 9.5623046875,
|
| 974 |
-
"epoch": 0.6975933031042902,
|
| 975 |
-
"grad_norm": 8.65544047927268,
|
| 976 |
-
"kl": 0.1734130859375,
|
| 977 |
-
"learning_rate": 2.1313622169397133e-07,
|
| 978 |
-
"loss": 0.005034286975860596,
|
| 979 |
-
"memory(GiB)": 50.01,
|
| 980 |
-
"response_clip_ratio": 0.0,
|
| 981 |
-
"reward": 1.376991518139839,
|
| 982 |
-
"reward_std": 0.08544575774110853,
|
| 983 |
-
"rewards/MultiModalMultiActionNoThinkNewAction": 1.376991518139839,
|
| 984 |
-
"step": 6000,
|
| 985 |
-
"train_speed(iter/s)": 0.09382
|
| 986 |
-
},
|
| 987 |
-
{
|
| 988 |
-
"clip_ratio": 0.0,
|
| 989 |
-
"completion_length": 9.5916796875,
|
| 990 |
-
"epoch": 0.7092198581560284,
|
| 991 |
-
"grad_norm": 10.832177815406336,
|
| 992 |
-
"kl": 0.1687158203125,
|
| 993 |
-
"learning_rate": 1.9822386780425242e-07,
|
| 994 |
-
"loss": 0.005690656900405884,
|
| 995 |
-
"memory(GiB)": 50.01,
|
| 996 |
-
"response_clip_ratio": 0.0,
|
| 997 |
-
"reward": 1.3775933372974396,
|
| 998 |
-
"reward_std": 0.09453948569484055,
|
| 999 |
-
"rewards/MultiModalMultiActionNoThinkNewAction": 1.3775933372974396,
|
| 1000 |
-
"step": 6100,
|
| 1001 |
-
"train_speed(iter/s)": 0.093779
|
| 1002 |
-
},
|
| 1003 |
-
{
|
| 1004 |
-
"clip_ratio": 0.0,
|
| 1005 |
-
"completion_length": 9.452265625,
|
| 1006 |
-
"epoch": 0.7208464132077665,
|
| 1007 |
-
"grad_norm": 9.5769032777987,
|
| 1008 |
-
"kl": 0.167353515625,
|
| 1009 |
-
"learning_rate": 1.8372234951054205e-07,
|
| 1010 |
-
"loss": 0.0064952802658081055,
|
| 1011 |
-
"memory(GiB)": 50.01,
|
| 1012 |
-
"response_clip_ratio": 0.0,
|
| 1013 |
-
"reward": 1.3917820090055466,
|
| 1014 |
-
"reward_std": 0.0956609710212797,
|
| 1015 |
-
"rewards/MultiModalMultiActionNoThinkNewAction": 1.3917820090055466,
|
| 1016 |
-
"step": 6200,
|
| 1017 |
-
"train_speed(iter/s)": 0.093802
|
| 1018 |
-
},
|
| 1019 |
-
{
|
| 1020 |
-
"clip_ratio": 0.0,
|
| 1021 |
-
"completion_length": 9.595390625,
|
| 1022 |
-
"epoch": 0.7324729682595047,
|
| 1023 |
-
"grad_norm": 13.490536213430035,
|
| 1024 |
-
"kl": 0.16693359375,
|
| 1025 |
-
"learning_rate": 1.696514090630571e-07,
|
| 1026 |
-
"loss": 0.006171835064888001,
|
| 1027 |
-
"memory(GiB)": 50.01,
|
| 1028 |
-
"response_clip_ratio": 0.0,
|
| 1029 |
-
"reward": 1.3438419443368912,
|
| 1030 |
-
"reward_std": 0.09881241522729396,
|
| 1031 |
-
"rewards/MultiModalMultiActionNoThinkNewAction": 1.3438419443368912,
|
| 1032 |
-
"step": 6300,
|
| 1033 |
-
"train_speed(iter/s)": 0.093826
|
| 1034 |
-
},
|
| 1035 |
-
{
|
| 1036 |
-
"clip_ratio": 0.0,
|
| 1037 |
-
"completion_length": 9.58125,
|
| 1038 |
-
"epoch": 0.7440995233112428,
|
| 1039 |
-
"grad_norm": 13.80425097201388,
|
| 1040 |
-
"kl": 0.1742724609375,
|
| 1041 |
-
"learning_rate": 1.560302025268016e-07,
|
| 1042 |
-
"loss": 0.004496442675590515,
|
| 1043 |
-
"memory(GiB)": 50.01,
|
| 1044 |
-
"response_clip_ratio": 0.0,
|
| 1045 |
-
"reward": 1.3792090493440627,
|
| 1046 |
-
"reward_std": 0.09534793998114764,
|
| 1047 |
-
"rewards/MultiModalMultiActionNoThinkNewAction": 1.3792090493440627,
|
| 1048 |
-
"step": 6400,
|
| 1049 |
-
"train_speed(iter/s)": 0.093832
|
| 1050 |
-
},
|
| 1051 |
-
{
|
| 1052 |
-
"clip_ratio": 0.0,
|
| 1053 |
-
"completion_length": 9.4980078125,
|
| 1054 |
-
"epoch": 0.755726078362981,
|
| 1055 |
-
"grad_norm": 12.228002609541004,
|
| 1056 |
-
"kl": 0.172158203125,
|
| 1057 |
-
"learning_rate": 1.4287727370265557e-07,
|
| 1058 |
-
"loss": 0.004531278908252716,
|
| 1059 |
-
"memory(GiB)": 50.01,
|
| 1060 |
-
"response_clip_ratio": 0.0,
|
| 1061 |
-
"reward": 1.3554240018129349,
|
| 1062 |
-
"reward_std": 0.0867443119455129,
|
| 1063 |
-
"rewards/MultiModalMultiActionNoThinkNewAction": 1.3554240018129349,
|
| 1064 |
-
"step": 6500,
|
| 1065 |
-
"train_speed(iter/s)": 0.093837
|
| 1066 |
-
},
|
| 1067 |
-
{
|
| 1068 |
-
"clip_ratio": 0.0,
|
| 1069 |
-
"completion_length": 9.415234375,
|
| 1070 |
-
"epoch": 0.7673526334147193,
|
| 1071 |
-
"grad_norm": 11.022738680450304,
|
| 1072 |
-
"kl": 0.164775390625,
|
| 1073 |
-
"learning_rate": 1.3021052888199273e-07,
|
| 1074 |
-
"loss": 0.006004486083984375,
|
| 1075 |
-
"memory(GiB)": 50.01,
|
| 1076 |
-
"response_clip_ratio": 0.0,
|
| 1077 |
-
"reward": 1.3786427634954452,
|
| 1078 |
-
"reward_std": 0.09436391741968692,
|
| 1079 |
-
"rewards/MultiModalMultiActionNoThinkNewAction": 1.3786427634954452,
|
| 1080 |
-
"step": 6600,
|
| 1081 |
-
"train_speed(iter/s)": 0.093792
|
| 1082 |
-
},
|
| 1083 |
-
{
|
| 1084 |
-
"clip_ratio": 0.0,
|
| 1085 |
-
"completion_length": 9.5644140625,
|
| 1086 |
-
"epoch": 0.7789791884664574,
|
| 1087 |
-
"grad_norm": 7.2361000289177095,
|
| 1088 |
-
"kl": 0.17333984375,
|
| 1089 |
-
"learning_rate": 1.1804721246919974e-07,
|
| 1090 |
-
"loss": 0.004695640504360199,
|
| 1091 |
-
"memory(GiB)": 50.01,
|
| 1092 |
-
"response_clip_ratio": 0.0,
|
| 1093 |
-
"reward": 1.383240016102791,
|
| 1094 |
-
"reward_std": 0.09347979573532939,
|
| 1095 |
-
"rewards/MultiModalMultiActionNoThinkNewAction": 1.383240016102791,
|
| 1096 |
-
"step": 6700,
|
| 1097 |
-
"train_speed(iter/s)": 0.093815
|
| 1098 |
-
},
|
| 1099 |
-
{
|
| 1100 |
-
"clip_ratio": 0.0,
|
| 1101 |
-
"completion_length": 9.3888671875,
|
| 1102 |
-
"epoch": 0.7906057435181956,
|
| 1103 |
-
"grad_norm": 10.627678410762373,
|
| 1104 |
-
"kl": 0.1702392578125,
|
| 1105 |
-
"learning_rate": 1.0640388350528345e-07,
|
| 1106 |
-
"loss": 0.006417045593261719,
|
| 1107 |
-
"memory(GiB)": 50.01,
|
| 1108 |
-
"response_clip_ratio": 0.0,
|
| 1109 |
-
"reward": 1.392098326086998,
|
| 1110 |
-
"reward_std": 0.08011089071165771,
|
| 1111 |
-
"rewards/MultiModalMultiActionNoThinkNewAction": 1.392098326086998,
|
| 1112 |
-
"step": 6800,
|
| 1113 |
-
"train_speed(iter/s)": 0.093838
|
| 1114 |
-
},
|
| 1115 |
-
{
|
| 1116 |
-
"clip_ratio": 0.0,
|
| 1117 |
-
"completion_length": 9.5668359375,
|
| 1118 |
-
"epoch": 0.8022322985699337,
|
| 1119 |
-
"grad_norm": 15.465333954982329,
|
| 1120 |
-
"kl": 0.170009765625,
|
| 1121 |
-
"learning_rate": 9.529639312452559e-08,
|
| 1122 |
-
"loss": 0.005341029167175293,
|
| 1123 |
-
"memory(GiB)": 50.01,
|
| 1124 |
-
"response_clip_ratio": 0.0,
|
| 1125 |
-
"reward": 1.4103717935085296,
|
| 1126 |
-
"reward_std": 0.09352740393951535,
|
| 1127 |
-
"rewards/MultiModalMultiActionNoThinkNewAction": 1.4103717935085296,
|
| 1128 |
-
"step": 6900,
|
| 1129 |
-
"train_speed(iter/s)": 0.093849
|
| 1130 |
-
},
|
| 1131 |
-
{
|
| 1132 |
-
"clip_ratio": 0.0,
|
| 1133 |
-
"completion_length": 9.5197265625,
|
| 1134 |
-
"epoch": 0.8138588536216719,
|
| 1135 |
-
"grad_norm": 14.110343993379132,
|
| 1136 |
-
"kl": 0.1599560546875,
|
| 1137 |
-
"learning_rate": 8.473986297487578e-08,
|
| 1138 |
-
"loss": 0.004906933009624481,
|
| 1139 |
-
"memory(GiB)": 50.01,
|
| 1140 |
-
"response_clip_ratio": 0.0,
|
| 1141 |
-
"reward": 1.4088983750343322,
|
| 1142 |
-
"reward_std": 0.08588065375573933,
|
| 1143 |
-
"rewards/MultiModalMultiActionNoThinkNewAction": 1.4088983750343322,
|
| 1144 |
-
"step": 7000,
|
| 1145 |
-
"train_speed(iter/s)": 0.093861
|
| 1146 |
-
},
|
| 1147 |
-
{
|
| 1148 |
-
"clip_ratio": 0.0,
|
| 1149 |
-
"completion_length": 9.4692578125,
|
| 1150 |
-
"epoch": 0.82548540867341,
|
| 1151 |
-
"grad_norm": 8.408536596023637,
|
| 1152 |
-
"kl": 0.1673828125,
|
| 1153 |
-
"learning_rate": 7.474866463146251e-08,
|
| 1154 |
-
"loss": 0.005006779432296753,
|
| 1155 |
-
"memory(GiB)": 50.01,
|
| 1156 |
-
"response_clip_ratio": 0.0,
|
| 1157 |
-
"reward": 1.4262621313333512,
|
| 1158 |
-
"reward_std": 0.08552982007153333,
|
| 1159 |
-
"rewards/MultiModalMultiActionNoThinkNewAction": 1.4262621313333512,
|
| 1160 |
-
"step": 7100,
|
| 1161 |
-
"train_speed(iter/s)": 0.093824
|
| 1162 |
-
},
|
| 1163 |
-
{
|
| 1164 |
-
"clip_ratio": 0.0,
|
| 1165 |
-
"completion_length": 9.4380859375,
|
| 1166 |
-
"epoch": 0.8371119637251483,
|
| 1167 |
-
"grad_norm": 15.805070345572961,
|
| 1168 |
-
"kl": 0.1696728515625,
|
| 1169 |
-
"learning_rate": 6.53364000312463e-08,
|
| 1170 |
-
"loss": 0.005111285448074341,
|
| 1171 |
-
"memory(GiB)": 50.01,
|
| 1172 |
-
"response_clip_ratio": 0.0,
|
| 1173 |
-
"reward": 1.4090092825889586,
|
| 1174 |
-
"reward_std": 0.08110610948875546,
|
| 1175 |
-
"rewards/MultiModalMultiActionNoThinkNewAction": 1.4090092825889586,
|
| 1176 |
-
"step": 7200,
|
| 1177 |
-
"train_speed(iter/s)": 0.093851
|
| 1178 |
-
},
|
| 1179 |
-
{
|
| 1180 |
-
"clip_ratio": 0.0,
|
| 1181 |
-
"completion_length": 9.5043359375,
|
| 1182 |
-
"epoch": 0.8487385187768864,
|
| 1183 |
-
"grad_norm": 9.57328394904971,
|
| 1184 |
-
"kl": 0.161328125,
|
| 1185 |
-
"learning_rate": 5.651588295545168e-08,
|
| 1186 |
-
"loss": 0.004713370501995087,
|
| 1187 |
-
"memory(GiB)": 50.01,
|
| 1188 |
-
"response_clip_ratio": 0.0,
|
| 1189 |
-
"reward": 1.3688128244876863,
|
| 1190 |
-
"reward_std": 0.08936454836279153,
|
| 1191 |
-
"rewards/MultiModalMultiActionNoThinkNewAction": 1.3688128244876863,
|
| 1192 |
-
"step": 7300,
|
| 1193 |
-
"train_speed(iter/s)": 0.093859
|
| 1194 |
-
},
|
| 1195 |
-
{
|
| 1196 |
-
"clip_ratio": 0.0,
|
| 1197 |
-
"completion_length": 9.4918359375,
|
| 1198 |
-
"epoch": 0.8603650738286246,
|
| 1199 |
-
"grad_norm": 13.141769298737369,
|
| 1200 |
-
"kl": 0.1631298828125,
|
| 1201 |
-
"learning_rate": 4.8299121584989445e-08,
|
| 1202 |
-
"loss": 0.005130698680877685,
|
| 1203 |
-
"memory(GiB)": 50.01,
|
| 1204 |
-
"response_clip_ratio": 0.0,
|
| 1205 |
-
"reward": 1.400255571603775,
|
| 1206 |
-
"reward_std": 0.09085271211341023,
|
| 1207 |
-
"rewards/MultiModalMultiActionNoThinkNewAction": 1.400255571603775,
|
| 1208 |
-
"step": 7400,
|
| 1209 |
-
"train_speed(iter/s)": 0.093861
|
| 1210 |
-
},
|
| 1211 |
-
{
|
| 1212 |
-
"clip_ratio": 0.0,
|
| 1213 |
-
"completion_length": 9.4555859375,
|
| 1214 |
-
"epoch": 0.8719916288803627,
|
| 1215 |
-
"grad_norm": 9.746601959968274,
|
| 1216 |
-
"kl": 0.1671728515625,
|
| 1217 |
-
"learning_rate": 4.069730215261552e-08,
|
| 1218 |
-
"loss": 0.005373966693878174,
|
| 1219 |
-
"memory(GiB)": 50.01,
|
| 1220 |
-
"response_clip_ratio": 0.0,
|
| 1221 |
-
"reward": 1.4106507396697998,
|
| 1222 |
-
"reward_std": 0.08433111377526074,
|
| 1223 |
-
"rewards/MultiModalMultiActionNoThinkNewAction": 1.4106507396697998,
|
| 1224 |
-
"step": 7500,
|
| 1225 |
-
"train_speed(iter/s)": 0.093864
|
| 1226 |
-
},
|
| 1227 |
-
{
|
| 1228 |
-
"clip_ratio": 0.0,
|
| 1229 |
-
"completion_length": 9.479375,
|
| 1230 |
-
"epoch": 0.8836181839321009,
|
| 1231 |
-
"grad_norm": 12.304794577190247,
|
| 1232 |
-
"kl": 0.166220703125,
|
| 1233 |
-
"learning_rate": 3.372077371408361e-08,
|
| 1234 |
-
"loss": 0.004605398774147034,
|
| 1235 |
-
"memory(GiB)": 50.01,
|
| 1236 |
-
"response_clip_ratio": 0.0,
|
| 1237 |
-
"reward": 1.4062111765146255,
|
| 1238 |
-
"reward_std": 0.08385045255534351,
|
| 1239 |
-
"rewards/MultiModalMultiActionNoThinkNewAction": 1.4062111765146255,
|
| 1240 |
-
"step": 7600,
|
| 1241 |
-
"train_speed(iter/s)": 0.093825
|
| 1242 |
-
},
|
| 1243 |
-
{
|
| 1244 |
-
"clip_ratio": 0.0,
|
| 1245 |
-
"completion_length": 9.4255078125,
|
| 1246 |
-
"epoch": 0.895244738983839,
|
| 1247 |
-
"grad_norm": 10.788517373431308,
|
| 1248 |
-
"kl": 0.1640380859375,
|
| 1249 |
-
"learning_rate": 2.7379034059024085e-08,
|
| 1250 |
-
"loss": 0.005341172218322754,
|
| 1251 |
-
"memory(GiB)": 50.01,
|
| 1252 |
-
"response_clip_ratio": 0.0,
|
| 1253 |
-
"reward": 1.400910457968712,
|
| 1254 |
-
"reward_std": 0.09175728805363179,
|
| 1255 |
-
"rewards/MultiModalMultiActionNoThinkNewAction": 1.400910457968712,
|
| 1256 |
-
"step": 7700,
|
| 1257 |
-
"train_speed(iter/s)": 0.093832
|
| 1258 |
-
},
|
| 1259 |
-
{
|
| 1260 |
-
"clip_ratio": 0.0,
|
| 1261 |
-
"completion_length": 9.557265625,
|
| 1262 |
-
"epoch": 0.9068712940355773,
|
| 1263 |
-
"grad_norm": 15.69745045067331,
|
| 1264 |
-
"kl": 0.1719189453125,
|
| 1265 |
-
"learning_rate": 2.1680716780730136e-08,
|
| 1266 |
-
"loss": 0.0057187509536743165,
|
| 1267 |
-
"memory(GiB)": 50.01,
|
| 1268 |
-
"response_clip_ratio": 0.0,
|
| 1269 |
-
"reward": 1.4035167163610458,
|
| 1270 |
-
"reward_std": 0.09192508255131543,
|
| 1271 |
-
"rewards/MultiModalMultiActionNoThinkNewAction": 1.4035167163610458,
|
| 1272 |
-
"step": 7800,
|
| 1273 |
-
"train_speed(iter/s)": 0.093839
|
| 1274 |
-
},
|
| 1275 |
-
{
|
| 1276 |
-
"clip_ratio": 0.0,
|
| 1277 |
-
"completion_length": 9.5982421875,
|
| 1278 |
-
"epoch": 0.9184978490873155,
|
| 1279 |
-
"grad_norm": 9.828379408967727,
|
| 1280 |
-
"kl": 0.167861328125,
|
| 1281 |
-
"learning_rate": 1.6633579522452923e-08,
|
| 1282 |
-
"loss": 0.004599595665931702,
|
| 1283 |
-
"memory(GiB)": 50.01,
|
| 1284 |
-
"response_clip_ratio": 0.0,
|
| 1285 |
-
"reward": 1.4050581753253937,
|
| 1286 |
-
"reward_std": 0.08899507346563042,
|
| 1287 |
-
"rewards/MultiModalMultiActionNoThinkNewAction": 1.4050581753253937,
|
| 1288 |
-
"step": 7900,
|
| 1289 |
-
"train_speed(iter/s)": 0.093848
|
| 1290 |
-
},
|
| 1291 |
-
{
|
| 1292 |
-
"clip_ratio": 0.0,
|
| 1293 |
-
"completion_length": 9.4523828125,
|
| 1294 |
-
"epoch": 0.9301244041390536,
|
| 1295 |
-
"grad_norm": 7.564846175327,
|
| 1296 |
-
"kl": 0.165205078125,
|
| 1297 |
-
"learning_rate": 1.2244493416208424e-08,
|
| 1298 |
-
"loss": 0.006511397361755371,
|
| 1299 |
-
"memory(GiB)": 50.01,
|
| 1300 |
-
"response_clip_ratio": 0.0,
|
| 1301 |
-
"reward": 1.3674461019039155,
|
| 1302 |
-
"reward_std": 0.09354917639866471,
|
| 1303 |
-
"rewards/MultiModalMultiActionNoThinkNewAction": 1.3674461019039155,
|
| 1304 |
-
"step": 8000,
|
| 1305 |
-
"train_speed(iter/s)": 0.093866
|
| 1306 |
-
},
|
| 1307 |
-
{
|
| 1308 |
-
"clip_ratio": 0.0,
|
| 1309 |
-
"completion_length": 9.441796875,
|
| 1310 |
-
"epoch": 0.9417509591907918,
|
| 1311 |
-
"grad_norm": 8.473570999026597,
|
| 1312 |
-
"kl": 0.166337890625,
|
| 1313 |
-
"learning_rate": 8.51943372847419e-09,
|
| 1314 |
-
"loss": 0.004957606792449951,
|
| 1315 |
-
"memory(GiB)": 50.01,
|
| 1316 |
-
"response_clip_ratio": 0.0,
|
| 1317 |
-
"reward": 1.3664843720197677,
|
| 1318 |
-
"reward_std": 0.09796678048558533,
|
| 1319 |
-
"rewards/MultiModalMultiActionNoThinkNewAction": 1.3664843720197677,
|
| 1320 |
-
"step": 8100,
|
| 1321 |
-
"train_speed(iter/s)": 0.093827
|
| 1322 |
-
},
|
| 1323 |
-
{
|
| 1324 |
-
"clip_ratio": 0.0,
|
| 1325 |
-
"completion_length": 9.4091796875,
|
| 1326 |
-
"epoch": 0.9533775142425299,
|
| 1327 |
-
"grad_norm": 17.38564208248951,
|
| 1328 |
-
"kl": 0.1664453125,
|
| 1329 |
-
"learning_rate": 5.463471725509206e-09,
|
| 1330 |
-
"loss": 0.005511881113052368,
|
| 1331 |
-
"memory(GiB)": 50.01,
|
| 1332 |
-
"response_clip_ratio": 0.0,
|
| 1333 |
-
"reward": 1.3864123088121414,
|
| 1334 |
-
"reward_std": 0.09022373898886143,
|
| 1335 |
-
"rewards/MultiModalMultiActionNoThinkNewAction": 1.3864123088121414,
|
| 1336 |
-
"step": 8200,
|
| 1337 |
-
"train_speed(iter/s)": 0.093846
|
| 1338 |
-
},
|
| 1339 |
-
{
|
| 1340 |
-
"clip_ratio": 0.0,
|
| 1341 |
-
"completion_length": 9.52640625,
|
| 1342 |
-
"epoch": 0.9650040692942681,
|
| 1343 |
-
"grad_norm": 12.073204680109393,
|
| 1344 |
-
"kl": 0.1724462890625,
|
| 1345 |
-
"learning_rate": 3.0807677693729385e-09,
|
| 1346 |
-
"loss": 0.006146684885025024,
|
| 1347 |
-
"memory(GiB)": 50.01,
|
| 1348 |
-
"response_clip_ratio": 0.0,
|
| 1349 |
-
"reward": 1.3671981352567673,
|
| 1350 |
-
"reward_std": 0.09982930341735483,
|
| 1351 |
-
"rewards/MultiModalMultiActionNoThinkNewAction": 1.3671981352567673,
|
| 1352 |
-
"step": 8300,
|
| 1353 |
-
"train_speed(iter/s)": 0.093858
|
| 1354 |
-
},
|
| 1355 |
-
{
|
| 1356 |
-
"clip_ratio": 0.0,
|
| 1357 |
-
"completion_length": 9.4923828125,
|
| 1358 |
-
"epoch": 0.9766306243460062,
|
| 1359 |
-
"grad_norm": 14.136069122546115,
|
| 1360 |
-
"kl": 0.164794921875,
|
| 1361 |
-
"learning_rate": 1.3745656540422036e-09,
|
| 1362 |
-
"loss": 0.0051899772882461545,
|
| 1363 |
-
"memory(GiB)": 50.01,
|
| 1364 |
-
"response_clip_ratio": 0.0,
|
| 1365 |
-
"reward": 1.3862751299142837,
|
| 1366 |
-
"reward_std": 0.09112708026543259,
|
| 1367 |
-
"rewards/MultiModalMultiActionNoThinkNewAction": 1.3862751299142837,
|
| 1368 |
-
"step": 8400,
|
| 1369 |
-
"train_speed(iter/s)": 0.093871
|
| 1370 |
-
},
|
| 1371 |
-
{
|
| 1372 |
-
"clip_ratio": 0.0,
|
| 1373 |
-
"completion_length": 9.5616015625,
|
| 1374 |
-
"epoch": 0.9882571793977445,
|
| 1375 |
-
"grad_norm": 7.965510618669221,
|
| 1376 |
-
"kl": 0.168076171875,
|
| 1377 |
-
"learning_rate": 3.4718818933582305e-10,
|
| 1378 |
-
"loss": 0.0061265045404434205,
|
| 1379 |
-
"memory(GiB)": 50.01,
|
| 1380 |
-
"response_clip_ratio": 0.0,
|
| 1381 |
-
"reward": 1.3907999232411385,
|
| 1382 |
-
"reward_std": 0.08998070719651878,
|
| 1383 |
-
"rewards/MultiModalMultiActionNoThinkNewAction": 1.3907999232411385,
|
| 1384 |
-
"step": 8500,
|
| 1385 |
-
"train_speed(iter/s)": 0.093882
|
| 1386 |
-
}
|
| 1387 |
-
],
|
| 1388 |
-
"logging_steps": 100,
|
| 1389 |
-
"max_steps": 8601,
|
| 1390 |
-
"num_input_tokens_seen": 0,
|
| 1391 |
-
"num_train_epochs": 1,
|
| 1392 |
-
"save_steps": 500,
|
| 1393 |
-
"stateful_callbacks": {
|
| 1394 |
-
"TrainerControl": {
|
| 1395 |
-
"args": {
|
| 1396 |
-
"should_epoch_stop": false,
|
| 1397 |
-
"should_evaluate": false,
|
| 1398 |
-
"should_log": false,
|
| 1399 |
-
"should_save": true,
|
| 1400 |
-
"should_training_stop": false
|
| 1401 |
-
},
|
| 1402 |
-
"attributes": {}
|
| 1403 |
-
}
|
| 1404 |
-
},
|
| 1405 |
-
"total_flos": 0.0,
|
| 1406 |
-
"train_batch_size": 8,
|
| 1407 |
-
"trial_name": null,
|
| 1408 |
-
"trial_params": null
|
| 1409 |
-
}
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|