Delete trainer_state.json
Browse files- trainer_state.json +0 -1465
trainer_state.json
DELETED
|
@@ -1,1465 +0,0 @@
|
|
| 1 |
-
{
|
| 2 |
-
"best_metric": null,
|
| 3 |
-
"best_model_checkpoint": null,
|
| 4 |
-
"epoch": 1.0,
|
| 5 |
-
"eval_steps": 500,
|
| 6 |
-
"global_step": 1000,
|
| 7 |
-
"is_hyper_param_search": false,
|
| 8 |
-
"is_local_process_zero": true,
|
| 9 |
-
"is_world_process_zero": true,
|
| 10 |
-
"log_history": [
|
| 11 |
-
{
|
| 12 |
-
"epoch": 0.0,
|
| 13 |
-
"learning_rate": 3.3333333333333334e-09,
|
| 14 |
-
"logits/chosen": -2.08943510055542,
|
| 15 |
-
"logits/rejected": -2.0892438888549805,
|
| 16 |
-
"logps/chosen": -0.08784887194633484,
|
| 17 |
-
"logps/rejected": -0.08645449578762054,
|
| 18 |
-
"loss": 0.6931,
|
| 19 |
-
"rewards/accuracies": 0.0,
|
| 20 |
-
"rewards/chosen": 0.0,
|
| 21 |
-
"rewards/margins": 0.0,
|
| 22 |
-
"rewards/rejected": 0.0,
|
| 23 |
-
"step": 1
|
| 24 |
-
},
|
| 25 |
-
{
|
| 26 |
-
"epoch": 0.01,
|
| 27 |
-
"learning_rate": 3.3333333333333334e-08,
|
| 28 |
-
"logits/chosen": -2.4944796562194824,
|
| 29 |
-
"logits/rejected": -2.4945976734161377,
|
| 30 |
-
"logps/chosen": -0.578792154788971,
|
| 31 |
-
"logps/rejected": -0.5907332897186279,
|
| 32 |
-
"loss": 0.6921,
|
| 33 |
-
"rewards/accuracies": 0.5555555820465088,
|
| 34 |
-
"rewards/chosen": 0.0014628150966018438,
|
| 35 |
-
"rewards/margins": 0.0020117585081607103,
|
| 36 |
-
"rewards/rejected": -0.0005489434115588665,
|
| 37 |
-
"step": 10
|
| 38 |
-
},
|
| 39 |
-
{
|
| 40 |
-
"epoch": 0.02,
|
| 41 |
-
"learning_rate": 6.666666666666667e-08,
|
| 42 |
-
"logits/chosen": -2.3721718788146973,
|
| 43 |
-
"logits/rejected": -2.468485116958618,
|
| 44 |
-
"logps/chosen": -1.28400719165802,
|
| 45 |
-
"logps/rejected": -12.984468460083008,
|
| 46 |
-
"loss": 0.6933,
|
| 47 |
-
"rewards/accuracies": 0.699999988079071,
|
| 48 |
-
"rewards/chosen": -0.0019450436811894178,
|
| 49 |
-
"rewards/margins": -0.0002596294798422605,
|
| 50 |
-
"rewards/rejected": -0.001685414114035666,
|
| 51 |
-
"step": 20
|
| 52 |
-
},
|
| 53 |
-
{
|
| 54 |
-
"epoch": 0.03,
|
| 55 |
-
"learning_rate": 1e-07,
|
| 56 |
-
"logits/chosen": -2.164445400238037,
|
| 57 |
-
"logits/rejected": -2.425915002822876,
|
| 58 |
-
"logps/chosen": -1.7286256551742554,
|
| 59 |
-
"logps/rejected": -25.388479232788086,
|
| 60 |
-
"loss": 0.6906,
|
| 61 |
-
"rewards/accuracies": 0.4000000059604645,
|
| 62 |
-
"rewards/chosen": -0.0019130244618281722,
|
| 63 |
-
"rewards/margins": 0.005339598748832941,
|
| 64 |
-
"rewards/rejected": -0.0072526223957538605,
|
| 65 |
-
"step": 30
|
| 66 |
-
},
|
| 67 |
-
{
|
| 68 |
-
"epoch": 0.04,
|
| 69 |
-
"learning_rate": 1.3333333333333334e-07,
|
| 70 |
-
"logits/chosen": -2.619248867034912,
|
| 71 |
-
"logits/rejected": -2.619331121444702,
|
| 72 |
-
"logps/chosen": -1.3749217987060547,
|
| 73 |
-
"logps/rejected": -1.3713041543960571,
|
| 74 |
-
"loss": 0.6937,
|
| 75 |
-
"rewards/accuracies": 0.5,
|
| 76 |
-
"rewards/chosen": 0.0021189216058701277,
|
| 77 |
-
"rewards/margins": -0.001049370621331036,
|
| 78 |
-
"rewards/rejected": 0.0031682923436164856,
|
| 79 |
-
"step": 40
|
| 80 |
-
},
|
| 81 |
-
{
|
| 82 |
-
"epoch": 0.05,
|
| 83 |
-
"learning_rate": 1.6666666666666665e-07,
|
| 84 |
-
"logits/chosen": -2.3369224071502686,
|
| 85 |
-
"logits/rejected": -2.431098461151123,
|
| 86 |
-
"logps/chosen": -1.0188841819763184,
|
| 87 |
-
"logps/rejected": -8.812786102294922,
|
| 88 |
-
"loss": 0.6923,
|
| 89 |
-
"rewards/accuracies": 0.5,
|
| 90 |
-
"rewards/chosen": 0.013968966901302338,
|
| 91 |
-
"rewards/margins": 0.001758316531777382,
|
| 92 |
-
"rewards/rejected": 0.01221065130084753,
|
| 93 |
-
"step": 50
|
| 94 |
-
},
|
| 95 |
-
{
|
| 96 |
-
"epoch": 0.06,
|
| 97 |
-
"learning_rate": 2e-07,
|
| 98 |
-
"logits/chosen": -2.284961700439453,
|
| 99 |
-
"logits/rejected": -2.494086742401123,
|
| 100 |
-
"logps/chosen": -0.683056116104126,
|
| 101 |
-
"logps/rejected": -19.27627182006836,
|
| 102 |
-
"loss": 0.6851,
|
| 103 |
-
"rewards/accuracies": 0.5,
|
| 104 |
-
"rewards/chosen": -0.0018667631084099412,
|
| 105 |
-
"rewards/margins": 0.016477219760417938,
|
| 106 |
-
"rewards/rejected": -0.018343981355428696,
|
| 107 |
-
"step": 60
|
| 108 |
-
},
|
| 109 |
-
{
|
| 110 |
-
"epoch": 0.07,
|
| 111 |
-
"learning_rate": 2.3333333333333333e-07,
|
| 112 |
-
"logits/chosen": -2.2826738357543945,
|
| 113 |
-
"logits/rejected": -2.384627103805542,
|
| 114 |
-
"logps/chosen": -0.2940216660499573,
|
| 115 |
-
"logps/rejected": -9.099252700805664,
|
| 116 |
-
"loss": 0.684,
|
| 117 |
-
"rewards/accuracies": 0.699999988079071,
|
| 118 |
-
"rewards/chosen": 0.005558961536735296,
|
| 119 |
-
"rewards/margins": 0.01876726932823658,
|
| 120 |
-
"rewards/rejected": -0.013208307325839996,
|
| 121 |
-
"step": 70
|
| 122 |
-
},
|
| 123 |
-
{
|
| 124 |
-
"epoch": 0.08,
|
| 125 |
-
"learning_rate": 2.6666666666666667e-07,
|
| 126 |
-
"logits/chosen": -2.477116107940674,
|
| 127 |
-
"logits/rejected": -2.613649606704712,
|
| 128 |
-
"logps/chosen": -1.0611478090286255,
|
| 129 |
-
"logps/rejected": -12.122793197631836,
|
| 130 |
-
"loss": 0.6836,
|
| 131 |
-
"rewards/accuracies": 0.4000000059604645,
|
| 132 |
-
"rewards/chosen": 0.018232952803373337,
|
| 133 |
-
"rewards/margins": 0.019915319979190826,
|
| 134 |
-
"rewards/rejected": -0.0016823671758174896,
|
| 135 |
-
"step": 80
|
| 136 |
-
},
|
| 137 |
-
{
|
| 138 |
-
"epoch": 0.09,
|
| 139 |
-
"learning_rate": 3e-07,
|
| 140 |
-
"logits/chosen": -2.6183485984802246,
|
| 141 |
-
"logits/rejected": -2.618346691131592,
|
| 142 |
-
"logps/chosen": -1.2046842575073242,
|
| 143 |
-
"logps/rejected": -1.2092653512954712,
|
| 144 |
-
"loss": 0.6924,
|
| 145 |
-
"rewards/accuracies": 0.4000000059604645,
|
| 146 |
-
"rewards/chosen": 0.012857586145401001,
|
| 147 |
-
"rewards/margins": 0.0014548657927662134,
|
| 148 |
-
"rewards/rejected": 0.011402720585465431,
|
| 149 |
-
"step": 90
|
| 150 |
-
},
|
| 151 |
-
{
|
| 152 |
-
"epoch": 0.1,
|
| 153 |
-
"learning_rate": 3.333333333333333e-07,
|
| 154 |
-
"logits/chosen": -2.152963161468506,
|
| 155 |
-
"logits/rejected": -2.5711560249328613,
|
| 156 |
-
"logps/chosen": -1.9411561489105225,
|
| 157 |
-
"logps/rejected": -47.01171112060547,
|
| 158 |
-
"loss": 0.6651,
|
| 159 |
-
"rewards/accuracies": 0.4000000059604645,
|
| 160 |
-
"rewards/chosen": -0.003847984131425619,
|
| 161 |
-
"rewards/margins": 0.0610489621758461,
|
| 162 |
-
"rewards/rejected": -0.06489694863557816,
|
| 163 |
-
"step": 100
|
| 164 |
-
},
|
| 165 |
-
{
|
| 166 |
-
"epoch": 0.11,
|
| 167 |
-
"learning_rate": 3.666666666666666e-07,
|
| 168 |
-
"logits/chosen": -2.353605270385742,
|
| 169 |
-
"logits/rejected": -2.5540683269500732,
|
| 170 |
-
"logps/chosen": -1.2066593170166016,
|
| 171 |
-
"logps/rejected": -30.185583114624023,
|
| 172 |
-
"loss": 0.682,
|
| 173 |
-
"rewards/accuracies": 0.4000000059604645,
|
| 174 |
-
"rewards/chosen": 0.003402869449928403,
|
| 175 |
-
"rewards/margins": 0.02602914348244667,
|
| 176 |
-
"rewards/rejected": -0.022626275196671486,
|
| 177 |
-
"step": 110
|
| 178 |
-
},
|
| 179 |
-
{
|
| 180 |
-
"epoch": 0.12,
|
| 181 |
-
"learning_rate": 4e-07,
|
| 182 |
-
"logits/chosen": -2.452103853225708,
|
| 183 |
-
"logits/rejected": -2.4946463108062744,
|
| 184 |
-
"logps/chosen": -0.47767752408981323,
|
| 185 |
-
"logps/rejected": -5.05490255355835,
|
| 186 |
-
"loss": 0.6859,
|
| 187 |
-
"rewards/accuracies": 0.699999988079071,
|
| 188 |
-
"rewards/chosen": 0.017536822706460953,
|
| 189 |
-
"rewards/margins": 0.014879658818244934,
|
| 190 |
-
"rewards/rejected": 0.0026571624912321568,
|
| 191 |
-
"step": 120
|
| 192 |
-
},
|
| 193 |
-
{
|
| 194 |
-
"epoch": 0.13,
|
| 195 |
-
"learning_rate": 4.3333333333333335e-07,
|
| 196 |
-
"logits/chosen": -2.2694547176361084,
|
| 197 |
-
"logits/rejected": -2.3567841053009033,
|
| 198 |
-
"logps/chosen": -1.3494905233383179,
|
| 199 |
-
"logps/rejected": -12.111764907836914,
|
| 200 |
-
"loss": 0.6465,
|
| 201 |
-
"rewards/accuracies": 0.5,
|
| 202 |
-
"rewards/chosen": 0.028263837099075317,
|
| 203 |
-
"rewards/margins": 0.11370788514614105,
|
| 204 |
-
"rewards/rejected": -0.08544405549764633,
|
| 205 |
-
"step": 130
|
| 206 |
-
},
|
| 207 |
-
{
|
| 208 |
-
"epoch": 0.14,
|
| 209 |
-
"learning_rate": 4.6666666666666666e-07,
|
| 210 |
-
"logits/chosen": -2.3862383365631104,
|
| 211 |
-
"logits/rejected": -2.522015333175659,
|
| 212 |
-
"logps/chosen": -0.9776835441589355,
|
| 213 |
-
"logps/rejected": -11.195852279663086,
|
| 214 |
-
"loss": 0.6435,
|
| 215 |
-
"rewards/accuracies": 0.4000000059604645,
|
| 216 |
-
"rewards/chosen": 0.016874078661203384,
|
| 217 |
-
"rewards/margins": 0.1495019793510437,
|
| 218 |
-
"rewards/rejected": -0.13262788951396942,
|
| 219 |
-
"step": 140
|
| 220 |
-
},
|
| 221 |
-
{
|
| 222 |
-
"epoch": 0.15,
|
| 223 |
-
"learning_rate": 5e-07,
|
| 224 |
-
"logits/chosen": -2.1118805408477783,
|
| 225 |
-
"logits/rejected": -2.4661762714385986,
|
| 226 |
-
"logps/chosen": -0.8852203488349915,
|
| 227 |
-
"logps/rejected": -36.519493103027344,
|
| 228 |
-
"loss": 0.594,
|
| 229 |
-
"rewards/accuracies": 0.8999999761581421,
|
| 230 |
-
"rewards/chosen": 0.016499249264597893,
|
| 231 |
-
"rewards/margins": 0.2467312514781952,
|
| 232 |
-
"rewards/rejected": -0.23023200035095215,
|
| 233 |
-
"step": 150
|
| 234 |
-
},
|
| 235 |
-
{
|
| 236 |
-
"epoch": 0.16,
|
| 237 |
-
"learning_rate": 4.941176470588235e-07,
|
| 238 |
-
"logits/chosen": -2.251331090927124,
|
| 239 |
-
"logits/rejected": -2.5972962379455566,
|
| 240 |
-
"logps/chosen": -1.6465107202529907,
|
| 241 |
-
"logps/rejected": -26.661739349365234,
|
| 242 |
-
"loss": 0.5775,
|
| 243 |
-
"rewards/accuracies": 0.6000000238418579,
|
| 244 |
-
"rewards/chosen": 0.06401287019252777,
|
| 245 |
-
"rewards/margins": 0.3227430284023285,
|
| 246 |
-
"rewards/rejected": -0.2587301731109619,
|
| 247 |
-
"step": 160
|
| 248 |
-
},
|
| 249 |
-
{
|
| 250 |
-
"epoch": 0.17,
|
| 251 |
-
"learning_rate": 4.88235294117647e-07,
|
| 252 |
-
"logits/chosen": -2.400430202484131,
|
| 253 |
-
"logits/rejected": -2.400381565093994,
|
| 254 |
-
"logps/chosen": -0.6768835783004761,
|
| 255 |
-
"logps/rejected": -0.677842915058136,
|
| 256 |
-
"loss": 0.6929,
|
| 257 |
-
"rewards/accuracies": 0.30000001192092896,
|
| 258 |
-
"rewards/chosen": 0.12107320874929428,
|
| 259 |
-
"rewards/margins": 0.0004425995866768062,
|
| 260 |
-
"rewards/rejected": 0.12063062191009521,
|
| 261 |
-
"step": 170
|
| 262 |
-
},
|
| 263 |
-
{
|
| 264 |
-
"epoch": 0.18,
|
| 265 |
-
"learning_rate": 4.823529411764705e-07,
|
| 266 |
-
"logits/chosen": -2.3488998413085938,
|
| 267 |
-
"logits/rejected": -2.3663113117218018,
|
| 268 |
-
"logps/chosen": -0.6453405618667603,
|
| 269 |
-
"logps/rejected": -2.275538444519043,
|
| 270 |
-
"loss": 0.67,
|
| 271 |
-
"rewards/accuracies": 0.5,
|
| 272 |
-
"rewards/chosen": 0.039185039699077606,
|
| 273 |
-
"rewards/margins": 0.05349387601017952,
|
| 274 |
-
"rewards/rejected": -0.014308829791843891,
|
| 275 |
-
"step": 180
|
| 276 |
-
},
|
| 277 |
-
{
|
| 278 |
-
"epoch": 0.19,
|
| 279 |
-
"learning_rate": 4.7647058823529405e-07,
|
| 280 |
-
"logits/chosen": -2.1920523643493652,
|
| 281 |
-
"logits/rejected": -2.5244088172912598,
|
| 282 |
-
"logps/chosen": -0.8233627080917358,
|
| 283 |
-
"logps/rejected": -37.117218017578125,
|
| 284 |
-
"loss": 0.5116,
|
| 285 |
-
"rewards/accuracies": 0.8999999761581421,
|
| 286 |
-
"rewards/chosen": 0.05563921853899956,
|
| 287 |
-
"rewards/margins": 0.5942346453666687,
|
| 288 |
-
"rewards/rejected": -0.53859543800354,
|
| 289 |
-
"step": 190
|
| 290 |
-
},
|
| 291 |
-
{
|
| 292 |
-
"epoch": 0.2,
|
| 293 |
-
"learning_rate": 4.705882352941176e-07,
|
| 294 |
-
"logits/chosen": -2.3363966941833496,
|
| 295 |
-
"logits/rejected": -2.336390495300293,
|
| 296 |
-
"logps/chosen": -0.5674120187759399,
|
| 297 |
-
"logps/rejected": -0.5631183385848999,
|
| 298 |
-
"loss": 0.6941,
|
| 299 |
-
"rewards/accuracies": 0.30000001192092896,
|
| 300 |
-
"rewards/chosen": 0.032920531928539276,
|
| 301 |
-
"rewards/margins": -0.0018675241153687239,
|
| 302 |
-
"rewards/rejected": 0.03478805348277092,
|
| 303 |
-
"step": 200
|
| 304 |
-
},
|
| 305 |
-
{
|
| 306 |
-
"epoch": 0.21,
|
| 307 |
-
"learning_rate": 4.647058823529412e-07,
|
| 308 |
-
"logits/chosen": -2.2569141387939453,
|
| 309 |
-
"logits/rejected": -2.256953477859497,
|
| 310 |
-
"logps/chosen": -0.3138192296028137,
|
| 311 |
-
"logps/rejected": -0.3109946846961975,
|
| 312 |
-
"loss": 0.6938,
|
| 313 |
-
"rewards/accuracies": 0.20000000298023224,
|
| 314 |
-
"rewards/chosen": 0.11521951854228973,
|
| 315 |
-
"rewards/margins": -0.0012991640251129866,
|
| 316 |
-
"rewards/rejected": 0.11651869118213654,
|
| 317 |
-
"step": 210
|
| 318 |
-
},
|
| 319 |
-
{
|
| 320 |
-
"epoch": 0.22,
|
| 321 |
-
"learning_rate": 4.5882352941176465e-07,
|
| 322 |
-
"logits/chosen": -2.047544479370117,
|
| 323 |
-
"logits/rejected": -2.4468607902526855,
|
| 324 |
-
"logps/chosen": -0.2669355571269989,
|
| 325 |
-
"logps/rejected": -38.318092346191406,
|
| 326 |
-
"loss": 0.4989,
|
| 327 |
-
"rewards/accuracies": 0.6000000238418579,
|
| 328 |
-
"rewards/chosen": 0.13280974328517914,
|
| 329 |
-
"rewards/margins": 1.0004825592041016,
|
| 330 |
-
"rewards/rejected": -0.8676728010177612,
|
| 331 |
-
"step": 220
|
| 332 |
-
},
|
| 333 |
-
{
|
| 334 |
-
"epoch": 0.23,
|
| 335 |
-
"learning_rate": 4.529411764705882e-07,
|
| 336 |
-
"logits/chosen": -1.9934049844741821,
|
| 337 |
-
"logits/rejected": -2.3484489917755127,
|
| 338 |
-
"logps/chosen": -0.6550347208976746,
|
| 339 |
-
"logps/rejected": -36.49030685424805,
|
| 340 |
-
"loss": 0.5546,
|
| 341 |
-
"rewards/accuracies": 0.6000000238418579,
|
| 342 |
-
"rewards/chosen": 0.23355157673358917,
|
| 343 |
-
"rewards/margins": 1.3387348651885986,
|
| 344 |
-
"rewards/rejected": -1.1051832437515259,
|
| 345 |
-
"step": 230
|
| 346 |
-
},
|
| 347 |
-
{
|
| 348 |
-
"epoch": 0.24,
|
| 349 |
-
"learning_rate": 4.470588235294118e-07,
|
| 350 |
-
"logits/chosen": -2.0561001300811768,
|
| 351 |
-
"logits/rejected": -2.314795970916748,
|
| 352 |
-
"logps/chosen": -0.24726076424121857,
|
| 353 |
-
"logps/rejected": -30.558258056640625,
|
| 354 |
-
"loss": 0.5707,
|
| 355 |
-
"rewards/accuracies": 0.4000000059604645,
|
| 356 |
-
"rewards/chosen": 0.10325585305690765,
|
| 357 |
-
"rewards/margins": 1.178950548171997,
|
| 358 |
-
"rewards/rejected": -1.0756947994232178,
|
| 359 |
-
"step": 240
|
| 360 |
-
},
|
| 361 |
-
{
|
| 362 |
-
"epoch": 0.25,
|
| 363 |
-
"learning_rate": 4.4117647058823526e-07,
|
| 364 |
-
"logits/chosen": -2.326530694961548,
|
| 365 |
-
"logits/rejected": -2.326476573944092,
|
| 366 |
-
"logps/chosen": -0.12658026814460754,
|
| 367 |
-
"logps/rejected": -0.12556590139865875,
|
| 368 |
-
"loss": 0.6929,
|
| 369 |
-
"rewards/accuracies": 0.6000000238418579,
|
| 370 |
-
"rewards/chosen": 0.11289018392562866,
|
| 371 |
-
"rewards/margins": 0.00047791999531909823,
|
| 372 |
-
"rewards/rejected": 0.11241227388381958,
|
| 373 |
-
"step": 250
|
| 374 |
-
},
|
| 375 |
-
{
|
| 376 |
-
"epoch": 0.26,
|
| 377 |
-
"learning_rate": 4.352941176470588e-07,
|
| 378 |
-
"logits/chosen": -2.2254791259765625,
|
| 379 |
-
"logits/rejected": -2.3039350509643555,
|
| 380 |
-
"logps/chosen": -0.24640262126922607,
|
| 381 |
-
"logps/rejected": -10.21367073059082,
|
| 382 |
-
"loss": 0.5724,
|
| 383 |
-
"rewards/accuracies": 0.4000000059604645,
|
| 384 |
-
"rewards/chosen": 0.14248333871364594,
|
| 385 |
-
"rewards/margins": 0.7252418994903564,
|
| 386 |
-
"rewards/rejected": -0.5827585458755493,
|
| 387 |
-
"step": 260
|
| 388 |
-
},
|
| 389 |
-
{
|
| 390 |
-
"epoch": 0.27,
|
| 391 |
-
"learning_rate": 4.294117647058823e-07,
|
| 392 |
-
"logits/chosen": -2.486194133758545,
|
| 393 |
-
"logits/rejected": -2.486302614212036,
|
| 394 |
-
"logps/chosen": -0.04246897995471954,
|
| 395 |
-
"logps/rejected": -0.04294817894697189,
|
| 396 |
-
"loss": 0.6925,
|
| 397 |
-
"rewards/accuracies": 0.699999988079071,
|
| 398 |
-
"rewards/chosen": 0.08347944170236588,
|
| 399 |
-
"rewards/margins": 0.0013219652464613318,
|
| 400 |
-
"rewards/rejected": 0.08215747773647308,
|
| 401 |
-
"step": 270
|
| 402 |
-
},
|
| 403 |
-
{
|
| 404 |
-
"epoch": 0.28,
|
| 405 |
-
"learning_rate": 4.2352941176470586e-07,
|
| 406 |
-
"logits/chosen": -2.258772373199463,
|
| 407 |
-
"logits/rejected": -2.350086212158203,
|
| 408 |
-
"logps/chosen": -0.19779756665229797,
|
| 409 |
-
"logps/rejected": -12.979776382446289,
|
| 410 |
-
"loss": 0.6264,
|
| 411 |
-
"rewards/accuracies": 0.6000000238418579,
|
| 412 |
-
"rewards/chosen": 0.15591862797737122,
|
| 413 |
-
"rewards/margins": 0.3447035253047943,
|
| 414 |
-
"rewards/rejected": -0.18878492712974548,
|
| 415 |
-
"step": 280
|
| 416 |
-
},
|
| 417 |
-
{
|
| 418 |
-
"epoch": 0.29,
|
| 419 |
-
"learning_rate": 4.1764705882352943e-07,
|
| 420 |
-
"logits/chosen": -2.400933265686035,
|
| 421 |
-
"logits/rejected": -2.4009623527526855,
|
| 422 |
-
"logps/chosen": -0.7456287741661072,
|
| 423 |
-
"logps/rejected": -0.7486740946769714,
|
| 424 |
-
"loss": 0.6928,
|
| 425 |
-
"rewards/accuracies": 0.4000000059604645,
|
| 426 |
-
"rewards/chosen": 0.20849373936653137,
|
| 427 |
-
"rewards/margins": 0.000649511581286788,
|
| 428 |
-
"rewards/rejected": 0.20784422755241394,
|
| 429 |
-
"step": 290
|
| 430 |
-
},
|
| 431 |
-
{
|
| 432 |
-
"epoch": 0.3,
|
| 433 |
-
"learning_rate": 4.117647058823529e-07,
|
| 434 |
-
"logits/chosen": -2.1085093021392822,
|
| 435 |
-
"logits/rejected": -2.419422149658203,
|
| 436 |
-
"logps/chosen": -0.23055700957775116,
|
| 437 |
-
"logps/rejected": -44.592384338378906,
|
| 438 |
-
"loss": 0.5007,
|
| 439 |
-
"rewards/accuracies": 0.5,
|
| 440 |
-
"rewards/chosen": 0.16842325031757355,
|
| 441 |
-
"rewards/margins": 1.4631106853485107,
|
| 442 |
-
"rewards/rejected": -1.2946875095367432,
|
| 443 |
-
"step": 300
|
| 444 |
-
},
|
| 445 |
-
{
|
| 446 |
-
"epoch": 0.31,
|
| 447 |
-
"learning_rate": 4.0588235294117646e-07,
|
| 448 |
-
"logits/chosen": -2.15910005569458,
|
| 449 |
-
"logits/rejected": -2.330493688583374,
|
| 450 |
-
"logps/chosen": -0.2197931557893753,
|
| 451 |
-
"logps/rejected": -15.632344245910645,
|
| 452 |
-
"loss": 0.6248,
|
| 453 |
-
"rewards/accuracies": 0.30000001192092896,
|
| 454 |
-
"rewards/chosen": 0.16517843306064606,
|
| 455 |
-
"rewards/margins": 0.4509655833244324,
|
| 456 |
-
"rewards/rejected": -0.2857871949672699,
|
| 457 |
-
"step": 310
|
| 458 |
-
},
|
| 459 |
-
{
|
| 460 |
-
"epoch": 0.32,
|
| 461 |
-
"learning_rate": 4e-07,
|
| 462 |
-
"logits/chosen": -2.0489771366119385,
|
| 463 |
-
"logits/rejected": -2.271772861480713,
|
| 464 |
-
"logps/chosen": -0.38775813579559326,
|
| 465 |
-
"logps/rejected": -15.12170696258545,
|
| 466 |
-
"loss": 0.5977,
|
| 467 |
-
"rewards/accuracies": 0.6000000238418579,
|
| 468 |
-
"rewards/chosen": 0.1870100200176239,
|
| 469 |
-
"rewards/margins": 0.35321617126464844,
|
| 470 |
-
"rewards/rejected": -0.16620615124702454,
|
| 471 |
-
"step": 320
|
| 472 |
-
},
|
| 473 |
-
{
|
| 474 |
-
"epoch": 0.33,
|
| 475 |
-
"learning_rate": 3.941176470588235e-07,
|
| 476 |
-
"logits/chosen": -2.066758632659912,
|
| 477 |
-
"logits/rejected": -2.2020325660705566,
|
| 478 |
-
"logps/chosen": -0.27063071727752686,
|
| 479 |
-
"logps/rejected": -16.94083595275879,
|
| 480 |
-
"loss": 0.5832,
|
| 481 |
-
"rewards/accuracies": 0.30000001192092896,
|
| 482 |
-
"rewards/chosen": 0.15286189317703247,
|
| 483 |
-
"rewards/margins": 0.5866376161575317,
|
| 484 |
-
"rewards/rejected": -0.43377572298049927,
|
| 485 |
-
"step": 330
|
| 486 |
-
},
|
| 487 |
-
{
|
| 488 |
-
"epoch": 0.34,
|
| 489 |
-
"learning_rate": 3.8823529411764707e-07,
|
| 490 |
-
"logits/chosen": -2.2944748401641846,
|
| 491 |
-
"logits/rejected": -2.29447603225708,
|
| 492 |
-
"logps/chosen": -0.3279079794883728,
|
| 493 |
-
"logps/rejected": -0.32968124747276306,
|
| 494 |
-
"loss": 0.6934,
|
| 495 |
-
"rewards/accuracies": 0.4000000059604645,
|
| 496 |
-
"rewards/chosen": 0.21330514550209045,
|
| 497 |
-
"rewards/margins": -0.0005384087562561035,
|
| 498 |
-
"rewards/rejected": 0.21384355425834656,
|
| 499 |
-
"step": 340
|
| 500 |
-
},
|
| 501 |
-
{
|
| 502 |
-
"epoch": 0.35,
|
| 503 |
-
"learning_rate": 3.8235294117647053e-07,
|
| 504 |
-
"logits/chosen": -2.189199924468994,
|
| 505 |
-
"logits/rejected": -2.2941532135009766,
|
| 506 |
-
"logps/chosen": -0.3213362693786621,
|
| 507 |
-
"logps/rejected": -17.19934844970703,
|
| 508 |
-
"loss": 0.5884,
|
| 509 |
-
"rewards/accuracies": 0.4000000059604645,
|
| 510 |
-
"rewards/chosen": 0.11426669359207153,
|
| 511 |
-
"rewards/margins": 1.0944595336914062,
|
| 512 |
-
"rewards/rejected": -0.9801927804946899,
|
| 513 |
-
"step": 350
|
| 514 |
-
},
|
| 515 |
-
{
|
| 516 |
-
"epoch": 0.36,
|
| 517 |
-
"learning_rate": 3.764705882352941e-07,
|
| 518 |
-
"logits/chosen": -2.1935980319976807,
|
| 519 |
-
"logits/rejected": -2.419231653213501,
|
| 520 |
-
"logps/chosen": -0.32695242762565613,
|
| 521 |
-
"logps/rejected": -21.586366653442383,
|
| 522 |
-
"loss": 0.5661,
|
| 523 |
-
"rewards/accuracies": 0.4000000059604645,
|
| 524 |
-
"rewards/chosen": 0.131852388381958,
|
| 525 |
-
"rewards/margins": 0.765614926815033,
|
| 526 |
-
"rewards/rejected": -0.633762538433075,
|
| 527 |
-
"step": 360
|
| 528 |
-
},
|
| 529 |
-
{
|
| 530 |
-
"epoch": 0.37,
|
| 531 |
-
"learning_rate": 3.705882352941176e-07,
|
| 532 |
-
"logits/chosen": -2.1839663982391357,
|
| 533 |
-
"logits/rejected": -2.184142827987671,
|
| 534 |
-
"logps/chosen": -0.0991433784365654,
|
| 535 |
-
"logps/rejected": -0.09875849634408951,
|
| 536 |
-
"loss": 0.694,
|
| 537 |
-
"rewards/accuracies": 0.30000001192092896,
|
| 538 |
-
"rewards/chosen": 0.07628553360700607,
|
| 539 |
-
"rewards/margins": -0.0017677659634500742,
|
| 540 |
-
"rewards/rejected": 0.0780533030629158,
|
| 541 |
-
"step": 370
|
| 542 |
-
},
|
| 543 |
-
{
|
| 544 |
-
"epoch": 0.38,
|
| 545 |
-
"learning_rate": 3.6470588235294114e-07,
|
| 546 |
-
"logits/chosen": -1.9649509191513062,
|
| 547 |
-
"logits/rejected": -2.1653668880462646,
|
| 548 |
-
"logps/chosen": -0.21254102885723114,
|
| 549 |
-
"logps/rejected": -22.567474365234375,
|
| 550 |
-
"loss": 0.6251,
|
| 551 |
-
"rewards/accuracies": 0.4000000059604645,
|
| 552 |
-
"rewards/chosen": 0.1867930144071579,
|
| 553 |
-
"rewards/margins": 0.7762778401374817,
|
| 554 |
-
"rewards/rejected": -0.5894848704338074,
|
| 555 |
-
"step": 380
|
| 556 |
-
},
|
| 557 |
-
{
|
| 558 |
-
"epoch": 0.39,
|
| 559 |
-
"learning_rate": 3.588235294117647e-07,
|
| 560 |
-
"logits/chosen": -2.1640498638153076,
|
| 561 |
-
"logits/rejected": -2.198374032974243,
|
| 562 |
-
"logps/chosen": -0.18704327940940857,
|
| 563 |
-
"logps/rejected": -9.075506210327148,
|
| 564 |
-
"loss": 0.5556,
|
| 565 |
-
"rewards/accuracies": 0.6000000238418579,
|
| 566 |
-
"rewards/chosen": 0.047349415719509125,
|
| 567 |
-
"rewards/margins": 1.023637056350708,
|
| 568 |
-
"rewards/rejected": -0.9762875437736511,
|
| 569 |
-
"step": 390
|
| 570 |
-
},
|
| 571 |
-
{
|
| 572 |
-
"epoch": 0.4,
|
| 573 |
-
"learning_rate": 3.529411764705882e-07,
|
| 574 |
-
"logits/chosen": -1.843392014503479,
|
| 575 |
-
"logits/rejected": -2.1060373783111572,
|
| 576 |
-
"logps/chosen": -0.3059827387332916,
|
| 577 |
-
"logps/rejected": -40.11814880371094,
|
| 578 |
-
"loss": 0.4882,
|
| 579 |
-
"rewards/accuracies": 0.699999988079071,
|
| 580 |
-
"rewards/chosen": 0.25154930353164673,
|
| 581 |
-
"rewards/margins": 2.534274101257324,
|
| 582 |
-
"rewards/rejected": -2.2827248573303223,
|
| 583 |
-
"step": 400
|
| 584 |
-
},
|
| 585 |
-
{
|
| 586 |
-
"epoch": 0.41,
|
| 587 |
-
"learning_rate": 3.4705882352941174e-07,
|
| 588 |
-
"logits/chosen": -1.9674293994903564,
|
| 589 |
-
"logits/rejected": -2.1082801818847656,
|
| 590 |
-
"logps/chosen": -0.11460398137569427,
|
| 591 |
-
"logps/rejected": -17.704038619995117,
|
| 592 |
-
"loss": 0.5573,
|
| 593 |
-
"rewards/accuracies": 0.4000000059604645,
|
| 594 |
-
"rewards/chosen": 0.17245104908943176,
|
| 595 |
-
"rewards/margins": 0.9313168525695801,
|
| 596 |
-
"rewards/rejected": -0.7588656544685364,
|
| 597 |
-
"step": 410
|
| 598 |
-
},
|
| 599 |
-
{
|
| 600 |
-
"epoch": 0.42,
|
| 601 |
-
"learning_rate": 3.411764705882353e-07,
|
| 602 |
-
"logits/chosen": -1.4523433446884155,
|
| 603 |
-
"logits/rejected": -2.156726121902466,
|
| 604 |
-
"logps/chosen": -0.15290628373622894,
|
| 605 |
-
"logps/rejected": -63.89361572265625,
|
| 606 |
-
"loss": 0.2949,
|
| 607 |
-
"rewards/accuracies": 0.800000011920929,
|
| 608 |
-
"rewards/chosen": 0.23414066433906555,
|
| 609 |
-
"rewards/margins": 4.00860071182251,
|
| 610 |
-
"rewards/rejected": -3.7744603157043457,
|
| 611 |
-
"step": 420
|
| 612 |
-
},
|
| 613 |
-
{
|
| 614 |
-
"epoch": 0.43,
|
| 615 |
-
"learning_rate": 3.352941176470588e-07,
|
| 616 |
-
"logits/chosen": -2.076636791229248,
|
| 617 |
-
"logits/rejected": -2.0995030403137207,
|
| 618 |
-
"logps/chosen": -0.09147205203771591,
|
| 619 |
-
"logps/rejected": -7.910800933837891,
|
| 620 |
-
"loss": 0.6241,
|
| 621 |
-
"rewards/accuracies": 0.30000001192092896,
|
| 622 |
-
"rewards/chosen": 0.11051849275827408,
|
| 623 |
-
"rewards/margins": 0.6908754110336304,
|
| 624 |
-
"rewards/rejected": -0.5803568959236145,
|
| 625 |
-
"step": 430
|
| 626 |
-
},
|
| 627 |
-
{
|
| 628 |
-
"epoch": 0.44,
|
| 629 |
-
"learning_rate": 3.2941176470588235e-07,
|
| 630 |
-
"logits/chosen": -2.1452138423919678,
|
| 631 |
-
"logits/rejected": -2.206829071044922,
|
| 632 |
-
"logps/chosen": -0.13904191553592682,
|
| 633 |
-
"logps/rejected": -10.327826499938965,
|
| 634 |
-
"loss": 0.6234,
|
| 635 |
-
"rewards/accuracies": 0.4000000059604645,
|
| 636 |
-
"rewards/chosen": 0.272757351398468,
|
| 637 |
-
"rewards/margins": 0.7720587849617004,
|
| 638 |
-
"rewards/rejected": -0.49930137395858765,
|
| 639 |
-
"step": 440
|
| 640 |
-
},
|
| 641 |
-
{
|
| 642 |
-
"epoch": 0.45,
|
| 643 |
-
"learning_rate": 3.2352941176470586e-07,
|
| 644 |
-
"logits/chosen": -2.090587854385376,
|
| 645 |
-
"logits/rejected": -2.090881824493408,
|
| 646 |
-
"logps/chosen": -0.17841731011867523,
|
| 647 |
-
"logps/rejected": -0.17845699191093445,
|
| 648 |
-
"loss": 0.6935,
|
| 649 |
-
"rewards/accuracies": 0.30000001192092896,
|
| 650 |
-
"rewards/chosen": 0.2406892031431198,
|
| 651 |
-
"rewards/margins": -0.0006073854747228324,
|
| 652 |
-
"rewards/rejected": 0.24129657447338104,
|
| 653 |
-
"step": 450
|
| 654 |
-
},
|
| 655 |
-
{
|
| 656 |
-
"epoch": 0.46,
|
| 657 |
-
"learning_rate": 3.176470588235294e-07,
|
| 658 |
-
"logits/chosen": -1.7833999395370483,
|
| 659 |
-
"logits/rejected": -2.0649399757385254,
|
| 660 |
-
"logps/chosen": -0.09775514900684357,
|
| 661 |
-
"logps/rejected": -44.72971725463867,
|
| 662 |
-
"loss": 0.503,
|
| 663 |
-
"rewards/accuracies": 0.6000000238418579,
|
| 664 |
-
"rewards/chosen": 0.25888940691947937,
|
| 665 |
-
"rewards/margins": 2.86980938911438,
|
| 666 |
-
"rewards/rejected": -2.610919952392578,
|
| 667 |
-
"step": 460
|
| 668 |
-
},
|
| 669 |
-
{
|
| 670 |
-
"epoch": 0.47,
|
| 671 |
-
"learning_rate": 3.1176470588235295e-07,
|
| 672 |
-
"logits/chosen": -1.9202232360839844,
|
| 673 |
-
"logits/rejected": -2.1586782932281494,
|
| 674 |
-
"logps/chosen": -0.4241597056388855,
|
| 675 |
-
"logps/rejected": -55.93566131591797,
|
| 676 |
-
"loss": 0.4373,
|
| 677 |
-
"rewards/accuracies": 0.699999988079071,
|
| 678 |
-
"rewards/chosen": 0.22231805324554443,
|
| 679 |
-
"rewards/margins": 4.350785255432129,
|
| 680 |
-
"rewards/rejected": -4.128467082977295,
|
| 681 |
-
"step": 470
|
| 682 |
-
},
|
| 683 |
-
{
|
| 684 |
-
"epoch": 0.48,
|
| 685 |
-
"learning_rate": 3.0588235294117647e-07,
|
| 686 |
-
"logits/chosen": -1.8953840732574463,
|
| 687 |
-
"logits/rejected": -2.0778305530548096,
|
| 688 |
-
"logps/chosen": -0.3732009530067444,
|
| 689 |
-
"logps/rejected": -32.9583740234375,
|
| 690 |
-
"loss": 0.4864,
|
| 691 |
-
"rewards/accuracies": 0.30000001192092896,
|
| 692 |
-
"rewards/chosen": 0.1983214169740677,
|
| 693 |
-
"rewards/margins": 3.323279857635498,
|
| 694 |
-
"rewards/rejected": -3.1249582767486572,
|
| 695 |
-
"step": 480
|
| 696 |
-
},
|
| 697 |
-
{
|
| 698 |
-
"epoch": 0.49,
|
| 699 |
-
"learning_rate": 3e-07,
|
| 700 |
-
"logits/chosen": -1.5736738443374634,
|
| 701 |
-
"logits/rejected": -2.016456365585327,
|
| 702 |
-
"logps/chosen": -0.012308413162827492,
|
| 703 |
-
"logps/rejected": -100.20381164550781,
|
| 704 |
-
"loss": 0.4155,
|
| 705 |
-
"rewards/accuracies": 0.8999999761581421,
|
| 706 |
-
"rewards/chosen": 0.23825187981128693,
|
| 707 |
-
"rewards/margins": 7.954277992248535,
|
| 708 |
-
"rewards/rejected": -7.716025352478027,
|
| 709 |
-
"step": 490
|
| 710 |
-
},
|
| 711 |
-
{
|
| 712 |
-
"epoch": 0.5,
|
| 713 |
-
"learning_rate": 2.941176470588235e-07,
|
| 714 |
-
"logits/chosen": -1.8802621364593506,
|
| 715 |
-
"logits/rejected": -2.047894239425659,
|
| 716 |
-
"logps/chosen": -0.06101088970899582,
|
| 717 |
-
"logps/rejected": -20.134754180908203,
|
| 718 |
-
"loss": 0.6235,
|
| 719 |
-
"rewards/accuracies": 0.5,
|
| 720 |
-
"rewards/chosen": 0.23272594809532166,
|
| 721 |
-
"rewards/margins": 1.6100852489471436,
|
| 722 |
-
"rewards/rejected": -1.37735915184021,
|
| 723 |
-
"step": 500
|
| 724 |
-
},
|
| 725 |
-
{
|
| 726 |
-
"epoch": 0.5,
|
| 727 |
-
"eval_logits/chosen": -1.9279996156692505,
|
| 728 |
-
"eval_logits/rejected": -2.0771560668945312,
|
| 729 |
-
"eval_logps/chosen": -0.13991177082061768,
|
| 730 |
-
"eval_logps/rejected": -20.526100158691406,
|
| 731 |
-
"eval_loss": 0.5603082776069641,
|
| 732 |
-
"eval_rewards/accuracies": 0.49000000953674316,
|
| 733 |
-
"eval_rewards/chosen": 0.2280687689781189,
|
| 734 |
-
"eval_rewards/margins": 1.3241862058639526,
|
| 735 |
-
"eval_rewards/rejected": -1.0961174964904785,
|
| 736 |
-
"eval_runtime": 22.8794,
|
| 737 |
-
"eval_samples_per_second": 4.371,
|
| 738 |
-
"eval_steps_per_second": 4.371,
|
| 739 |
-
"step": 500
|
| 740 |
-
},
|
| 741 |
-
{
|
| 742 |
-
"epoch": 0.51,
|
| 743 |
-
"learning_rate": 2.88235294117647e-07,
|
| 744 |
-
"logits/chosen": -1.881134271621704,
|
| 745 |
-
"logits/rejected": -1.9070965051651,
|
| 746 |
-
"logps/chosen": -0.02279834821820259,
|
| 747 |
-
"logps/rejected": -5.849534034729004,
|
| 748 |
-
"loss": 0.6284,
|
| 749 |
-
"rewards/accuracies": 0.30000001192092896,
|
| 750 |
-
"rewards/chosen": 0.13564881682395935,
|
| 751 |
-
"rewards/margins": 0.31011906266212463,
|
| 752 |
-
"rewards/rejected": -0.1744702160358429,
|
| 753 |
-
"step": 510
|
| 754 |
-
},
|
| 755 |
-
{
|
| 756 |
-
"epoch": 0.52,
|
| 757 |
-
"learning_rate": 2.823529411764706e-07,
|
| 758 |
-
"logits/chosen": -1.9525129795074463,
|
| 759 |
-
"logits/rejected": -1.960097074508667,
|
| 760 |
-
"logps/chosen": -0.05416996404528618,
|
| 761 |
-
"logps/rejected": -6.789488792419434,
|
| 762 |
-
"loss": 0.6232,
|
| 763 |
-
"rewards/accuracies": 0.5,
|
| 764 |
-
"rewards/chosen": 0.19361652433872223,
|
| 765 |
-
"rewards/margins": 0.9131872057914734,
|
| 766 |
-
"rewards/rejected": -0.7195707559585571,
|
| 767 |
-
"step": 520
|
| 768 |
-
},
|
| 769 |
-
{
|
| 770 |
-
"epoch": 0.53,
|
| 771 |
-
"learning_rate": 2.764705882352941e-07,
|
| 772 |
-
"logits/chosen": -2.0894522666931152,
|
| 773 |
-
"logits/rejected": -2.089811325073242,
|
| 774 |
-
"logps/chosen": -0.017659317702054977,
|
| 775 |
-
"logps/rejected": -0.0175788514316082,
|
| 776 |
-
"loss": 0.6929,
|
| 777 |
-
"rewards/accuracies": 0.5,
|
| 778 |
-
"rewards/chosen": 0.17115317285060883,
|
| 779 |
-
"rewards/margins": 0.0004329962539486587,
|
| 780 |
-
"rewards/rejected": 0.170720174908638,
|
| 781 |
-
"step": 530
|
| 782 |
-
},
|
| 783 |
-
{
|
| 784 |
-
"epoch": 0.54,
|
| 785 |
-
"learning_rate": 2.705882352941176e-07,
|
| 786 |
-
"logits/chosen": -2.1036479473114014,
|
| 787 |
-
"logits/rejected": -2.1614184379577637,
|
| 788 |
-
"logps/chosen": -0.1067914217710495,
|
| 789 |
-
"logps/rejected": -12.363987922668457,
|
| 790 |
-
"loss": 0.5895,
|
| 791 |
-
"rewards/accuracies": 0.6000000238418579,
|
| 792 |
-
"rewards/chosen": 0.19864901900291443,
|
| 793 |
-
"rewards/margins": 0.9721519351005554,
|
| 794 |
-
"rewards/rejected": -0.7735029458999634,
|
| 795 |
-
"step": 540
|
| 796 |
-
},
|
| 797 |
-
{
|
| 798 |
-
"epoch": 0.55,
|
| 799 |
-
"learning_rate": 2.6470588235294114e-07,
|
| 800 |
-
"logits/chosen": -1.9976797103881836,
|
| 801 |
-
"logits/rejected": -1.9976022243499756,
|
| 802 |
-
"logps/chosen": -0.04686691239476204,
|
| 803 |
-
"logps/rejected": -0.046861741691827774,
|
| 804 |
-
"loss": 0.693,
|
| 805 |
-
"rewards/accuracies": 0.4000000059604645,
|
| 806 |
-
"rewards/chosen": 0.23503179848194122,
|
| 807 |
-
"rewards/margins": 0.0003768761525861919,
|
| 808 |
-
"rewards/rejected": 0.23465490341186523,
|
| 809 |
-
"step": 550
|
| 810 |
-
},
|
| 811 |
-
{
|
| 812 |
-
"epoch": 0.56,
|
| 813 |
-
"learning_rate": 2.588235294117647e-07,
|
| 814 |
-
"logits/chosen": -1.6446807384490967,
|
| 815 |
-
"logits/rejected": -2.0919888019561768,
|
| 816 |
-
"logps/chosen": -0.08167463541030884,
|
| 817 |
-
"logps/rejected": -50.200626373291016,
|
| 818 |
-
"loss": 0.4924,
|
| 819 |
-
"rewards/accuracies": 0.6000000238418579,
|
| 820 |
-
"rewards/chosen": 0.27800416946411133,
|
| 821 |
-
"rewards/margins": 2.7979769706726074,
|
| 822 |
-
"rewards/rejected": -2.519972801208496,
|
| 823 |
-
"step": 560
|
| 824 |
-
},
|
| 825 |
-
{
|
| 826 |
-
"epoch": 0.57,
|
| 827 |
-
"learning_rate": 2.5294117647058823e-07,
|
| 828 |
-
"logits/chosen": -1.6458909511566162,
|
| 829 |
-
"logits/rejected": -2.06436824798584,
|
| 830 |
-
"logps/chosen": -0.004592462442815304,
|
| 831 |
-
"logps/rejected": -85.18163299560547,
|
| 832 |
-
"loss": 0.4162,
|
| 833 |
-
"rewards/accuracies": 0.6000000238418579,
|
| 834 |
-
"rewards/chosen": 0.23764410614967346,
|
| 835 |
-
"rewards/margins": 7.737754821777344,
|
| 836 |
-
"rewards/rejected": -7.500109672546387,
|
| 837 |
-
"step": 570
|
| 838 |
-
},
|
| 839 |
-
{
|
| 840 |
-
"epoch": 0.58,
|
| 841 |
-
"learning_rate": 2.4705882352941175e-07,
|
| 842 |
-
"logits/chosen": -1.5587949752807617,
|
| 843 |
-
"logits/rejected": -1.8453130722045898,
|
| 844 |
-
"logps/chosen": -0.04938965290784836,
|
| 845 |
-
"logps/rejected": -76.6133804321289,
|
| 846 |
-
"loss": 0.4181,
|
| 847 |
-
"rewards/accuracies": 0.800000011920929,
|
| 848 |
-
"rewards/chosen": 0.27002671360969543,
|
| 849 |
-
"rewards/margins": 6.7017316818237305,
|
| 850 |
-
"rewards/rejected": -6.431704521179199,
|
| 851 |
-
"step": 580
|
| 852 |
-
},
|
| 853 |
-
{
|
| 854 |
-
"epoch": 0.59,
|
| 855 |
-
"learning_rate": 2.4117647058823526e-07,
|
| 856 |
-
"logits/chosen": -2.0468690395355225,
|
| 857 |
-
"logits/rejected": -2.046814441680908,
|
| 858 |
-
"logps/chosen": -0.01457706093788147,
|
| 859 |
-
"logps/rejected": -0.014673275873064995,
|
| 860 |
-
"loss": 0.6931,
|
| 861 |
-
"rewards/accuracies": 0.4000000059604645,
|
| 862 |
-
"rewards/chosen": 0.1107385978102684,
|
| 863 |
-
"rewards/margins": -4.159938725933898e-06,
|
| 864 |
-
"rewards/rejected": 0.11074274778366089,
|
| 865 |
-
"step": 590
|
| 866 |
-
},
|
| 867 |
-
{
|
| 868 |
-
"epoch": 0.6,
|
| 869 |
-
"learning_rate": 2.352941176470588e-07,
|
| 870 |
-
"logits/chosen": -2.068368673324585,
|
| 871 |
-
"logits/rejected": -2.087043285369873,
|
| 872 |
-
"logps/chosen": -0.007376219145953655,
|
| 873 |
-
"logps/rejected": -8.290468215942383,
|
| 874 |
-
"loss": 0.6237,
|
| 875 |
-
"rewards/accuracies": 0.4000000059604645,
|
| 876 |
-
"rewards/chosen": 0.15371395647525787,
|
| 877 |
-
"rewards/margins": 0.8146413564682007,
|
| 878 |
-
"rewards/rejected": -0.660927414894104,
|
| 879 |
-
"step": 600
|
| 880 |
-
},
|
| 881 |
-
{
|
| 882 |
-
"epoch": 0.61,
|
| 883 |
-
"learning_rate": 2.2941176470588233e-07,
|
| 884 |
-
"logits/chosen": -1.8815044164657593,
|
| 885 |
-
"logits/rejected": -1.9951889514923096,
|
| 886 |
-
"logps/chosen": -0.12692415714263916,
|
| 887 |
-
"logps/rejected": -25.312314987182617,
|
| 888 |
-
"loss": 0.6251,
|
| 889 |
-
"rewards/accuracies": 0.4000000059604645,
|
| 890 |
-
"rewards/chosen": 0.2646506428718567,
|
| 891 |
-
"rewards/margins": 2.2484025955200195,
|
| 892 |
-
"rewards/rejected": -1.983751893043518,
|
| 893 |
-
"step": 610
|
| 894 |
-
},
|
| 895 |
-
{
|
| 896 |
-
"epoch": 0.62,
|
| 897 |
-
"learning_rate": 2.235294117647059e-07,
|
| 898 |
-
"logits/chosen": -1.907518744468689,
|
| 899 |
-
"logits/rejected": -1.9557015895843506,
|
| 900 |
-
"logps/chosen": -0.0008927548187784851,
|
| 901 |
-
"logps/rejected": -10.989554405212402,
|
| 902 |
-
"loss": 0.624,
|
| 903 |
-
"rewards/accuracies": 0.20000000298023224,
|
| 904 |
-
"rewards/chosen": 0.047840237617492676,
|
| 905 |
-
"rewards/margins": 0.6518641114234924,
|
| 906 |
-
"rewards/rejected": -0.6040239334106445,
|
| 907 |
-
"step": 620
|
| 908 |
-
},
|
| 909 |
-
{
|
| 910 |
-
"epoch": 0.63,
|
| 911 |
-
"learning_rate": 2.176470588235294e-07,
|
| 912 |
-
"logits/chosen": -1.7601346969604492,
|
| 913 |
-
"logits/rejected": -1.9046695232391357,
|
| 914 |
-
"logps/chosen": -0.07976426184177399,
|
| 915 |
-
"logps/rejected": -29.652408599853516,
|
| 916 |
-
"loss": 0.6233,
|
| 917 |
-
"rewards/accuracies": 0.30000001192092896,
|
| 918 |
-
"rewards/chosen": 0.28871995210647583,
|
| 919 |
-
"rewards/margins": 2.4649205207824707,
|
| 920 |
-
"rewards/rejected": -2.1762008666992188,
|
| 921 |
-
"step": 630
|
| 922 |
-
},
|
| 923 |
-
{
|
| 924 |
-
"epoch": 0.64,
|
| 925 |
-
"learning_rate": 2.1176470588235293e-07,
|
| 926 |
-
"logits/chosen": -1.8790966272354126,
|
| 927 |
-
"logits/rejected": -1.8792489767074585,
|
| 928 |
-
"logps/chosen": -0.0008465357241220772,
|
| 929 |
-
"logps/rejected": -0.0007601732504554093,
|
| 930 |
-
"loss": 0.6931,
|
| 931 |
-
"rewards/accuracies": 0.4000000059604645,
|
| 932 |
-
"rewards/chosen": 0.06679291278123856,
|
| 933 |
-
"rewards/margins": 0.00011558351252460852,
|
| 934 |
-
"rewards/rejected": 0.06667732447385788,
|
| 935 |
-
"step": 640
|
| 936 |
-
},
|
| 937 |
-
{
|
| 938 |
-
"epoch": 0.65,
|
| 939 |
-
"learning_rate": 2.0588235294117645e-07,
|
| 940 |
-
"logits/chosen": -1.8315871953964233,
|
| 941 |
-
"logits/rejected": -1.8316688537597656,
|
| 942 |
-
"logps/chosen": -0.26076430082321167,
|
| 943 |
-
"logps/rejected": -0.2667176425457001,
|
| 944 |
-
"loss": 0.6922,
|
| 945 |
-
"rewards/accuracies": 0.6000000238418579,
|
| 946 |
-
"rewards/chosen": 0.11814270913600922,
|
| 947 |
-
"rewards/margins": 0.001997224520891905,
|
| 948 |
-
"rewards/rejected": 0.11614549160003662,
|
| 949 |
-
"step": 650
|
| 950 |
-
},
|
| 951 |
-
{
|
| 952 |
-
"epoch": 0.66,
|
| 953 |
-
"learning_rate": 2e-07,
|
| 954 |
-
"logits/chosen": -1.969988226890564,
|
| 955 |
-
"logits/rejected": -1.97628915309906,
|
| 956 |
-
"logps/chosen": -0.14302313327789307,
|
| 957 |
-
"logps/rejected": -1.4409807920455933,
|
| 958 |
-
"loss": 0.6544,
|
| 959 |
-
"rewards/accuracies": 0.6000000238418579,
|
| 960 |
-
"rewards/chosen": 0.33637282252311707,
|
| 961 |
-
"rewards/margins": 0.10214731842279434,
|
| 962 |
-
"rewards/rejected": 0.23422551155090332,
|
| 963 |
-
"step": 660
|
| 964 |
-
},
|
| 965 |
-
{
|
| 966 |
-
"epoch": 0.67,
|
| 967 |
-
"learning_rate": 1.9411764705882353e-07,
|
| 968 |
-
"logits/chosen": -1.6727327108383179,
|
| 969 |
-
"logits/rejected": -1.9902541637420654,
|
| 970 |
-
"logps/chosen": -0.024806343019008636,
|
| 971 |
-
"logps/rejected": -38.02031707763672,
|
| 972 |
-
"loss": 0.5553,
|
| 973 |
-
"rewards/accuracies": 0.800000011920929,
|
| 974 |
-
"rewards/chosen": 0.27485889196395874,
|
| 975 |
-
"rewards/margins": 2.909092426300049,
|
| 976 |
-
"rewards/rejected": -2.634232997894287,
|
| 977 |
-
"step": 670
|
| 978 |
-
},
|
| 979 |
-
{
|
| 980 |
-
"epoch": 0.68,
|
| 981 |
-
"learning_rate": 1.8823529411764705e-07,
|
| 982 |
-
"logits/chosen": -1.9072097539901733,
|
| 983 |
-
"logits/rejected": -1.9726791381835938,
|
| 984 |
-
"logps/chosen": -0.4985428750514984,
|
| 985 |
-
"logps/rejected": -13.495981216430664,
|
| 986 |
-
"loss": 0.6256,
|
| 987 |
-
"rewards/accuracies": 0.4000000059604645,
|
| 988 |
-
"rewards/chosen": 0.23625138401985168,
|
| 989 |
-
"rewards/margins": 0.47252601385116577,
|
| 990 |
-
"rewards/rejected": -0.2362746298313141,
|
| 991 |
-
"step": 680
|
| 992 |
-
},
|
| 993 |
-
{
|
| 994 |
-
"epoch": 0.69,
|
| 995 |
-
"learning_rate": 1.8235294117647057e-07,
|
| 996 |
-
"logits/chosen": -1.6648544073104858,
|
| 997 |
-
"logits/rejected": -1.88479483127594,
|
| 998 |
-
"logps/chosen": -0.012854146771132946,
|
| 999 |
-
"logps/rejected": -43.54383850097656,
|
| 1000 |
-
"loss": 0.5551,
|
| 1001 |
-
"rewards/accuracies": 0.4000000059604645,
|
| 1002 |
-
"rewards/chosen": 0.15211549401283264,
|
| 1003 |
-
"rewards/margins": 3.2219433784484863,
|
| 1004 |
-
"rewards/rejected": -3.0698275566101074,
|
| 1005 |
-
"step": 690
|
| 1006 |
-
},
|
| 1007 |
-
{
|
| 1008 |
-
"epoch": 0.7,
|
| 1009 |
-
"learning_rate": 1.764705882352941e-07,
|
| 1010 |
-
"logits/chosen": -1.8432905673980713,
|
| 1011 |
-
"logits/rejected": -1.843271017074585,
|
| 1012 |
-
"logps/chosen": -0.06649269163608551,
|
| 1013 |
-
"logps/rejected": -0.06794900447130203,
|
| 1014 |
-
"loss": 0.6921,
|
| 1015 |
-
"rewards/accuracies": 0.6000000238418579,
|
| 1016 |
-
"rewards/chosen": 0.17590782046318054,
|
| 1017 |
-
"rewards/margins": 0.002167056081816554,
|
| 1018 |
-
"rewards/rejected": 0.17374077439308167,
|
| 1019 |
-
"step": 700
|
| 1020 |
-
},
|
| 1021 |
-
{
|
| 1022 |
-
"epoch": 0.71,
|
| 1023 |
-
"learning_rate": 1.7058823529411766e-07,
|
| 1024 |
-
"logits/chosen": -1.782904028892517,
|
| 1025 |
-
"logits/rejected": -2.027859687805176,
|
| 1026 |
-
"logps/chosen": -0.027702972292900085,
|
| 1027 |
-
"logps/rejected": -63.74357223510742,
|
| 1028 |
-
"loss": 0.5547,
|
| 1029 |
-
"rewards/accuracies": 0.5,
|
| 1030 |
-
"rewards/chosen": 0.3461371064186096,
|
| 1031 |
-
"rewards/margins": 5.373135566711426,
|
| 1032 |
-
"rewards/rejected": -5.026998519897461,
|
| 1033 |
-
"step": 710
|
| 1034 |
-
},
|
| 1035 |
-
{
|
| 1036 |
-
"epoch": 0.72,
|
| 1037 |
-
"learning_rate": 1.6470588235294117e-07,
|
| 1038 |
-
"logits/chosen": -1.8250236511230469,
|
| 1039 |
-
"logits/rejected": -1.8965476751327515,
|
| 1040 |
-
"logps/chosen": -0.25108733773231506,
|
| 1041 |
-
"logps/rejected": -14.722552299499512,
|
| 1042 |
-
"loss": 0.4899,
|
| 1043 |
-
"rewards/accuracies": 0.6000000238418579,
|
| 1044 |
-
"rewards/chosen": 0.05279017239809036,
|
| 1045 |
-
"rewards/margins": 1.3603849411010742,
|
| 1046 |
-
"rewards/rejected": -1.307594656944275,
|
| 1047 |
-
"step": 720
|
| 1048 |
-
},
|
| 1049 |
-
{
|
| 1050 |
-
"epoch": 0.73,
|
| 1051 |
-
"learning_rate": 1.588235294117647e-07,
|
| 1052 |
-
"logits/chosen": -1.6344257593154907,
|
| 1053 |
-
"logits/rejected": -1.8426809310913086,
|
| 1054 |
-
"logps/chosen": -0.048086974769830704,
|
| 1055 |
-
"logps/rejected": -26.516277313232422,
|
| 1056 |
-
"loss": 0.5546,
|
| 1057 |
-
"rewards/accuracies": 0.30000001192092896,
|
| 1058 |
-
"rewards/chosen": 0.18577437102794647,
|
| 1059 |
-
"rewards/margins": 2.011146068572998,
|
| 1060 |
-
"rewards/rejected": -1.8253717422485352,
|
| 1061 |
-
"step": 730
|
| 1062 |
-
},
|
| 1063 |
-
{
|
| 1064 |
-
"epoch": 0.74,
|
| 1065 |
-
"learning_rate": 1.5294117647058823e-07,
|
| 1066 |
-
"logits/chosen": -1.5386077165603638,
|
| 1067 |
-
"logits/rejected": -1.8887131214141846,
|
| 1068 |
-
"logps/chosen": -0.2795502543449402,
|
| 1069 |
-
"logps/rejected": -46.01464080810547,
|
| 1070 |
-
"loss": 0.4861,
|
| 1071 |
-
"rewards/accuracies": 0.699999988079071,
|
| 1072 |
-
"rewards/chosen": 0.20460227131843567,
|
| 1073 |
-
"rewards/margins": 3.9117095470428467,
|
| 1074 |
-
"rewards/rejected": -3.7071070671081543,
|
| 1075 |
-
"step": 740
|
| 1076 |
-
},
|
| 1077 |
-
{
|
| 1078 |
-
"epoch": 0.75,
|
| 1079 |
-
"learning_rate": 1.4705882352941175e-07,
|
| 1080 |
-
"logits/chosen": -2.0386643409729004,
|
| 1081 |
-
"logits/rejected": -2.0383331775665283,
|
| 1082 |
-
"logps/chosen": -0.1892136037349701,
|
| 1083 |
-
"logps/rejected": -0.18968766927719116,
|
| 1084 |
-
"loss": 0.6934,
|
| 1085 |
-
"rewards/accuracies": 0.30000001192092896,
|
| 1086 |
-
"rewards/chosen": 0.21108309924602509,
|
| 1087 |
-
"rewards/margins": -0.0005298491450957954,
|
| 1088 |
-
"rewards/rejected": 0.2116129845380783,
|
| 1089 |
-
"step": 750
|
| 1090 |
-
},
|
| 1091 |
-
{
|
| 1092 |
-
"epoch": 0.76,
|
| 1093 |
-
"learning_rate": 1.411764705882353e-07,
|
| 1094 |
-
"logits/chosen": -1.850419282913208,
|
| 1095 |
-
"logits/rejected": -1.8857438564300537,
|
| 1096 |
-
"logps/chosen": -0.3321111798286438,
|
| 1097 |
-
"logps/rejected": -4.540987014770508,
|
| 1098 |
-
"loss": 0.6396,
|
| 1099 |
-
"rewards/accuracies": 0.30000001192092896,
|
| 1100 |
-
"rewards/chosen": 0.15547718107700348,
|
| 1101 |
-
"rewards/margins": 0.18077853322029114,
|
| 1102 |
-
"rewards/rejected": -0.02530134655535221,
|
| 1103 |
-
"step": 760
|
| 1104 |
-
},
|
| 1105 |
-
{
|
| 1106 |
-
"epoch": 0.77,
|
| 1107 |
-
"learning_rate": 1.352941176470588e-07,
|
| 1108 |
-
"logits/chosen": -1.9080699682235718,
|
| 1109 |
-
"logits/rejected": -1.9150307178497314,
|
| 1110 |
-
"logps/chosen": -0.008314008824527264,
|
| 1111 |
-
"logps/rejected": -5.123701095581055,
|
| 1112 |
-
"loss": 0.6239,
|
| 1113 |
-
"rewards/accuracies": 0.699999988079071,
|
| 1114 |
-
"rewards/chosen": 0.09305348992347717,
|
| 1115 |
-
"rewards/margins": 0.6812450289726257,
|
| 1116 |
-
"rewards/rejected": -0.5881915092468262,
|
| 1117 |
-
"step": 770
|
| 1118 |
-
},
|
| 1119 |
-
{
|
| 1120 |
-
"epoch": 0.78,
|
| 1121 |
-
"learning_rate": 1.2941176470588236e-07,
|
| 1122 |
-
"logits/chosen": -1.6551542282104492,
|
| 1123 |
-
"logits/rejected": -1.853063941001892,
|
| 1124 |
-
"logps/chosen": -0.054176896810531616,
|
| 1125 |
-
"logps/rejected": -45.885746002197266,
|
| 1126 |
-
"loss": 0.4863,
|
| 1127 |
-
"rewards/accuracies": 0.4000000059604645,
|
| 1128 |
-
"rewards/chosen": 0.1725289672613144,
|
| 1129 |
-
"rewards/margins": 4.1526994705200195,
|
| 1130 |
-
"rewards/rejected": -3.980170488357544,
|
| 1131 |
-
"step": 780
|
| 1132 |
-
},
|
| 1133 |
-
{
|
| 1134 |
-
"epoch": 0.79,
|
| 1135 |
-
"learning_rate": 1.2352941176470587e-07,
|
| 1136 |
-
"logits/chosen": -1.7695019245147705,
|
| 1137 |
-
"logits/rejected": -2.03476881980896,
|
| 1138 |
-
"logps/chosen": -0.31666165590286255,
|
| 1139 |
-
"logps/rejected": -50.392616271972656,
|
| 1140 |
-
"loss": 0.5548,
|
| 1141 |
-
"rewards/accuracies": 0.5,
|
| 1142 |
-
"rewards/chosen": 0.15865585207939148,
|
| 1143 |
-
"rewards/margins": 4.350362777709961,
|
| 1144 |
-
"rewards/rejected": -4.191707134246826,
|
| 1145 |
-
"step": 790
|
| 1146 |
-
},
|
| 1147 |
-
{
|
| 1148 |
-
"epoch": 0.8,
|
| 1149 |
-
"learning_rate": 1.176470588235294e-07,
|
| 1150 |
-
"logits/chosen": -1.8488401174545288,
|
| 1151 |
-
"logits/rejected": -2.141782283782959,
|
| 1152 |
-
"logps/chosen": -0.3054754436016083,
|
| 1153 |
-
"logps/rejected": -34.490135192871094,
|
| 1154 |
-
"loss": 0.4853,
|
| 1155 |
-
"rewards/accuracies": 0.699999988079071,
|
| 1156 |
-
"rewards/chosen": 0.23360002040863037,
|
| 1157 |
-
"rewards/margins": 2.897139072418213,
|
| 1158 |
-
"rewards/rejected": -2.663539171218872,
|
| 1159 |
-
"step": 800
|
| 1160 |
-
},
|
| 1161 |
-
{
|
| 1162 |
-
"epoch": 0.81,
|
| 1163 |
-
"learning_rate": 1.1176470588235295e-07,
|
| 1164 |
-
"logits/chosen": -1.8038222789764404,
|
| 1165 |
-
"logits/rejected": -1.8816568851470947,
|
| 1166 |
-
"logps/chosen": -0.08031131327152252,
|
| 1167 |
-
"logps/rejected": -10.264053344726562,
|
| 1168 |
-
"loss": 0.623,
|
| 1169 |
-
"rewards/accuracies": 0.699999988079071,
|
| 1170 |
-
"rewards/chosen": 0.24331481754779816,
|
| 1171 |
-
"rewards/margins": 1.0161577463150024,
|
| 1172 |
-
"rewards/rejected": -0.7728429436683655,
|
| 1173 |
-
"step": 810
|
| 1174 |
-
},
|
| 1175 |
-
{
|
| 1176 |
-
"epoch": 0.82,
|
| 1177 |
-
"learning_rate": 1.0588235294117647e-07,
|
| 1178 |
-
"logits/chosen": -1.9077497720718384,
|
| 1179 |
-
"logits/rejected": -1.9633766412734985,
|
| 1180 |
-
"logps/chosen": -0.0009624619269743562,
|
| 1181 |
-
"logps/rejected": -12.475212097167969,
|
| 1182 |
-
"loss": 0.6238,
|
| 1183 |
-
"rewards/accuracies": 0.5,
|
| 1184 |
-
"rewards/chosen": 0.14481396973133087,
|
| 1185 |
-
"rewards/margins": 1.0594360828399658,
|
| 1186 |
-
"rewards/rejected": -0.9146221280097961,
|
| 1187 |
-
"step": 820
|
| 1188 |
-
},
|
| 1189 |
-
{
|
| 1190 |
-
"epoch": 0.83,
|
| 1191 |
-
"learning_rate": 1e-07,
|
| 1192 |
-
"logits/chosen": -1.6920783519744873,
|
| 1193 |
-
"logits/rejected": -1.9319671392440796,
|
| 1194 |
-
"logps/chosen": -0.013360400684177876,
|
| 1195 |
-
"logps/rejected": -46.24522018432617,
|
| 1196 |
-
"loss": 0.554,
|
| 1197 |
-
"rewards/accuracies": 0.6000000238418579,
|
| 1198 |
-
"rewards/chosen": 0.21019235253334045,
|
| 1199 |
-
"rewards/margins": 3.6845498085021973,
|
| 1200 |
-
"rewards/rejected": -3.4743571281433105,
|
| 1201 |
-
"step": 830
|
| 1202 |
-
},
|
| 1203 |
-
{
|
| 1204 |
-
"epoch": 0.84,
|
| 1205 |
-
"learning_rate": 9.411764705882353e-08,
|
| 1206 |
-
"logits/chosen": -1.7484924793243408,
|
| 1207 |
-
"logits/rejected": -2.0209856033325195,
|
| 1208 |
-
"logps/chosen": -0.1407555192708969,
|
| 1209 |
-
"logps/rejected": -43.27410125732422,
|
| 1210 |
-
"loss": 0.5543,
|
| 1211 |
-
"rewards/accuracies": 0.5,
|
| 1212 |
-
"rewards/chosen": 0.32391467690467834,
|
| 1213 |
-
"rewards/margins": 3.707659959793091,
|
| 1214 |
-
"rewards/rejected": -3.3837451934814453,
|
| 1215 |
-
"step": 840
|
| 1216 |
-
},
|
| 1217 |
-
{
|
| 1218 |
-
"epoch": 0.85,
|
| 1219 |
-
"learning_rate": 8.823529411764706e-08,
|
| 1220 |
-
"logits/chosen": -1.5334498882293701,
|
| 1221 |
-
"logits/rejected": -1.8631703853607178,
|
| 1222 |
-
"logps/chosen": -0.033944085240364075,
|
| 1223 |
-
"logps/rejected": -70.71894836425781,
|
| 1224 |
-
"loss": 0.4852,
|
| 1225 |
-
"rewards/accuracies": 0.5,
|
| 1226 |
-
"rewards/chosen": 0.23922660946846008,
|
| 1227 |
-
"rewards/margins": 6.69946813583374,
|
| 1228 |
-
"rewards/rejected": -6.46024227142334,
|
| 1229 |
-
"step": 850
|
| 1230 |
-
},
|
| 1231 |
-
{
|
| 1232 |
-
"epoch": 0.86,
|
| 1233 |
-
"learning_rate": 8.235294117647059e-08,
|
| 1234 |
-
"logits/chosen": -1.7201436758041382,
|
| 1235 |
-
"logits/rejected": -1.857613205909729,
|
| 1236 |
-
"logps/chosen": -0.10154436528682709,
|
| 1237 |
-
"logps/rejected": -27.654672622680664,
|
| 1238 |
-
"loss": 0.6237,
|
| 1239 |
-
"rewards/accuracies": 0.5,
|
| 1240 |
-
"rewards/chosen": 0.21049244701862335,
|
| 1241 |
-
"rewards/margins": 2.7311854362487793,
|
| 1242 |
-
"rewards/rejected": -2.520692825317383,
|
| 1243 |
-
"step": 860
|
| 1244 |
-
},
|
| 1245 |
-
{
|
| 1246 |
-
"epoch": 0.87,
|
| 1247 |
-
"learning_rate": 7.647058823529412e-08,
|
| 1248 |
-
"logits/chosen": -1.8255866765975952,
|
| 1249 |
-
"logits/rejected": -2.007643461227417,
|
| 1250 |
-
"logps/chosen": -0.12339305877685547,
|
| 1251 |
-
"logps/rejected": -23.465425491333008,
|
| 1252 |
-
"loss": 0.5546,
|
| 1253 |
-
"rewards/accuracies": 0.6000000238418579,
|
| 1254 |
-
"rewards/chosen": 0.37787729501724243,
|
| 1255 |
-
"rewards/margins": 2.0081450939178467,
|
| 1256 |
-
"rewards/rejected": -1.630267858505249,
|
| 1257 |
-
"step": 870
|
| 1258 |
-
},
|
| 1259 |
-
{
|
| 1260 |
-
"epoch": 0.88,
|
| 1261 |
-
"learning_rate": 7.058823529411765e-08,
|
| 1262 |
-
"logits/chosen": -1.675270676612854,
|
| 1263 |
-
"logits/rejected": -1.7470086812973022,
|
| 1264 |
-
"logps/chosen": -0.042650602757930756,
|
| 1265 |
-
"logps/rejected": -39.634857177734375,
|
| 1266 |
-
"loss": 0.5666,
|
| 1267 |
-
"rewards/accuracies": 0.6000000238418579,
|
| 1268 |
-
"rewards/chosen": 0.2249513566493988,
|
| 1269 |
-
"rewards/margins": 4.4023356437683105,
|
| 1270 |
-
"rewards/rejected": -4.177384376525879,
|
| 1271 |
-
"step": 880
|
| 1272 |
-
},
|
| 1273 |
-
{
|
| 1274 |
-
"epoch": 0.89,
|
| 1275 |
-
"learning_rate": 6.470588235294118e-08,
|
| 1276 |
-
"logits/chosen": -1.6610826253890991,
|
| 1277 |
-
"logits/rejected": -1.8155969381332397,
|
| 1278 |
-
"logps/chosen": -0.013693208806216717,
|
| 1279 |
-
"logps/rejected": -12.216829299926758,
|
| 1280 |
-
"loss": 0.6252,
|
| 1281 |
-
"rewards/accuracies": 0.4000000059604645,
|
| 1282 |
-
"rewards/chosen": 0.09875396639108658,
|
| 1283 |
-
"rewards/margins": 0.4886496961116791,
|
| 1284 |
-
"rewards/rejected": -0.3898957371711731,
|
| 1285 |
-
"step": 890
|
| 1286 |
-
},
|
| 1287 |
-
{
|
| 1288 |
-
"epoch": 0.9,
|
| 1289 |
-
"learning_rate": 5.88235294117647e-08,
|
| 1290 |
-
"logits/chosen": -2.034945487976074,
|
| 1291 |
-
"logits/rejected": -2.0349366664886475,
|
| 1292 |
-
"logps/chosen": -0.5003259181976318,
|
| 1293 |
-
"logps/rejected": -0.5002902746200562,
|
| 1294 |
-
"loss": 0.6941,
|
| 1295 |
-
"rewards/accuracies": 0.4000000059604645,
|
| 1296 |
-
"rewards/chosen": 0.26111483573913574,
|
| 1297 |
-
"rewards/margins": -0.0018190573900938034,
|
| 1298 |
-
"rewards/rejected": 0.2629338800907135,
|
| 1299 |
-
"step": 900
|
| 1300 |
-
},
|
| 1301 |
-
{
|
| 1302 |
-
"epoch": 0.91,
|
| 1303 |
-
"learning_rate": 5.294117647058823e-08,
|
| 1304 |
-
"logits/chosen": -1.7996795177459717,
|
| 1305 |
-
"logits/rejected": -1.7996826171875,
|
| 1306 |
-
"logps/chosen": -0.18381330370903015,
|
| 1307 |
-
"logps/rejected": -0.1839282214641571,
|
| 1308 |
-
"loss": 0.6927,
|
| 1309 |
-
"rewards/accuracies": 0.5,
|
| 1310 |
-
"rewards/chosen": 0.16678330302238464,
|
| 1311 |
-
"rewards/margins": 0.0009464531322009861,
|
| 1312 |
-
"rewards/rejected": 0.16583684086799622,
|
| 1313 |
-
"step": 910
|
| 1314 |
-
},
|
| 1315 |
-
{
|
| 1316 |
-
"epoch": 0.92,
|
| 1317 |
-
"learning_rate": 4.705882352941176e-08,
|
| 1318 |
-
"logits/chosen": -1.8732879161834717,
|
| 1319 |
-
"logits/rejected": -1.9507211446762085,
|
| 1320 |
-
"logps/chosen": -0.04379552602767944,
|
| 1321 |
-
"logps/rejected": -11.802661895751953,
|
| 1322 |
-
"loss": 0.624,
|
| 1323 |
-
"rewards/accuracies": 0.699999988079071,
|
| 1324 |
-
"rewards/chosen": 0.16234037280082703,
|
| 1325 |
-
"rewards/margins": 0.9312402009963989,
|
| 1326 |
-
"rewards/rejected": -0.76889967918396,
|
| 1327 |
-
"step": 920
|
| 1328 |
-
},
|
| 1329 |
-
{
|
| 1330 |
-
"epoch": 0.93,
|
| 1331 |
-
"learning_rate": 4.1176470588235293e-08,
|
| 1332 |
-
"logits/chosen": -1.722858190536499,
|
| 1333 |
-
"logits/rejected": -1.8730077743530273,
|
| 1334 |
-
"logps/chosen": -0.03575155884027481,
|
| 1335 |
-
"logps/rejected": -18.92257308959961,
|
| 1336 |
-
"loss": 0.5551,
|
| 1337 |
-
"rewards/accuracies": 0.5,
|
| 1338 |
-
"rewards/chosen": 0.16664153337478638,
|
| 1339 |
-
"rewards/margins": 1.2264705896377563,
|
| 1340 |
-
"rewards/rejected": -1.0598289966583252,
|
| 1341 |
-
"step": 930
|
| 1342 |
-
},
|
| 1343 |
-
{
|
| 1344 |
-
"epoch": 0.94,
|
| 1345 |
-
"learning_rate": 3.5294117647058824e-08,
|
| 1346 |
-
"logits/chosen": -1.4443856477737427,
|
| 1347 |
-
"logits/rejected": -1.968544363975525,
|
| 1348 |
-
"logps/chosen": -0.005307979416102171,
|
| 1349 |
-
"logps/rejected": -73.07327270507812,
|
| 1350 |
-
"loss": 0.2777,
|
| 1351 |
-
"rewards/accuracies": 0.800000011920929,
|
| 1352 |
-
"rewards/chosen": 0.241426020860672,
|
| 1353 |
-
"rewards/margins": 5.943356513977051,
|
| 1354 |
-
"rewards/rejected": -5.701930046081543,
|
| 1355 |
-
"step": 940
|
| 1356 |
-
},
|
| 1357 |
-
{
|
| 1358 |
-
"epoch": 0.95,
|
| 1359 |
-
"learning_rate": 2.941176470588235e-08,
|
| 1360 |
-
"logits/chosen": -1.917584776878357,
|
| 1361 |
-
"logits/rejected": -1.9409071207046509,
|
| 1362 |
-
"logps/chosen": -0.0067234160378575325,
|
| 1363 |
-
"logps/rejected": -7.634241580963135,
|
| 1364 |
-
"loss": 0.6239,
|
| 1365 |
-
"rewards/accuracies": 0.6000000238418579,
|
| 1366 |
-
"rewards/chosen": 0.07056023925542831,
|
| 1367 |
-
"rewards/margins": 1.0394136905670166,
|
| 1368 |
-
"rewards/rejected": -0.9688535928726196,
|
| 1369 |
-
"step": 950
|
| 1370 |
-
},
|
| 1371 |
-
{
|
| 1372 |
-
"epoch": 0.96,
|
| 1373 |
-
"learning_rate": 2.352941176470588e-08,
|
| 1374 |
-
"logits/chosen": -1.659624457359314,
|
| 1375 |
-
"logits/rejected": -1.8984715938568115,
|
| 1376 |
-
"logps/chosen": -0.6621147394180298,
|
| 1377 |
-
"logps/rejected": -41.00469207763672,
|
| 1378 |
-
"loss": 0.5561,
|
| 1379 |
-
"rewards/accuracies": 0.5,
|
| 1380 |
-
"rewards/chosen": 0.24302108585834503,
|
| 1381 |
-
"rewards/margins": 3.9856536388397217,
|
| 1382 |
-
"rewards/rejected": -3.7426326274871826,
|
| 1383 |
-
"step": 960
|
| 1384 |
-
},
|
| 1385 |
-
{
|
| 1386 |
-
"epoch": 0.97,
|
| 1387 |
-
"learning_rate": 1.7647058823529412e-08,
|
| 1388 |
-
"logits/chosen": -2.0915744304656982,
|
| 1389 |
-
"logits/rejected": -2.0915427207946777,
|
| 1390 |
-
"logps/chosen": -0.09594538807868958,
|
| 1391 |
-
"logps/rejected": -0.09524588286876678,
|
| 1392 |
-
"loss": 0.6935,
|
| 1393 |
-
"rewards/accuracies": 0.30000001192092896,
|
| 1394 |
-
"rewards/chosen": 0.11800364404916763,
|
| 1395 |
-
"rewards/margins": -0.000787623692303896,
|
| 1396 |
-
"rewards/rejected": 0.11879126727581024,
|
| 1397 |
-
"step": 970
|
| 1398 |
-
},
|
| 1399 |
-
{
|
| 1400 |
-
"epoch": 0.98,
|
| 1401 |
-
"learning_rate": 1.176470588235294e-08,
|
| 1402 |
-
"logits/chosen": -1.8284385204315186,
|
| 1403 |
-
"logits/rejected": -2.00864839553833,
|
| 1404 |
-
"logps/chosen": -0.2936480641365051,
|
| 1405 |
-
"logps/rejected": -42.53371810913086,
|
| 1406 |
-
"loss": 0.5554,
|
| 1407 |
-
"rewards/accuracies": 0.5,
|
| 1408 |
-
"rewards/chosen": 0.20751357078552246,
|
| 1409 |
-
"rewards/margins": 4.451984405517578,
|
| 1410 |
-
"rewards/rejected": -4.244471073150635,
|
| 1411 |
-
"step": 980
|
| 1412 |
-
},
|
| 1413 |
-
{
|
| 1414 |
-
"epoch": 0.99,
|
| 1415 |
-
"learning_rate": 5.88235294117647e-09,
|
| 1416 |
-
"logits/chosen": -1.9873313903808594,
|
| 1417 |
-
"logits/rejected": -1.9968726634979248,
|
| 1418 |
-
"logps/chosen": -0.013185739517211914,
|
| 1419 |
-
"logps/rejected": -2.775447368621826,
|
| 1420 |
-
"loss": 0.6283,
|
| 1421 |
-
"rewards/accuracies": 0.4000000059604645,
|
| 1422 |
-
"rewards/chosen": 0.1859985888004303,
|
| 1423 |
-
"rewards/margins": 0.2977893352508545,
|
| 1424 |
-
"rewards/rejected": -0.1117907166481018,
|
| 1425 |
-
"step": 990
|
| 1426 |
-
},
|
| 1427 |
-
{
|
| 1428 |
-
"epoch": 1.0,
|
| 1429 |
-
"learning_rate": 0.0,
|
| 1430 |
-
"logits/chosen": -1.8508293628692627,
|
| 1431 |
-
"logits/rejected": -2.014707088470459,
|
| 1432 |
-
"logps/chosen": -0.03528173640370369,
|
| 1433 |
-
"logps/rejected": -29.331050872802734,
|
| 1434 |
-
"loss": 0.5542,
|
| 1435 |
-
"rewards/accuracies": 0.699999988079071,
|
| 1436 |
-
"rewards/chosen": 0.2999701201915741,
|
| 1437 |
-
"rewards/margins": 2.2009005546569824,
|
| 1438 |
-
"rewards/rejected": -1.900930404663086,
|
| 1439 |
-
"step": 1000
|
| 1440 |
-
},
|
| 1441 |
-
{
|
| 1442 |
-
"epoch": 1.0,
|
| 1443 |
-
"eval_logits/chosen": -1.8047094345092773,
|
| 1444 |
-
"eval_logits/rejected": -1.954627513885498,
|
| 1445 |
-
"eval_logps/chosen": -0.11939573287963867,
|
| 1446 |
-
"eval_logps/rejected": -22.671850204467773,
|
| 1447 |
-
"eval_loss": 0.5534428954124451,
|
| 1448 |
-
"eval_rewards/accuracies": 0.5099999904632568,
|
| 1449 |
-
"eval_rewards/chosen": 0.23217196762561798,
|
| 1450 |
-
"eval_rewards/margins": 1.7574390172958374,
|
| 1451 |
-
"eval_rewards/rejected": -1.5252670049667358,
|
| 1452 |
-
"eval_runtime": 20.0707,
|
| 1453 |
-
"eval_samples_per_second": 4.982,
|
| 1454 |
-
"eval_steps_per_second": 4.982,
|
| 1455 |
-
"step": 1000
|
| 1456 |
-
}
|
| 1457 |
-
],
|
| 1458 |
-
"logging_steps": 10,
|
| 1459 |
-
"max_steps": 1000,
|
| 1460 |
-
"num_train_epochs": 1,
|
| 1461 |
-
"save_steps": 500,
|
| 1462 |
-
"total_flos": 0.0,
|
| 1463 |
-
"trial_name": null,
|
| 1464 |
-
"trial_params": null
|
| 1465 |
-
}
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|