File size: 12,523 Bytes
eca7497 |
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65 66 67 68 69 70 71 72 73 74 75 76 77 78 79 80 81 82 83 84 85 86 87 88 89 90 91 92 93 94 95 96 97 98 99 100 101 102 103 104 105 106 107 108 109 110 111 112 113 114 115 116 117 118 119 120 121 122 123 124 125 126 127 128 129 130 131 132 133 134 135 136 137 138 139 140 141 142 143 144 145 146 147 148 149 150 151 152 153 154 155 156 157 158 159 160 161 162 163 164 165 166 167 168 169 170 171 172 173 174 175 176 177 178 179 180 181 182 183 184 185 186 187 188 189 190 191 192 193 194 195 196 197 198 199 200 201 202 203 204 205 206 207 208 209 210 211 212 213 214 215 216 217 218 219 220 221 222 223 224 225 226 227 228 229 230 231 232 233 234 235 236 237 238 239 240 241 242 243 244 245 246 247 248 249 250 251 252 253 254 255 256 257 258 259 260 261 262 263 264 265 266 267 268 269 270 271 272 273 274 275 276 277 278 279 280 281 282 283 284 285 286 287 288 289 290 291 292 293 294 295 296 297 298 299 300 301 302 303 304 305 306 307 308 309 310 311 312 313 314 315 316 317 318 319 320 321 322 323 324 325 326 327 328 329 330 331 332 333 334 335 336 337 338 339 340 341 342 343 344 345 346 347 348 349 350 351 352 353 354 355 356 357 358 359 360 361 362 363 364 365 366 |
{
"best_metric": null,
"best_model_checkpoint": null,
"epoch": 1.4925373134328357,
"eval_steps": 50,
"global_step": 100,
"is_hyper_param_search": false,
"is_local_process_zero": true,
"is_world_process_zero": true,
"log_history": [
{
"epoch": 0.07462686567164178,
"grad_norm": 60.46895766728989,
"learning_rate": 5e-07,
"logits/chosen": -2.737339735031128,
"logits/rejected": -2.717646837234497,
"logps/chosen": -269.46966552734375,
"logps/rejected": -218.519287109375,
"loss": 0.6897,
"rewards/accuracies": 0.33125001192092896,
"rewards/chosen": 0.016166506335139275,
"rewards/margins": 0.005314634181559086,
"rewards/rejected": 0.010851874016225338,
"step": 5
},
{
"epoch": 0.14925373134328357,
"grad_norm": 47.55077346828281,
"learning_rate": 1e-06,
"logits/chosen": -2.707444190979004,
"logits/rejected": -2.6988959312438965,
"logps/chosen": -259.7918395996094,
"logps/rejected": -198.8335723876953,
"loss": 0.6419,
"rewards/accuracies": 0.668749988079071,
"rewards/chosen": 0.5464529991149902,
"rewards/margins": 0.22305770218372345,
"rewards/rejected": 0.323395311832428,
"step": 10
},
{
"epoch": 0.22388059701492538,
"grad_norm": 49.794624138997065,
"learning_rate": 9.983100718730718e-07,
"logits/chosen": -2.5312893390655518,
"logits/rejected": -2.528350353240967,
"logps/chosen": -250.9792022705078,
"logps/rejected": -215.8843231201172,
"loss": 0.6325,
"rewards/accuracies": 0.668749988079071,
"rewards/chosen": 1.2436325550079346,
"rewards/margins": 0.6625381708145142,
"rewards/rejected": 0.5810943841934204,
"step": 15
},
{
"epoch": 0.29850746268656714,
"grad_norm": 45.89015113492354,
"learning_rate": 9.932517109205849e-07,
"logits/chosen": -2.345860719680786,
"logits/rejected": -2.330219030380249,
"logps/chosen": -261.6305236816406,
"logps/rejected": -205.9855499267578,
"loss": 0.5741,
"rewards/accuracies": 0.731249988079071,
"rewards/chosen": 1.5929079055786133,
"rewards/margins": 0.922720730304718,
"rewards/rejected": 0.6701873540878296,
"step": 20
},
{
"epoch": 0.373134328358209,
"grad_norm": 63.53673896186562,
"learning_rate": 9.848591102083375e-07,
"logits/chosen": -2.2251970767974854,
"logits/rejected": -2.210585832595825,
"logps/chosen": -256.94158935546875,
"logps/rejected": -198.90554809570312,
"loss": 0.5846,
"rewards/accuracies": 0.699999988079071,
"rewards/chosen": 1.4611073732376099,
"rewards/margins": 0.925146758556366,
"rewards/rejected": 0.5359605550765991,
"step": 25
},
{
"epoch": 0.44776119402985076,
"grad_norm": 40.6998829965568,
"learning_rate": 9.731890013043367e-07,
"logits/chosen": -2.203108310699463,
"logits/rejected": -2.188232898712158,
"logps/chosen": -220.20474243164062,
"logps/rejected": -174.1934356689453,
"loss": 0.5755,
"rewards/accuracies": 0.731249988079071,
"rewards/chosen": 1.1598186492919922,
"rewards/margins": 1.0439460277557373,
"rewards/rejected": 0.11587252467870712,
"step": 30
},
{
"epoch": 0.5223880597014925,
"grad_norm": 39.86530018922445,
"learning_rate": 9.583202707897073e-07,
"logits/chosen": -2.159541606903076,
"logits/rejected": -2.146280288696289,
"logps/chosen": -271.1134033203125,
"logps/rejected": -222.50732421875,
"loss": 0.5581,
"rewards/accuracies": 0.78125,
"rewards/chosen": 1.6402053833007812,
"rewards/margins": 1.4621944427490234,
"rewards/rejected": 0.1780107319355011,
"step": 35
},
{
"epoch": 0.5970149253731343,
"grad_norm": 47.764041993597665,
"learning_rate": 9.403534270080829e-07,
"logits/chosen": -2.097519874572754,
"logits/rejected": -2.0528969764709473,
"logps/chosen": -246.8362274169922,
"logps/rejected": -208.1300048828125,
"loss": 0.5329,
"rewards/accuracies": 0.668749988079071,
"rewards/chosen": 1.0587269067764282,
"rewards/margins": 1.0248701572418213,
"rewards/rejected": 0.033856604248285294,
"step": 40
},
{
"epoch": 0.6716417910447762,
"grad_norm": 33.33992129726437,
"learning_rate": 9.19409920658098e-07,
"logits/chosen": -2.1117985248565674,
"logits/rejected": -2.0575757026672363,
"logps/chosen": -238.47335815429688,
"logps/rejected": -232.6268310546875,
"loss": 0.5597,
"rewards/accuracies": 0.706250011920929,
"rewards/chosen": 1.020668387413025,
"rewards/margins": 1.3137915134429932,
"rewards/rejected": -0.2931232154369354,
"step": 45
},
{
"epoch": 0.746268656716418,
"grad_norm": 36.75599018498795,
"learning_rate": 8.956313238215823e-07,
"logits/chosen": -2.192502021789551,
"logits/rejected": -2.1830239295959473,
"logps/chosen": -246.92733764648438,
"logps/rejected": -219.84817504882812,
"loss": 0.5242,
"rewards/accuracies": 0.7437499761581421,
"rewards/chosen": 0.8840469121932983,
"rewards/margins": 1.3445603847503662,
"rewards/rejected": -0.4605133533477783,
"step": 50
},
{
"epoch": 0.746268656716418,
"eval_logits/chosen": -2.2182116508483887,
"eval_logits/rejected": -2.195751190185547,
"eval_logps/chosen": -256.5732727050781,
"eval_logps/rejected": -219.44544982910156,
"eval_loss": 0.5641670823097229,
"eval_rewards/accuracies": 0.7583333253860474,
"eval_rewards/chosen": 0.8116810321807861,
"eval_rewards/margins": 1.5396502017974854,
"eval_rewards/rejected": -0.7279691696166992,
"eval_runtime": 125.8473,
"eval_samples_per_second": 15.098,
"eval_steps_per_second": 0.238,
"step": 50
},
{
"epoch": 0.8208955223880597,
"grad_norm": 33.66480221016764,
"learning_rate": 8.691783729769873e-07,
"logits/chosen": -2.252897024154663,
"logits/rejected": -2.210509777069092,
"logps/chosen": -273.36102294921875,
"logps/rejected": -232.4573974609375,
"loss": 0.5359,
"rewards/accuracies": 0.793749988079071,
"rewards/chosen": 1.0710394382476807,
"rewards/margins": 1.8447446823120117,
"rewards/rejected": -0.7737051248550415,
"step": 55
},
{
"epoch": 0.8955223880597015,
"grad_norm": 42.22753576523594,
"learning_rate": 8.402298824670029e-07,
"logits/chosen": -2.2703890800476074,
"logits/rejected": -2.252363443374634,
"logps/chosen": -274.8329162597656,
"logps/rejected": -254.45663452148438,
"loss": 0.5811,
"rewards/accuracies": 0.699999988079071,
"rewards/chosen": 0.9601489901542664,
"rewards/margins": 1.3747169971466064,
"rewards/rejected": -0.4145679473876953,
"step": 60
},
{
"epoch": 0.9701492537313433,
"grad_norm": 38.82616914526827,
"learning_rate": 8.089815357650089e-07,
"logits/chosen": -2.237731456756592,
"logits/rejected": -2.245671510696411,
"logps/chosen": -264.57989501953125,
"logps/rejected": -241.2982940673828,
"loss": 0.5166,
"rewards/accuracies": 0.7562500238418579,
"rewards/chosen": 1.0662319660186768,
"rewards/margins": 1.8653961420059204,
"rewards/rejected": -0.799164354801178,
"step": 65
},
{
"epoch": 1.044776119402985,
"grad_norm": 21.744667598719886,
"learning_rate": 7.756445627110522e-07,
"logits/chosen": -2.2960398197174072,
"logits/rejected": -2.2724010944366455,
"logps/chosen": -264.32720947265625,
"logps/rejected": -239.6345977783203,
"loss": 0.346,
"rewards/accuracies": 0.856249988079071,
"rewards/chosen": 2.158931255340576,
"rewards/margins": 2.908942222595215,
"rewards/rejected": -0.7500110864639282,
"step": 70
},
{
"epoch": 1.1194029850746268,
"grad_norm": 19.678947962162013,
"learning_rate": 7.404443116588547e-07,
"logits/chosen": -2.396239995956421,
"logits/rejected": -2.362546443939209,
"logps/chosen": -246.3358154296875,
"logps/rejected": -195.57357788085938,
"loss": 0.2283,
"rewards/accuracies": 0.918749988079071,
"rewards/chosen": 2.2872486114501953,
"rewards/margins": 2.971306562423706,
"rewards/rejected": -0.6840583086013794,
"step": 75
},
{
"epoch": 1.1940298507462686,
"grad_norm": 20.2993371431914,
"learning_rate": 7.036187261857288e-07,
"logits/chosen": -2.4099860191345215,
"logits/rejected": -2.384648084640503,
"logps/chosen": -260.5028381347656,
"logps/rejected": -223.2133331298828,
"loss": 0.2312,
"rewards/accuracies": 0.9125000238418579,
"rewards/chosen": 2.834780693054199,
"rewards/margins": 3.5776984691619873,
"rewards/rejected": -0.7429176568984985,
"step": 80
},
{
"epoch": 1.2686567164179103,
"grad_norm": 25.4573275034557,
"learning_rate": 6.654167366624008e-07,
"logits/chosen": -2.36556339263916,
"logits/rejected": -2.349255084991455,
"logps/chosen": -231.6350860595703,
"logps/rejected": -221.1610870361328,
"loss": 0.2307,
"rewards/accuracies": 0.9437500238418579,
"rewards/chosen": 1.9625335931777954,
"rewards/margins": 3.2484028339385986,
"rewards/rejected": -1.2858693599700928,
"step": 85
},
{
"epoch": 1.3432835820895521,
"grad_norm": 29.79365948091661,
"learning_rate": 6.260965775552713e-07,
"logits/chosen": -2.374537944793701,
"logits/rejected": -2.348567485809326,
"logps/chosen": -247.57614135742188,
"logps/rejected": -238.37899780273438,
"loss": 0.2523,
"rewards/accuracies": 0.925000011920929,
"rewards/chosen": 1.934003233909607,
"rewards/margins": 3.8235995769500732,
"rewards/rejected": -1.8895965814590454,
"step": 90
},
{
"epoch": 1.417910447761194,
"grad_norm": 22.109779497676236,
"learning_rate": 5.859240418356614e-07,
"logits/chosen": -2.3896210193634033,
"logits/rejected": -2.3736777305603027,
"logps/chosen": -257.51898193359375,
"logps/rejected": -228.7509765625,
"loss": 0.2453,
"rewards/accuracies": 0.9125000238418579,
"rewards/chosen": 2.010754108428955,
"rewards/margins": 3.4373581409454346,
"rewards/rejected": -1.4266036748886108,
"step": 95
},
{
"epoch": 1.4925373134328357,
"grad_norm": 18.59821752381129,
"learning_rate": 5.451706842957421e-07,
"logits/chosen": -2.3828063011169434,
"logits/rejected": -2.3802871704101562,
"logps/chosen": -238.89547729492188,
"logps/rejected": -224.2349395751953,
"loss": 0.2523,
"rewards/accuracies": 0.918749988079071,
"rewards/chosen": 2.049046039581299,
"rewards/margins": 3.4148106575012207,
"rewards/rejected": -1.3657643795013428,
"step": 100
},
{
"epoch": 1.4925373134328357,
"eval_logits/chosen": -2.3702290058135986,
"eval_logits/rejected": -2.352924108505249,
"eval_logps/chosen": -250.75942993164062,
"eval_logps/rejected": -224.2054901123047,
"eval_loss": 0.5878031253814697,
"eval_rewards/accuracies": 0.8041666746139526,
"eval_rewards/chosen": 1.393066644668579,
"eval_rewards/margins": 2.597040891647339,
"eval_rewards/rejected": -1.2039742469787598,
"eval_runtime": 125.6186,
"eval_samples_per_second": 15.125,
"eval_steps_per_second": 0.239,
"step": 100
}
],
"logging_steps": 5,
"max_steps": 201,
"num_input_tokens_seen": 0,
"num_train_epochs": 3,
"save_steps": 100,
"stateful_callbacks": {
"TrainerControl": {
"args": {
"should_epoch_stop": false,
"should_evaluate": false,
"should_log": false,
"should_save": true,
"should_training_stop": false
},
"attributes": {}
}
},
"total_flos": 1178822762299392.0,
"train_batch_size": 8,
"trial_name": null,
"trial_params": null
}
|