File size: 12,596 Bytes
cd83cf6 |
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": 0.3418803418803419,
"eval_steps": 50,
"global_step": 100,
"is_hyper_param_search": false,
"is_local_process_zero": true,
"is_world_process_zero": true,
"log_history": [
{
"epoch": 0.017094017094017096,
"grad_norm": 35.00202204303521,
"learning_rate": 5e-07,
"logits/chosen": -2.7455849647521973,
"logits/rejected": -2.7442612648010254,
"logps/chosen": -164.2725830078125,
"logps/rejected": -170.57113647460938,
"loss": 0.6934,
"rewards/accuracies": 0.23749999701976776,
"rewards/chosen": 0.0026612328365445137,
"rewards/margins": -0.001539617427624762,
"rewards/rejected": 0.004200850613415241,
"step": 5
},
{
"epoch": 0.03418803418803419,
"grad_norm": 36.29266486314593,
"learning_rate": 1e-06,
"logits/chosen": -2.709902763366699,
"logits/rejected": -2.7155404090881348,
"logps/chosen": -171.80032348632812,
"logps/rejected": -165.20169067382812,
"loss": 0.6879,
"rewards/accuracies": 0.5062500238418579,
"rewards/chosen": 0.012009668163955212,
"rewards/margins": 0.0021203968208283186,
"rewards/rejected": 0.009889272041618824,
"step": 10
},
{
"epoch": 0.05128205128205128,
"grad_norm": 33.83921269470837,
"learning_rate": 9.999177507263144e-07,
"logits/chosen": -2.6502068042755127,
"logits/rejected": -2.628007411956787,
"logps/chosen": -174.082275390625,
"logps/rejected": -174.13429260253906,
"loss": 0.6698,
"rewards/accuracies": 0.643750011920929,
"rewards/chosen": 0.23495244979858398,
"rewards/margins": 0.1125468835234642,
"rewards/rejected": 0.12240554392337799,
"step": 15
},
{
"epoch": 0.06837606837606838,
"grad_norm": 34.14427373918799,
"learning_rate": 9.996710299650301e-07,
"logits/chosen": -2.473665714263916,
"logits/rejected": -2.4469008445739746,
"logps/chosen": -158.2163848876953,
"logps/rejected": -158.0710906982422,
"loss": 0.661,
"rewards/accuracies": 0.581250011920929,
"rewards/chosen": 0.4233472943305969,
"rewards/margins": 0.1434161365032196,
"rewards/rejected": 0.2799311578273773,
"step": 20
},
{
"epoch": 0.08547008547008547,
"grad_norm": 33.2696083475879,
"learning_rate": 9.992599188865604e-07,
"logits/chosen": -2.314507007598877,
"logits/rejected": -2.3168132305145264,
"logps/chosen": -150.67019653320312,
"logps/rejected": -156.8417510986328,
"loss": 0.6501,
"rewards/accuracies": 0.543749988079071,
"rewards/chosen": 0.4975205063819885,
"rewards/margins": 0.15743504464626312,
"rewards/rejected": 0.3400854766368866,
"step": 25
},
{
"epoch": 0.10256410256410256,
"grad_norm": 34.42253361988952,
"learning_rate": 9.98684552745256e-07,
"logits/chosen": -2.243194103240967,
"logits/rejected": -2.251340866088867,
"logps/chosen": -161.2266845703125,
"logps/rejected": -161.32298278808594,
"loss": 0.6289,
"rewards/accuracies": 0.6187499761581421,
"rewards/chosen": 0.4243805408477783,
"rewards/margins": 0.2635195851325989,
"rewards/rejected": 0.16086098551750183,
"step": 30
},
{
"epoch": 0.11965811965811966,
"grad_norm": 31.414296706456245,
"learning_rate": 9.979451208349055e-07,
"logits/chosen": -2.30315899848938,
"logits/rejected": -2.289762496948242,
"logps/chosen": -171.71713256835938,
"logps/rejected": -174.50900268554688,
"loss": 0.6296,
"rewards/accuracies": 0.65625,
"rewards/chosen": -0.019384615123271942,
"rewards/margins": 0.318477988243103,
"rewards/rejected": -0.3378625512123108,
"step": 35
},
{
"epoch": 0.13675213675213677,
"grad_norm": 32.071830655862556,
"learning_rate": 9.970418664264595e-07,
"logits/chosen": -2.3935599327087402,
"logits/rejected": -2.3812546730041504,
"logps/chosen": -171.0698699951172,
"logps/rejected": -176.58578491210938,
"loss": 0.5991,
"rewards/accuracies": 0.675000011920929,
"rewards/chosen": -0.26089853048324585,
"rewards/margins": 0.5235068202018738,
"rewards/rejected": -0.7844053506851196,
"step": 40
},
{
"epoch": 0.15384615384615385,
"grad_norm": 36.19466541168301,
"learning_rate": 9.95975086687994e-07,
"logits/chosen": -2.4914021492004395,
"logits/rejected": -2.4973323345184326,
"logps/chosen": -163.68099975585938,
"logps/rejected": -167.174072265625,
"loss": 0.6141,
"rewards/accuracies": 0.7250000238418579,
"rewards/chosen": -0.2962096929550171,
"rewards/margins": 0.4588828682899475,
"rewards/rejected": -0.7550925016403198,
"step": 45
},
{
"epoch": 0.17094017094017094,
"grad_norm": 31.16276115760231,
"learning_rate": 9.947451325869439e-07,
"logits/chosen": -2.5575203895568848,
"logits/rejected": -2.557717800140381,
"logps/chosen": -172.04318237304688,
"logps/rejected": -177.67672729492188,
"loss": 0.5777,
"rewards/accuracies": 0.731249988079071,
"rewards/chosen": -0.2069791853427887,
"rewards/margins": 0.6018465757369995,
"rewards/rejected": -0.808825671672821,
"step": 50
},
{
"epoch": 0.17094017094017094,
"eval_logits/chosen": -2.5221025943756104,
"eval_logits/rejected": -2.5152711868286133,
"eval_logps/chosen": -163.01820373535156,
"eval_logps/rejected": -169.54832458496094,
"eval_loss": 0.5812540650367737,
"eval_rewards/accuracies": 0.6682692170143127,
"eval_rewards/chosen": -0.45408713817596436,
"eval_rewards/margins": 0.6127156615257263,
"eval_rewards/rejected": -1.0668028593063354,
"eval_runtime": 510.3361,
"eval_samples_per_second": 16.291,
"eval_steps_per_second": 0.255,
"step": 50
},
{
"epoch": 0.18803418803418803,
"grad_norm": 31.575578721339145,
"learning_rate": 9.933524087746347e-07,
"logits/chosen": -2.490377426147461,
"logits/rejected": -2.4825081825256348,
"logps/chosen": -168.06161499023438,
"logps/rejected": -175.0494384765625,
"loss": 0.5706,
"rewards/accuracies": 0.737500011920929,
"rewards/chosen": -0.5060762763023376,
"rewards/margins": 0.7589826583862305,
"rewards/rejected": -1.2650587558746338,
"step": 55
},
{
"epoch": 0.20512820512820512,
"grad_norm": 30.171745273288415,
"learning_rate": 9.917973734531549e-07,
"logits/chosen": -2.48228120803833,
"logits/rejected": -2.4833157062530518,
"logps/chosen": -159.47142028808594,
"logps/rejected": -170.63671875,
"loss": 0.5753,
"rewards/accuracies": 0.675000011920929,
"rewards/chosen": -0.35752761363983154,
"rewards/margins": 0.5991309881210327,
"rewards/rejected": -0.9566585421562195,
"step": 60
},
{
"epoch": 0.2222222222222222,
"grad_norm": 32.13878319029882,
"learning_rate": 9.90080538224607e-07,
"logits/chosen": -2.585407018661499,
"logits/rejected": -2.5767769813537598,
"logps/chosen": -157.43936157226562,
"logps/rejected": -166.13589477539062,
"loss": 0.566,
"rewards/accuracies": 0.6937500238418579,
"rewards/chosen": -0.02057185396552086,
"rewards/margins": 0.47568243741989136,
"rewards/rejected": -0.4962543547153473,
"step": 65
},
{
"epoch": 0.23931623931623933,
"grad_norm": 29.494674721856043,
"learning_rate": 9.882024679227938e-07,
"logits/chosen": -2.6504979133605957,
"logits/rejected": -2.6398470401763916,
"logps/chosen": -178.0801239013672,
"logps/rejected": -179.46328735351562,
"loss": 0.5444,
"rewards/accuracies": 0.706250011920929,
"rewards/chosen": -0.43436694145202637,
"rewards/margins": 0.8427752256393433,
"rewards/rejected": -1.27714204788208,
"step": 70
},
{
"epoch": 0.2564102564102564,
"grad_norm": 28.856733948308104,
"learning_rate": 9.861637804273881e-07,
"logits/chosen": -2.660489082336426,
"logits/rejected": -2.655539035797119,
"logps/chosen": -162.1233673095703,
"logps/rejected": -170.16131591796875,
"loss": 0.5568,
"rewards/accuracies": 0.699999988079071,
"rewards/chosen": -0.4032784402370453,
"rewards/margins": 0.6959114074707031,
"rewards/rejected": -1.0991899967193604,
"step": 75
},
{
"epoch": 0.27350427350427353,
"grad_norm": 26.646061534818323,
"learning_rate": 9.83965146460653e-07,
"logits/chosen": -2.6391615867614746,
"logits/rejected": -2.628577709197998,
"logps/chosen": -168.58099365234375,
"logps/rejected": -179.22805786132812,
"loss": 0.5448,
"rewards/accuracies": 0.6812499761581421,
"rewards/chosen": -0.6665827035903931,
"rewards/margins": 0.8240470886230469,
"rewards/rejected": -1.4906299114227295,
"step": 80
},
{
"epoch": 0.2905982905982906,
"grad_norm": 36.04159750418885,
"learning_rate": 9.816072893667758e-07,
"logits/chosen": -2.6322970390319824,
"logits/rejected": -2.6053385734558105,
"logps/chosen": -174.82640075683594,
"logps/rejected": -186.0735626220703,
"loss": 0.5579,
"rewards/accuracies": 0.7562500238418579,
"rewards/chosen": -1.0639268159866333,
"rewards/margins": 1.0258175134658813,
"rewards/rejected": -2.0897443294525146,
"step": 85
},
{
"epoch": 0.3076923076923077,
"grad_norm": 26.922939193632168,
"learning_rate": 9.790909848738904e-07,
"logits/chosen": -2.60801362991333,
"logits/rejected": -2.6101624965667725,
"logps/chosen": -176.20538330078125,
"logps/rejected": -184.7812957763672,
"loss": 0.5215,
"rewards/accuracies": 0.6312500238418579,
"rewards/chosen": -0.9929834604263306,
"rewards/margins": 0.8646324276924133,
"rewards/rejected": -1.8576160669326782,
"step": 90
},
{
"epoch": 0.3247863247863248,
"grad_norm": 30.3564450245371,
"learning_rate": 9.764170608388647e-07,
"logits/chosen": -2.6054036617279053,
"logits/rejected": -2.5733799934387207,
"logps/chosen": -168.037109375,
"logps/rejected": -174.51144409179688,
"loss": 0.5197,
"rewards/accuracies": 0.762499988079071,
"rewards/chosen": -0.6652337312698364,
"rewards/margins": 1.060430884361267,
"rewards/rejected": -1.725664734840393,
"step": 95
},
{
"epoch": 0.3418803418803419,
"grad_norm": 28.936164680674203,
"learning_rate": 9.735863969749371e-07,
"logits/chosen": -2.5255179405212402,
"logits/rejected": -2.4874520301818848,
"logps/chosen": -177.73861694335938,
"logps/rejected": -189.82369995117188,
"loss": 0.4982,
"rewards/accuracies": 0.706250011920929,
"rewards/chosen": -0.8511013984680176,
"rewards/margins": 1.1354777812957764,
"rewards/rejected": -1.986579179763794,
"step": 100
},
{
"epoch": 0.3418803418803419,
"eval_logits/chosen": -2.484687328338623,
"eval_logits/rejected": -2.460559368133545,
"eval_logps/chosen": -168.28323364257812,
"eval_logps/rejected": -180.8539276123047,
"eval_loss": 0.5161151885986328,
"eval_rewards/accuracies": 0.7211538553237915,
"eval_rewards/chosen": -0.9805887937545776,
"eval_rewards/margins": 1.2167747020721436,
"eval_rewards/rejected": -2.1973636150360107,
"eval_runtime": 510.3447,
"eval_samples_per_second": 16.291,
"eval_steps_per_second": 0.255,
"step": 100
}
],
"logging_steps": 5,
"max_steps": 876,
"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
}
|