zerozeroz/CodeLlama-7b-hf_1000rl_cpp
015
1{2 "best_metric": null,3 "best_model_checkpoint": null,4 "epoch": 2.9734513274336285,5 "eval_steps": 500,6 "global_step": 169,7 "is_hyper_param_search": false,8 "is_local_process_zero": true,9 "is_world_process_zero": true,10 "log_history": [11 {12 "completion_length": 186.12500381469727,13 "epoch": 0.017699115044247787,14 "grad_norm": 2.5752858647583037,15 "kl": 0.0,16 "learning_rate": 8.333333333333333e-08,17 "loss": 0.0,18 "reward": 0.48444899916648865,19 "reward_std": 0.46400773525238037,20 "rewards/correct_code_reward_func": 0.3125,21 "rewards/len_reward_func": 0.17194896191358566,22 "step": 123 },24 {25 "completion_length": 164.375,26 "epoch": 0.035398230088495575,27 "grad_norm": 10.223894549317365,28 "kl": 0.0,29 "learning_rate": 1.6666666666666665e-07,30 "loss": 0.0,31 "reward": 0.47450903058052063,32 "reward_std": 0.40167468786239624,33 "rewards/correct_code_reward_func": 0.2500000149011612,34 "rewards/len_reward_func": 0.22450900077819824,35 "step": 236 },37 {38 "completion_length": 164.4791717529297,39 "epoch": 0.05309734513274336,40 "grad_norm": 2.2899995314753827,41 "kl": 4.023313522338867e-05,42 "learning_rate": 2.5e-07,43 "loss": 0.0,44 "reward": 0.37769100069999695,45 "reward_std": 0.5044284015893936,46 "rewards/correct_code_reward_func": 0.2083333432674408,47 "rewards/len_reward_func": 0.16935766488313675,48 "step": 349 },50 {51 "completion_length": 190.12500762939453,52 "epoch": 0.07079646017699115,53 "grad_norm": 6.009103649538728,54 "kl": 8.815526962280273e-05,55 "learning_rate": 3.333333333333333e-07,56 "loss": 0.0,57 "reward": 0.37105801701545715,58 "reward_std": 0.47604119777679443,59 "rewards/correct_code_reward_func": 0.2083333395421505,60 "rewards/len_reward_func": 0.16272468864917755,61 "step": 462 },63 {64 "completion_length": 126.10416793823242,65 "epoch": 0.08849557522123894,66 "grad_norm": 2.3931977443517964,67 "kl": 3.49879264831543e-05,68 "learning_rate": 4.1666666666666667e-07,69 "loss": 0.0,70 "reward": 0.5486156344413757,71 "reward_std": 0.40741610527038574,72 "rewards/correct_code_reward_func": 0.3958333358168602,73 "rewards/len_reward_func": 0.15278229862451553,74 "step": 575 },76 {77 "completion_length": 166.45833587646484,78 "epoch": 0.10619469026548672,79 "grad_norm": 1.5486991749328145,80 "kl": 5.221366882324219e-05,81 "learning_rate": 5e-07,82 "loss": 0.0,83 "reward": 0.5515251755714417,84 "reward_std": 0.47483181953430176,85 "rewards/correct_code_reward_func": 0.2916666716337204,86 "rewards/len_reward_func": 0.25985850393772125,87 "step": 688 },89 {90 "completion_length": 161.58333587646484,91 "epoch": 0.12389380530973451,92 "grad_norm": 5.714527026195051,93 "kl": 5.5789947509765625e-05,94 "learning_rate": 4.999535676028337e-07,95 "loss": 0.0,96 "reward": 0.46977874636650085,97 "reward_std": 0.461082324385643,98 "rewards/correct_code_reward_func": 0.229166679084301,99 "rewards/len_reward_func": 0.24061208218336105,100 "step": 7101 },102 {103 "completion_length": 161.89583587646484,104 "epoch": 0.1415929203539823,105 "grad_norm": 3.6484747523940286,106 "kl": 0.00010442733764648438,107 "learning_rate": 4.998142876590749e-07,108 "loss": 0.0,109 "reward": 0.2943377196788788,110 "reward_std": 0.3771718442440033,111 "rewards/correct_code_reward_func": 0.1666666679084301,112 "rewards/len_reward_func": 0.12767105549573898,113 "step": 8114 },115 {116 "completion_length": 147.7916717529297,117 "epoch": 0.1592920353982301,118 "grad_norm": 14.356223088928283,119 "kl": 0.00017464160919189453,120 "learning_rate": 4.99582211905537e-07,121 "loss": 0.0,122 "reward": 0.6274224072694778,123 "reward_std": 0.43979279696941376,124 "rewards/correct_code_reward_func": 0.4166666865348816,125 "rewards/len_reward_func": 0.21075571328401566,126 "step": 9127 },128 {129 "completion_length": 188.2916717529297,130 "epoch": 0.17699115044247787,131 "grad_norm": 3.0299854001506494,132 "kl": 8.487701416015625e-05,133 "learning_rate": 4.992574265488882e-07,134 "loss": 0.0,135 "reward": 0.502624899148941,136 "reward_std": 0.35237132012844086,137 "rewards/correct_code_reward_func": 0.3125000149011612,138 "rewards/len_reward_func": 0.19012490659952164,139 "step": 10140 },141 {142 "completion_length": 109.1875,143 "epoch": 0.19469026548672566,144 "grad_norm": 1.8891984084011832,145 "kl": 0.00060272216796875,146 "learning_rate": 4.988400522336303e-07,147 "loss": 0.0,148 "reward": 0.6938449144363403,149 "reward_std": 0.5790197551250458,150 "rewards/correct_code_reward_func": 0.5208333432674408,151 "rewards/len_reward_func": 0.1730116307735443,152 "step": 11153 },154 {155 "completion_length": 171.2916717529297,156 "epoch": 0.21238938053097345,157 "grad_norm": 2.7849518039157664,158 "kl": 0.0006923675537109375,159 "learning_rate": 4.983302439972828e-07,160 "loss": 0.0,161 "reward": 0.4239259362220764,162 "reward_std": 0.48799367249011993,163 "rewards/correct_code_reward_func": 0.20833333395421505,164 "rewards/len_reward_func": 0.21559256315231323,165 "step": 12166 },167 {168 "completion_length": 196.2291717529297,169 "epoch": 0.23008849557522124,170 "grad_norm": 1.392005725111761,171 "kl": 0.000652313232421875,172 "learning_rate": 4.977281912127938e-07,173 "loss": 0.0,174 "reward": 0.5957696735858917,175 "reward_std": 0.5921532511711121,176 "rewards/correct_code_reward_func": 0.4166666716337204,177 "rewards/len_reward_func": 0.17910300940275192,178 "step": 13179 },180 {181 "completion_length": 158.64583587646484,182 "epoch": 0.24778761061946902,183 "grad_norm": 1.5701172166666861,184 "kl": 0.000614166259765625,185 "learning_rate": 4.970341175181955e-07,186 "loss": 0.0,187 "reward": 0.46809761226177216,188 "reward_std": 0.4979168623685837,189 "rewards/correct_code_reward_func": 0.2708333432674408,190 "rewards/len_reward_func": 0.19726425409317017,191 "step": 14192 },193 {194 "completion_length": 147.1041717529297,195 "epoch": 0.26548672566371684,196 "grad_norm": 2.032633618646926,197 "kl": 0.0015106201171875,198 "learning_rate": 4.962482807335314e-07,199 "loss": 0.0,200 "reward": 0.5865300893783569,201 "reward_std": 0.5591276288032532,202 "rewards/correct_code_reward_func": 0.4166666716337204,203 "rewards/len_reward_func": 0.16986342146992683,204 "step": 15205 },206 {207 "completion_length": 177.9791717529297,208 "epoch": 0.2831858407079646,209 "grad_norm": 53.72073405864756,210 "kl": 0.018054962158203125,211 "learning_rate": 4.95370972765087e-07,212 "loss": 0.0,213 "reward": 0.8554503321647644,214 "reward_std": 0.48426124453544617,215 "rewards/correct_code_reward_func": 0.5625000298023224,216 "rewards/len_reward_func": 0.2929503321647644,217 "step": 16218 },219 {220 "completion_length": 187.31250762939453,221 "epoch": 0.3008849557522124,222 "grad_norm": 2.4065750408913034,223 "kl": 0.001964569091796875,224 "learning_rate": 4.944025194969586e-07,225 "loss": 0.0,226 "reward": 0.36077703535556793,227 "reward_std": 0.35280461609363556,228 "rewards/correct_code_reward_func": 0.1041666716337204,229 "rewards/len_reward_func": 0.25661035627126694,230 "step": 17231 },232 {233 "completion_length": 186.14583587646484,234 "epoch": 0.3185840707964602,235 "grad_norm": 20.459541697505053,236 "kl": 0.021259307861328125,237 "learning_rate": 4.933432806700003e-07,238 "loss": 0.0,239 "reward": 0.5004578679800034,240 "reward_std": 0.42624031007289886,241 "rewards/correct_code_reward_func": 0.2708333432674408,242 "rewards/len_reward_func": 0.22962453216314316,243 "step": 18244 },245 {246 "completion_length": 184.2916717529297,247 "epoch": 0.336283185840708,248 "grad_norm": 3.4167543405430436,249 "kl": 0.008358001708984375,250 "learning_rate": 4.921936497481956e-07,251 "loss": 0.0,252 "reward": 0.3476429730653763,253 "reward_std": 0.43119728565216064,254 "rewards/correct_code_reward_func": 0.0833333358168602,255 "rewards/len_reward_func": 0.2643096297979355,256 "step": 19257 },258 {259 "completion_length": 179.87500762939453,260 "epoch": 0.35398230088495575,261 "grad_norm": 2.8250290981037574,262 "kl": 0.0052490234375,263 "learning_rate": 4.909540537725006e-07,264 "loss": 0.0,265 "reward": 0.476805180311203,266 "reward_std": 0.36166805028915405,267 "rewards/correct_code_reward_func": 0.2708333432674408,268 "rewards/len_reward_func": 0.2059718370437622,269 "step": 20270 },271 {272 "completion_length": 148.9375,273 "epoch": 0.37168141592920356,274 "grad_norm": 1.9459373869466972,275 "kl": 0.002933502197265625,276 "learning_rate": 4.896249532022171e-07,277 "loss": 0.0,278 "reward": 0.5851579010486603,279 "reward_std": 0.3626042455434799,280 "rewards/correct_code_reward_func": 0.3333333358168602,281 "rewards/len_reward_func": 0.2518245801329613,282 "step": 21283 },284 {285 "completion_length": 213.8541717529297,286 "epoch": 0.3893805309734513,287 "grad_norm": 0.5828214607615373,288 "kl": 0.0029754638671875,289 "learning_rate": 4.882068417439492e-07,290 "loss": 0.0,291 "reward": 0.5553887784481049,292 "reward_std": 0.5428541898727417,293 "rewards/correct_code_reward_func": 0.291666679084301,294 "rewards/len_reward_func": 0.26372211426496506,295 "step": 22296 },297 {298 "completion_length": 113.43750381469727,299 "epoch": 0.40707964601769914,300 "grad_norm": 1.0373945570716487,301 "kl": 0.0034637451171875,302 "learning_rate": 4.867002461682128e-07,303 "loss": 0.0,304 "reward": 0.49457328021526337,305 "reward_std": 0.38711363077163696,306 "rewards/correct_code_reward_func": 0.2291666716337204,307 "rewards/len_reward_func": 0.26540662348270416,308 "step": 23309 },310 {311 "completion_length": 145.5416717529297,312 "epoch": 0.4247787610619469,313 "grad_norm": 0.8868802158083684,314 "kl": 0.00489044189453125,315 "learning_rate": 4.851057261137608e-07,316 "loss": 0.0,317 "reward": 0.7876249551773071,318 "reward_std": 0.4593852013349533,319 "rewards/correct_code_reward_func": 0.5208333432674408,320 "rewards/len_reward_func": 0.26679155230522156,321 "step": 24322 },323 {324 "completion_length": 223.50000762939453,325 "epoch": 0.4424778761061947,326 "grad_norm": 2.7373544353808152,327 "kl": 0.005214691162109375,328 "learning_rate": 4.83423873879701e-07,329 "loss": 0.0,330 "reward": 0.45352864265441895,331 "reward_std": 0.5192392915487289,332 "rewards/correct_code_reward_func": 0.2291666679084301,333 "rewards/len_reward_func": 0.22436197102069855,334 "step": 25335 },336 {337 "completion_length": 172.56250762939453,338 "epoch": 0.46017699115044247,339 "grad_norm": 1.5744378292792975,340 "kl": 0.0035400390625,341 "learning_rate": 4.816553142054805e-07,342 "loss": 0.0,343 "reward": 0.4348207265138626,344 "reward_std": 0.40823744237422943,345 "rewards/correct_code_reward_func": 0.14583333395421505,346 "rewards/len_reward_func": 0.2889874130487442,347 "step": 26348 },349 {350 "completion_length": 163.5416717529297,351 "epoch": 0.4778761061946903,352 "grad_norm": 37.42131429307232,353 "kl": 0.0279998779296875,354 "learning_rate": 4.798007040388211e-07,355 "loss": 0.0,356 "reward": 0.44616006314754486,357 "reward_std": 0.41116151213645935,358 "rewards/correct_code_reward_func": 0.18750000558793545,359 "rewards/len_reward_func": 0.25866006314754486,360 "step": 27361 },362 {363 "completion_length": 147.1458396911621,364 "epoch": 0.49557522123893805,365 "grad_norm": 3.226533563373661,366 "kl": 0.013214111328125,367 "learning_rate": 4.778607322916895e-07,368 "loss": 0.0,369 "reward": 0.5665057748556137,370 "reward_std": 0.3880573436617851,371 "rewards/correct_code_reward_func": 0.2708333432674408,372 "rewards/len_reward_func": 0.29567238688468933,373 "step": 28374 },375 {376 "completion_length": 127.56250381469727,377 "epoch": 0.5132743362831859,378 "grad_norm": 4.362166865135135,379 "kl": 0.01708984375,380 "learning_rate": 4.758361195843951e-07,381 "loss": 0.0,382 "reward": 0.8003825545310974,383 "reward_std": 0.510110080242157,384 "rewards/correct_code_reward_func": 0.4583333432674408,385 "rewards/len_reward_func": 0.34204918146133423,386 "step": 29387 },388 {389 "completion_length": 173.37500381469727,390 "epoch": 0.5309734513274337,391 "grad_norm": 0.8500372624160994,392 "kl": 0.00797271728515625,393 "learning_rate": 4.7372761797790825e-07,394 "loss": 0.0,395 "reward": 0.37487032264471054,396 "reward_std": 0.32572413980960846,397 "rewards/correct_code_reward_func": 0.1458333432674408,398 "rewards/len_reward_func": 0.22903696447610855,399 "step": 30400 },401 {402 "completion_length": 119.14583587646484,403 "epoch": 0.5486725663716814,404 "grad_norm": 2.2352274664870624,405 "kl": 0.014129638671875,406 "learning_rate": 4.7153601069450146e-07,407 "loss": 0.0,408 "reward": 0.5202069282531738,409 "reward_std": 0.5186183750629425,410 "rewards/correct_code_reward_func": 0.291666679084301,411 "rewards/len_reward_func": 0.22854027152061462,412 "step": 31413 },414 {415 "completion_length": 111.31250381469727,416 "epoch": 0.5663716814159292,417 "grad_norm": 1.1283808844332623,418 "kl": 0.015625,419 "learning_rate": 4.6926211182681294e-07,420 "loss": 0.0,421 "reward": 0.4815867692232132,422 "reward_std": 0.294177770614624,423 "rewards/correct_code_reward_func": 0.1458333432674408,424 "rewards/len_reward_func": 0.3357534259557724,425 "step": 32426 },427 {428 "completion_length": 111.87500381469727,429 "epoch": 0.584070796460177,430 "grad_norm": 15.382447167413178,431 "kl": 0.0740966796875,432 "learning_rate": 4.6690676603544553e-07,433 "loss": 0.0001,434 "reward": 0.30563782900571823,435 "reward_std": 0.5501309335231781,436 "rewards/correct_code_reward_func": 0.1458333395421505,437 "rewards/len_reward_func": 0.15980449318885803,438 "step": 33439 },440 {441 "completion_length": 74.31250190734863,442 "epoch": 0.6017699115044248,443 "grad_norm": 6.16835237121908,444 "kl": 0.038818359375,445 "learning_rate": 4.644708482352092e-07,446 "loss": 0.0,447 "reward": 0.7065277397632599,448 "reward_std": 0.4479583501815796,449 "rewards/correct_code_reward_func": 0.3958333358168602,450 "rewards/len_reward_func": 0.3106943815946579,451 "step": 34452 },453 {454 "completion_length": 96.45833587646484,455 "epoch": 0.6194690265486725,456 "grad_norm": 6.369953012669968,457 "kl": 0.035186767578125,458 "learning_rate": 4.6195526327012623e-07,459 "loss": 0.0,460 "reward": 0.3565668910741806,461 "reward_std": 0.39397968351840973,462 "rewards/correct_code_reward_func": 0.08333333395421505,463 "rewards/len_reward_func": 0.273233562707901,464 "step": 35465 },466 {467 "completion_length": 112.14583587646484,468 "epoch": 0.6371681415929203,469 "grad_norm": 3.3487214552142164,470 "kl": 0.0289306640625,471 "learning_rate": 4.5936094557731806e-07,472 "loss": 0.0,473 "reward": 0.7201341390609741,474 "reward_std": 0.3427349328994751,475 "rewards/correct_code_reward_func": 0.3541666865348816,476 "rewards/len_reward_func": 0.3659674823284149,477 "step": 36478 },479 {480 "completion_length": 101.64583587646484,481 "epoch": 0.6548672566371682,482 "grad_norm": 5.120880381398817,483 "kl": 0.05126953125,484 "learning_rate": 4.566888588399006e-07,485 "loss": 0.0001,486 "reward": 0.5204495191574097,487 "reward_std": 0.3199263662099838,488 "rewards/correct_code_reward_func": 0.1250000037252903,489 "rewards/len_reward_func": 0.39544953405857086,490 "step": 37491 },492 {493 "completion_length": 120.5625,494 "epoch": 0.672566371681416,495 "grad_norm": 10.359850576348157,496 "kl": 0.09923553466796875,497 "learning_rate": 4.5393999562901517e-07,498 "loss": 0.0001,499 "reward": 0.6964526474475861,500 "reward_std": 0.5385148227214813,501 "rewards/correct_code_reward_func": 0.4166666865348816,502 "rewards/len_reward_func": 0.27978597581386566,503 "step": 38504 },505 {506 "completion_length": 110.85417175292969,507 "epoch": 0.6902654867256637,508 "grad_norm": 12.273703285519217,509 "kl": 0.0859375,510 "learning_rate": 4.511153770351288e-07,511 "loss": 0.0001,512 "reward": 0.551011860370636,513 "reward_std": 0.36342084407806396,514 "rewards/correct_code_reward_func": 0.1875000111758709,515 "rewards/len_reward_func": 0.3635118901729584,516 "step": 39517 },518 {519 "completion_length": 68.77083587646484,520 "epoch": 0.7079646017699115,521 "grad_norm": 2.2127761337332514,522 "kl": 0.03765869140625,523 "learning_rate": 4.482160522887403e-07,524 "loss": 0.0,525 "reward": 0.7378278374671936,526 "reward_std": 0.39128121733665466,527 "rewards/correct_code_reward_func": 0.4375000149011612,528 "rewards/len_reward_func": 0.3003278076648712,529 "step": 40530 },531 {532 "completion_length": 135.33333587646484,533 "epoch": 0.7256637168141593,534 "grad_norm": 7.600755671861986,535 "kl": 0.1025390625,536 "learning_rate": 4.4524309837063504e-07,537 "loss": 0.0001,538 "reward": 0.6550733149051666,539 "reward_std": 0.5409570336341858,540 "rewards/correct_code_reward_func": 0.3333333432674408,541 "rewards/len_reward_func": 0.32173997163772583,542 "step": 41543 },544 {545 "completion_length": 87.35416793823242,546 "epoch": 0.7433628318584071,547 "grad_norm": 1.6503719126782672,548 "kl": 0.0264892578125,549 "learning_rate": 4.4219761961182965e-07,550 "loss": 0.0,551 "reward": 0.6292450726032257,552 "reward_std": 0.4537566006183624,553 "rewards/correct_code_reward_func": 0.2708333432674408,554 "rewards/len_reward_func": 0.3584117591381073,555 "step": 42556 },557 {558 "completion_length": 65.29166793823242,559 "epoch": 0.7610619469026548,560 "grad_norm": 4.327974078868625,561 "kl": 0.1019287109375,562 "learning_rate": 4.390807472833584e-07,563 "loss": 0.0001,564 "reward": 0.5142460465431213,565 "reward_std": 0.41541849076747894,566 "rewards/correct_code_reward_func": 0.1875000074505806,567 "rewards/len_reward_func": 0.32674603164196014,568 "step": 43569 },570 {571 "completion_length": 88.62500381469727,572 "epoch": 0.7787610619469026,573 "grad_norm": 2.8856078275542814,574 "kl": 0.068115234375,575 "learning_rate": 4.3589363917605234e-07,576 "loss": 0.0001,577 "reward": 0.48976023495197296,578 "reward_std": 0.39692020416259766,579 "rewards/correct_code_reward_func": 0.1875,580 "rewards/len_reward_func": 0.30226024240255356,581 "step": 44582 },583 {584 "completion_length": 78.75000381469727,585 "epoch": 0.7964601769911505,586 "grad_norm": 1.6857493365819893,587 "kl": 0.04052734375,588 "learning_rate": 4.3263747917046697e-07,589 "loss": 0.0,590 "reward": 0.5636806637048721,591 "reward_std": 0.5081076323986053,592 "rewards/correct_code_reward_func": 0.2500000111758709,593 "rewards/len_reward_func": 0.3136806935071945,594 "step": 45595 },596 {597 "completion_length": 78.50000190734863,598 "epoch": 0.8141592920353983,599 "grad_norm": 1.2755984783224172,600 "kl": 0.0218505859375,601 "learning_rate": 4.2931347679711924e-07,602 "loss": 0.0,603 "reward": 0.6265330016613007,604 "reward_std": 0.38107292354106903,605 "rewards/correct_code_reward_func": 0.2916666716337204,606 "rewards/len_reward_func": 0.33486635982990265,607 "step": 46608 },609 {610 "completion_length": 80.29166793823242,611 "epoch": 0.831858407079646,612 "grad_norm": 1.1240225012044924,613 "kl": 0.0335693359375,614 "learning_rate": 4.259228667871962e-07,615 "loss": 0.0,616 "reward": 0.5936128199100494,617 "reward_std": 0.4675884544849396,618 "rewards/correct_code_reward_func": 0.25,619 "rewards/len_reward_func": 0.34361281991004944,620 "step": 47621 },622 {623 "completion_length": 50.0,624 "epoch": 0.8495575221238938,625 "grad_norm": 8.693611957424073,626 "kl": 0.2470703125,627 "learning_rate": 4.224669086139029e-07,628 "loss": 0.0002,629 "reward": 0.49966277182102203,630 "reward_std": 0.3804493248462677,631 "rewards/correct_code_reward_func": 0.08333333395421505,632 "rewards/len_reward_func": 0.41632944345474243,633 "step": 48634 },635 {636 "completion_length": 92.20833587646484,637 "epoch": 0.8672566371681416,638 "grad_norm": 1.2744927278879081,639 "kl": 0.02978515625,640 "learning_rate": 4.1894688602461913e-07,641 "loss": 0.0,642 "reward": 0.7709261626005173,643 "reward_std": 0.3614697605371475,644 "rewards/correct_code_reward_func": 0.3958333544433117,645 "rewards/len_reward_func": 0.3750927746295929,646 "step": 49647 },648 {649 "completion_length": 127.62500381469727,650 "epoch": 0.8849557522123894,651 "grad_norm": 0.566625300559752,652 "kl": 0.016326904296875,653 "learning_rate": 4.153641065640402e-07,654 "loss": 0.0,655 "reward": 0.5525736510753632,656 "reward_std": 0.42078813910484314,657 "rewards/correct_code_reward_func": 0.20833333395421505,658 "rewards/len_reward_func": 0.34424029290676117,659 "step": 50660 },661 {662 "completion_length": 65.375,663 "epoch": 0.9026548672566371,664 "grad_norm": 12.046039464547933,665 "kl": 0.081298828125,666 "learning_rate": 4.11719901088477e-07,667 "loss": 0.0001,668 "reward": 0.6643719673156738,669 "reward_std": 0.4581465870141983,670 "rewards/correct_code_reward_func": 0.3333333432674408,671 "rewards/len_reward_func": 0.3310386538505554,672 "step": 51673 },674 {675 "completion_length": 103.58333587646484,676 "epoch": 0.9203539823008849,677 "grad_norm": 10.540579232965062,678 "kl": 0.0799560546875,679 "learning_rate": 4.080156232714976e-07,680 "loss": 0.0001,681 "reward": 0.6433776319026947,682 "reward_std": 0.3912515640258789,683 "rewards/correct_code_reward_func": 0.29166667722165585,684 "rewards/len_reward_func": 0.3517109453678131,685 "step": 52686 },687 {688 "completion_length": 62.375003814697266,689 "epoch": 0.9380530973451328,690 "grad_norm": 2.819698340947531,691 "kl": 0.115966796875,692 "learning_rate": 4.0425264910109233e-07,693 "loss": 0.0001,694 "reward": 0.7070481181144714,695 "reward_std": 0.44986239075660706,696 "rewards/correct_code_reward_func": 0.3125,697 "rewards/len_reward_func": 0.39454813301563263,698 "step": 53699 },700 {701 "completion_length": 59.0625,702 "epoch": 0.9557522123893806,703 "grad_norm": 3.22095354432732,704 "kl": 0.1131591796875,705 "learning_rate": 4.004323763685511e-07,706 "loss": 0.0001,707 "reward": 0.5839295089244843,708 "reward_std": 0.26084955781698227,709 "rewards/correct_code_reward_func": 0.14583333395421505,710 "rewards/len_reward_func": 0.43809613585472107,711 "step": 54712 },713 {714 "completion_length": 125.58333969116211,715 "epoch": 0.9734513274336283,716 "grad_norm": 8.551645640447115,717 "kl": 0.15509033203125,718 "learning_rate": 3.9655622414924007e-07,719 "loss": 0.0002,720 "reward": 0.528711274266243,721 "reward_std": 0.3454602509737015,722 "rewards/correct_code_reward_func": 0.1666666716337204,723 "rewards/len_reward_func": 0.3620445877313614,724 "step": 55725 },726 {727 "completion_length": 60.000003814697266,728 "epoch": 0.9911504424778761,729 "grad_norm": 38.727062970707685,730 "kl": 0.20166015625,731 "learning_rate": 3.92625632275474e-07,732 "loss": 0.0002,733 "reward": 0.891117125749588,734 "reward_std": 0.36153580248355865,735 "rewards/correct_code_reward_func": 0.5000000149011612,736 "rewards/len_reward_func": 0.39111708104610443,737 "step": 56738 },739 {740 "completion_length": 91.54167175292969,741 "epoch": 1.0,742 "grad_norm": 38.727062970707685,743 "kl": 0.1865234375,744 "learning_rate": 3.886420608016766e-07,745 "loss": 0.0001,746 "reward": 0.5094150304794312,747 "reward_std": 0.4194638133049011,748 "rewards/correct_code_reward_func": 0.1666666716337204,749 "rewards/len_reward_func": 0.3427482843399048,750 "step": 57751 },752 {753 "completion_length": 96.41666793823242,754 "epoch": 1.0176991150442478,755 "grad_norm": 13.425409986987928,756 "kl": 0.04998779296875,757 "learning_rate": 3.846069894620305e-07,758 "loss": 0.0001,759 "reward": 0.5195242762565613,760 "reward_std": 0.3198210597038269,761 "rewards/correct_code_reward_func": 0.125,762 "rewards/len_reward_func": 0.3945242464542389,763 "step": 58764 },765 {766 "completion_length": 79.3125,767 "epoch": 1.0353982300884956,768 "grad_norm": 1.7429704138905477,769 "kl": 0.0509033203125,770 "learning_rate": 3.8052191712081595e-07,771 "loss": 0.0001,772 "reward": 0.655539482831955,773 "reward_std": 0.2987876646220684,774 "rewards/correct_code_reward_func": 0.1875,775 "rewards/len_reward_func": 0.46803948283195496,776 "step": 59777 },778 {779 "completion_length": 61.187503814697266,780 "epoch": 1.0530973451327434,781 "grad_norm": 7.538322601947877,782 "kl": 0.1343994140625,783 "learning_rate": 3.763883612156441e-07,784 "loss": 0.0001,785 "reward": 0.7049243450164795,786 "reward_std": 0.5336888134479523,787 "rewards/correct_code_reward_func": 0.3125000149011612,788 "rewards/len_reward_func": 0.3924243599176407,789 "step": 60790 },791 {792 "completion_length": 71.45833587646484,793 "epoch": 1.0707964601769913,794 "grad_norm": 1.628031440232102,795 "kl": 0.053955078125,796 "learning_rate": 3.7220785719379095e-07,797 "loss": 0.0001,798 "reward": 0.5856566280126572,799 "reward_std": 0.31656239926815033,800 "rewards/correct_code_reward_func": 0.2083333395421505,801 "rewards/len_reward_func": 0.37732329964637756,802 "step": 61803 },804 {805 "completion_length": 80.64583587646484,806 "epoch": 1.0884955752212389,807 "grad_norm": 0.8807120675882707,808 "kl": 0.05035400390625,809 "learning_rate": 3.6798195794184134e-07,810 "loss": 0.0001,811 "reward": 0.5311767309904099,812 "reward_std": 0.2926383763551712,813 "rewards/correct_code_reward_func": 0.1041666716337204,814 "rewards/len_reward_func": 0.42701007425785065,815 "step": 62816 },817 {818 "completion_length": 112.29167175292969,819 "epoch": 1.1061946902654867,820 "grad_norm": 1.0053639016562932,821 "kl": 0.07916259765625,822 "learning_rate": 3.6371223320885484e-07,823 "loss": 0.0001,824 "reward": 0.4587331861257553,825 "reward_std": 0.33583496510982513,826 "rewards/correct_code_reward_func": 0.08333333395421505,827 "rewards/len_reward_func": 0.3753998577594757,828 "step": 63829 },830 {831 "completion_length": 83.20833587646484,832 "epoch": 1.1238938053097345,833 "grad_norm": 1.7713543449970612,834 "kl": 0.08245849609375,835 "learning_rate": 3.594002690232682e-07,836 "loss": 0.0001,837 "reward": 0.6741357147693634,838 "reward_std": 0.43141913414001465,839 "rewards/correct_code_reward_func": 0.2708333432674408,840 "rewards/len_reward_func": 0.4033023715019226,841 "step": 64842 },843 {844 "completion_length": 84.10416984558105,845 "epoch": 1.1415929203539823,846 "grad_norm": 25.050556888869668,847 "kl": 0.12774658203125,848 "learning_rate": 3.5504766710375047e-07,849 "loss": 0.0001,850 "reward": 0.6237881779670715,851 "reward_std": 0.28890369832515717,852 "rewards/correct_code_reward_func": 0.2083333432674408,853 "rewards/len_reward_func": 0.41545483469963074,854 "step": 65855 },856 {857 "completion_length": 82.27083587646484,858 "epoch": 1.1592920353982301,859 "grad_norm": 1.3229069926746502,860 "kl": 0.0810546875,861 "learning_rate": 3.506560442642299e-07,862 "loss": 0.0001,863 "reward": 0.7232638597488403,864 "reward_std": 0.37123511731624603,865 "rewards/correct_code_reward_func": 0.3541666716337204,866 "rewards/len_reward_func": 0.36909720301628113,867 "step": 66868 },869 {870 "completion_length": 58.00000190734863,871 "epoch": 1.176991150442478,872 "grad_norm": 14.439064098343355,873 "kl": 0.1383056640625,874 "learning_rate": 3.462270318133136e-07,875 "loss": 0.0001,876 "reward": 0.9435772597789764,877 "reward_std": 0.2863286882638931,878 "rewards/correct_code_reward_func": 0.520833358168602,879 "rewards/len_reward_func": 0.42274393141269684,880 "step": 67881 },882 {883 "completion_length": 71.10416984558105,884 "epoch": 1.1946902654867257,885 "grad_norm": 7.987124238678158,886 "kl": 0.122314453125,887 "learning_rate": 3.41762274948323e-07,888 "loss": 0.0001,889 "reward": 0.7681821584701538,890 "reward_std": 0.3728492707014084,891 "rewards/correct_code_reward_func": 0.3541666865348816,892 "rewards/len_reward_func": 0.41401544213294983,893 "step": 68894 },895 {896 "completion_length": 60.66666793823242,897 "epoch": 1.2123893805309733,898 "grad_norm": 2.9983488159761817,899 "kl": 0.0537109375,900 "learning_rate": 3.372634321441702e-07,901 "loss": 0.0001,902 "reward": 0.7418330609798431,903 "reward_std": 0.38279734551906586,904 "rewards/correct_code_reward_func": 0.3541666865348816,905 "rewards/len_reward_func": 0.38766637444496155,906 "step": 69907 },908 {909 "completion_length": 80.62500381469727,910 "epoch": 1.2300884955752212,911 "grad_norm": 3.023771746536167,912 "kl": 0.08154296875,913 "learning_rate": 3.3273217453730205e-07,914 "loss": 0.0001,915 "reward": 0.6804832518100739,916 "reward_std": 0.4174923449754715,917 "rewards/correct_code_reward_func": 0.2708333432674408,918 "rewards/len_reward_func": 0.40964992344379425,919 "step": 70920 },921 {922 "completion_length": 69.5,923 "epoch": 1.247787610619469,924 "grad_norm": 1.4050656113052118,925 "kl": 0.0853271484375,926 "learning_rate": 3.281701853049416e-07,927 "loss": 0.0001,928 "reward": 0.7291666865348816,929 "reward_std": 0.32618680596351624,930 "rewards/correct_code_reward_func": 0.2708333432674408,931 "rewards/len_reward_func": 0.4583333432674408,932 "step": 71933 },934 {935 "completion_length": 74.47916793823242,936 "epoch": 1.2654867256637168,937 "grad_norm": 2.8179580865899894,938 "kl": 0.0390625,939 "learning_rate": 3.23579159039856e-07,940 "loss": 0.0,941 "reward": 0.75091952085495,942 "reward_std": 0.4079275578260422,943 "rewards/correct_code_reward_func": 0.2916666716337204,944 "rewards/len_reward_func": 0.45925286412239075,945 "step": 72946 },947 {948 "completion_length": 65.52083587646484,949 "epoch": 1.2831858407079646,950 "grad_norm": 1.1299435463555585,951 "kl": 0.0399169921875,952 "learning_rate": 3.1896080112088477e-07,953 "loss": 0.0,954 "reward": 0.7291666865348816,955 "reward_std": 0.40629878640174866,956 "rewards/correct_code_reward_func": 0.2708333395421505,957 "rewards/len_reward_func": 0.4583333432674408,958 "step": 73959 },960 {961 "completion_length": 57.000003814697266,962 "epoch": 1.3008849557522124,963 "grad_norm": 24.705642459397488,964 "kl": 0.17333984375,965 "learning_rate": 3.1431682707946117e-07,966 "loss": 0.0002,967 "reward": 0.5623736083507538,968 "reward_std": 0.24504420161247253,969 "rewards/correct_code_reward_func": 0.12500000558793545,970 "rewards/len_reward_func": 0.4373735785484314,971 "step": 74972 },973 {974 "completion_length": 78.33333587646484,975 "epoch": 1.3185840707964602,976 "grad_norm": 0.7830442669221305,977 "kl": 0.0224609375,978 "learning_rate": 3.096489619623621e-07,979 "loss": 0.0,980 "reward": 0.7488380074501038,981 "reward_std": 0.30931201577186584,982 "rewards/correct_code_reward_func": 0.2500000149011612,983 "rewards/len_reward_func": 0.49883799254894257,984 "step": 75985 },986 {987 "completion_length": 87.60417175292969,988 "epoch": 1.336283185840708,989 "grad_norm": 2.1746791001290044,990 "kl": 0.04168701171875,991 "learning_rate": 3.049589396909239e-07,992 "loss": 0.0,993 "reward": 0.5558712333440781,994 "reward_std": 0.20058585330843925,995 "rewards/correct_code_reward_func": 0.1041666716337204,996 "rewards/len_reward_func": 0.45170454680919647,997 "step": 76998 },999 {1000 "completion_length": 72.14583587646484,1001 "epoch": 1.3539823008849559,1002 "grad_norm": 0.41651138876768207,1003 "kl": 0.0919189453125,1004 "learning_rate": 3.0024850241696126e-07,1005 "loss": 0.0001,1006 "reward": 0.6021915674209595,1007 "reward_std": 0.22243696451187134,1008 "rewards/correct_code_reward_func": 0.1458333358168602,1009 "rewards/len_reward_func": 0.4563582092523575,1010 "step": 771011 },1012 {1013 "completion_length": 54.916666984558105,1014 "epoch": 1.3716814159292037,1015 "grad_norm": 8.88612982008444,1016 "kl": 0.106689453125,1017 "learning_rate": 2.9551939987562865e-07,1018 "loss": 0.0001,1019 "reward": 0.6803071200847626,1020 "reward_std": 0.2964301370084286,1021 "rewards/correct_code_reward_func": 0.2291666679084301,1022 "rewards/len_reward_func": 0.4511404484510422,1023 "step": 781024 },1025 {1026 "completion_length": 93.91666793823242,1027 "epoch": 1.3893805309734513,1028 "grad_norm": 3.7969039322604865,1029 "kl": 0.086761474609375,1030 "learning_rate": 2.9077338873546564e-07,1031 "loss": 0.0001,1032 "reward": 0.7916666865348816,1033 "reward_std": 0.34503278136253357,1034 "rewards/correct_code_reward_func": 0.2916666716337204,1035 "rewards/len_reward_func": 0.5,1036 "step": 791037 },1038 {1039 "completion_length": 56.08333396911621,1040 "epoch": 1.407079646017699,1041 "grad_norm": 3.0759876398616743,1042 "kl": 0.076171875,1043 "learning_rate": 2.860122319458661e-07,1044 "loss": 0.0001,1045 "reward": 0.8750000596046448,1046 "reward_std": 0.36751919984817505,1047 "rewards/correct_code_reward_func": 0.3958333432674408,1048 "rewards/len_reward_func": 0.4791666716337204,1049 "step": 801050 },1051 {1052 "completion_length": 77.58333587646484,1053 "epoch": 1.424778761061947,1054 "grad_norm": 5.267849973823214,1055 "kl": 0.0604248046875,1056 "learning_rate": 2.8123769808221403e-07,1057 "loss": 0.0001,1058 "reward": 0.9166666865348816,1059 "reward_std": 0.4720035046339035,1060 "rewards/correct_code_reward_func": 0.4375000298023224,1061 "rewards/len_reward_func": 0.4791666716337204,1062 "step": 811063 },1064 {1065 "completion_length": 86.18750190734863,1066 "epoch": 1.4424778761061947,1067 "grad_norm": 16.256704067320886,1068 "kl": 0.05712890625,1069 "learning_rate": 2.764515606889307e-07,1070 "loss": 0.0001,1071 "reward": 0.8077563941478729,1072 "reward_std": 0.3568233400583267,1073 "rewards/correct_code_reward_func": 0.3958333432674408,1074 "rewards/len_reward_func": 0.41192303597927094,1075 "step": 821076 },1077 {1078 "completion_length": 106.89583587646484,1079 "epoch": 1.4601769911504425,1080 "grad_norm": 2.2341312424056214,1081 "kl": 0.021087646484375,1082 "learning_rate": 2.716555976206747e-07,1083 "loss": 0.0,1084 "reward": 0.7582251131534576,1085 "reward_std": 0.3638576865196228,1086 "rewards/correct_code_reward_func": 0.2916666716337204,1087 "rewards/len_reward_func": 0.46655844151973724,1088 "step": 831089 },1090 {1091 "completion_length": 48.41666793823242,1092 "epoch": 1.4778761061946903,1093 "grad_norm": 1.3490761289725859,1094 "kl": 0.05279541015625,1095 "learning_rate": 2.66851590381942e-07,1096 "loss": 0.0001,1097 "reward": 0.7612736523151398,1098 "reward_std": 0.23646536469459534,1099 "rewards/correct_code_reward_func": 0.2708333358168602,1100 "rewards/len_reward_func": 0.490440309047699,1101 "step": 841102 },1103 {1104 "completion_length": 60.187503814697266,1105 "epoch": 1.495575221238938,1106 "grad_norm": 4.538439276716702,1107 "kl": 0.0347900390625,1108 "learning_rate": 2.620413234653093e-07,1109 "loss": 0.0,1110 "reward": 0.8333333432674408,1111 "reward_std": 0.438393235206604,1112 "rewards/correct_code_reward_func": 0.3541666716337204,1113 "rewards/len_reward_func": 0.4791666716337204,1114 "step": 851115 },1116 {1117 "completion_length": 138.50000762939453,1118 "epoch": 1.5132743362831858,1119 "grad_norm": 0.9112971760232713,1120 "kl": 0.021392822265625,1121 "learning_rate": 2.5722658368856814e-07,1122 "loss": 0.0,1123 "reward": 0.6596881151199341,1124 "reward_std": 0.4098925292491913,1125 "rewards/correct_code_reward_func": 0.1875000074505806,1126 "rewards/len_reward_func": 0.4721881300210953,1127 "step": 861128 },1129 {1130 "completion_length": 67.89583587646484,1131 "epoch": 1.5309734513274336,1132 "grad_norm": 1.5750222049346805,1133 "kl": 0.033447265625,1134 "learning_rate": 2.5240915953099515e-07,1135 "loss": 0.0,1136 "reward": 0.9335185885429382,1137 "reward_std": 0.32536589354276657,1138 "rewards/correct_code_reward_func": 0.4375000149011612,1139 "rewards/len_reward_func": 0.49601852893829346,1140 "step": 871141 },1142 {1143 "completion_length": 56.395835876464844,1144 "epoch": 1.5486725663716814,1145 "grad_norm": 0.8224337332016152,1146 "kl": 0.0401611328125,1147 "learning_rate": 2.4759084046900483e-07,1148 "loss": 0.0,1149 "reward": 0.7500000298023224,1150 "reward_std": 0.2342708334326744,1151 "rewards/correct_code_reward_func": 0.3125000149011612,1152 "rewards/len_reward_func": 0.4375000149011612,1153 "step": 881154 },1155 {1156 "completion_length": 55.250003814697266,1157 "epoch": 1.5663716814159292,1158 "grad_norm": 1.7919585934332407,1159 "kl": 0.046142578125,1160 "learning_rate": 2.427734163114319e-07,1161 "loss": 0.0,1162 "reward": 0.8125000298023224,1163 "reward_std": 0.3177132308483124,1164 "rewards/correct_code_reward_func": 0.3125000149011612,1165 "rewards/len_reward_func": 0.5,1166 "step": 891167 },1168 {1169 "completion_length": 78.27083587646484,1170 "epoch": 1.584070796460177,1171 "grad_norm": 1.6894819978109317,1172 "kl": 0.0758056640625,1173 "learning_rate": 2.3795867653469067e-07,1174 "loss": 0.0001,1175 "reward": 0.8326875865459442,1176 "reward_std": 0.27240147814154625,1177 "rewards/correct_code_reward_func": 0.35416667722165585,1178 "rewards/len_reward_func": 0.47852087020874023,1179 "step": 901180 },1181 {1182 "completion_length": 102.83333969116211,1183 "epoch": 1.6017699115044248,1184 "grad_norm": 22.765338134366964,1185 "kl": 0.07025146484375,1186 "learning_rate": 2.3314840961805802e-07,1187 "loss": 0.0001,1188 "reward": 0.686832070350647,1189 "reward_std": 0.2984095811843872,1190 "rewards/correct_code_reward_func": 0.2083333432674408,1191 "rewards/len_reward_func": 0.47849874198436737,1192 "step": 911193 },1194 {1195 "completion_length": 54.31250190734863,1196 "epoch": 1.6194690265486726,1197 "grad_norm": 1.6877966926685302,1198 "kl": 0.0550537109375,1199 "learning_rate": 2.2834440237932533e-07,1200 "loss": 0.0001,