{ "best_metric": null, "best_model_checkpoint": null, "epoch": 0.6736609607951408, "eval_steps": 500, "global_step": 2440, "is_hyper_param_search": false, "is_local_process_zero": true, "is_world_process_zero": true, "log_history": [ { "completion_length": 39.4375, "epoch": 0.0027609055770292656, "grad_norm": 0.384765625, "kl": 0.0011762777163918752, "learning_rate": 9.97239094422971e-06, "loss": 0.0001, "reward": 1.0288461538461537, "reward_std": 0.09615384615384616, "rewards/accuracy_reward": 0.057692307692307696, "rewards/format_reward": 0.9711538461538461, "step": 10 }, { "completion_length": 52.55, "epoch": 0.005521811154058531, "grad_norm": 0.201171875, "kl": 0.010907393763773143, "learning_rate": 9.944781888459416e-06, "loss": 0.0004, "reward": 1.0, "reward_std": 0.0, "rewards/accuracy_reward": 0.0, "rewards/format_reward": 1.0, "step": 20 }, { "completion_length": 55.275, "epoch": 0.008282716731087797, "grad_norm": 0.34765625, "kl": 0.009688957687467336, "learning_rate": 9.917172832689123e-06, "loss": 0.0004, "reward": 0.95, "reward_std": 0.1, "rewards/accuracy_reward": 0.0, "rewards/format_reward": 0.95, "step": 30 }, { "completion_length": 60.35, "epoch": 0.011043622308117063, "grad_norm": 0.125, "kl": 0.01061137397773564, "learning_rate": 9.889563776918831e-06, "loss": 0.0004, "reward": 0.975, "reward_std": 0.05, "rewards/accuracy_reward": 0.0, "rewards/format_reward": 0.975, "step": 40 }, { "completion_length": 35.725, "epoch": 0.013804527885146328, "grad_norm": 0.205078125, "kl": 0.02767656436190009, "learning_rate": 9.861954721148538e-06, "loss": 0.0011, "reward": 1.025, "reward_std": 0.05, "rewards/accuracy_reward": 0.025, "rewards/format_reward": 1.0, "step": 50 }, { "completion_length": 63.925, "epoch": 0.016565433462175594, "grad_norm": 0.1923828125, "kl": 0.018191399239003658, "learning_rate": 9.834345665378246e-06, "loss": 0.0007, "reward": 0.975, "reward_std": 0.16547005176544188, "rewards/accuracy_reward": 0.05, "rewards/format_reward": 0.925, "step": 60 }, { "completion_length": 62.65, "epoch": 0.01932633903920486, "grad_norm": 0.408203125, "kl": 0.014432385563850403, "learning_rate": 9.806736609607953e-06, "loss": 0.0006, "reward": 1.025, "reward_std": 0.05, "rewards/accuracy_reward": 0.025, "rewards/format_reward": 1.0, "step": 70 }, { "completion_length": 51.15, "epoch": 0.022087244616234125, "grad_norm": 0.3671875, "kl": 0.01872744457796216, "learning_rate": 9.77912755383766e-06, "loss": 0.0007, "reward": 1.025, "reward_std": 0.15, "rewards/accuracy_reward": 0.05, "rewards/format_reward": 0.975, "step": 80 }, { "completion_length": 40.325, "epoch": 0.02484815019326339, "grad_norm": 0.734375, "kl": 0.015993439219892026, "learning_rate": 9.751518498067368e-06, "loss": 0.0006, "reward": 1.025, "reward_std": 0.05, "rewards/accuracy_reward": 0.025, "rewards/format_reward": 1.0, "step": 90 }, { "completion_length": 40.85, "epoch": 0.027609055770292656, "grad_norm": 0.66015625, "kl": 0.015149969141930342, "learning_rate": 9.723909442297074e-06, "loss": 0.0006, "reward": 1.0, "reward_std": 0.0, "rewards/accuracy_reward": 0.0, "rewards/format_reward": 1.0, "step": 100 }, { "completion_length": 73.125, "epoch": 0.030369961347321922, "grad_norm": 19.25, "kl": 0.011897594993934035, "learning_rate": 9.696300386526781e-06, "loss": 0.0005, "reward": 0.925, "reward_std": 0.18938468694686889, "rewards/accuracy_reward": 0.025, "rewards/format_reward": 0.9, "step": 110 }, { "completion_length": 60.65, "epoch": 0.03313086692435119, "grad_norm": 0.1494140625, "kl": 0.01287351930513978, "learning_rate": 9.66869133075649e-06, "loss": 0.0005, "reward": 0.975, "reward_std": 0.05, "rewards/accuracy_reward": 0.0, "rewards/format_reward": 0.975, "step": 120 }, { "completion_length": 50.525, "epoch": 0.03589177250138045, "grad_norm": 0.37890625, "kl": 0.013646931713446974, "learning_rate": 9.641082274986196e-06, "loss": 0.0005, "reward": 1.025, "reward_std": 0.05, "rewards/accuracy_reward": 0.025, "rewards/format_reward": 1.0, "step": 130 }, { "completion_length": 45.275, "epoch": 0.03865267807840972, "grad_norm": 0.37109375, "kl": 0.011697723204270006, "learning_rate": 9.613473219215904e-06, "loss": 0.0005, "reward": 1.0, "reward_std": 0.0, "rewards/accuracy_reward": 0.0, "rewards/format_reward": 1.0, "step": 140 }, { "completion_length": 44.725, "epoch": 0.041413583655438985, "grad_norm": 0.25, "kl": 0.008990359771996737, "learning_rate": 9.585864163445611e-06, "loss": 0.0004, "reward": 1.0, "reward_std": 0.0, "rewards/accuracy_reward": 0.0, "rewards/format_reward": 1.0, "step": 150 }, { "completion_length": 45.575, "epoch": 0.04417448923246825, "grad_norm": 0.3125, "kl": 0.008469818299636245, "learning_rate": 9.558255107675318e-06, "loss": 0.0003, "reward": 0.975, "reward_std": 0.05, "rewards/accuracy_reward": 0.0, "rewards/format_reward": 0.975, "step": 160 }, { "completion_length": 57.625, "epoch": 0.046935394809497516, "grad_norm": 0.5625, "kl": 0.0104887580499053, "learning_rate": 9.530646051905026e-06, "loss": 0.0004, "reward": 1.025, "reward_std": 0.05, "rewards/accuracy_reward": 0.025, "rewards/format_reward": 1.0, "step": 170 }, { "completion_length": 50.475, "epoch": 0.04969630038652678, "grad_norm": 0.53515625, "kl": 0.01001789947040379, "learning_rate": 9.503036996134733e-06, "loss": 0.0004, "reward": 1.0, "reward_std": 0.0, "rewards/accuracy_reward": 0.0, "rewards/format_reward": 1.0, "step": 180 }, { "completion_length": 48.725, "epoch": 0.05245720596355605, "grad_norm": 0.224609375, "kl": 0.006566637242212891, "learning_rate": 9.47542794036444e-06, "loss": 0.0003, "reward": 1.0, "reward_std": 0.0, "rewards/accuracy_reward": 0.0, "rewards/format_reward": 1.0, "step": 190 }, { "completion_length": 50.125, "epoch": 0.05521811154058531, "grad_norm": 0.271484375, "kl": 0.0062376999529078605, "learning_rate": 9.447818884594148e-06, "loss": 0.0002, "reward": 1.0, "reward_std": 0.1, "rewards/accuracy_reward": 0.025, "rewards/format_reward": 0.975, "step": 200 }, { "completion_length": 63.85, "epoch": 0.05797901711761458, "grad_norm": 0.2734375, "kl": 0.01781983794644475, "learning_rate": 9.420209828823854e-06, "loss": 0.0007, "reward": 1.025, "reward_std": 0.10773502588272095, "rewards/accuracy_reward": 0.05, "rewards/format_reward": 0.975, "step": 210 }, { "completion_length": 46.025, "epoch": 0.060739922694643844, "grad_norm": 0.54296875, "kl": 0.009719261107966304, "learning_rate": 9.392600773053563e-06, "loss": 0.0004, "reward": 1.025, "reward_std": 0.05, "rewards/accuracy_reward": 0.025, "rewards/format_reward": 1.0, "step": 220 }, { "completion_length": 50.175, "epoch": 0.06350082827167311, "grad_norm": 0.2490234375, "kl": 0.006523344293236732, "learning_rate": 9.36499171728327e-06, "loss": 0.0003, "reward": 1.0, "reward_std": 0.0, "rewards/accuracy_reward": 0.0, "rewards/format_reward": 1.0, "step": 230 }, { "completion_length": 52.375, "epoch": 0.06626173384870238, "grad_norm": 0.5546875, "kl": 0.008632810413837433, "learning_rate": 9.337382661512978e-06, "loss": 0.0003, "reward": 1.0, "reward_std": 0.0, "rewards/accuracy_reward": 0.0, "rewards/format_reward": 1.0, "step": 240 }, { "completion_length": 38.75, "epoch": 0.06902263942573164, "grad_norm": 0.126953125, "kl": 0.007552449638023972, "learning_rate": 9.309773605742684e-06, "loss": 0.0003, "reward": 1.05, "reward_std": 0.1, "rewards/accuracy_reward": 0.05, "rewards/format_reward": 1.0, "step": 250 }, { "completion_length": 48.375, "epoch": 0.0717835450027609, "grad_norm": 0.142578125, "kl": 0.01255230070091784, "learning_rate": 9.282164549972393e-06, "loss": 0.0005, "reward": 1.0, "reward_std": 0.0, "rewards/accuracy_reward": 0.0, "rewards/format_reward": 1.0, "step": 260 }, { "completion_length": 63.85, "epoch": 0.07454445057979017, "grad_norm": 0.1513671875, "kl": 0.008702394971624016, "learning_rate": 9.2545554942021e-06, "loss": 0.0003, "reward": 1.0, "reward_std": 0.0, "rewards/accuracy_reward": 0.0, "rewards/format_reward": 1.0, "step": 270 }, { "completion_length": 55.275, "epoch": 0.07730535615681944, "grad_norm": 0.09716796875, "kl": 0.010681362450122833, "learning_rate": 9.226946438431806e-06, "loss": 0.0004, "reward": 1.0, "reward_std": 0.0, "rewards/accuracy_reward": 0.0, "rewards/format_reward": 1.0, "step": 280 }, { "completion_length": 62.7, "epoch": 0.0800662617338487, "grad_norm": 9.5625, "kl": 0.007968542701564729, "learning_rate": 9.199337382661515e-06, "loss": 0.0003, "reward": 1.0, "reward_std": 0.2, "rewards/accuracy_reward": 0.05, "rewards/format_reward": 0.95, "step": 290 }, { "completion_length": 49.3, "epoch": 0.08282716731087797, "grad_norm": 8.75, "kl": 0.019733481900766493, "learning_rate": 9.171728326891221e-06, "loss": 0.0008, "reward": 1.05, "reward_std": 0.15773502588272095, "rewards/accuracy_reward": 0.075, "rewards/format_reward": 0.975, "step": 300 }, { "completion_length": 58.925, "epoch": 0.08558807288790723, "grad_norm": 0.267578125, "kl": 0.01582272765226662, "learning_rate": 9.144119271120928e-06, "loss": 0.0006, "reward": 1.075, "reward_std": 0.10773502588272095, "rewards/accuracy_reward": 0.075, "rewards/format_reward": 1.0, "step": 310 }, { "completion_length": 36.35, "epoch": 0.0883489784649365, "grad_norm": 10.6875, "kl": 0.01768846157938242, "learning_rate": 9.116510215350636e-06, "loss": 0.0007, "reward": 1.05, "reward_std": 0.1, "rewards/accuracy_reward": 0.05, "rewards/format_reward": 1.0, "step": 320 }, { "completion_length": 36.1, "epoch": 0.09110988404196577, "grad_norm": 0.2373046875, "kl": 0.024144751764833928, "learning_rate": 9.088901159580343e-06, "loss": 0.001, "reward": 1.05, "reward_std": 0.1, "rewards/accuracy_reward": 0.05, "rewards/format_reward": 1.0, "step": 330 }, { "completion_length": 35.225, "epoch": 0.09387078961899503, "grad_norm": 0.19140625, "kl": 0.021723289880901574, "learning_rate": 9.061292103810051e-06, "loss": 0.0009, "reward": 1.025, "reward_std": 0.05, "rewards/accuracy_reward": 0.025, "rewards/format_reward": 1.0, "step": 340 }, { "completion_length": 41.35, "epoch": 0.0966316951960243, "grad_norm": 0.1962890625, "kl": 0.020848556840792298, "learning_rate": 9.033683048039758e-06, "loss": 0.0008, "reward": 1.075, "reward_std": 0.10773502588272095, "rewards/accuracy_reward": 0.075, "rewards/format_reward": 1.0, "step": 350 }, { "completion_length": 42.825, "epoch": 0.09939260077305356, "grad_norm": 0.26171875, "kl": 0.01832999554462731, "learning_rate": 9.006073992269465e-06, "loss": 0.0007, "reward": 1.05, "reward_std": 0.18164966106414795, "rewards/accuracy_reward": 0.075, "rewards/format_reward": 0.975, "step": 360 }, { "completion_length": 50.575, "epoch": 0.10215350635008283, "grad_norm": 0.57421875, "kl": 0.017559885466471314, "learning_rate": 8.978464936499173e-06, "loss": 0.0007, "reward": 1.05, "reward_std": 0.05773502588272095, "rewards/accuracy_reward": 0.05, "rewards/format_reward": 1.0, "step": 370 }, { "completion_length": 40.825, "epoch": 0.1049144119271121, "grad_norm": 0.2041015625, "kl": 0.017281288420781492, "learning_rate": 8.95085588072888e-06, "loss": 0.0007, "reward": 1.025, "reward_std": 0.05, "rewards/accuracy_reward": 0.025, "rewards/format_reward": 1.0, "step": 380 }, { "completion_length": 52.85, "epoch": 0.10767531750414136, "grad_norm": 16.875, "kl": 0.021487308479845525, "learning_rate": 8.923246824958586e-06, "loss": 0.0009, "reward": 1.125, "reward_std": 0.15, "rewards/accuracy_reward": 0.175, "rewards/format_reward": 0.95, "step": 390 }, { "completion_length": 40.625, "epoch": 0.11043622308117063, "grad_norm": 0.388671875, "kl": 0.017221461795270444, "learning_rate": 8.895637769188295e-06, "loss": 0.0007, "reward": 0.975, "reward_std": 0.05, "rewards/accuracy_reward": 0.0, "rewards/format_reward": 0.975, "step": 400 }, { "completion_length": 38.725, "epoch": 0.11319712865819989, "grad_norm": 29.125, "kl": 0.017092665936797858, "learning_rate": 8.868028713418001e-06, "loss": 0.0007, "reward": 0.95, "reward_std": 0.1, "rewards/accuracy_reward": 0.0, "rewards/format_reward": 0.95, "step": 410 }, { "completion_length": 36.15, "epoch": 0.11595803423522916, "grad_norm": 0.3671875, "kl": 0.012849510973319411, "learning_rate": 8.84041965764771e-06, "loss": 0.0005, "reward": 0.975, "reward_std": 0.05, "rewards/accuracy_reward": 0.0, "rewards/format_reward": 0.975, "step": 420 }, { "completion_length": 44.075, "epoch": 0.11871893981225842, "grad_norm": 0.34375, "kl": 0.01729746237397194, "learning_rate": 8.812810601877416e-06, "loss": 0.0007, "reward": 1.0, "reward_std": 0.1, "rewards/accuracy_reward": 0.025, "rewards/format_reward": 0.975, "step": 430 }, { "completion_length": 47.05, "epoch": 0.12147984538928769, "grad_norm": 0.232421875, "kl": 0.012639089813455939, "learning_rate": 8.785201546107123e-06, "loss": 0.0005, "reward": 1.05, "reward_std": 0.1, "rewards/accuracy_reward": 0.05, "rewards/format_reward": 1.0, "step": 440 }, { "completion_length": 62.525, "epoch": 0.12424075096631695, "grad_norm": 0.5234375, "kl": 0.014472203236073256, "learning_rate": 8.757592490336831e-06, "loss": 0.0006, "reward": 1.0, "reward_std": 0.0, "rewards/accuracy_reward": 0.0, "rewards/format_reward": 1.0, "step": 450 }, { "completion_length": 69.5, "epoch": 0.12700165654334622, "grad_norm": 0.267578125, "kl": 0.011864464683458208, "learning_rate": 8.72998343456654e-06, "loss": 0.0005, "reward": 1.1, "reward_std": 0.1, "rewards/accuracy_reward": 0.1, "rewards/format_reward": 1.0, "step": 460 }, { "completion_length": 47.625, "epoch": 0.12976256212037549, "grad_norm": 0.2041015625, "kl": 0.01089612185023725, "learning_rate": 8.702374378796245e-06, "loss": 0.0004, "reward": 1.05, "reward_std": 0.05773502588272095, "rewards/accuracy_reward": 0.05, "rewards/format_reward": 1.0, "step": 470 }, { "completion_length": 48.025, "epoch": 0.13252346769740475, "grad_norm": 0.1865234375, "kl": 0.011578870052471758, "learning_rate": 8.674765323025953e-06, "loss": 0.0005, "reward": 1.075, "reward_std": 0.10773502588272095, "rewards/accuracy_reward": 0.075, "rewards/format_reward": 1.0, "step": 480 }, { "completion_length": 48.675, "epoch": 0.13528437327443402, "grad_norm": 0.15234375, "kl": 0.015454969368875027, "learning_rate": 8.647156267255661e-06, "loss": 0.0006, "reward": 1.0, "reward_std": 0.1, "rewards/accuracy_reward": 0.025, "rewards/format_reward": 0.975, "step": 490 }, { "completion_length": 51.525, "epoch": 0.13804527885146328, "grad_norm": 0.341796875, "kl": 0.011392440786585212, "learning_rate": 8.619547211485368e-06, "loss": 0.0005, "reward": 1.15, "reward_std": 0.1, "rewards/accuracy_reward": 0.15, "rewards/format_reward": 1.0, "step": 500 }, { "completion_length": 51.85, "epoch": 0.14080618442849255, "grad_norm": 0.302734375, "kl": 0.010139884939417244, "learning_rate": 8.591938155715075e-06, "loss": 0.0004, "reward": 1.075, "reward_std": 0.10773502588272095, "rewards/accuracy_reward": 0.075, "rewards/format_reward": 1.0, "step": 510 }, { "completion_length": 49.175, "epoch": 0.1435670900055218, "grad_norm": 0.486328125, "kl": 0.013384641613811255, "learning_rate": 8.564329099944783e-06, "loss": 0.0005, "reward": 1.075, "reward_std": 0.10773502588272095, "rewards/accuracy_reward": 0.075, "rewards/format_reward": 1.0, "step": 520 }, { "completion_length": 58.45, "epoch": 0.14632799558255108, "grad_norm": 0.3046875, "kl": 0.014279201999306679, "learning_rate": 8.53672004417449e-06, "loss": 0.0006, "reward": 1.05, "reward_std": 0.2, "rewards/accuracy_reward": 0.075, "rewards/format_reward": 0.975, "step": 530 }, { "completion_length": 46.65, "epoch": 0.14908890115958034, "grad_norm": 0.462890625, "kl": 0.017445506248623132, "learning_rate": 8.509110988404198e-06, "loss": 0.0007, "reward": 1.05, "reward_std": 0.05773502588272095, "rewards/accuracy_reward": 0.05, "rewards/format_reward": 1.0, "step": 540 }, { "completion_length": 43.45, "epoch": 0.1518498067366096, "grad_norm": 0.6875, "kl": 0.029544192645698786, "learning_rate": 8.481501932633905e-06, "loss": 0.0012, "reward": 0.975, "reward_std": 0.05, "rewards/accuracy_reward": 0.0, "rewards/format_reward": 0.975, "step": 550 }, { "completion_length": 52.625, "epoch": 0.15461071231363888, "grad_norm": 0.052001953125, "kl": 0.0027885228249942885, "learning_rate": 8.453892876863612e-06, "loss": 0.0001, "reward": 0.975, "reward_std": 0.05, "rewards/accuracy_reward": 0.0, "rewards/format_reward": 0.975, "step": 560 }, { "completion_length": 69.475, "epoch": 0.15737161789066814, "grad_norm": 0.126953125, "kl": 0.0009036411240231246, "learning_rate": 8.42628382109332e-06, "loss": 0.0, "reward": 1.0, "reward_std": 0.1, "rewards/accuracy_reward": 0.025, "rewards/format_reward": 0.975, "step": 570 }, { "completion_length": 46.75, "epoch": 0.1601325234676974, "grad_norm": 0.1201171875, "kl": 0.0012649911106564105, "learning_rate": 8.398674765323027e-06, "loss": 0.0001, "reward": 1.025, "reward_std": 0.05, "rewards/accuracy_reward": 0.025, "rewards/format_reward": 1.0, "step": 580 }, { "completion_length": 57.325, "epoch": 0.16289342904472667, "grad_norm": 0.11669921875, "kl": 0.0012186030071461573, "learning_rate": 8.371065709552733e-06, "loss": 0.0, "reward": 1.1, "reward_std": 0.0, "rewards/accuracy_reward": 0.1, "rewards/format_reward": 1.0, "step": 590 }, { "completion_length": 54.925, "epoch": 0.16565433462175594, "grad_norm": 0.10595703125, "kl": 0.0020359954505693167, "learning_rate": 8.343456653782442e-06, "loss": 0.0001, "reward": 1.075, "reward_std": 0.05, "rewards/accuracy_reward": 0.075, "rewards/format_reward": 1.0, "step": 600 }, { "completion_length": 45.775, "epoch": 0.1684152401987852, "grad_norm": 0.1396484375, "kl": 0.0013287237728945912, "learning_rate": 8.315847598012148e-06, "loss": 0.0001, "reward": 1.0, "reward_std": 0.0, "rewards/accuracy_reward": 0.0, "rewards/format_reward": 1.0, "step": 610 }, { "completion_length": 57.35, "epoch": 0.17117614577581447, "grad_norm": 13.3125, "kl": 0.0009392780877533369, "learning_rate": 8.288238542241857e-06, "loss": 0.0, "reward": 1.05, "reward_std": 0.05773502588272095, "rewards/accuracy_reward": 0.05, "rewards/format_reward": 1.0, "step": 620 }, { "completion_length": 61.025, "epoch": 0.17393705135284374, "grad_norm": 0.1376953125, "kl": 0.0012591792619787157, "learning_rate": 8.260629486471563e-06, "loss": 0.0001, "reward": 1.075, "reward_std": 0.10773502588272095, "rewards/accuracy_reward": 0.075, "rewards/format_reward": 1.0, "step": 630 }, { "completion_length": 42.275, "epoch": 0.176697956929873, "grad_norm": 0.1533203125, "kl": 0.0016204892890527844, "learning_rate": 8.23302043070127e-06, "loss": 0.0001, "reward": 1.0, "reward_std": 0.0, "rewards/accuracy_reward": 0.0, "rewards/format_reward": 1.0, "step": 640 }, { "completion_length": 57.25, "epoch": 0.17945886250690227, "grad_norm": 0.19140625, "kl": 0.0028853721683844923, "learning_rate": 8.205411374930978e-06, "loss": 0.0001, "reward": 0.975, "reward_std": 0.15, "rewards/accuracy_reward": 0.025, "rewards/format_reward": 0.95, "step": 650 }, { "completion_length": 50.125, "epoch": 0.18221976808393153, "grad_norm": 0.189453125, "kl": 0.002419785177335143, "learning_rate": 8.177802319160687e-06, "loss": 0.0001, "reward": 1.025, "reward_std": 0.05, "rewards/accuracy_reward": 0.025, "rewards/format_reward": 1.0, "step": 660 }, { "completion_length": 55.85, "epoch": 0.1849806736609608, "grad_norm": 0.1474609375, "kl": 0.0022337425500154494, "learning_rate": 8.150193263390392e-06, "loss": 0.0001, "reward": 1.025, "reward_std": 0.05, "rewards/accuracy_reward": 0.025, "rewards/format_reward": 1.0, "step": 670 }, { "completion_length": 51.05, "epoch": 0.18774157923799006, "grad_norm": 0.2099609375, "kl": 0.0033827337785623966, "learning_rate": 8.1225842076201e-06, "loss": 0.0001, "reward": 1.0, "reward_std": 0.0, "rewards/accuracy_reward": 0.0, "rewards/format_reward": 1.0, "step": 680 }, { "completion_length": 43.275, "epoch": 0.19050248481501933, "grad_norm": 0.5078125, "kl": 0.01497054547071457, "learning_rate": 8.094975151849808e-06, "loss": 0.0006, "reward": 1.025, "reward_std": 0.05, "rewards/accuracy_reward": 0.025, "rewards/format_reward": 1.0, "step": 690 }, { "completion_length": 42.1, "epoch": 0.1932633903920486, "grad_norm": 12.0625, "kl": 0.013852290622889996, "learning_rate": 8.067366096079515e-06, "loss": 0.0006, "reward": 1.05, "reward_std": 0.15773502588272095, "rewards/accuracy_reward": 0.075, "rewards/format_reward": 0.975, "step": 700 }, { "completion_length": 53.15, "epoch": 0.19602429596907786, "grad_norm": 19.625, "kl": 0.013076470326632261, "learning_rate": 8.039757040309222e-06, "loss": 0.0005, "reward": 0.975, "reward_std": 0.05, "rewards/accuracy_reward": 0.0, "rewards/format_reward": 0.975, "step": 710 }, { "completion_length": 56.225, "epoch": 0.19878520154610713, "grad_norm": 0.1669921875, "kl": 0.014432823657989502, "learning_rate": 8.01214798453893e-06, "loss": 0.0006, "reward": 1.0, "reward_std": 0.1, "rewards/accuracy_reward": 0.025, "rewards/format_reward": 0.975, "step": 720 }, { "completion_length": 46.8, "epoch": 0.2015461071231364, "grad_norm": 16.5, "kl": 0.013125162106007338, "learning_rate": 7.984538928768637e-06, "loss": 0.0005, "reward": 1.05, "reward_std": 0.05773502588272095, "rewards/accuracy_reward": 0.05, "rewards/format_reward": 1.0, "step": 730 }, { "completion_length": 47.425, "epoch": 0.20430701270016566, "grad_norm": 17.125, "kl": 0.010656389640644193, "learning_rate": 7.956929872998345e-06, "loss": 0.0004, "reward": 1.0, "reward_std": 0.2, "rewards/accuracy_reward": 0.05, "rewards/format_reward": 0.95, "step": 740 }, { "completion_length": 44.45, "epoch": 0.20706791827719492, "grad_norm": 0.2734375, "kl": 0.014079915639013053, "learning_rate": 7.929320817228052e-06, "loss": 0.0006, "reward": 1.1, "reward_std": 0.0, "rewards/accuracy_reward": 0.1, "rewards/format_reward": 1.0, "step": 750 }, { "completion_length": 58.25, "epoch": 0.2098288238542242, "grad_norm": 0.1123046875, "kl": 0.013165965024381877, "learning_rate": 7.901711761457758e-06, "loss": 0.0005, "reward": 0.975, "reward_std": 0.05, "rewards/accuracy_reward": 0.0, "rewards/format_reward": 0.975, "step": 760 }, { "completion_length": 50.35, "epoch": 0.21258972943125345, "grad_norm": 0.53125, "kl": 0.02708628848195076, "learning_rate": 7.874102705687467e-06, "loss": 0.0011, "reward": 1.075, "reward_std": 0.05, "rewards/accuracy_reward": 0.1, "rewards/format_reward": 0.975, "step": 770 }, { "completion_length": 49.125, "epoch": 0.21535063500828272, "grad_norm": 0.302734375, "kl": 0.006959504983387888, "learning_rate": 7.846493649917174e-06, "loss": 0.0003, "reward": 1.05, "reward_std": 0.05773502588272095, "rewards/accuracy_reward": 0.05, "rewards/format_reward": 1.0, "step": 780 }, { "completion_length": 53.575, "epoch": 0.21811154058531199, "grad_norm": 0.1923828125, "kl": 0.008690130175091326, "learning_rate": 7.81888459414688e-06, "loss": 0.0003, "reward": 1.0, "reward_std": 0.1, "rewards/accuracy_reward": 0.025, "rewards/format_reward": 0.975, "step": 790 }, { "completion_length": 48.7, "epoch": 0.22087244616234125, "grad_norm": 0.234375, "kl": 0.01659679235890508, "learning_rate": 7.791275538376589e-06, "loss": 0.0007, "reward": 1.0, "reward_std": 0.0, "rewards/accuracy_reward": 0.0, "rewards/format_reward": 1.0, "step": 800 }, { "completion_length": 39.625, "epoch": 0.22363335173937052, "grad_norm": 0.244140625, "kl": 0.010390093945898116, "learning_rate": 7.763666482606295e-06, "loss": 0.0004, "reward": 1.0, "reward_std": 0.0, "rewards/accuracy_reward": 0.0, "rewards/format_reward": 1.0, "step": 810 }, { "completion_length": 62.025, "epoch": 0.22639425731639978, "grad_norm": 0.154296875, "kl": 0.009211322665214539, "learning_rate": 7.736057426836004e-06, "loss": 0.0004, "reward": 1.0, "reward_std": 0.0, "rewards/accuracy_reward": 0.0, "rewards/format_reward": 1.0, "step": 820 }, { "completion_length": 53.2, "epoch": 0.22915516289342905, "grad_norm": 0.263671875, "kl": 0.011441088351421057, "learning_rate": 7.70844837106571e-06, "loss": 0.0005, "reward": 1.0, "reward_std": 0.0, "rewards/accuracy_reward": 0.0, "rewards/format_reward": 1.0, "step": 830 }, { "completion_length": 51.25, "epoch": 0.23191606847045831, "grad_norm": 0.2578125, "kl": 0.00715010860003531, "learning_rate": 7.680839315295417e-06, "loss": 0.0003, "reward": 1.075, "reward_std": 0.10773502588272095, "rewards/accuracy_reward": 0.075, "rewards/format_reward": 1.0, "step": 840 }, { "completion_length": 52.925, "epoch": 0.23467697404748758, "grad_norm": 0.34765625, "kl": 0.010182457882910967, "learning_rate": 7.653230259525125e-06, "loss": 0.0004, "reward": 1.05, "reward_std": 0.1, "rewards/accuracy_reward": 0.05, "rewards/format_reward": 1.0, "step": 850 }, { "completion_length": 51.9, "epoch": 0.23743787962451685, "grad_norm": 0.279296875, "kl": 0.009262909227982164, "learning_rate": 7.625621203754833e-06, "loss": 0.0004, "reward": 0.975, "reward_std": 0.05, "rewards/accuracy_reward": 0.0, "rewards/format_reward": 0.975, "step": 860 }, { "completion_length": 47.1, "epoch": 0.2401987852015461, "grad_norm": 0.1201171875, "kl": 0.014673679834231734, "learning_rate": 7.5980121479845395e-06, "loss": 0.0006, "reward": 1.075, "reward_std": 0.05, "rewards/accuracy_reward": 0.075, "rewards/format_reward": 1.0, "step": 870 }, { "completion_length": 43.425, "epoch": 0.24295969077857538, "grad_norm": 0.291015625, "kl": 0.019366781413555145, "learning_rate": 7.570403092214247e-06, "loss": 0.0008, "reward": 1.025, "reward_std": 0.05, "rewards/accuracy_reward": 0.025, "rewards/format_reward": 1.0, "step": 880 }, { "completion_length": 59.625, "epoch": 0.24572059635560464, "grad_norm": 0.5625, "kl": 0.011405015597119927, "learning_rate": 7.5427940364439545e-06, "loss": 0.0005, "reward": 1.0, "reward_std": 0.1, "rewards/accuracy_reward": 0.025, "rewards/format_reward": 0.975, "step": 890 }, { "completion_length": 56.85, "epoch": 0.2484815019326339, "grad_norm": 0.3515625, "kl": 0.012598360935226083, "learning_rate": 7.515184980673662e-06, "loss": 0.0005, "reward": 1.025, "reward_std": 0.05, "rewards/accuracy_reward": 0.025, "rewards/format_reward": 1.0, "step": 900 }, { "completion_length": 42.475, "epoch": 0.25124240750966315, "grad_norm": 19.125, "kl": 0.013188354019075633, "learning_rate": 7.487575924903369e-06, "loss": 0.0005, "reward": 1.15, "reward_std": 0.15773502588272095, "rewards/accuracy_reward": 0.15, "rewards/format_reward": 1.0, "step": 910 }, { "completion_length": 58.5, "epoch": 0.25400331308669244, "grad_norm": 10.625, "kl": 0.024910295475274326, "learning_rate": 7.459966869133076e-06, "loss": 0.001, "reward": 1.0, "reward_std": 0.1, "rewards/accuracy_reward": 0.025, "rewards/format_reward": 0.975, "step": 920 }, { "completion_length": 47.925, "epoch": 0.2567642186637217, "grad_norm": 9.125, "kl": 0.013725518900901078, "learning_rate": 7.432357813362784e-06, "loss": 0.0005, "reward": 1.05, "reward_std": 0.05773502588272095, "rewards/accuracy_reward": 0.05, "rewards/format_reward": 1.0, "step": 930 }, { "completion_length": 42.125, "epoch": 0.25952512424075097, "grad_norm": 0.50390625, "kl": 0.01066298601217568, "learning_rate": 7.404748757592491e-06, "loss": 0.0004, "reward": 1.0, "reward_std": 0.1, "rewards/accuracy_reward": 0.025, "rewards/format_reward": 0.975, "step": 940 }, { "completion_length": 44.525, "epoch": 0.2622860298177802, "grad_norm": 0.2578125, "kl": 0.010468068346381188, "learning_rate": 7.377139701822198e-06, "loss": 0.0004, "reward": 1.0, "reward_std": 0.0, "rewards/accuracy_reward": 0.0, "rewards/format_reward": 1.0, "step": 950 }, { "completion_length": 53.225, "epoch": 0.2650469353948095, "grad_norm": 0.09521484375, "kl": 0.009685284015722573, "learning_rate": 7.3495306460519054e-06, "loss": 0.0004, "reward": 1.0, "reward_std": 0.08164966106414795, "rewards/accuracy_reward": 0.025, "rewards/format_reward": 0.975, "step": 960 }, { "completion_length": 45.075, "epoch": 0.26780784097183874, "grad_norm": 0.3046875, "kl": 0.01133408104069531, "learning_rate": 7.321921590281613e-06, "loss": 0.0005, "reward": 1.05, "reward_std": 0.05773502588272095, "rewards/accuracy_reward": 0.05, "rewards/format_reward": 1.0, "step": 970 }, { "completion_length": 53.85, "epoch": 0.27056874654886803, "grad_norm": 8.4375, "kl": 0.009008155239280313, "learning_rate": 7.2943125345113205e-06, "loss": 0.0004, "reward": 1.0, "reward_std": 0.1, "rewards/accuracy_reward": 0.025, "rewards/format_reward": 0.975, "step": 980 }, { "completion_length": 55.875, "epoch": 0.27332965212589727, "grad_norm": 0.2734375, "kl": 0.01266736425459385, "learning_rate": 7.266703478741027e-06, "loss": 0.0005, "reward": 1.0, "reward_std": 0.0, "rewards/accuracy_reward": 0.0, "rewards/format_reward": 1.0, "step": 990 }, { "completion_length": 60.4, "epoch": 0.27609055770292656, "grad_norm": 0.33984375, "kl": 0.009570520836859941, "learning_rate": 7.239094422970735e-06, "loss": 0.0004, "reward": 1.0, "reward_std": 0.2, "rewards/accuracy_reward": 0.05, "rewards/format_reward": 0.95, "step": 1000 }, { "completion_length": 41.35, "epoch": 0.2788514632799558, "grad_norm": 0.21484375, "kl": 0.017147039994597435, "learning_rate": 7.211485367200442e-06, "loss": 0.0007, "reward": 0.975, "reward_std": 0.05, "rewards/accuracy_reward": 0.0, "rewards/format_reward": 0.975, "step": 1010 }, { "completion_length": 56.6, "epoch": 0.2816123688569851, "grad_norm": 0.248046875, "kl": 0.010259191761724651, "learning_rate": 7.1838763114301505e-06, "loss": 0.0004, "reward": 0.975, "reward_std": 0.05, "rewards/accuracy_reward": 0.0, "rewards/format_reward": 0.975, "step": 1020 }, { "completion_length": 46.375, "epoch": 0.28437327443401433, "grad_norm": 14.25, "kl": 0.012054991768673062, "learning_rate": 7.156267255659856e-06, "loss": 0.0005, "reward": 1.125, "reward_std": 0.10773502588272095, "rewards/accuracy_reward": 0.15, "rewards/format_reward": 0.975, "step": 1030 }, { "completion_length": 59.975, "epoch": 0.2871341800110436, "grad_norm": 0.1884765625, "kl": 0.00799824194982648, "learning_rate": 7.128658199889565e-06, "loss": 0.0003, "reward": 0.975, "reward_std": 0.05, "rewards/accuracy_reward": 0.0, "rewards/format_reward": 0.975, "step": 1040 }, { "completion_length": 48.4, "epoch": 0.28989508558807286, "grad_norm": 0.12451171875, "kl": 0.00863449857570231, "learning_rate": 7.101049144119272e-06, "loss": 0.0003, "reward": 1.025, "reward_std": 0.05, "rewards/accuracy_reward": 0.025, "rewards/format_reward": 1.0, "step": 1050 }, { "completion_length": 46.525, "epoch": 0.29265599116510216, "grad_norm": 0.443359375, "kl": 0.015027077682316304, "learning_rate": 7.07344008834898e-06, "loss": 0.0006, "reward": 1.025, "reward_std": 0.05, "rewards/accuracy_reward": 0.025, "rewards/format_reward": 1.0, "step": 1060 }, { "completion_length": 52.5, "epoch": 0.2954168967421314, "grad_norm": 0.216796875, "kl": 0.023942758096382022, "learning_rate": 9.97239094422971e-06, "loss": 0.001, "reward": 1.0, "reward_std": 0.0, "rewards/accuracy_reward": 0.0, "rewards/format_reward": 1.0, "step": 1070 }, { "completion_length": 43.475, "epoch": 0.2981778023191607, "grad_norm": 2.5625, "kl": 0.022519207373261452, "learning_rate": 9.944781888459416e-06, "loss": 0.0009, "reward": 1.125, "reward_std": 0.05, "rewards/accuracy_reward": 0.125, "rewards/format_reward": 1.0, "step": 1080 }, { "completion_length": 39.875, "epoch": 0.3009387078961899, "grad_norm": 0.173828125, "kl": 0.015989596396684645, "learning_rate": 9.917172832689123e-06, "loss": 0.0006, "reward": 1.0, "reward_std": 0.0, "rewards/accuracy_reward": 0.0, "rewards/format_reward": 1.0, "step": 1090 }, { "completion_length": 53.975, "epoch": 0.3036996134732192, "grad_norm": 25.125, "kl": 0.013253817660734057, "learning_rate": 9.889563776918831e-06, "loss": 0.0005, "reward": 1.075, "reward_std": 0.10773502588272095, "rewards/accuracy_reward": 0.075, "rewards/format_reward": 1.0, "step": 1100 }, { "completion_length": 46.475, "epoch": 0.30646051905024846, "grad_norm": 0.193359375, "kl": 0.015551709942519665, "learning_rate": 9.861954721148538e-06, "loss": 0.0006, "reward": 1.05, "reward_std": 0.1, "rewards/accuracy_reward": 0.05, "rewards/format_reward": 1.0, "step": 1110 }, { "completion_length": 41.15, "epoch": 0.30922142462727775, "grad_norm": 0.6796875, "kl": 0.01951191038824618, "learning_rate": 9.834345665378246e-06, "loss": 0.0008, "reward": 1.0, "reward_std": 0.0, "rewards/accuracy_reward": 0.0, "rewards/format_reward": 1.0, "step": 1120 }, { "completion_length": 54.9, "epoch": 0.311982330204307, "grad_norm": 14.4375, "kl": 0.020035648718476295, "learning_rate": 9.806736609607953e-06, "loss": 0.0008, "reward": 1.15, "reward_std": 0.1, "rewards/accuracy_reward": 0.15, "rewards/format_reward": 1.0, "step": 1130 }, { "completion_length": 47.95, "epoch": 0.3147432357813363, "grad_norm": 11.375, "kl": 0.014760071225464345, "learning_rate": 9.77912755383766e-06, "loss": 0.0006, "reward": 1.025, "reward_std": 0.05, "rewards/accuracy_reward": 0.025, "rewards/format_reward": 1.0, "step": 1140 }, { "completion_length": 51.55, "epoch": 0.3175041413583655, "grad_norm": 0.30859375, "kl": 0.013024270394816994, "learning_rate": 9.751518498067368e-06, "loss": 0.0005, "reward": 0.975, "reward_std": 0.05, "rewards/accuracy_reward": 0.0, "rewards/format_reward": 0.975, "step": 1150 }, { "completion_length": 42.075, "epoch": 0.3202650469353948, "grad_norm": 0.08837890625, "kl": 0.009475133288651705, "learning_rate": 9.723909442297074e-06, "loss": 0.0004, "reward": 1.0, "reward_std": 0.0, "rewards/accuracy_reward": 0.0, "rewards/format_reward": 1.0, "step": 1160 }, { "completion_length": 54.35, "epoch": 0.32302595251242405, "grad_norm": 14.0625, "kl": 0.012639770423993469, "learning_rate": 9.696300386526781e-06, "loss": 0.0005, "reward": 1.0, "reward_std": 0.1154700517654419, "rewards/accuracy_reward": 0.05, "rewards/format_reward": 0.95, "step": 1170 }, { "completion_length": 43.6, "epoch": 0.32578685808945335, "grad_norm": 0.1875, "kl": 0.02179276868700981, "learning_rate": 9.66869133075649e-06, "loss": 0.0009, "reward": 1.025, "reward_std": 0.05, "rewards/accuracy_reward": 0.025, "rewards/format_reward": 1.0, "step": 1180 }, { "completion_length": 43.725, "epoch": 0.3285477636664826, "grad_norm": 0.1689453125, "kl": 0.014609583420678973, "learning_rate": 9.641082274986196e-06, "loss": 0.0006, "reward": 1.075, "reward_std": 0.10773502588272095, "rewards/accuracy_reward": 0.075, "rewards/format_reward": 1.0, "step": 1190 }, { "completion_length": 56.45, "epoch": 0.3313086692435119, "grad_norm": 0.205078125, "kl": 0.01842534001916647, "learning_rate": 9.613473219215904e-06, "loss": 0.0007, "reward": 0.975, "reward_std": 0.05, "rewards/accuracy_reward": 0.0, "rewards/format_reward": 0.975, "step": 1200 }, { "completion_length": 42.975, "epoch": 0.3340695748205411, "grad_norm": 0.330078125, "kl": 0.02300509326159954, "learning_rate": 9.585864163445611e-06, "loss": 0.0009, "reward": 0.975, "reward_std": 0.05, "rewards/accuracy_reward": 0.0, "rewards/format_reward": 0.975, "step": 1210 }, { "completion_length": 41.375, "epoch": 0.3368304803975704, "grad_norm": 0.2421875, "kl": 0.011269495077431201, "learning_rate": 9.558255107675318e-06, "loss": 0.0005, "reward": 1.05, "reward_std": 0.05773502588272095, "rewards/accuracy_reward": 0.05, "rewards/format_reward": 1.0, "step": 1220 }, { "completion_length": 47.8, "epoch": 0.33959138597459965, "grad_norm": 0.34375, "kl": 0.019025292107835413, "learning_rate": 9.530646051905026e-06, "loss": 0.0008, "reward": 0.95, "reward_std": 0.1, "rewards/accuracy_reward": 0.0, "rewards/format_reward": 0.95, "step": 1230 }, { "completion_length": 39.875, "epoch": 0.34235229155162894, "grad_norm": 0.60546875, "kl": 0.016004760656505822, "learning_rate": 9.503036996134733e-06, "loss": 0.0006, "reward": 1.025, "reward_std": 0.05, "rewards/accuracy_reward": 0.025, "rewards/format_reward": 1.0, "step": 1240 }, { "completion_length": 47.925, "epoch": 0.3451131971286582, "grad_norm": 19.375, "kl": 0.021373574901372195, "learning_rate": 9.47542794036444e-06, "loss": 0.0009, "reward": 1.025, "reward_std": 0.05, "rewards/accuracy_reward": 0.025, "rewards/format_reward": 1.0, "step": 1250 }, { "completion_length": 53.9, "epoch": 0.34787410270568747, "grad_norm": 0.16015625, "kl": 0.013111930852755905, "learning_rate": 9.447818884594148e-06, "loss": 0.0005, "reward": 1.075, "reward_std": 0.05, "rewards/accuracy_reward": 0.1, "rewards/format_reward": 0.975, "step": 1260 }, { "completion_length": 53.8, "epoch": 0.3506350082827167, "grad_norm": 0.2236328125, "kl": 0.017734084371477367, "learning_rate": 9.420209828823854e-06, "loss": 0.0007, "reward": 1.05, "reward_std": 0.05773502588272095, "rewards/accuracy_reward": 0.05, "rewards/format_reward": 1.0, "step": 1270 }, { "completion_length": 45.125, "epoch": 0.353395913859746, "grad_norm": 0.326171875, "kl": 0.018635424505919217, "learning_rate": 9.392600773053563e-06, "loss": 0.0007, "reward": 1.0, "reward_std": 0.0, "rewards/accuracy_reward": 0.0, "rewards/format_reward": 1.0, "step": 1280 }, { "completion_length": 37.35, "epoch": 0.35615681943677524, "grad_norm": 0.330078125, "kl": 0.009210627479478716, "learning_rate": 9.36499171728327e-06, "loss": 0.0004, "reward": 1.05, "reward_std": 0.1, "rewards/accuracy_reward": 0.05, "rewards/format_reward": 1.0, "step": 1290 }, { "completion_length": 50.85, "epoch": 0.35891772501380453, "grad_norm": 0.470703125, "kl": 0.028195956349372865, "learning_rate": 9.337382661512978e-06, "loss": 0.0011, "reward": 1.0, "reward_std": 0.0, "rewards/accuracy_reward": 0.0, "rewards/format_reward": 1.0, "step": 1300 }, { "completion_length": 64.925, "epoch": 0.36167863059083377, "grad_norm": 0.173828125, "kl": 0.01299036997370422, "learning_rate": 9.309773605742684e-06, "loss": 0.0005, "reward": 0.975, "reward_std": 0.05, "rewards/accuracy_reward": 0.0, "rewards/format_reward": 0.975, "step": 1310 }, { "completion_length": 48.3, "epoch": 0.36443953616786307, "grad_norm": 0.2314453125, "kl": 0.013318869331851602, "learning_rate": 9.282164549972393e-06, "loss": 0.0005, "reward": 1.0, "reward_std": 0.0, "rewards/accuracy_reward": 0.0, "rewards/format_reward": 1.0, "step": 1320 }, { "completion_length": 43.675, "epoch": 0.3672004417448923, "grad_norm": 12.375, "kl": 0.021023075561970472, "learning_rate": 9.2545554942021e-06, "loss": 0.0008, "reward": 1.05, "reward_std": 0.1, "rewards/accuracy_reward": 0.05, "rewards/format_reward": 1.0, "step": 1330 }, { "completion_length": 50.0, "epoch": 0.3699613473219216, "grad_norm": 0.1865234375, "kl": 0.018457670323550703, "learning_rate": 9.226946438431806e-06, "loss": 0.0007, "reward": 0.975, "reward_std": 0.05, "rewards/accuracy_reward": 0.0, "rewards/format_reward": 0.975, "step": 1340 }, { "completion_length": 38.625, "epoch": 0.37272225289895083, "grad_norm": 0.265625, "kl": 0.03328995956107974, "learning_rate": 9.199337382661515e-06, "loss": 0.0013, "reward": 1.0, "reward_std": 0.0, "rewards/accuracy_reward": 0.0, "rewards/format_reward": 1.0, "step": 1350 }, { "completion_length": 55.825, "epoch": 0.3754831584759801, "grad_norm": 0.29296875, "kl": 0.013746723486110567, "learning_rate": 9.171728326891221e-06, "loss": 0.0005, "reward": 0.975, "reward_std": 0.10773502588272095, "rewards/accuracy_reward": 0.025, "rewards/format_reward": 0.95, "step": 1360 }, { "completion_length": 47.65, "epoch": 0.37824406405300937, "grad_norm": 0.21484375, "kl": 0.01800147187896073, "learning_rate": 9.144119271120928e-06, "loss": 0.0007, "reward": 1.0, "reward_std": 0.0, "rewards/accuracy_reward": 0.0, "rewards/format_reward": 1.0, "step": 1370 }, { "completion_length": 44.6, "epoch": 0.38100496963003866, "grad_norm": 0.275390625, "kl": 0.01803777916356921, "learning_rate": 9.116510215350636e-06, "loss": 0.0007, "reward": 0.975, "reward_std": 0.05, "rewards/accuracy_reward": 0.0, "rewards/format_reward": 0.975, "step": 1380 }, { "completion_length": 54.575, "epoch": 0.3837658752070679, "grad_norm": 0.46484375, "kl": 0.010938380844891071, "learning_rate": 9.088901159580343e-06, "loss": 0.0004, "reward": 1.0, "reward_std": 0.0, "rewards/accuracy_reward": 0.0, "rewards/format_reward": 1.0, "step": 1390 }, { "completion_length": 47.55, "epoch": 0.3865267807840972, "grad_norm": 0.1748046875, "kl": 0.01793263861909509, "learning_rate": 9.061292103810051e-06, "loss": 0.0007, "reward": 1.025, "reward_std": 0.05, "rewards/accuracy_reward": 0.025, "rewards/format_reward": 1.0, "step": 1400 }, { "completion_length": 42.2, "epoch": 0.38928768636112643, "grad_norm": 0.27734375, "kl": 0.015857995301485062, "learning_rate": 9.033683048039758e-06, "loss": 0.0006, "reward": 1.025, "reward_std": 0.05, "rewards/accuracy_reward": 0.025, "rewards/format_reward": 1.0, "step": 1410 }, { "completion_length": 50.9, "epoch": 0.3920485919381557, "grad_norm": 0.1875, "kl": 0.01180995129980147, "learning_rate": 9.006073992269465e-06, "loss": 0.0005, "reward": 1.05, "reward_std": 0.05773502588272095, "rewards/accuracy_reward": 0.05, "rewards/format_reward": 1.0, "step": 1420 }, { "completion_length": 45.775, "epoch": 0.39480949751518496, "grad_norm": 15.75, "kl": 0.012839672528207302, "learning_rate": 8.978464936499173e-06, "loss": 0.0005, "reward": 1.025, "reward_std": 0.05, "rewards/accuracy_reward": 0.025, "rewards/format_reward": 1.0, "step": 1430 }, { "completion_length": 60.45, "epoch": 0.39757040309221425, "grad_norm": 0.50390625, "kl": 0.014541534706950188, "learning_rate": 8.95085588072888e-06, "loss": 0.0006, "reward": 1.0, "reward_std": 0.0, "rewards/accuracy_reward": 0.0, "rewards/format_reward": 1.0, "step": 1440 }, { "completion_length": 48.25, "epoch": 0.4003313086692435, "grad_norm": 0.435546875, "kl": 0.013137860433198511, "learning_rate": 8.923246824958586e-06, "loss": 0.0005, "reward": 1.0, "reward_std": 0.0, "rewards/accuracy_reward": 0.0, "rewards/format_reward": 1.0, "step": 1450 }, { "completion_length": 55.175, "epoch": 0.4030922142462728, "grad_norm": 18.875, "kl": 0.018580901995301248, "learning_rate": 8.895637769188295e-06, "loss": 0.0007, "reward": 1.125, "reward_std": 0.20773502588272094, "rewards/accuracy_reward": 0.15, "rewards/format_reward": 0.975, "step": 1460 }, { "completion_length": 45.825, "epoch": 0.405853119823302, "grad_norm": 0.203125, "kl": 0.019587689847685396, "learning_rate": 8.868028713418001e-06, "loss": 0.0008, "reward": 1.0, "reward_std": 0.0, "rewards/accuracy_reward": 0.0, "rewards/format_reward": 1.0, "step": 1470 }, { "completion_length": 44.4, "epoch": 0.4086140254003313, "grad_norm": 0.28515625, "kl": 0.015912714693695307, "learning_rate": 8.84041965764771e-06, "loss": 0.0006, "reward": 1.025, "reward_std": 0.05, "rewards/accuracy_reward": 0.025, "rewards/format_reward": 1.0, "step": 1480 }, { "completion_length": 61.825, "epoch": 0.41137493097736055, "grad_norm": 0.38671875, "kl": 0.015045941760763525, "learning_rate": 8.812810601877416e-06, "loss": 0.0006, "reward": 1.05, "reward_std": 0.1, "rewards/accuracy_reward": 0.05, "rewards/format_reward": 1.0, "step": 1490 }, { "completion_length": 40.4, "epoch": 0.41413583655438985, "grad_norm": 14.25, "kl": 0.015176839521154761, "learning_rate": 8.785201546107123e-06, "loss": 0.0006, "reward": 1.025, "reward_std": 0.15, "rewards/accuracy_reward": 0.05, "rewards/format_reward": 0.975, "step": 1500 }, { "completion_length": 66.775, "epoch": 0.4168967421314191, "grad_norm": 0.373046875, "kl": 0.013549512764438987, "learning_rate": 8.757592490336831e-06, "loss": 0.0005, "reward": 1.0, "reward_std": 0.1, "rewards/accuracy_reward": 0.025, "rewards/format_reward": 0.975, "step": 1510 }, { "completion_length": 58.625, "epoch": 0.4196576477084484, "grad_norm": 0.224609375, "kl": 0.019057274051010608, "learning_rate": 8.72998343456654e-06, "loss": 0.0008, "reward": 1.075, "reward_std": 0.15, "rewards/accuracy_reward": 0.1, "rewards/format_reward": 0.975, "step": 1520 }, { "completion_length": 61.15, "epoch": 0.4224185532854776, "grad_norm": 0.7265625, "kl": 0.02888321438804269, "learning_rate": 8.702374378796245e-06, "loss": 0.0012, "reward": 1.05, "reward_std": 0.1393846869468689, "rewards/accuracy_reward": 0.075, "rewards/format_reward": 0.975, "step": 1530 }, { "completion_length": 62.375, "epoch": 0.4251794588625069, "grad_norm": 0.1630859375, "kl": 0.014266025042161345, "learning_rate": 8.674765323025953e-06, "loss": 0.0006, "reward": 1.025, "reward_std": 0.05, "rewards/accuracy_reward": 0.025, "rewards/format_reward": 1.0, "step": 1540 }, { "completion_length": 69.3, "epoch": 0.42794036443953615, "grad_norm": 0.28125, "kl": 0.015173655841499567, "learning_rate": 8.647156267255661e-06, "loss": 0.0006, "reward": 0.975, "reward_std": 0.05, "rewards/accuracy_reward": 0.0, "rewards/format_reward": 0.975, "step": 1550 }, { "completion_length": 49.8, "epoch": 0.43070127001656544, "grad_norm": 0.189453125, "kl": 0.015002421289682388, "learning_rate": 8.619547211485368e-06, "loss": 0.0006, "reward": 1.0, "reward_std": 0.0, "rewards/accuracy_reward": 0.0, "rewards/format_reward": 1.0, "step": 1560 }, { "completion_length": 53.725, "epoch": 0.4334621755935947, "grad_norm": 14.875, "kl": 0.0250665825791657, "learning_rate": 8.591938155715075e-06, "loss": 0.001, "reward": 1.0, "reward_std": 0.15773502588272095, "rewards/accuracy_reward": 0.05, "rewards/format_reward": 0.95, "step": 1570 }, { "completion_length": 37.4, "epoch": 0.43622308117062397, "grad_norm": 0.44140625, "kl": 0.020354483788833023, "learning_rate": 8.564329099944783e-06, "loss": 0.0008, "reward": 1.1, "reward_std": 0.1, "rewards/accuracy_reward": 0.1, "rewards/format_reward": 1.0, "step": 1580 }, { "completion_length": 62.6, "epoch": 0.4389839867476532, "grad_norm": 14.6875, "kl": 0.014426431572064757, "learning_rate": 8.53672004417449e-06, "loss": 0.0006, "reward": 1.075, "reward_std": 0.10773502588272095, "rewards/accuracy_reward": 0.075, "rewards/format_reward": 1.0, "step": 1590 }, { "completion_length": 41.65, "epoch": 0.4417448923246825, "grad_norm": 0.4609375, "kl": 0.013243564590811729, "learning_rate": 8.509110988404198e-06, "loss": 0.0005, "reward": 1.225, "reward_std": 0.10773502588272095, "rewards/accuracy_reward": 0.225, "rewards/format_reward": 1.0, "step": 1600 }, { "completion_length": 52.45, "epoch": 0.44450579790171174, "grad_norm": 1.03125, "kl": 0.01361188031733036, "learning_rate": 8.481501932633905e-06, "loss": 0.0005, "reward": 1.0, "reward_std": 0.0, "rewards/accuracy_reward": 0.0, "rewards/format_reward": 1.0, "step": 1610 }, { "completion_length": 50.15, "epoch": 0.44726670347874103, "grad_norm": 17.375, "kl": 0.018944200593978167, "learning_rate": 8.453892876863612e-06, "loss": 0.0008, "reward": 1.075, "reward_std": 0.15, "rewards/accuracy_reward": 0.075, "rewards/format_reward": 1.0, "step": 1620 }, { "completion_length": 49.65, "epoch": 0.4500276090557703, "grad_norm": 16.75, "kl": 0.015870027616620062, "learning_rate": 8.42628382109332e-06, "loss": 0.0006, "reward": 1.075, "reward_std": 0.10773502588272095, "rewards/accuracy_reward": 0.075, "rewards/format_reward": 1.0, "step": 1630 }, { "completion_length": 78.3, "epoch": 0.45278851463279957, "grad_norm": 0.06494140625, "kl": 0.017166434228420256, "learning_rate": 8.398674765323027e-06, "loss": 0.0007, "reward": 1.025, "reward_std": 0.10773502588272095, "rewards/accuracy_reward": 0.05, "rewards/format_reward": 0.975, "step": 1640 }, { "completion_length": 58.325, "epoch": 0.4555494202098288, "grad_norm": 0.318359375, "kl": 0.012359794043004513, "learning_rate": 8.371065709552733e-06, "loss": 0.0005, "reward": 1.05, "reward_std": 0.05773502588272095, "rewards/accuracy_reward": 0.05, "rewards/format_reward": 1.0, "step": 1650 }, { "completion_length": 56.5, "epoch": 0.4583103257868581, "grad_norm": 0.244140625, "kl": 0.011522294348105788, "learning_rate": 8.343456653782442e-06, "loss": 0.0005, "reward": 1.1, "reward_std": 0.1154700517654419, "rewards/accuracy_reward": 0.1, "rewards/format_reward": 1.0, "step": 1660 }, { "completion_length": 59.55, "epoch": 0.46107123136388733, "grad_norm": 0.3125, "kl": 0.013884695107117295, "learning_rate": 8.315847598012148e-06, "loss": 0.0006, "reward": 1.05, "reward_std": 0.1, "rewards/accuracy_reward": 0.1, "rewards/format_reward": 0.95, "step": 1670 }, { "completion_length": 44.025, "epoch": 0.46383213694091663, "grad_norm": 0.380859375, "kl": 0.012195908557623624, "learning_rate": 8.288238542241857e-06, "loss": 0.0005, "reward": 1.025, "reward_std": 0.05, "rewards/accuracy_reward": 0.025, "rewards/format_reward": 1.0, "step": 1680 }, { "completion_length": 54.825, "epoch": 0.46659304251794587, "grad_norm": 0.54296875, "kl": 0.01730495747178793, "learning_rate": 8.260629486471563e-06, "loss": 0.0007, "reward": 1.025, "reward_std": 0.05, "rewards/accuracy_reward": 0.025, "rewards/format_reward": 1.0, "step": 1690 }, { "completion_length": 47.55, "epoch": 0.46935394809497516, "grad_norm": 0.283203125, "kl": 0.015025637298822402, "learning_rate": 8.23302043070127e-06, "loss": 0.0006, "reward": 1.0, "reward_std": 0.1, "rewards/accuracy_reward": 0.025, "rewards/format_reward": 0.975, "step": 1700 }, { "completion_length": 52.875, "epoch": 0.4721148536720044, "grad_norm": 7.78125, "kl": 0.012533561326563358, "learning_rate": 8.205411374930978e-06, "loss": 0.0005, "reward": 1.075, "reward_std": 0.10773502588272095, "rewards/accuracy_reward": 0.075, "rewards/format_reward": 1.0, "step": 1710 }, { "completion_length": 57.725, "epoch": 0.4748757592490337, "grad_norm": 0.2294921875, "kl": 0.015987796289846302, "learning_rate": 8.177802319160687e-06, "loss": 0.0006, "reward": 1.025, "reward_std": 0.05, "rewards/accuracy_reward": 0.025, "rewards/format_reward": 1.0, "step": 1720 }, { "completion_length": 35.375, "epoch": 0.47763666482606293, "grad_norm": 0.2041015625, "kl": 0.01583628305234015, "learning_rate": 8.150193263390392e-06, "loss": 0.0006, "reward": 1.075, "reward_std": 0.10773502588272095, "rewards/accuracy_reward": 0.075, "rewards/format_reward": 1.0, "step": 1730 }, { "completion_length": 57.15, "epoch": 0.4803975704030922, "grad_norm": 0.349609375, "kl": 0.013509547524154186, "learning_rate": 8.1225842076201e-06, "loss": 0.0005, "reward": 0.925, "reward_std": 0.10773502588272095, "rewards/accuracy_reward": 0.0, "rewards/format_reward": 0.925, "step": 1740 }, { "completion_length": 43.875, "epoch": 0.48315847598012146, "grad_norm": 0.38671875, "kl": 0.01573203131556511, "learning_rate": 8.094975151849808e-06, "loss": 0.0006, "reward": 1.0, "reward_std": 0.0, "rewards/accuracy_reward": 0.0, "rewards/format_reward": 1.0, "step": 1750 }, { "completion_length": 38.725, "epoch": 0.48591938155715075, "grad_norm": 0.3984375, "kl": 0.015004007937386632, "learning_rate": 8.067366096079515e-06, "loss": 0.0006, "reward": 1.0, "reward_std": 0.0, "rewards/accuracy_reward": 0.0, "rewards/format_reward": 1.0, "step": 1760 }, { "completion_length": 51.575, "epoch": 0.48868028713418, "grad_norm": 0.3359375, "kl": 0.012508704885840416, "learning_rate": 8.039757040309222e-06, "loss": 0.0005, "reward": 0.975, "reward_std": 0.05, "rewards/accuracy_reward": 0.0, "rewards/format_reward": 0.975, "step": 1770 }, { "completion_length": 39.95, "epoch": 0.4914411927112093, "grad_norm": 0.294921875, "kl": 0.011702212109230458, "learning_rate": 8.01214798453893e-06, "loss": 0.0005, "reward": 1.0, "reward_std": 0.0, "rewards/accuracy_reward": 0.0, "rewards/format_reward": 1.0, "step": 1780 }, { "completion_length": 52.575, "epoch": 0.4942020982882385, "grad_norm": 0.265625, "kl": 0.011571823060512543, "learning_rate": 7.984538928768637e-06, "loss": 0.0005, "reward": 1.0, "reward_std": 0.0, "rewards/accuracy_reward": 0.0, "rewards/format_reward": 1.0, "step": 1790 }, { "completion_length": 65.675, "epoch": 0.4969630038652678, "grad_norm": 9.1875, "kl": 0.009807331394404173, "learning_rate": 7.956929872998345e-06, "loss": 0.0004, "reward": 1.0, "reward_std": 0.1, "rewards/accuracy_reward": 0.025, "rewards/format_reward": 0.975, "step": 1800 }, { "completion_length": 41.25, "epoch": 0.49972390944229705, "grad_norm": 10.125, "kl": 0.012691410770639778, "learning_rate": 7.929320817228052e-06, "loss": 0.0005, "reward": 1.05, "reward_std": 0.1, "rewards/accuracy_reward": 0.05, "rewards/format_reward": 1.0, "step": 1810 }, { "completion_length": 39.975, "epoch": 0.5024848150193263, "grad_norm": 0.388671875, "kl": 0.012763157673180103, "learning_rate": 7.901711761457758e-06, "loss": 0.0005, "reward": 1.1, "reward_std": 0.0, "rewards/accuracy_reward": 0.1, "rewards/format_reward": 1.0, "step": 1820 }, { "completion_length": 59.4, "epoch": 0.5052457205963556, "grad_norm": 0.2890625, "kl": 0.012727226875722409, "learning_rate": 7.874102705687467e-06, "loss": 0.0005, "reward": 1.075, "reward_std": 0.10773502588272095, "rewards/accuracy_reward": 0.075, "rewards/format_reward": 1.0, "step": 1830 }, { "completion_length": 59.375, "epoch": 0.5080066261733849, "grad_norm": 9.1875, "kl": 0.011757147964090109, "learning_rate": 7.846493649917174e-06, "loss": 0.0005, "reward": 1.075, "reward_std": 0.25, "rewards/accuracy_reward": 0.1, "rewards/format_reward": 0.975, "step": 1840 }, { "completion_length": 42.075, "epoch": 0.5107675317504141, "grad_norm": 0.1552734375, "kl": 0.011294585885480046, "learning_rate": 7.81888459414688e-06, "loss": 0.0005, "reward": 1.0, "reward_std": 0.0, "rewards/accuracy_reward": 0.0, "rewards/format_reward": 1.0, "step": 1850 }, { "completion_length": 51.15, "epoch": 0.5135284373274434, "grad_norm": 11.9375, "kl": 0.011790570057928561, "learning_rate": 7.791275538376589e-06, "loss": 0.0005, "reward": 1.025, "reward_std": 0.10773502588272095, "rewards/accuracy_reward": 0.05, "rewards/format_reward": 0.975, "step": 1860 }, { "completion_length": 44.25, "epoch": 0.5162893429044727, "grad_norm": 1.234375, "kl": 0.023779562208801508, "learning_rate": 7.763666482606295e-06, "loss": 0.001, "reward": 1.075, "reward_std": 0.05, "rewards/accuracy_reward": 0.1, "rewards/format_reward": 0.975, "step": 1870 }, { "completion_length": 44.45, "epoch": 0.5190502484815019, "grad_norm": 0.265625, "kl": 0.018847981560975315, "learning_rate": 7.736057426836004e-06, "loss": 0.0008, "reward": 1.025, "reward_std": 0.05, "rewards/accuracy_reward": 0.025, "rewards/format_reward": 1.0, "step": 1880 }, { "completion_length": 67.5, "epoch": 0.5218111540585312, "grad_norm": 0.423828125, "kl": 0.010858609480783344, "learning_rate": 7.70844837106571e-06, "loss": 0.0004, "reward": 1.025, "reward_std": 0.10773502588272095, "rewards/accuracy_reward": 0.05, "rewards/format_reward": 0.975, "step": 1890 }, { "completion_length": 57.375, "epoch": 0.5245720596355604, "grad_norm": 0.359375, "kl": 0.012408159440383315, "learning_rate": 7.680839315295417e-06, "loss": 0.0005, "reward": 0.9, "reward_std": 0.1, "rewards/accuracy_reward": 0.025, "rewards/format_reward": 0.875, "step": 1900 }, { "completion_length": 41.075, "epoch": 0.5273329652125898, "grad_norm": 11.375, "kl": 0.01915076021105051, "learning_rate": 7.653230259525125e-06, "loss": 0.0008, "reward": 1.1, "reward_std": 0.15773502588272095, "rewards/accuracy_reward": 0.1, "rewards/format_reward": 1.0, "step": 1910 }, { "completion_length": 44.75, "epoch": 0.530093870789619, "grad_norm": 0.134765625, "kl": 0.013074626727029682, "learning_rate": 7.625621203754833e-06, "loss": 0.0005, "reward": 1.0, "reward_std": 0.0, "rewards/accuracy_reward": 0.0, "rewards/format_reward": 1.0, "step": 1920 }, { "completion_length": 47.875, "epoch": 0.5328547763666482, "grad_norm": 0.1728515625, "kl": 0.022233243193477394, "learning_rate": 7.5980121479845395e-06, "loss": 0.0009, "reward": 1.05, "reward_std": 0.1, "rewards/accuracy_reward": 0.05, "rewards/format_reward": 1.0, "step": 1930 }, { "completion_length": 40.375, "epoch": 0.5356156819436775, "grad_norm": 0.33203125, "kl": 0.013019495364278554, "learning_rate": 7.570403092214247e-06, "loss": 0.0005, "reward": 1.1, "reward_std": 0.2, "rewards/accuracy_reward": 0.1, "rewards/format_reward": 1.0, "step": 1940 }, { "completion_length": 50.2, "epoch": 0.5383765875207068, "grad_norm": 0.265625, "kl": 0.013900231663137674, "learning_rate": 7.5427940364439545e-06, "loss": 0.0006, "reward": 1.025, "reward_std": 0.10773502588272095, "rewards/accuracy_reward": 0.05, "rewards/format_reward": 0.975, "step": 1950 }, { "completion_length": 44.625, "epoch": 0.5411374930977361, "grad_norm": 14.125, "kl": 0.015454199723899364, "learning_rate": 7.515184980673662e-06, "loss": 0.0006, "reward": 1.05, "reward_std": 0.1, "rewards/accuracy_reward": 0.05, "rewards/format_reward": 1.0, "step": 1960 }, { "completion_length": 52.725, "epoch": 0.5438983986747653, "grad_norm": 0.1259765625, "kl": 0.012598886806517839, "learning_rate": 7.487575924903369e-06, "loss": 0.0005, "reward": 0.975, "reward_std": 0.05, "rewards/accuracy_reward": 0.0, "rewards/format_reward": 0.975, "step": 1970 }, { "completion_length": 42.975, "epoch": 0.5466593042517945, "grad_norm": 0.1806640625, "kl": 0.01452862734440714, "learning_rate": 7.459966869133076e-06, "loss": 0.0006, "reward": 1.2, "reward_std": 0.1, "rewards/accuracy_reward": 0.2, "rewards/format_reward": 1.0, "step": 1980 }, { "completion_length": 46.725, "epoch": 0.5494202098288239, "grad_norm": 0.2236328125, "kl": 0.011979705933481456, "learning_rate": 7.432357813362784e-06, "loss": 0.0005, "reward": 1.0, "reward_std": 0.0, "rewards/accuracy_reward": 0.0, "rewards/format_reward": 1.0, "step": 1990 }, { "completion_length": 62.225, "epoch": 0.5521811154058531, "grad_norm": 13.8125, "kl": 0.009113723039627075, "learning_rate": 7.404748757592491e-06, "loss": 0.0004, "reward": 1.05, "reward_std": 0.05773502588272095, "rewards/accuracy_reward": 0.05, "rewards/format_reward": 1.0, "step": 2000 }, { "completion_length": 50.425, "epoch": 0.5549420209828824, "grad_norm": 0.2333984375, "kl": 0.012480269372463226, "learning_rate": 7.377139701822198e-06, "loss": 0.0005, "reward": 1.05, "reward_std": 0.05773502588272095, "rewards/accuracy_reward": 0.05, "rewards/format_reward": 1.0, "step": 2010 }, { "completion_length": 45.65, "epoch": 0.5577029265599116, "grad_norm": 0.24609375, "kl": 0.010359317273832857, "learning_rate": 7.3495306460519054e-06, "loss": 0.0004, "reward": 1.0, "reward_std": 0.0, "rewards/accuracy_reward": 0.0, "rewards/format_reward": 1.0, "step": 2020 }, { "completion_length": 54.025, "epoch": 0.560463832136941, "grad_norm": 0.068359375, "kl": 0.010447596153244375, "learning_rate": 7.321921590281613e-06, "loss": 0.0004, "reward": 1.1, "reward_std": 0.15773502588272095, "rewards/accuracy_reward": 0.125, "rewards/format_reward": 0.975, "step": 2030 }, { "completion_length": 51.575, "epoch": 0.5632247377139702, "grad_norm": 0.296875, "kl": 0.01125028906390071, "learning_rate": 7.2943125345113205e-06, "loss": 0.0005, "reward": 1.075, "reward_std": 0.05, "rewards/accuracy_reward": 0.1, "rewards/format_reward": 0.975, "step": 2040 }, { "completion_length": 57.05, "epoch": 0.5659856432909994, "grad_norm": 5.34375, "kl": 0.014221622655168176, "learning_rate": 7.266703478741027e-06, "loss": 0.0006, "reward": 1.075, "reward_std": 0.13164966106414794, "rewards/accuracy_reward": 0.1, "rewards/format_reward": 0.975, "step": 2050 }, { "completion_length": 47.65, "epoch": 0.5687465488680287, "grad_norm": 0.2099609375, "kl": 0.011002461006864906, "learning_rate": 7.239094422970735e-06, "loss": 0.0004, "reward": 1.125, "reward_std": 0.10773502588272095, "rewards/accuracy_reward": 0.125, "rewards/format_reward": 1.0, "step": 2060 }, { "completion_length": 40.75, "epoch": 0.571507454445058, "grad_norm": 3.65625, "kl": 0.02065484719350934, "learning_rate": 7.211485367200442e-06, "loss": 0.0008, "reward": 1.125, "reward_std": 0.16547005176544188, "rewards/accuracy_reward": 0.125, "rewards/format_reward": 1.0, "step": 2070 }, { "completion_length": 54.775, "epoch": 0.5742683600220873, "grad_norm": 9.75, "kl": 0.012140319403260946, "learning_rate": 7.1838763114301505e-06, "loss": 0.0005, "reward": 1.1, "reward_std": 0.2, "rewards/accuracy_reward": 0.1, "rewards/format_reward": 1.0, "step": 2080 }, { "completion_length": 47.2, "epoch": 0.5770292655991165, "grad_norm": 15.3125, "kl": 0.015356668597087264, "learning_rate": 7.156267255659856e-06, "loss": 0.0006, "reward": 1.1, "reward_std": 0.1, "rewards/accuracy_reward": 0.1, "rewards/format_reward": 1.0, "step": 2090 }, { "completion_length": 56.45, "epoch": 0.5797901711761457, "grad_norm": 0.4609375, "kl": 0.014882310247048735, "learning_rate": 7.128658199889565e-06, "loss": 0.0006, "reward": 1.0, "reward_std": 0.1, "rewards/accuracy_reward": 0.025, "rewards/format_reward": 0.975, "step": 2100 }, { "completion_length": 51.45, "epoch": 0.5825510767531751, "grad_norm": 12.1875, "kl": 0.013292353972792626, "learning_rate": 7.101049144119272e-06, "loss": 0.0005, "reward": 1.05, "reward_std": 0.05773502588272095, "rewards/accuracy_reward": 0.05, "rewards/format_reward": 1.0, "step": 2110 }, { "completion_length": 34.5, "epoch": 0.5853119823302043, "grad_norm": 0.2373046875, "kl": 0.013515590690076351, "learning_rate": 7.07344008834898e-06, "loss": 0.0005, "reward": 1.05, "reward_std": 0.1, "rewards/accuracy_reward": 0.05, "rewards/format_reward": 1.0, "step": 2120 }, { "completion_length": 45.725, "epoch": 0.5880728879072336, "grad_norm": 0.1611328125, "kl": 0.01667518252506852, "learning_rate": 7.045831032578686e-06, "loss": 0.0007, "reward": 1.025, "reward_std": 0.05, "rewards/accuracy_reward": 0.025, "rewards/format_reward": 1.0, "step": 2130 }, { "completion_length": 55.575, "epoch": 0.5908337934842628, "grad_norm": 0.75, "kl": 0.016457650857046247, "learning_rate": 7.018221976808394e-06, "loss": 0.0007, "reward": 1.0, "reward_std": 0.0, "rewards/accuracy_reward": 0.0, "rewards/format_reward": 1.0, "step": 2140 }, { "completion_length": 43.8, "epoch": 0.5935946990612921, "grad_norm": 0.2255859375, "kl": 0.012675747368484735, "learning_rate": 6.9906129210381015e-06, "loss": 0.0005, "reward": 1.1, "reward_std": 0.1154700517654419, "rewards/accuracy_reward": 0.1, "rewards/format_reward": 1.0, "step": 2150 }, { "completion_length": 50.85, "epoch": 0.5963556046383214, "grad_norm": 0.4296875, "kl": 0.013116476451978087, "learning_rate": 6.963003865267809e-06, "loss": 0.0005, "reward": 0.975, "reward_std": 0.05, "rewards/accuracy_reward": 0.0, "rewards/format_reward": 0.975, "step": 2160 }, { "completion_length": 67.65, "epoch": 0.5991165102153506, "grad_norm": 0.197265625, "kl": 0.011186759034171701, "learning_rate": 6.935394809497516e-06, "loss": 0.0004, "reward": 0.95, "reward_std": 0.14574271440505981, "rewards/accuracy_reward": 0.025, "rewards/format_reward": 0.925, "step": 2170 }, { "completion_length": 66.6, "epoch": 0.6018774157923799, "grad_norm": 0.259765625, "kl": 0.014577974379062653, "learning_rate": 6.907785753727223e-06, "loss": 0.0006, "reward": 1.0, "reward_std": 0.0, "rewards/accuracy_reward": 0.0, "rewards/format_reward": 1.0, "step": 2180 }, { "completion_length": 47.9, "epoch": 0.6046383213694092, "grad_norm": 0.2021484375, "kl": 0.012043908704072237, "learning_rate": 6.880176697956931e-06, "loss": 0.0005, "reward": 1.0, "reward_std": 0.0, "rewards/accuracy_reward": 0.0, "rewards/format_reward": 1.0, "step": 2190 }, { "completion_length": 52.125, "epoch": 0.6073992269464384, "grad_norm": 0.1630859375, "kl": 0.008721905946731567, "learning_rate": 6.852567642186638e-06, "loss": 0.0003, "reward": 1.075, "reward_std": 0.05, "rewards/accuracy_reward": 0.075, "rewards/format_reward": 1.0, "step": 2200 }, { "completion_length": 46.925, "epoch": 0.6101601325234677, "grad_norm": 0.31640625, "kl": 0.010289201140403747, "learning_rate": 6.824958586416345e-06, "loss": 0.0004, "reward": 1.075, "reward_std": 0.05, "rewards/accuracy_reward": 0.075, "rewards/format_reward": 1.0, "step": 2210 }, { "completion_length": 48.95, "epoch": 0.6129210381004969, "grad_norm": 0.2421875, "kl": 0.01192173978779465, "learning_rate": 6.797349530646052e-06, "loss": 0.0005, "reward": 1.05, "reward_std": 0.1, "rewards/accuracy_reward": 0.05, "rewards/format_reward": 1.0, "step": 2220 }, { "completion_length": 58.65, "epoch": 0.6156819436775263, "grad_norm": 12.125, "kl": 0.009246612852439284, "learning_rate": 6.76974047487576e-06, "loss": 0.0004, "reward": 1.025, "reward_std": 0.15, "rewards/accuracy_reward": 0.05, "rewards/format_reward": 0.975, "step": 2230 }, { "completion_length": 42.225, "epoch": 0.6184428492545555, "grad_norm": 0.44140625, "kl": 0.016198016284033657, "learning_rate": 6.742131419105467e-06, "loss": 0.0006, "reward": 1.075, "reward_std": 0.10773502588272095, "rewards/accuracy_reward": 0.075, "rewards/format_reward": 1.0, "step": 2240 }, { "completion_length": 64.425, "epoch": 0.6212037548315847, "grad_norm": 0.1640625, "kl": 0.008573147933930159, "learning_rate": 6.714522363335174e-06, "loss": 0.0003, "reward": 0.975, "reward_std": 0.05, "rewards/accuracy_reward": 0.0, "rewards/format_reward": 0.975, "step": 2250 }, { "completion_length": 42.85, "epoch": 0.623964660408614, "grad_norm": 0.2431640625, "kl": 0.018018354568630456, "learning_rate": 6.686913307564882e-06, "loss": 0.0007, "reward": 1.05, "reward_std": 0.05773502588272095, "rewards/accuracy_reward": 0.05, "rewards/format_reward": 1.0, "step": 2260 }, { "completion_length": 44.275, "epoch": 0.6267255659856433, "grad_norm": 0.2470703125, "kl": 0.01138524515554309, "learning_rate": 6.659304251794589e-06, "loss": 0.0005, "reward": 1.0, "reward_std": 0.0, "rewards/accuracy_reward": 0.0, "rewards/format_reward": 1.0, "step": 2270 }, { "completion_length": 56.425, "epoch": 0.6294864715626726, "grad_norm": 0.166015625, "kl": 0.007961477618664503, "learning_rate": 6.631695196024297e-06, "loss": 0.0003, "reward": 1.05, "reward_std": 0.05773502588272095, "rewards/accuracy_reward": 0.05, "rewards/format_reward": 1.0, "step": 2280 }, { "completion_length": 40.1, "epoch": 0.6322473771397018, "grad_norm": 0.1474609375, "kl": 0.011783124506473541, "learning_rate": 6.604086140254003e-06, "loss": 0.0005, "reward": 1.0, "reward_std": 0.0, "rewards/accuracy_reward": 0.0, "rewards/format_reward": 1.0, "step": 2290 }, { "completion_length": 40.425, "epoch": 0.635008282716731, "grad_norm": 12.0, "kl": 0.010805513197556137, "learning_rate": 6.576477084483711e-06, "loss": 0.0004, "reward": 1.05, "reward_std": 0.05773502588272095, "rewards/accuracy_reward": 0.05, "rewards/format_reward": 1.0, "step": 2300 }, { "completion_length": 50.275, "epoch": 0.6377691882937604, "grad_norm": 0.484375, "kl": 0.012207349389791488, "learning_rate": 6.548868028713419e-06, "loss": 0.0005, "reward": 1.025, "reward_std": 0.15, "rewards/accuracy_reward": 0.05, "rewards/format_reward": 0.975, "step": 2310 }, { "completion_length": 49.575, "epoch": 0.6405300938707896, "grad_norm": 0.11767578125, "kl": 0.014282464934512972, "learning_rate": 6.521258972943127e-06, "loss": 0.0006, "reward": 1.0, "reward_std": 0.0, "rewards/accuracy_reward": 0.0, "rewards/format_reward": 1.0, "step": 2320 }, { "completion_length": 50.975, "epoch": 0.6432909994478189, "grad_norm": 0.23828125, "kl": 0.010307052405551076, "learning_rate": 6.4936499171728325e-06, "loss": 0.0004, "reward": 0.975, "reward_std": 0.05, "rewards/accuracy_reward": 0.0, "rewards/format_reward": 0.975, "step": 2330 }, { "completion_length": 43.45, "epoch": 0.6460519050248481, "grad_norm": 0.283203125, "kl": 0.009153263340704144, "learning_rate": 6.466040861402541e-06, "loss": 0.0004, "reward": 1.05, "reward_std": 0.1, "rewards/accuracy_reward": 0.05, "rewards/format_reward": 1.0, "step": 2340 }, { "completion_length": 42.45, "epoch": 0.6488128106018775, "grad_norm": 0.1669921875, "kl": 0.012787223979830741, "learning_rate": 6.438431805632248e-06, "loss": 0.0005, "reward": 1.025, "reward_std": 0.05, "rewards/accuracy_reward": 0.025, "rewards/format_reward": 1.0, "step": 2350 }, { "completion_length": 49.15, "epoch": 0.6515737161789067, "grad_norm": 0.306640625, "kl": 0.011894195433706045, "learning_rate": 6.410822749861956e-06, "loss": 0.0005, "reward": 1.05, "reward_std": 0.1, "rewards/accuracy_reward": 0.05, "rewards/format_reward": 1.0, "step": 2360 }, { "completion_length": 40.175, "epoch": 0.6543346217559359, "grad_norm": 0.2177734375, "kl": 0.014734916668385268, "learning_rate": 6.3832136940916626e-06, "loss": 0.0006, "reward": 1.0, "reward_std": 0.0, "rewards/accuracy_reward": 0.0, "rewards/format_reward": 1.0, "step": 2370 }, { "completion_length": 45.1, "epoch": 0.6570955273329652, "grad_norm": 0.27734375, "kl": 0.01835952061228454, "learning_rate": 6.35560463832137e-06, "loss": 0.0007, "reward": 1.0, "reward_std": 0.0, "rewards/accuracy_reward": 0.0, "rewards/format_reward": 1.0, "step": 2380 }, { "completion_length": 49.15, "epoch": 0.6598564329099945, "grad_norm": 16.625, "kl": 0.011911587277427315, "learning_rate": 6.327995582551078e-06, "loss": 0.0005, "reward": 1.25, "reward_std": 0.25773502588272096, "rewards/accuracy_reward": 0.275, "rewards/format_reward": 0.975, "step": 2390 }, { "completion_length": 56.825, "epoch": 0.6626173384870238, "grad_norm": 0.1357421875, "kl": 0.006976808630861342, "learning_rate": 6.300386526780785e-06, "loss": 0.0003, "reward": 1.0, "reward_std": 0.0, "rewards/accuracy_reward": 0.0, "rewards/format_reward": 1.0, "step": 2400 }, { "completion_length": 60.475, "epoch": 0.665378244064053, "grad_norm": 0.2353515625, "kl": 0.010746273957192898, "learning_rate": 6.272777471010492e-06, "loss": 0.0004, "reward": 1.0, "reward_std": 0.0, "rewards/accuracy_reward": 0.0, "rewards/format_reward": 1.0, "step": 2410 }, { "completion_length": 39.75, "epoch": 0.6681391496410822, "grad_norm": 0.37109375, "kl": 0.01077571688219905, "learning_rate": 6.245168415240199e-06, "loss": 0.0004, "reward": 1.0, "reward_std": 0.0, "rewards/accuracy_reward": 0.0, "rewards/format_reward": 1.0, "step": 2420 }, { "completion_length": 49.2, "epoch": 0.6709000552181116, "grad_norm": 0.27734375, "kl": 0.010110868490301073, "learning_rate": 6.217559359469907e-06, "loss": 0.0004, "reward": 1.0, "reward_std": 0.0, "rewards/accuracy_reward": 0.0, "rewards/format_reward": 1.0, "step": 2430 }, { "completion_length": 58.1, "epoch": 0.6736609607951408, "grad_norm": 11.125, "kl": 0.007148551405407488, "learning_rate": 6.189950303699614e-06, "loss": 0.0003, "reward": 1.025, "reward_std": 0.05, "rewards/accuracy_reward": 0.025, "rewards/format_reward": 1.0, "step": 2440 } ], "logging_steps": 10, "max_steps": 3622, "num_input_tokens_seen": 0, "num_train_epochs": 1, "save_steps": 10, "stateful_callbacks": { "TrainerControl": { "args": { "should_epoch_stop": false, "should_evaluate": false, "should_log": false, "should_save": true, "should_training_stop": false }, "attributes": {} } }, "total_flos": 0.0, "train_batch_size": 4, "trial_name": null, "trial_params": null }