{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 0.9965010496850945,
  "eval_steps": 100,
  "global_step": 178,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "completion_length": 621.5997104644775,
      "epoch": 0.005598320503848845,
      "grad_norm": 0.018829375505447388,
      "learning_rate": 1e-06,
      "loss": 0.0,
      "reward": 0.26314485468901694,
      "reward_std": 0.09563918481580913,
      "rewards/semantic_entropy_math_reward": 0.26314485468901694,
      "step": 1
    },
    {
      "completion_length": 600.1235256195068,
      "epoch": 0.01119664100769769,
      "grad_norm": 0.021261176094412804,
      "learning_rate": 1e-06,
      "loss": 0.0,
      "reward": 0.2620287789031863,
      "reward_std": 0.09228823287412524,
      "rewards/semantic_entropy_math_reward": 0.2620287789031863,
      "step": 2
    },
    {
      "completion_length": 640.598970413208,
      "epoch": 0.016794961511546535,
      "grad_norm": 0.0211239755153656,
      "learning_rate": 1e-06,
      "loss": 0.0,
      "reward": 0.2310267963912338,
      "reward_std": 0.08434095047414303,
      "rewards/semantic_entropy_math_reward": 0.2310267963912338,
      "step": 3
    },
    {
      "completion_length": 643.0141410827637,
      "epoch": 0.02239328201539538,
      "grad_norm": 0.021212320774793625,
      "learning_rate": 1e-06,
      "loss": 0.0,
      "reward": 0.22259425604715943,
      "reward_std": 0.09978899103589356,
      "rewards/semantic_entropy_math_reward": 0.22259425604715943,
      "step": 4
    },
    {
      "completion_length": 623.704626083374,
      "epoch": 0.02799160251924423,
      "grad_norm": 0.01992746628820896,
      "learning_rate": 1e-06,
      "loss": 0.0,
      "reward": 0.3144841371104121,
      "reward_std": 0.09517159080132842,
      "rewards/semantic_entropy_math_reward": 0.3144841371104121,
      "step": 5
    },
    {
      "completion_length": 602.6532821655273,
      "epoch": 0.03358992302309307,
      "grad_norm": 0.027432788163423538,
      "learning_rate": 1e-06,
      "loss": 0.0,
      "reward": 0.3053075410425663,
      "reward_std": 0.11108083021827042,
      "rewards/semantic_entropy_math_reward": 0.3053075410425663,
      "step": 6
    },
    {
      "completion_length": 632.2589473724365,
      "epoch": 0.03918824352694192,
      "grad_norm": 0.024943526834249496,
      "learning_rate": 1e-06,
      "loss": 0.0,
      "reward": 0.2508680587634444,
      "reward_std": 0.06983336608391255,
      "rewards/semantic_entropy_math_reward": 0.2508680587634444,
      "step": 7
    },
    {
      "completion_length": 578.0238265991211,
      "epoch": 0.04478656403079076,
      "grad_norm": 0.030510740354657173,
      "learning_rate": 1e-06,
      "loss": 0.0,
      "reward": 0.3060516007244587,
      "reward_std": 0.09902600082568824,
      "rewards/semantic_entropy_math_reward": 0.3060516007244587,
      "step": 8
    },
    {
      "completion_length": 656.7507629394531,
      "epoch": 0.05038488453463961,
      "grad_norm": 0.03022916242480278,
      "learning_rate": 1e-06,
      "loss": 0.0,
      "reward": 0.2726934589445591,
      "reward_std": 0.08185443677939475,
      "rewards/semantic_entropy_math_reward": 0.2726934589445591,
      "step": 9
    },
    {
      "completion_length": 687.5104351043701,
      "epoch": 0.05598320503848846,
      "grad_norm": 0.03694581612944603,
      "learning_rate": 1e-06,
      "loss": 0.0,
      "reward": 0.24987599975429475,
      "reward_std": 0.08242391713429242,
      "rewards/semantic_entropy_math_reward": 0.24987599975429475,
      "step": 10
    },
    {
      "completion_length": 726.5305213928223,
      "epoch": 0.0615815255423373,
      "grad_norm": 0.033508434891700745,
      "learning_rate": 1e-06,
      "loss": 0.0,
      "reward": 0.21453373739495873,
      "reward_std": 0.08386916620656848,
      "rewards/semantic_entropy_math_reward": 0.21453373739495873,
      "step": 11
    },
    {
      "completion_length": 744.8936157226562,
      "epoch": 0.06717984604618614,
      "grad_norm": 0.030294176191091537,
      "learning_rate": 1e-06,
      "loss": 0.0,
      "reward": 0.196924609830603,
      "reward_std": 0.08265715674497187,
      "rewards/semantic_entropy_math_reward": 0.196924609830603,
      "step": 12
    },
    {
      "completion_length": 744.174861907959,
      "epoch": 0.072778166550035,
      "grad_norm": 0.03118356317281723,
      "learning_rate": 1e-06,
      "loss": 0.0,
      "reward": 0.19890873844269663,
      "reward_std": 0.08003655297216028,
      "rewards/semantic_entropy_math_reward": 0.19890873844269663,
      "step": 13
    },
    {
      "completion_length": 765.6265029907227,
      "epoch": 0.07837648705388384,
      "grad_norm": 0.02831243723630905,
      "learning_rate": 1e-06,
      "loss": 0.0,
      "reward": 0.18092759023420513,
      "reward_std": 0.06603722460567951,
      "rewards/semantic_entropy_math_reward": 0.18092759023420513,
      "step": 14
    },
    {
      "completion_length": 758.721736907959,
      "epoch": 0.08397480755773268,
      "grad_norm": 0.03191084787249565,
      "learning_rate": 1e-06,
      "loss": 0.0,
      "reward": 0.18340774439275265,
      "reward_std": 0.07208729872945696,
      "rewards/semantic_entropy_math_reward": 0.18340774439275265,
      "step": 15
    },
    {
      "completion_length": 662.6317081451416,
      "epoch": 0.08957312806158152,
      "grad_norm": 0.03559156879782677,
      "learning_rate": 1e-06,
      "loss": 0.0,
      "reward": 0.26698909467086196,
      "reward_std": 0.10032190917991102,
      "rewards/semantic_entropy_math_reward": 0.26698909467086196,
      "step": 16
    },
    {
      "completion_length": 690.2068557739258,
      "epoch": 0.09517144856543037,
      "grad_norm": 0.040907151997089386,
      "learning_rate": 1e-06,
      "loss": 0.0,
      "reward": 0.22792659618426114,
      "reward_std": 0.07990038132993504,
      "rewards/semantic_entropy_math_reward": 0.22792659618426114,
      "step": 17
    },
    {
      "completion_length": 714.0669784545898,
      "epoch": 0.10076976906927922,
      "grad_norm": 0.07471004873514175,
      "learning_rate": 1e-06,
      "loss": 0.0,
      "reward": 0.21217758627608418,
      "reward_std": 0.09499240922741592,
      "rewards/semantic_entropy_math_reward": 0.21217758627608418,
      "step": 18
    },
    {
      "completion_length": 671.5535850524902,
      "epoch": 0.10636808957312806,
      "grad_norm": 0.07692024111747742,
      "learning_rate": 1e-06,
      "loss": 0.0,
      "reward": 0.21106151002459228,
      "reward_std": 0.09371245314832777,
      "rewards/semantic_entropy_math_reward": 0.21106151002459228,
      "step": 19
    },
    {
      "completion_length": 599.3058185577393,
      "epoch": 0.11196641007697691,
      "grad_norm": 0.09223277866840363,
      "learning_rate": 1e-06,
      "loss": 0.0,
      "reward": 0.2771577457897365,
      "reward_std": 0.10236490913666785,
      "rewards/semantic_entropy_math_reward": 0.2771577457897365,
      "step": 20
    },
    {
      "completion_length": 512.9620628356934,
      "epoch": 0.11756473058082575,
      "grad_norm": 0.13926267623901367,
      "learning_rate": 1e-06,
      "loss": 0.0,
      "reward": 0.34102183766663074,
      "reward_std": 0.1208265540190041,
      "rewards/semantic_entropy_math_reward": 0.34102183766663074,
      "step": 21
    },
    {
      "completion_length": 479.3913764953613,
      "epoch": 0.1231630510846746,
      "grad_norm": 0.3304852247238159,
      "learning_rate": 1e-06,
      "loss": 0.0,
      "reward": 0.2891865149140358,
      "reward_std": 0.11506026284769177,
      "rewards/semantic_entropy_math_reward": 0.2891865149140358,
      "step": 22
    },
    {
      "completion_length": 389.78051376342773,
      "epoch": 0.12876137158852344,
      "grad_norm": 0.6791955232620239,
      "learning_rate": 1e-06,
      "loss": 0.0,
      "reward": 0.3836805671453476,
      "reward_std": 0.12086579436436296,
      "rewards/semantic_entropy_math_reward": 0.3836805671453476,
      "step": 23
    },
    {
      "completion_length": 364.2626552581787,
      "epoch": 0.13435969209237228,
      "grad_norm": 1.3652504682540894,
      "learning_rate": 1e-06,
      "loss": 0.0,
      "reward": 0.32800100184977055,
      "reward_std": 0.11237839260138571,
      "rewards/semantic_entropy_math_reward": 0.32800100184977055,
      "step": 24
    },
    {
      "completion_length": 343.3311061859131,
      "epoch": 0.13995801259622112,
      "grad_norm": 1.9864494800567627,
      "learning_rate": 1e-06,
      "loss": 0.0,
      "reward": 0.27318949438631535,
      "reward_std": 0.10484480275772512,
      "rewards/semantic_entropy_math_reward": 0.27318949438631535,
      "step": 25
    },
    {
      "completion_length": 363.51934814453125,
      "epoch": 0.14555633310007,
      "grad_norm": 5.861116886138916,
      "learning_rate": 1e-06,
      "loss": 0.0,
      "reward": 0.18601191136986017,
      "reward_std": 0.07814609340857714,
      "rewards/semantic_entropy_math_reward": 0.18601191136986017,
      "step": 26
    },
    {
      "completion_length": 637.8519458770752,
      "epoch": 0.15115465360391883,
      "grad_norm": 7.510809898376465,
      "learning_rate": 1e-06,
      "loss": 0.0,
      "reward": 0.07031250384170562,
      "reward_std": 0.03756278438959271,
      "rewards/semantic_entropy_math_reward": 0.07031250384170562,
      "step": 27
    },
    {
      "completion_length": 925.8988342285156,
      "epoch": 0.15675297410776767,
      "grad_norm": 3.905956745147705,
      "learning_rate": 1e-06,
      "loss": 0.0,
      "reward": 0.0363343273056671,
      "reward_std": 0.018063054973026738,
      "rewards/semantic_entropy_math_reward": 0.0363343273056671,
      "step": 28
    },
    {
      "completion_length": 894.2656402587891,
      "epoch": 0.16235129461161651,
      "grad_norm": 3.104583501815796,
      "learning_rate": 1e-06,
      "loss": 0.0,
      "reward": 0.08692956482991576,
      "reward_std": 0.043345644196961075,
      "rewards/semantic_entropy_math_reward": 0.08692956482991576,
      "step": 29
    },
    {
      "completion_length": 774.4866256713867,
      "epoch": 0.16794961511546536,
      "grad_norm": 17.407346725463867,
      "learning_rate": 1e-06,
      "loss": 0.0,
      "reward": 0.13442460726946592,
      "reward_std": 0.07532014499884099,
      "rewards/semantic_entropy_math_reward": 0.13442460726946592,
      "step": 30
    },
    {
      "completion_length": 434.0379524230957,
      "epoch": 0.1735479356193142,
      "grad_norm": 7.19148588180542,
      "learning_rate": 1e-06,
      "loss": 0.0,
      "reward": 0.2876984179019928,
      "reward_std": 0.108892708318308,
      "rewards/semantic_entropy_math_reward": 0.2876984179019928,
      "step": 31
    },
    {
      "completion_length": 405.40477180480957,
      "epoch": 0.17914625612316304,
      "grad_norm": 1.139600157737732,
      "learning_rate": 1e-06,
      "loss": 0.0,
      "reward": 0.34771826211363077,
      "reward_std": 0.13636958738788962,
      "rewards/semantic_entropy_math_reward": 0.34771826211363077,
      "step": 32
    },
    {
      "completion_length": 463.8638458251953,
      "epoch": 0.1847445766270119,
      "grad_norm": 1.6658934354782104,
      "learning_rate": 1e-06,
      "loss": 0.0,
      "reward": 0.35949901584535837,
      "reward_std": 0.12283765664324164,
      "rewards/semantic_entropy_math_reward": 0.35949901584535837,
      "step": 33
    },
    {
      "completion_length": 486.4196491241455,
      "epoch": 0.19034289713086075,
      "grad_norm": 1.7289444208145142,
      "learning_rate": 1e-06,
      "loss": 0.0,
      "reward": 0.3540426651015878,
      "reward_std": 0.10729875811375678,
      "rewards/semantic_entropy_math_reward": 0.3540426651015878,
      "step": 34
    },
    {
      "completion_length": 529.6004600524902,
      "epoch": 0.1959412176347096,
      "grad_norm": 2.8688220977783203,
      "learning_rate": 1e-06,
      "loss": 0.0,
      "reward": 0.3628472303971648,
      "reward_std": 0.11101956386119127,
      "rewards/semantic_entropy_math_reward": 0.3628472303971648,
      "step": 35
    },
    {
      "completion_length": 517.907751083374,
      "epoch": 0.20153953813855843,
      "grad_norm": 2.089806318283081,
      "learning_rate": 1e-06,
      "loss": 0.0,
      "reward": 0.3950892863795161,
      "reward_std": 0.11218473897315562,
      "rewards/semantic_entropy_math_reward": 0.3950892863795161,
      "step": 36
    },
    {
      "completion_length": 558.5573024749756,
      "epoch": 0.20713785864240727,
      "grad_norm": 11.361101150512695,
      "learning_rate": 1e-06,
      "loss": 0.0,
      "reward": 0.3583829468116164,
      "reward_std": 0.11303460272029042,
      "rewards/semantic_entropy_math_reward": 0.3583829468116164,
      "step": 37
    },
    {
      "completion_length": 640.2663822174072,
      "epoch": 0.21273617914625612,
      "grad_norm": 9.671446800231934,
      "learning_rate": 1e-06,
      "loss": 0.0,
      "reward": 0.30580358393490314,
      "reward_std": 0.09917579509783536,
      "rewards/semantic_entropy_math_reward": 0.30580358393490314,
      "step": 38
    },
    {
      "completion_length": 656.6250057220459,
      "epoch": 0.21833449965010496,
      "grad_norm": 4.101109981536865,
      "learning_rate": 1e-06,
      "loss": 0.0,
      "reward": 0.25942460889928043,
      "reward_std": 0.08745632297359407,
      "rewards/semantic_entropy_math_reward": 0.25942460889928043,
      "step": 39
    },
    {
      "completion_length": 694.5468864440918,
      "epoch": 0.22393282015395383,
      "grad_norm": 1.9280970096588135,
      "learning_rate": 1e-06,
      "loss": 0.0,
      "reward": 0.21515377657487988,
      "reward_std": 0.10014541284181178,
      "rewards/semantic_entropy_math_reward": 0.21515377657487988,
      "step": 40
    },
    {
      "completion_length": 682.7931613922119,
      "epoch": 0.22953114065780267,
      "grad_norm": 1.6009750366210938,
      "learning_rate": 1e-06,
      "loss": 0.0,
      "reward": 0.24652778403833508,
      "reward_std": 0.09196118649560958,
      "rewards/semantic_entropy_math_reward": 0.24652778403833508,
      "step": 41
    },
    {
      "completion_length": 691.9077568054199,
      "epoch": 0.2351294611616515,
      "grad_norm": 0.7951090335845947,
      "learning_rate": 1e-06,
      "loss": 0.0,
      "reward": 0.24776786682195961,
      "reward_std": 0.09380459249950945,
      "rewards/semantic_entropy_math_reward": 0.24776786682195961,
      "step": 42
    },
    {
      "completion_length": 655.6860256195068,
      "epoch": 0.24072778166550035,
      "grad_norm": 0.602069079875946,
      "learning_rate": 1e-06,
      "loss": 0.0,
      "reward": 0.25818453216925263,
      "reward_std": 0.08527852385304868,
      "rewards/semantic_entropy_math_reward": 0.25818453216925263,
      "step": 43
    },
    {
      "completion_length": 615.9233703613281,
      "epoch": 0.2463261021693492,
      "grad_norm": 4.384374618530273,
      "learning_rate": 1e-06,
      "loss": 0.0,
      "reward": 0.3162202490493655,
      "reward_std": 0.09647805755957961,
      "rewards/semantic_entropy_math_reward": 0.3162202490493655,
      "step": 44
    },
    {
      "completion_length": 572.6949520111084,
      "epoch": 0.25192442267319803,
      "grad_norm": 3.9872536659240723,
      "learning_rate": 1e-06,
      "loss": 0.0,
      "reward": 0.3339533871039748,
      "reward_std": 0.10575204784981906,
      "rewards/semantic_entropy_math_reward": 0.3339533871039748,
      "step": 45
    },
    {
      "completion_length": 504.4799213409424,
      "epoch": 0.2575227431770469,
      "grad_norm": 0.722204327583313,
      "learning_rate": 1e-06,
      "loss": 0.0,
      "reward": 0.3819444477558136,
      "reward_std": 0.12700667465105653,
      "rewards/semantic_entropy_math_reward": 0.3819444477558136,
      "step": 46
    },
    {
      "completion_length": 444.5267906188965,
      "epoch": 0.2631210636808957,
      "grad_norm": 1.130468487739563,
      "learning_rate": 1e-06,
      "loss": 0.0,
      "reward": 0.37996032927185297,
      "reward_std": 0.11253955401480198,
      "rewards/semantic_entropy_math_reward": 0.37996032927185297,
      "step": 47
    },
    {
      "completion_length": 446.88840103149414,
      "epoch": 0.26871938418474456,
      "grad_norm": 0.16186973452568054,
      "learning_rate": 1e-06,
      "loss": 0.0,
      "reward": 0.38616071827709675,
      "reward_std": 0.12775122001767159,
      "rewards/semantic_entropy_math_reward": 0.38616071827709675,
      "step": 48
    },
    {
      "completion_length": 465.04167556762695,
      "epoch": 0.2743177046885934,
      "grad_norm": 0.16776132583618164,
      "learning_rate": 1e-06,
      "loss": 0.0,
      "reward": 0.33407739456743,
      "reward_std": 0.08897499740123749,
      "rewards/semantic_entropy_math_reward": 0.33407739456743,
      "step": 49
    },
    {
      "completion_length": 447.424861907959,
      "epoch": 0.27991602519244224,
      "grad_norm": 0.3498786687850952,
      "learning_rate": 1e-06,
      "loss": 0.0,
      "reward": 0.3691716268658638,
      "reward_std": 0.11830427148379385,
      "rewards/semantic_entropy_math_reward": 0.3691716268658638,
      "step": 50
    },
    {
      "completion_length": 418.9003047943115,
      "epoch": 0.28551434569629114,
      "grad_norm": 0.4594447612762451,
      "learning_rate": 1e-06,
      "loss": 0.0,
      "reward": 0.36669147573411465,
      "reward_std": 0.11925438744947314,
      "rewards/semantic_entropy_math_reward": 0.36669147573411465,
      "step": 51
    },
    {
      "completion_length": 423.0595302581787,
      "epoch": 0.29111266620014,
      "grad_norm": 1.4399406909942627,
      "learning_rate": 1e-06,
      "loss": 0.0,
      "reward": 0.36941965389996767,
      "reward_std": 0.11310782423242927,
      "rewards/semantic_entropy_math_reward": 0.36941965389996767,
      "step": 52
    },
    {
      "completion_length": 548.479923248291,
      "epoch": 0.2967109867039888,
      "grad_norm": 46.641822814941406,
      "learning_rate": 1e-06,
      "loss": 0.0,
      "reward": 0.36706350184977055,
      "reward_std": 0.10590148437768221,
      "rewards/semantic_entropy_math_reward": 0.36706350184977055,
      "step": 53
    },
    {
      "completion_length": 583.7537307739258,
      "epoch": 0.30230930720783766,
      "grad_norm": 3.891123056411743,
      "learning_rate": 1e-06,
      "loss": 0.0,
      "reward": 0.3219246044754982,
      "reward_std": 0.11164773069322109,
      "rewards/semantic_entropy_math_reward": 0.3219246044754982,
      "step": 54
    },
    {
      "completion_length": 569.5848350524902,
      "epoch": 0.3079076277116865,
      "grad_norm": 5.51741886138916,
      "learning_rate": 1e-06,
      "loss": 0.0,
      "reward": 0.3412698497995734,
      "reward_std": 0.11707511683925986,
      "rewards/semantic_entropy_math_reward": 0.3412698497995734,
      "step": 55
    },
    {
      "completion_length": 589.2455463409424,
      "epoch": 0.31350594821553535,
      "grad_norm": 2.6841273307800293,
      "learning_rate": 1e-06,
      "loss": 0.0,
      "reward": 0.33382937777787447,
      "reward_std": 0.109538255026564,
      "rewards/semantic_entropy_math_reward": 0.33382937777787447,
      "step": 56
    },
    {
      "completion_length": 536.4256038665771,
      "epoch": 0.3191042687193842,
      "grad_norm": 17.737791061401367,
      "learning_rate": 1e-06,
      "loss": 0.0,
      "reward": 0.4088541707023978,
      "reward_std": 0.12808961933478713,
      "rewards/semantic_entropy_math_reward": 0.4088541707023978,
      "step": 57
    },
    {
      "completion_length": 424.6919708251953,
      "epoch": 0.32470258922323303,
      "grad_norm": 1.9888161420822144,
      "learning_rate": 1e-06,
      "loss": 0.0,
      "reward": 0.3665674636140466,
      "reward_std": 0.12649336433969438,
      "rewards/semantic_entropy_math_reward": 0.3665674636140466,
      "step": 58
    },
    {
      "completion_length": 412.2708396911621,
      "epoch": 0.33030090972708187,
      "grad_norm": 0.3387698531150818,
      "learning_rate": 1e-06,
      "loss": 0.0,
      "reward": 0.3368055559694767,
      "reward_std": 0.1261643674224615,
      "rewards/semantic_entropy_math_reward": 0.3368055559694767,
      "step": 59
    },
    {
      "completion_length": 390.2321529388428,
      "epoch": 0.3358992302309307,
      "grad_norm": 0.5467884540557861,
      "learning_rate": 1e-06,
      "loss": 0.0,
      "reward": 0.4055059552192688,
      "reward_std": 0.13236648589372635,
      "rewards/semantic_entropy_math_reward": 0.4055059552192688,
      "step": 60
    },
    {
      "completion_length": 407.4642925262451,
      "epoch": 0.34149755073477955,
      "grad_norm": 0.5977727770805359,
      "learning_rate": 1e-06,
      "loss": 0.0,
      "reward": 0.385416685603559,
      "reward_std": 0.12455174163915217,
      "rewards/semantic_entropy_math_reward": 0.385416685603559,
      "step": 61
    },
    {
      "completion_length": 368.8139953613281,
      "epoch": 0.3470958712386284,
      "grad_norm": 0.7032583355903625,
      "learning_rate": 1e-06,
      "loss": 0.0,
      "reward": 0.3903769990429282,
      "reward_std": 0.14396332437172532,
      "rewards/semantic_entropy_math_reward": 0.3903769990429282,
      "step": 62
    },
    {
      "completion_length": 399.3407802581787,
      "epoch": 0.35269419174247724,
      "grad_norm": 1.6613659858703613,
      "learning_rate": 1e-06,
      "loss": 0.0,
      "reward": 0.3821924654766917,
      "reward_std": 0.13627444114536047,
      "rewards/semantic_entropy_math_reward": 0.3821924654766917,
      "step": 63
    },
    {
      "completion_length": 386.2135524749756,
      "epoch": 0.3582925122463261,
      "grad_norm": 0.297186940908432,
      "learning_rate": 1e-06,
      "loss": 0.0,
      "reward": 0.4055059542879462,
      "reward_std": 0.1174642383120954,
      "rewards/semantic_entropy_math_reward": 0.4055059542879462,
      "step": 64
    },
    {
      "completion_length": 414.6153335571289,
      "epoch": 0.363890832750175,
      "grad_norm": 0.2781192660331726,
      "learning_rate": 1e-06,
      "loss": 0.0,
      "reward": 0.37723215017467737,
      "reward_std": 0.1255739361513406,
      "rewards/semantic_entropy_math_reward": 0.37723215017467737,
      "step": 65
    },
    {
      "completion_length": 402.4501533508301,
      "epoch": 0.3694891532540238,
      "grad_norm": 0.2593112587928772,
      "learning_rate": 1e-06,
      "loss": 0.0,
      "reward": 0.3885168796405196,
      "reward_std": 0.12460076110437512,
      "rewards/semantic_entropy_math_reward": 0.3885168796405196,
      "step": 66
    },
    {
      "completion_length": 417.7031364440918,
      "epoch": 0.37508747375787266,
      "grad_norm": 0.24701371788978577,
      "learning_rate": 1e-06,
      "loss": 0.0,
      "reward": 0.37400794960558414,
      "reward_std": 0.1167642290238291,
      "rewards/semantic_entropy_math_reward": 0.37400794960558414,
      "step": 67
    },
    {
      "completion_length": 397.74777603149414,
      "epoch": 0.3806857942617215,
      "grad_norm": 0.2930716872215271,
      "learning_rate": 1e-06,
      "loss": 0.0,
      "reward": 0.4553571604192257,
      "reward_std": 0.11352779855951667,
      "rewards/semantic_entropy_math_reward": 0.4553571604192257,
      "step": 68
    },
    {
      "completion_length": 415.2507495880127,
      "epoch": 0.38628411476557034,
      "grad_norm": 0.28370365500450134,
      "learning_rate": 1e-06,
      "loss": 0.0,
      "reward": 0.3591269953176379,
      "reward_std": 0.1410878817550838,
      "rewards/semantic_entropy_math_reward": 0.3591269953176379,
      "step": 69
    },
    {
      "completion_length": 415.7284297943115,
      "epoch": 0.3918824352694192,
      "grad_norm": 0.25602635741233826,
      "learning_rate": 1e-06,
      "loss": 0.0,
      "reward": 0.36582342255860567,
      "reward_std": 0.12171732308343053,
      "rewards/semantic_entropy_math_reward": 0.36582342255860567,
      "step": 70
    },
    {
      "completion_length": 399.8206958770752,
      "epoch": 0.397480755773268,
      "grad_norm": 0.2549217641353607,
      "learning_rate": 1e-06,
      "loss": 0.0,
      "reward": 0.462301604449749,
      "reward_std": 0.13242140784859657,
      "rewards/semantic_entropy_math_reward": 0.462301604449749,
      "step": 71
    },
    {
      "completion_length": 405.83185386657715,
      "epoch": 0.40307907627711687,
      "grad_norm": 0.26034626364707947,
      "learning_rate": 1e-06,
      "loss": 0.0,
      "reward": 0.38876488897949457,
      "reward_std": 0.12143889861181378,
      "rewards/semantic_entropy_math_reward": 0.38876488897949457,
      "step": 72
    },
    {
      "completion_length": 389.92634773254395,
      "epoch": 0.4086773967809657,
      "grad_norm": 0.3175303041934967,
      "learning_rate": 1e-06,
      "loss": 0.0,
      "reward": 0.3635912863537669,
      "reward_std": 0.1250197091139853,
      "rewards/semantic_entropy_math_reward": 0.3635912863537669,
      "step": 73
    },
    {
      "completion_length": 405.5067024230957,
      "epoch": 0.41427571728481455,
      "grad_norm": 0.3240727484226227,
      "learning_rate": 1e-06,
      "loss": 0.0,
      "reward": 0.3521825484931469,
      "reward_std": 0.1221736806910485,
      "rewards/semantic_entropy_math_reward": 0.3521825484931469,
      "step": 74
    },
    {
      "completion_length": 395.6436080932617,
      "epoch": 0.4198740377886634,
      "grad_norm": 0.6251773834228516,
      "learning_rate": 1e-06,
      "loss": 0.0,
      "reward": 0.39942957274615765,
      "reward_std": 0.13623498589731753,
      "rewards/semantic_entropy_math_reward": 0.39942957274615765,
      "step": 75
    },
    {
      "completion_length": 390.2894401550293,
      "epoch": 0.42547235829251223,
      "grad_norm": 0.8764235377311707,
      "learning_rate": 1e-06,
      "loss": 0.0,
      "reward": 0.428075410425663,
      "reward_std": 0.12979253917001188,
      "rewards/semantic_entropy_math_reward": 0.428075410425663,
      "step": 76
    },
    {
      "completion_length": 400.01861000061035,
      "epoch": 0.4310706787963611,
      "grad_norm": 0.8677442669868469,
      "learning_rate": 1e-06,
      "loss": 0.0,
      "reward": 0.37165179289877415,
      "reward_std": 0.12289492227137089,
      "rewards/semantic_entropy_math_reward": 0.37165179289877415,
      "step": 77
    },
    {
      "completion_length": 377.8936080932617,
      "epoch": 0.4366689993002099,
      "grad_norm": 0.9437581896781921,
      "learning_rate": 1e-06,
      "loss": 0.0,
      "reward": 0.373139888048172,
      "reward_std": 0.1251775654964149,
      "rewards/semantic_entropy_math_reward": 0.373139888048172,
      "step": 78
    },
    {
      "completion_length": 340.2700939178467,
      "epoch": 0.44226731980405876,
      "grad_norm": 2.5648627281188965,
      "learning_rate": 1e-06,
      "loss": 0.0,
      "reward": 0.3627232192084193,
      "reward_std": 0.15555713046342134,
      "rewards/semantic_entropy_math_reward": 0.3627232192084193,
      "step": 79
    },
    {
      "completion_length": 268.25521659851074,
      "epoch": 0.44786564030790765,
      "grad_norm": 4.964110851287842,
      "learning_rate": 1e-06,
      "loss": 0.0,
      "reward": 0.32589286658912897,
      "reward_std": 0.1293578795157373,
      "rewards/semantic_entropy_math_reward": 0.32589286658912897,
      "step": 80
    },
    {
      "completion_length": 245.03795051574707,
      "epoch": 0.4534639608117565,
      "grad_norm": 2.0234344005584717,
      "learning_rate": 1e-06,
      "loss": 0.0,
      "reward": 0.34858631901443005,
      "reward_std": 0.1360030840151012,
      "rewards/semantic_entropy_math_reward": 0.34858631901443005,
      "step": 81
    },
    {
      "completion_length": 201.16071891784668,
      "epoch": 0.45906228131560534,
      "grad_norm": 10.089159965515137,
      "learning_rate": 1e-06,
      "loss": 0.0,
      "reward": 0.34126985911279917,
      "reward_std": 0.14876779448240995,
      "rewards/semantic_entropy_math_reward": 0.34126985911279917,
      "step": 82
    },
    {
      "completion_length": 163.12500381469727,
      "epoch": 0.4646606018194542,
      "grad_norm": 7.054829120635986,
      "learning_rate": 1e-06,
      "loss": 0.0,
      "reward": 0.33816965017467737,
      "reward_std": 0.14539710199460387,
      "rewards/semantic_entropy_math_reward": 0.33816965017467737,
      "step": 83
    },
    {
      "completion_length": 172.52455711364746,
      "epoch": 0.470258922323303,
      "grad_norm": 3.74849796295166,
      "learning_rate": 1e-06,
      "loss": 0.0,
      "reward": 0.3324652798473835,
      "reward_std": 0.14292793814092875,
      "rewards/semantic_entropy_math_reward": 0.3324652798473835,
      "step": 84
    },
    {
      "completion_length": 190.39881229400635,
      "epoch": 0.47585724282715186,
      "grad_norm": 3.7803595066070557,
      "learning_rate": 1e-06,
      "loss": 0.0,
      "reward": 0.36879961006343365,
      "reward_std": 0.14196715434081852,
      "rewards/semantic_entropy_math_reward": 0.36879961006343365,
      "step": 85
    },
    {
      "completion_length": 202.71949863433838,
      "epoch": 0.4814555633310007,
      "grad_norm": 7.561800479888916,
      "learning_rate": 1e-06,
      "loss": 0.0,
      "reward": 0.3539186557754874,
      "reward_std": 0.13645589677616954,
      "rewards/semantic_entropy_math_reward": 0.3539186557754874,
      "step": 86
    },
    {
      "completion_length": 192.92336750030518,
      "epoch": 0.48705388383484954,
      "grad_norm": 6.106121063232422,
      "learning_rate": 1e-06,
      "loss": 0.0,
      "reward": 0.41753472574055195,
      "reward_std": 0.14698259718716145,
      "rewards/semantic_entropy_math_reward": 0.41753472574055195,
      "step": 87
    },
    {
      "completion_length": 192.75223636627197,
      "epoch": 0.4926522043386984,
      "grad_norm": 7.5285186767578125,
      "learning_rate": 1e-06,
      "loss": 0.0,
      "reward": 0.3844246091321111,
      "reward_std": 0.154494424816221,
      "rewards/semantic_entropy_math_reward": 0.3844246091321111,
      "step": 88
    },
    {
      "completion_length": 196.7128028869629,
      "epoch": 0.4982505248425472,
      "grad_norm": 5.367832183837891,
      "learning_rate": 1e-06,
      "loss": 0.0,
      "reward": 0.40811011102050543,
      "reward_std": 0.14661366492509842,
      "rewards/semantic_entropy_math_reward": 0.40811011102050543,
      "step": 89
    },
    {
      "completion_length": 200.94940757751465,
      "epoch": 0.5038488453463961,
      "grad_norm": 18.42447853088379,
      "learning_rate": 1e-06,
      "loss": 0.0,
      "reward": 0.4272073404863477,
      "reward_std": 0.14780932199209929,
      "rewards/semantic_entropy_math_reward": 0.4272073404863477,
      "step": 90
    },
    {
      "completion_length": 204.03497314453125,
      "epoch": 0.509447165850245,
      "grad_norm": 7.817193984985352,
      "learning_rate": 1e-06,
      "loss": 0.0,
      "reward": 0.418278768658638,
      "reward_std": 0.13225925154983997,
      "rewards/semantic_entropy_math_reward": 0.418278768658638,
      "step": 91
    },
    {
      "completion_length": 192.34896183013916,
      "epoch": 0.5150454863540938,
      "grad_norm": 6.789738655090332,
      "learning_rate": 1e-06,
      "loss": 0.0,
      "reward": 0.47457837499678135,
      "reward_std": 0.14893107814714313,
      "rewards/semantic_entropy_math_reward": 0.47457837499678135,
      "step": 92
    },
    {
      "completion_length": 210.90179061889648,
      "epoch": 0.5206438068579426,
      "grad_norm": 8.528867721557617,
      "learning_rate": 1e-06,
      "loss": 0.0,
      "reward": 0.41455854661762714,
      "reward_std": 0.1263797995634377,
      "rewards/semantic_entropy_math_reward": 0.41455854661762714,
      "step": 93
    },
    {
      "completion_length": 222.63169956207275,
      "epoch": 0.5262421273617914,
      "grad_norm": 10.166954040527344,
      "learning_rate": 1e-06,
      "loss": 0.0,
      "reward": 0.41530259046703577,
      "reward_std": 0.1387968505732715,
      "rewards/semantic_entropy_math_reward": 0.41530259046703577,
      "step": 94
    },
    {
      "completion_length": 208.9508981704712,
      "epoch": 0.5318404478656403,
      "grad_norm": 8.634490966796875,
      "learning_rate": 1e-06,
      "loss": 0.0,
      "reward": 0.45324901584535837,
      "reward_std": 0.1519172815605998,
      "rewards/semantic_entropy_math_reward": 0.45324901584535837,
      "step": 95
    },
    {
      "completion_length": 240.5915231704712,
      "epoch": 0.5374387683694891,
      "grad_norm": 15.801514625549316,
      "learning_rate": 1e-06,
      "loss": 0.0,
      "reward": 0.4201389057561755,
      "reward_std": 0.15532761020585895,
      "rewards/semantic_entropy_math_reward": 0.4201389057561755,
      "step": 96
    },
    {
      "completion_length": 288.0282793045044,
      "epoch": 0.543037088873338,
      "grad_norm": 7.69626522064209,
      "learning_rate": 1e-06,
      "loss": 0.0,
      "reward": 0.49342758767306805,
      "reward_std": 0.1336272112093866,
      "rewards/semantic_entropy_math_reward": 0.49342758767306805,
      "step": 97
    },
    {
      "completion_length": 570.5766410827637,
      "epoch": 0.5486354093771868,
      "grad_norm": 9.325577735900879,
      "learning_rate": 1e-06,
      "loss": 0.0,
      "reward": 0.4603174738585949,
      "reward_std": 0.14188312110491097,
      "rewards/semantic_entropy_math_reward": 0.4603174738585949,
      "step": 98
    },
    {
      "completion_length": 924.8459968566895,
      "epoch": 0.5542337298810357,
      "grad_norm": 4.683976650238037,
      "learning_rate": 1e-06,
      "loss": 0.0,
      "reward": 0.4341518022119999,
      "reward_std": 0.15863031102344394,
      "rewards/semantic_entropy_math_reward": 0.4341518022119999,
      "step": 99
    },
    {
      "completion_length": 975.7284393310547,
      "epoch": 0.5598320503848845,
      "grad_norm": 6.508670806884766,
      "learning_rate": 1e-06,
      "loss": 0.0,
      "reward": 0.43588790111243725,
      "reward_std": 0.1634892332367599,
      "rewards/semantic_entropy_math_reward": 0.43588790111243725,
      "step": 100
    },
    {
      "completion_length": 979.6651954650879,
      "epoch": 0.5654303708887334,
      "grad_norm": 6.556097507476807,
      "learning_rate": 1e-06,
      "loss": 0.0,
      "reward": 0.4521329505369067,
      "reward_std": 0.16528919152915478,
      "rewards/semantic_entropy_math_reward": 0.4521329505369067,
      "step": 101
    },
    {
      "completion_length": 966.6339378356934,
      "epoch": 0.5710286913925823,
      "grad_norm": 1.9696649312973022,
      "learning_rate": 1e-06,
      "loss": 0.0,
      "reward": 0.45213295705616474,
      "reward_std": 0.1434005326591432,
      "rewards/semantic_entropy_math_reward": 0.45213295705616474,
      "step": 102
    },
    {
      "completion_length": 933.5625114440918,
      "epoch": 0.5766270118964311,
      "grad_norm": 2.5271127223968506,
      "learning_rate": 1e-06,
      "loss": 0.0,
      "reward": 0.4157986231148243,
      "reward_std": 0.15396162681281567,
      "rewards/semantic_entropy_math_reward": 0.4157986231148243,
      "step": 103
    },
    {
      "completion_length": 609.2589454650879,
      "epoch": 0.58222533240028,
      "grad_norm": 2.110628366470337,
      "learning_rate": 1e-06,
      "loss": 0.0,
      "reward": 0.5197172742336988,
      "reward_std": 0.1446234486065805,
      "rewards/semantic_entropy_math_reward": 0.5197172742336988,
      "step": 104
    },
    {
      "completion_length": 299.1971788406372,
      "epoch": 0.5878236529041287,
      "grad_norm": 1.4388288259506226,
      "learning_rate": 1e-06,
      "loss": 0.0,
      "reward": 0.49330358020961285,
      "reward_std": 0.15013255970552564,
      "rewards/semantic_entropy_math_reward": 0.49330358020961285,
      "step": 105
    },
    {
      "completion_length": 281.92931747436523,
      "epoch": 0.5934219734079776,
      "grad_norm": 1.1156909465789795,
      "learning_rate": 1e-06,
      "loss": 0.0,
      "reward": 0.5328621100634336,
      "reward_std": 0.16049590334296227,
      "rewards/semantic_entropy_math_reward": 0.5328621100634336,
      "step": 106
    },
    {
      "completion_length": 277.87053871154785,
      "epoch": 0.5990202939118264,
      "grad_norm": 1.5050427913665771,
      "learning_rate": 1e-06,
      "loss": 0.0,
      "reward": 0.4991319552063942,
      "reward_std": 0.13887666864320636,
      "rewards/semantic_entropy_math_reward": 0.4991319552063942,
      "step": 107
    },
    {
      "completion_length": 244.03050994873047,
      "epoch": 0.6046186144156753,
      "grad_norm": 2.5947532653808594,
      "learning_rate": 1e-06,
      "loss": 0.0,
      "reward": 0.5357143003493547,
      "reward_std": 0.15269030537456274,
      "rewards/semantic_entropy_math_reward": 0.5357143003493547,
      "step": 108
    },
    {
      "completion_length": 137.79092693328857,
      "epoch": 0.6102169349195241,
      "grad_norm": 4.043277263641357,
      "learning_rate": 1e-06,
      "loss": 0.0,
      "reward": 0.5419146958738565,
      "reward_std": 0.1557740089483559,
      "rewards/semantic_entropy_math_reward": 0.5419146958738565,
      "step": 109
    },
    {
      "completion_length": 107.52009153366089,
      "epoch": 0.615815255423373,
      "grad_norm": 1.8183155059814453,
      "learning_rate": 1e-06,
      "loss": 0.0,
      "reward": 0.5765129253268242,
      "reward_std": 0.12702399701811373,
      "rewards/semantic_entropy_math_reward": 0.5765129253268242,
      "step": 110
    },
    {
      "completion_length": 96.88541841506958,
      "epoch": 0.6214135759272218,
      "grad_norm": 2.091888904571533,
      "learning_rate": 1e-06,
      "loss": 0.0,
      "reward": 0.6276041902601719,
      "reward_std": 0.13468103110790253,
      "rewards/semantic_entropy_math_reward": 0.6276041902601719,
      "step": 111
    },
    {
      "completion_length": 138.0691990852356,
      "epoch": 0.6270118964310707,
      "grad_norm": 1.2132967710494995,
      "learning_rate": 1e-06,
      "loss": 0.0,
      "reward": 0.5658482331782579,
      "reward_std": 0.14125833846628666,
      "rewards/semantic_entropy_math_reward": 0.5658482331782579,
      "step": 112
    },
    {
      "completion_length": 168.62128067016602,
      "epoch": 0.6326102169349195,
      "grad_norm": 1.3657702207565308,
      "learning_rate": 1e-06,
      "loss": 0.0,
      "reward": 0.5846974309533834,
      "reward_std": 0.12009491049684584,
      "rewards/semantic_entropy_math_reward": 0.5846974309533834,
      "step": 113
    },
    {
      "completion_length": 172.63616180419922,
      "epoch": 0.6382085374387684,
      "grad_norm": 4.615123748779297,
      "learning_rate": 1e-06,
      "loss": 0.0,
      "reward": 0.6726190745830536,
      "reward_std": 0.14088664995506406,
      "rewards/semantic_entropy_math_reward": 0.6726190745830536,
      "step": 114
    },
    {
      "completion_length": 184.50372314453125,
      "epoch": 0.6438068579426172,
      "grad_norm": 7.566853046417236,
      "learning_rate": 1e-06,
      "loss": 0.0,
      "reward": 0.623387910425663,
      "reward_std": 0.1348249651491642,
      "rewards/semantic_entropy_math_reward": 0.623387910425663,
      "step": 115
    },
    {
      "completion_length": 268.1763458251953,
      "epoch": 0.6494051784464661,
      "grad_norm": 3.787036657333374,
      "learning_rate": 1e-06,
      "loss": 0.0,
      "reward": 0.5989583469927311,
      "reward_std": 0.14310514647513628,
      "rewards/semantic_entropy_math_reward": 0.5989583469927311,
      "step": 116
    },
    {
      "completion_length": 424.9270887374878,
      "epoch": 0.655003498950315,
      "grad_norm": 1.769083857536316,
      "learning_rate": 1e-06,
      "loss": 0.0,
      "reward": 0.6443452537059784,
      "reward_std": 0.1348385107703507,
      "rewards/semantic_entropy_math_reward": 0.6443452537059784,
      "step": 117
    },
    {
      "completion_length": 627.7061138153076,
      "epoch": 0.6606018194541637,
      "grad_norm": 7.572482109069824,
      "learning_rate": 1e-06,
      "loss": 0.0,
      "reward": 0.6216518022119999,
      "reward_std": 0.12173205520957708,
      "rewards/semantic_entropy_math_reward": 0.6216518022119999,
      "step": 118
    },
    {
      "completion_length": 597.2857189178467,
      "epoch": 0.6662001399580126,
      "grad_norm": 4.5395965576171875,
      "learning_rate": 1e-06,
      "loss": 0.0,
      "reward": 0.6326885130256414,
      "reward_std": 0.11796635785140097,
      "rewards/semantic_entropy_math_reward": 0.6326885130256414,
      "step": 119
    },
    {
      "completion_length": 499.34674072265625,
      "epoch": 0.6717984604618614,
      "grad_norm": 9.828434944152832,
      "learning_rate": 1e-06,
      "loss": 0.0,
      "reward": 0.6607142873108387,
      "reward_std": 0.12943885754793882,
      "rewards/semantic_entropy_math_reward": 0.6607142873108387,
      "step": 120
    },
    {
      "completion_length": 416.2641429901123,
      "epoch": 0.6773967809657103,
      "grad_norm": 4.24314022064209,
      "learning_rate": 1e-06,
      "loss": 0.0,
      "reward": 0.7173859365284443,
      "reward_std": 0.10959133435972035,
      "rewards/semantic_entropy_math_reward": 0.7173859365284443,
      "step": 121
    },
    {
      "completion_length": 133.405508518219,
      "epoch": 0.6829951014695591,
      "grad_norm": 5.226346969604492,
      "learning_rate": 1e-06,
      "loss": 0.0,
      "reward": 0.6529017835855484,
      "reward_std": 0.13599514495581388,
      "rewards/semantic_entropy_math_reward": 0.6529017835855484,
      "step": 122
    },
    {
      "completion_length": 108.45610427856445,
      "epoch": 0.688593421973408,
      "grad_norm": 4.747099876403809,
      "learning_rate": 1e-06,
      "loss": 0.0,
      "reward": 0.6951884962618351,
      "reward_std": 0.1218845120165497,
      "rewards/semantic_entropy_math_reward": 0.6951884962618351,
      "step": 123
    },
    {
      "completion_length": 116.35714483261108,
      "epoch": 0.6941917424772568,
      "grad_norm": 5.495078086853027,
      "learning_rate": 1e-06,
      "loss": 0.0,
      "reward": 0.7322668917477131,
      "reward_std": 0.10345165804028511,
      "rewards/semantic_entropy_math_reward": 0.7322668917477131,
      "step": 124
    },
    {
      "completion_length": 190.98735427856445,
      "epoch": 0.6997900629811057,
      "grad_norm": 4.758305072784424,
      "learning_rate": 1e-06,
      "loss": 0.0,
      "reward": 0.6990327574312687,
      "reward_std": 0.1072727048303932,
      "rewards/semantic_entropy_math_reward": 0.6990327574312687,
      "step": 125
    },
    {
      "completion_length": 154.2269356250763,
      "epoch": 0.7053883834849545,
      "grad_norm": 4.895815372467041,
      "learning_rate": 1e-06,
      "loss": 0.0,
      "reward": 0.7566964458674192,
      "reward_std": 0.10861217533238232,
      "rewards/semantic_entropy_math_reward": 0.7566964458674192,
      "step": 126
    },
    {
      "completion_length": 219.0520896911621,
      "epoch": 0.7109867039888034,
      "grad_norm": 5.073260307312012,
      "learning_rate": 1e-06,
      "loss": 0.0,
      "reward": 0.7007688824087381,
      "reward_std": 0.10281824320554733,
      "rewards/semantic_entropy_math_reward": 0.7007688824087381,
      "step": 127
    },
    {
      "completion_length": 150.2425627708435,
      "epoch": 0.7165850244926522,
      "grad_norm": 6.930428981781006,
      "learning_rate": 1e-06,
      "loss": 0.0,
      "reward": 0.7514881230890751,
      "reward_std": 0.11211708444170654,
      "rewards/semantic_entropy_math_reward": 0.7514881230890751,
      "step": 128
    },
    {
      "completion_length": 150.25967574119568,
      "epoch": 0.722183344996501,
      "grad_norm": 4.981149196624756,
      "learning_rate": 1e-06,
      "loss": 0.0,
      "reward": 0.7196180820465088,
      "reward_std": 0.1129911975003779,
      "rewards/semantic_entropy_math_reward": 0.7196180820465088,
      "step": 129
    },
    {
      "completion_length": 129.3750023841858,
      "epoch": 0.72778166550035,
      "grad_norm": 5.079310417175293,
      "learning_rate": 1e-06,
      "loss": 0.0,
      "reward": 0.7373512163758278,
      "reward_std": 0.10283095168415457,
      "rewards/semantic_entropy_math_reward": 0.7373512163758278,
      "step": 130
    },
    {
      "completion_length": 134.2485146522522,
      "epoch": 0.7333799860041987,
      "grad_norm": 5.917436599731445,
      "learning_rate": 1e-06,
      "loss": 0.0,
      "reward": 0.7232142984867096,
      "reward_std": 0.1203189252410084,
      "rewards/semantic_entropy_math_reward": 0.7232142984867096,
      "step": 131
    },
    {
      "completion_length": 107.0967288017273,
      "epoch": 0.7389783065080476,
      "grad_norm": 8.360037803649902,
      "learning_rate": 1e-06,
      "loss": 0.0,
      "reward": 0.7357390988618135,
      "reward_std": 0.10952205723151565,
      "rewards/semantic_entropy_math_reward": 0.7357390988618135,
      "step": 132
    },
    {
      "completion_length": 128.16592502593994,
      "epoch": 0.7445766270118964,
      "grad_norm": 7.823781490325928,
      "learning_rate": 1e-06,
      "loss": 0.0,
      "reward": 0.7218502275645733,
      "reward_std": 0.11286869831383228,
      "rewards/semantic_entropy_math_reward": 0.7218502275645733,
      "step": 133
    },
    {
      "completion_length": 106.69270944595337,
      "epoch": 0.7501749475157453,
      "grad_norm": 7.141363143920898,
      "learning_rate": 1e-06,
      "loss": 0.0,
      "reward": 0.7478918954730034,
      "reward_std": 0.10708213225007057,
      "rewards/semantic_entropy_math_reward": 0.7478918954730034,
      "step": 134
    },
    {
      "completion_length": 93.68006038665771,
      "epoch": 0.7557732680195941,
      "grad_norm": 5.968332767486572,
      "learning_rate": 1e-06,
      "loss": 0.0,
      "reward": 0.7960069701075554,
      "reward_std": 0.08689452323596925,
      "rewards/semantic_entropy_math_reward": 0.7960069701075554,
      "step": 135
    },
    {
      "completion_length": 90.8816967010498,
      "epoch": 0.761371588523443,
      "grad_norm": 4.925920486450195,
      "learning_rate": 1e-06,
      "loss": 0.0,
      "reward": 0.7625248245894909,
      "reward_std": 0.10826884070411325,
      "rewards/semantic_entropy_math_reward": 0.7625248245894909,
      "step": 136
    },
    {
      "completion_length": 106.96056652069092,
      "epoch": 0.7669699090272918,
      "grad_norm": 6.21066427230835,
      "learning_rate": 1e-06,
      "loss": 0.0,
      "reward": 0.7124256156384945,
      "reward_std": 0.12395848939195275,
      "rewards/semantic_entropy_math_reward": 0.7124256156384945,
      "step": 137
    },
    {
      "completion_length": 113.4546160697937,
      "epoch": 0.7725682295311407,
      "grad_norm": 4.43303108215332,
      "learning_rate": 1e-06,
      "loss": 0.0,
      "reward": 0.7115575522184372,
      "reward_std": 0.12078050547279418,
      "rewards/semantic_entropy_math_reward": 0.7115575522184372,
      "step": 138
    },
    {
      "completion_length": 140.69047927856445,
      "epoch": 0.7781665500349895,
      "grad_norm": 3.2451171875,
      "learning_rate": 1e-06,
      "loss": 0.0,
      "reward": 0.694320447742939,
      "reward_std": 0.1306492048315704,
      "rewards/semantic_entropy_math_reward": 0.694320447742939,
      "step": 139
    },
    {
      "completion_length": 142.95833444595337,
      "epoch": 0.7837648705388384,
      "grad_norm": 6.136082649230957,
      "learning_rate": 1e-06,
      "loss": 0.0,
      "reward": 0.7182539887726307,
      "reward_std": 0.15217532915994525,
      "rewards/semantic_entropy_math_reward": 0.7182539887726307,
      "step": 140
    },
    {
      "completion_length": 142.6287226676941,
      "epoch": 0.7893631910426872,
      "grad_norm": 2.892977476119995,
      "learning_rate": 1e-06,
      "loss": 0.0,
      "reward": 0.7118055783212185,
      "reward_std": 0.13839181000366807,
      "rewards/semantic_entropy_math_reward": 0.7118055783212185,
      "step": 141
    },
    {
      "completion_length": 157.30431842803955,
      "epoch": 0.794961511546536,
      "grad_norm": 4.3441619873046875,
      "learning_rate": 1e-06,
      "loss": 0.0,
      "reward": 0.6943204514682293,
      "reward_std": 0.14929893938824534,
      "rewards/semantic_entropy_math_reward": 0.6943204514682293,
      "step": 142
    },
    {
      "completion_length": 168.61533069610596,
      "epoch": 0.8005598320503848,
      "grad_norm": 1.7150957584381104,
      "learning_rate": 1e-06,
      "loss": 0.0,
      "reward": 0.686507947742939,
      "reward_std": 0.1514957039617002,
      "rewards/semantic_entropy_math_reward": 0.686507947742939,
      "step": 143
    },
    {
      "completion_length": 162.0543179512024,
      "epoch": 0.8061581525542337,
      "grad_norm": 1.2334315776824951,
      "learning_rate": 1e-06,
      "loss": 0.0,
      "reward": 0.6346726343035698,
      "reward_std": 0.14308730140328407,
      "rewards/semantic_entropy_math_reward": 0.6346726343035698,
      "step": 144
    },
    {
      "completion_length": 265.3921184539795,
      "epoch": 0.8117564730580826,
      "grad_norm": 2.8437411785125732,
      "learning_rate": 1e-06,
      "loss": 0.0,
      "reward": 0.5422867126762867,
      "reward_std": 0.17818555515259504,
      "rewards/semantic_entropy_math_reward": 0.5422867126762867,
      "step": 145
    },
    {
      "completion_length": 358.7314052581787,
      "epoch": 0.8173547935619314,
      "grad_norm": 1.9524171352386475,
      "learning_rate": 1e-06,
      "loss": 0.0,
      "reward": 0.4378720261156559,
      "reward_std": 0.194162187166512,
      "rewards/semantic_entropy_math_reward": 0.4378720261156559,
      "step": 146
    },
    {
      "completion_length": 268.26190662384033,
      "epoch": 0.8229531140657803,
      "grad_norm": 0.9712325930595398,
      "learning_rate": 1e-06,
      "loss": 0.0,
      "reward": 0.5739087481051683,
      "reward_std": 0.19231022708117962,
      "rewards/semantic_entropy_math_reward": 0.5739087481051683,
      "step": 147
    },
    {
      "completion_length": 247.8772373199463,
      "epoch": 0.8285514345696291,
      "grad_norm": 0.29546335339546204,
      "learning_rate": 1e-06,
      "loss": 0.0,
      "reward": 0.5668402947485447,
      "reward_std": 0.14728267351165414,
      "rewards/semantic_entropy_math_reward": 0.5668402947485447,
      "step": 148
    },
    {
      "completion_length": 286.6562547683716,
      "epoch": 0.834149755073478,
      "grad_norm": 0.19167956709861755,
      "learning_rate": 1e-06,
      "loss": 0.0,
      "reward": 0.5652281939983368,
      "reward_std": 0.16519550560042262,
      "rewards/semantic_entropy_math_reward": 0.5652281939983368,
      "step": 149
    },
    {
      "completion_length": 299.7745580673218,
      "epoch": 0.8397480755773268,
      "grad_norm": 0.16220510005950928,
      "learning_rate": 1e-06,
      "loss": 0.0,
      "reward": 0.563988110050559,
      "reward_std": 0.16206924617290497,
      "rewards/semantic_entropy_math_reward": 0.563988110050559,
      "step": 150
    },
    {
      "completion_length": 323.96875381469727,
      "epoch": 0.8453463960811757,
      "grad_norm": 0.11583292484283447,
      "learning_rate": 1e-06,
      "loss": 0.0,
      "reward": 0.5203373022377491,
      "reward_std": 0.17140463134273887,
      "rewards/semantic_entropy_math_reward": 0.5203373022377491,
      "step": 151
    },
    {
      "completion_length": 333.31994819641113,
      "epoch": 0.8509447165850245,
      "grad_norm": 0.09664779901504517,
      "learning_rate": 1e-06,
      "loss": 0.0,
      "reward": 0.503100199624896,
      "reward_std": 0.17030652752146125,
      "rewards/semantic_entropy_math_reward": 0.503100199624896,
      "step": 152
    },
    {
      "completion_length": 319.05804443359375,
      "epoch": 0.8565430370888734,
      "grad_norm": 0.08762703090906143,
      "learning_rate": 1e-06,
      "loss": 0.0,
      "reward": 0.5343502033501863,
      "reward_std": 0.1655228640884161,
      "rewards/semantic_entropy_math_reward": 0.5343502033501863,
      "step": 153
    },
    {
      "completion_length": 346.2261962890625,
      "epoch": 0.8621413575927221,
      "grad_norm": 0.07872086763381958,
      "learning_rate": 1e-06,
      "loss": 0.0,
      "reward": 0.5121527835726738,
      "reward_std": 0.15845369640737772,
      "rewards/semantic_entropy_math_reward": 0.5121527835726738,
      "step": 154
    },
    {
      "completion_length": 357.5372085571289,
      "epoch": 0.867739678096571,
      "grad_norm": 0.0757875069975853,
      "learning_rate": 1e-06,
      "loss": 0.0,
      "reward": 0.5648561716079712,
      "reward_std": 0.1613363642245531,
      "rewards/semantic_entropy_math_reward": 0.5648561716079712,
      "step": 155
    },
    {
      "completion_length": 353.3415222167969,
      "epoch": 0.8733379986004198,
      "grad_norm": 0.09012839198112488,
      "learning_rate": 1e-06,
      "loss": 0.0,
      "reward": 0.49231152422726154,
      "reward_std": 0.1589321019127965,
      "rewards/semantic_entropy_math_reward": 0.49231152422726154,
      "step": 156
    },
    {
      "completion_length": 354.72173500061035,
      "epoch": 0.8789363191042687,
      "grad_norm": 0.24965420365333557,
      "learning_rate": 1e-06,
      "loss": 0.0,
      "reward": 0.5529513861984015,
      "reward_std": 0.16521730786189437,
      "rewards/semantic_entropy_math_reward": 0.5529513861984015,
      "step": 157
    },
    {
      "completion_length": 348.0320014953613,
      "epoch": 0.8845346396081175,
      "grad_norm": 0.06840746104717255,
      "learning_rate": 1e-06,
      "loss": 0.0,
      "reward": 0.5613839365541935,
      "reward_std": 0.1481709172949195,
      "rewards/semantic_entropy_math_reward": 0.5613839365541935,
      "step": 158
    },
    {
      "completion_length": 350.35715103149414,
      "epoch": 0.8901329601119664,
      "grad_norm": 0.13703550398349762,
      "learning_rate": 1e-06,
      "loss": 0.0,
      "reward": 0.5834573470056057,
      "reward_std": 0.164536252617836,
      "rewards/semantic_entropy_math_reward": 0.5834573470056057,
      "step": 159
    },
    {
      "completion_length": 358.2440528869629,
      "epoch": 0.8957312806158153,
      "grad_norm": 0.09526026993989944,
      "learning_rate": 1e-06,
      "loss": 0.0,
      "reward": 0.5683283749967813,
      "reward_std": 0.14862215495668352,
      "rewards/semantic_entropy_math_reward": 0.5683283749967813,
      "step": 160
    },
    {
      "completion_length": 374.41518211364746,
      "epoch": 0.9013296011196641,
      "grad_norm": 0.09629662334918976,
      "learning_rate": 1e-06,
      "loss": 0.0,
      "reward": 0.5731646921485662,
      "reward_std": 0.15710427099838853,
      "rewards/semantic_entropy_math_reward": 0.5731646921485662,
      "step": 161
    },
    {
      "completion_length": 359.19643211364746,
      "epoch": 0.906927921623513,
      "grad_norm": 0.10354594886302948,
      "learning_rate": 1e-06,
      "loss": 0.0,
      "reward": 0.5668402947485447,
      "reward_std": 0.1442330675199628,
      "rewards/semantic_entropy_math_reward": 0.5668402947485447,
      "step": 162
    },
    {
      "completion_length": 363.5342330932617,
      "epoch": 0.9125262421273618,
      "grad_norm": 0.09924127161502838,
      "learning_rate": 1e-06,
      "loss": 0.0,
      "reward": 0.6071428768336773,
      "reward_std": 0.15955356042832136,
      "rewards/semantic_entropy_math_reward": 0.6071428768336773,
      "step": 163
    },
    {
      "completion_length": 344.59301376342773,
      "epoch": 0.9181245626312107,
      "grad_norm": 0.10809105634689331,
      "learning_rate": 1e-06,
      "loss": 0.0,
      "reward": 0.6684027947485447,
      "reward_std": 0.14918948477134109,
      "rewards/semantic_entropy_math_reward": 0.6684027947485447,
      "step": 164
    },
    {
      "completion_length": 361.6019401550293,
      "epoch": 0.9237228831350595,
      "grad_norm": 0.08925337344408035,
      "learning_rate": 1e-06,
      "loss": 0.0,
      "reward": 0.713541679084301,
      "reward_std": 0.14462757343426347,
      "rewards/semantic_entropy_math_reward": 0.713541679084301,
      "step": 165
    },
    {
      "completion_length": 358.32738876342773,
      "epoch": 0.9293212036389084,
      "grad_norm": 0.10002636164426804,
      "learning_rate": 1e-06,
      "loss": 0.0,
      "reward": 0.8019593358039856,
      "reward_std": 0.101166230160743,
      "rewards/semantic_entropy_math_reward": 0.8019593358039856,
      "step": 166
    },
    {
      "completion_length": 349.5200939178467,
      "epoch": 0.9349195241427571,
      "grad_norm": 0.0706295296549797,
      "learning_rate": 1e-06,
      "loss": 0.0,
      "reward": 0.8885168805718422,
      "reward_std": 0.0957920125219971,
      "rewards/semantic_entropy_math_reward": 0.8885168805718422,
      "step": 167
    },
    {
      "completion_length": 357.4538764953613,
      "epoch": 0.940517844646606,
      "grad_norm": 0.04306298866868019,
      "learning_rate": 1e-06,
      "loss": 0.0,
      "reward": 0.9391121119260788,
      "reward_std": 0.04490397102199495,
      "rewards/semantic_entropy_math_reward": 0.9391121119260788,
      "step": 168
    },
    {
      "completion_length": 344.33408546447754,
      "epoch": 0.9461161651504548,
      "grad_norm": 0.05653183534741402,
      "learning_rate": 1e-06,
      "loss": 0.0,
      "reward": 0.9515129216015339,
      "reward_std": 0.03180735826026648,
      "rewards/semantic_entropy_math_reward": 0.9515129216015339,
      "step": 169
    },
    {
      "completion_length": 344.59078216552734,
      "epoch": 0.9517144856543037,
      "grad_norm": 0.033216774463653564,
      "learning_rate": 1e-06,
      "loss": 0.0,
      "reward": 0.9871031977236271,
      "reward_std": 0.008495657471939921,
      "rewards/semantic_entropy_math_reward": 0.9871031977236271,
      "step": 170
    },
    {
      "completion_length": 346.92560386657715,
      "epoch": 0.9573128061581525,
      "grad_norm": 0.0,
      "learning_rate": 1e-06,
      "loss": 0.0,
      "reward": 1.0,
      "reward_std": 0.0,
      "rewards/semantic_entropy_math_reward": 1.0,
      "step": 171
    },
    {
      "completion_length": 357.2983684539795,
      "epoch": 0.9629111266620014,
      "grad_norm": 0.001862065983004868,
      "learning_rate": 1e-06,
      "loss": 0.0,
      "reward": 0.9986359179019928,
      "reward_std": 0.0021478806156665087,
      "rewards/semantic_entropy_math_reward": 0.9986359179019928,
      "step": 172
    },
    {
      "completion_length": 338.4322986602783,
      "epoch": 0.9685094471658502,
      "grad_norm": 0.0,
      "learning_rate": 1e-06,
      "loss": 0.0,
      "reward": 1.0,
      "reward_std": 0.0,
      "rewards/semantic_entropy_math_reward": 1.0,
      "step": 173
    },
    {
      "completion_length": 340.3586368560791,
      "epoch": 0.9741077676696991,
      "grad_norm": 0.0,
      "learning_rate": 1e-06,
      "loss": 0.0,
      "reward": 1.0,
      "reward_std": 0.0,
      "rewards/semantic_entropy_math_reward": 1.0,
      "step": 174
    },
    {
      "completion_length": 351.07218170166016,
      "epoch": 0.979706088173548,
      "grad_norm": 0.005613424815237522,
      "learning_rate": 1e-06,
      "loss": 0.0,
      "reward": 0.9986359179019928,
      "reward_std": 0.0021478806156665087,
      "rewards/semantic_entropy_math_reward": 0.9986359179019928,
      "step": 175
    },
    {
      "completion_length": 346.41890716552734,
      "epoch": 0.9853044086773968,
      "grad_norm": 0.0,
      "learning_rate": 1e-06,
      "loss": 0.0,
      "reward": 1.0,
      "reward_std": 0.0,
      "rewards/semantic_entropy_math_reward": 1.0,
      "step": 176
    },
    {
      "completion_length": 352.03422927856445,
      "epoch": 0.9909027291812457,
      "grad_norm": 0.0,
      "learning_rate": 1e-06,
      "loss": 0.0,
      "reward": 1.0,
      "reward_std": 0.0,
      "rewards/semantic_entropy_math_reward": 1.0,
      "step": 177
    },
    {
      "completion_length": 361.8452491760254,
      "epoch": 0.9965010496850945,
      "grad_norm": 0.008886351250112057,
      "learning_rate": 1e-06,
      "loss": 0.0,
      "reward": 0.9972718358039856,
      "reward_std": 0.004295761231333017,
      "rewards/semantic_entropy_math_reward": 0.9972718358039856,
      "step": 178
    },
    {
      "epoch": 0.9965010496850945,
      "step": 178,
      "total_flos": 0.0,
      "train_loss": 0.0,
      "train_runtime": 5.6459,
      "train_samples_per_second": 3542.367,
      "train_steps_per_second": 31.527
    }
  ],
  "logging_steps": 1,
  "max_steps": 178,
  "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": true
      },
      "attributes": {}
    }
  },
  "total_flos": 0.0,
  "train_batch_size": 1,
  "trial_name": null,
  "trial_params": null
}