{
  "best_global_step": null,
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 2.0,
  "eval_steps": 500,
  "global_step": 212,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "completion_length": 295.42578125,
      "epoch": 0.009481481481481481,
      "grad_norm": 7.500036498446221,
      "kl": 1.150369644165039e-05,
      "learning_rate": 0.0,
      "loss": 0.0,
      "reward": 3.017578125,
      "reward_std": 0.5333664314821362,
      "rewards/accuracy_reward": 0.810546875,
      "rewards/format_reward": 0.841796875,
      "rewards/influence_reward": 0.732421875,
      "rewards/len_reward": 0.6328125,
      "step": 1
    },
    {
      "completion_length": 294.02685546875,
      "epoch": 0.047407407407407405,
      "grad_norm": 7.92642971938252,
      "kl": 0.004460334777832031,
      "learning_rate": 1.818181818181818e-07,
      "loss": 0.0002,
      "reward": 2.873046875,
      "reward_std": 0.5104024603497237,
      "rewards/accuracy_reward": 0.794921875,
      "rewards/format_reward": 0.8486328125,
      "rewards/influence_reward": 0.71923828125,
      "rewards/len_reward": 0.51025390625,
      "step": 5
    },
    {
      "completion_length": 288.498828125,
      "epoch": 0.09481481481481481,
      "grad_norm": 7.020840931953974,
      "kl": 0.04069175720214844,
      "learning_rate": 4.090909090909091e-07,
      "loss": 0.0016,
      "reward": 2.8578125,
      "reward_std": 0.5119308672845364,
      "rewards/accuracy_reward": 0.76484375,
      "rewards/format_reward": 0.85859375,
      "rewards/influence_reward": 0.701171875,
      "rewards/len_reward": 0.533203125,
      "step": 10
    },
    {
      "completion_length": 292.551171875,
      "epoch": 0.14222222222222222,
      "grad_norm": 4.823838531105677,
      "kl": 0.40450439453125,
      "learning_rate": 6.363636363636363e-07,
      "loss": 0.0162,
      "reward": 2.818359375,
      "reward_std": 0.47696277797222136,
      "rewards/accuracy_reward": 0.75078125,
      "rewards/format_reward": 0.876953125,
      "rewards/influence_reward": 0.69375,
      "rewards/len_reward": 0.496875,
      "step": 15
    },
    {
      "completion_length": 292.27890625,
      "epoch": 0.18962962962962962,
      "grad_norm": 3.9824205036861335,
      "kl": 1.21318359375,
      "learning_rate": 8.636363636363636e-07,
      "loss": 0.0485,
      "reward": 2.9734375,
      "reward_std": 0.4506275236606598,
      "rewards/accuracy_reward": 0.770703125,
      "rewards/format_reward": 0.915625,
      "rewards/influence_reward": 0.723828125,
      "rewards/len_reward": 0.56328125,
      "step": 20
    },
    {
      "completion_length": 272.598828125,
      "epoch": 0.23703703703703705,
      "grad_norm": 11.519849853989422,
      "kl": 1.022705078125,
      "learning_rate": 9.99726628670463e-07,
      "loss": 0.0409,
      "reward": 3.033203125,
      "reward_std": 0.43844289109110834,
      "rewards/accuracy_reward": 0.793359375,
      "rewards/format_reward": 0.929296875,
      "rewards/influence_reward": 0.749609375,
      "rewards/len_reward": 0.5609375,
      "step": 25
    },
    {
      "completion_length": 258.675390625,
      "epoch": 0.28444444444444444,
      "grad_norm": 4.266007081111176,
      "kl": 0.8201416015625,
      "learning_rate": 9.966546331768192e-07,
      "loss": 0.0328,
      "reward": 3.074609375,
      "reward_std": 0.404691843315959,
      "rewards/accuracy_reward": 0.785546875,
      "rewards/format_reward": 0.930859375,
      "rewards/influence_reward": 0.74453125,
      "rewards/len_reward": 0.613671875,
      "step": 30
    },
    {
      "completion_length": 254.255859375,
      "epoch": 0.33185185185185184,
      "grad_norm": 6.382198551084978,
      "kl": 0.82421875,
      "learning_rate": 9.901899829374047e-07,
      "loss": 0.033,
      "reward": 3.066015625,
      "reward_std": 0.4163201114162803,
      "rewards/accuracy_reward": 0.77265625,
      "rewards/format_reward": 0.9390625,
      "rewards/influence_reward": 0.738671875,
      "rewards/len_reward": 0.615625,
      "step": 35
    },
    {
      "completion_length": 243.677734375,
      "epoch": 0.37925925925925924,
      "grad_norm": 3.662204783931497,
      "kl": 0.63624267578125,
      "learning_rate": 9.803768380684242e-07,
      "loss": 0.0255,
      "reward": 3.06640625,
      "reward_std": 0.42637087870389223,
      "rewards/accuracy_reward": 0.75625,
      "rewards/format_reward": 0.944921875,
      "rewards/influence_reward": 0.717578125,
      "rewards/len_reward": 0.64765625,
      "step": 40
    },
    {
      "completion_length": 245.52734375,
      "epoch": 0.4266666666666667,
      "grad_norm": 2.7957198867659767,
      "kl": 0.6301025390625,
      "learning_rate": 9.672822322997304e-07,
      "loss": 0.0252,
      "reward": 3.121484375,
      "reward_std": 0.3937599007040262,
      "rewards/accuracy_reward": 0.783984375,
      "rewards/format_reward": 0.94765625,
      "rewards/influence_reward": 0.758984375,
      "rewards/len_reward": 0.630859375,
      "step": 45
    },
    {
      "completion_length": 244.16171875,
      "epoch": 0.4740740740740741,
      "grad_norm": 4.651687815378263,
      "kl": 0.61866455078125,
      "learning_rate": 9.509956150664795e-07,
      "loss": 0.0248,
      "reward": 3.048828125,
      "reward_std": 0.3970143750309944,
      "rewards/accuracy_reward": 0.753125,
      "rewards/format_reward": 0.946484375,
      "rewards/influence_reward": 0.722265625,
      "rewards/len_reward": 0.626953125,
      "step": 50
    },
    {
      "completion_length": 246.1078125,
      "epoch": 0.5214814814814814,
      "grad_norm": 2.5901378575830027,
      "kl": 0.6184326171875,
      "learning_rate": 9.316282404787869e-07,
      "loss": 0.0247,
      "reward": 3.086328125,
      "reward_std": 0.39846538882702587,
      "rewards/accuracy_reward": 0.76640625,
      "rewards/format_reward": 0.958203125,
      "rewards/influence_reward": 0.74375,
      "rewards/len_reward": 0.61796875,
      "step": 55
    },
    {
      "completion_length": 254.9,
      "epoch": 0.5688888888888889,
      "grad_norm": 2.3739337593214387,
      "kl": 0.63668212890625,
      "learning_rate": 9.093124073433462e-07,
      "loss": 0.0255,
      "reward": 3.06328125,
      "reward_std": 0.38780092503875496,
      "rewards/accuracy_reward": 0.768359375,
      "rewards/format_reward": 0.959765625,
      "rewards/influence_reward": 0.73125,
      "rewards/len_reward": 0.60390625,
      "step": 60
    },
    {
      "completion_length": 259.262890625,
      "epoch": 0.6162962962962963,
      "grad_norm": 2.587101111118058,
      "kl": 0.7432861328125,
      "learning_rate": 8.842005554284295e-07,
      "loss": 0.0297,
      "reward": 3.07578125,
      "reward_std": 0.3625458924099803,
      "rewards/accuracy_reward": 0.791796875,
      "rewards/format_reward": 0.962890625,
      "rewards/influence_reward": 0.762890625,
      "rewards/len_reward": 0.558203125,
      "step": 65
    },
    {
      "completion_length": 258.9421875,
      "epoch": 0.6637037037037037,
      "grad_norm": 5.206145849848888,
      "kl": 0.8478271484375,
      "learning_rate": 8.564642241456986e-07,
      "loss": 0.0339,
      "reward": 3.125,
      "reward_std": 0.3845095796510577,
      "rewards/accuracy_reward": 0.8015625,
      "rewards/format_reward": 0.9671875,
      "rewards/influence_reward": 0.76484375,
      "rewards/len_reward": 0.59140625,
      "step": 70
    },
    {
      "completion_length": 255.653515625,
      "epoch": 0.7111111111111111,
      "grad_norm": 5.593671379711057,
      "kl": 0.7322998046875,
      "learning_rate": 8.262928807620843e-07,
      "loss": 0.0293,
      "reward": 3.069921875,
      "reward_std": 0.365702610835433,
      "rewards/accuracy_reward": 0.7671875,
      "rewards/format_reward": 0.96015625,
      "rewards/influence_reward": 0.743359375,
      "rewards/len_reward": 0.59921875,
      "step": 75
    },
    {
      "completion_length": 255.75703125,
      "epoch": 0.7585185185185185,
      "grad_norm": 3.480405307563463,
      "kl": 0.75357666015625,
      "learning_rate": 7.938926261462365e-07,
      "loss": 0.0301,
      "reward": 3.03203125,
      "reward_std": 0.37990338783711197,
      "rewards/accuracy_reward": 0.76640625,
      "rewards/format_reward": 0.958984375,
      "rewards/influence_reward": 0.734375,
      "rewards/len_reward": 0.572265625,
      "step": 80
    },
    {
      "completion_length": 254.5109375,
      "epoch": 0.8059259259259259,
      "grad_norm": 3.488858481117881,
      "kl": 0.90634765625,
      "learning_rate": 7.594847868906076e-07,
      "loss": 0.0362,
      "reward": 3.072265625,
      "reward_std": 0.3591079205274582,
      "rewards/accuracy_reward": 0.75859375,
      "rewards/format_reward": 0.96796875,
      "rewards/influence_reward": 0.73203125,
      "rewards/len_reward": 0.613671875,
      "step": 85
    },
    {
      "completion_length": 248.31796875,
      "epoch": 0.8533333333333334,
      "grad_norm": 2.6180252046561625,
      "kl": 0.9040771484375,
      "learning_rate": 7.233044034264033e-07,
      "loss": 0.0362,
      "reward": 3.08515625,
      "reward_std": 0.36243810392916204,
      "rewards/accuracy_reward": 0.778125,
      "rewards/format_reward": 0.956640625,
      "rewards/influence_reward": 0.75078125,
      "rewards/len_reward": 0.599609375,
      "step": 90
    },
    {
      "completion_length": 251.9234375,
      "epoch": 0.9007407407407407,
      "grad_norm": 2.22240578233429,
      "kl": 0.810595703125,
      "learning_rate": 6.855986244591103e-07,
      "loss": 0.0324,
      "reward": 3.100390625,
      "reward_std": 0.3691500350832939,
      "rewards/accuracy_reward": 0.7828125,
      "rewards/format_reward": 0.962109375,
      "rewards/influence_reward": 0.75390625,
      "rewards/len_reward": 0.6015625,
      "step": 95
    },
    {
      "completion_length": 253.403125,
      "epoch": 0.9481481481481482,
      "grad_norm": 2.023848581916178,
      "kl": 0.7365478515625,
      "learning_rate": 6.466250186922324e-07,
      "loss": 0.0295,
      "reward": 3.122265625,
      "reward_std": 0.37429424840956926,
      "rewards/accuracy_reward": 0.7859375,
      "rewards/format_reward": 0.961328125,
      "rewards/influence_reward": 0.7578125,
      "rewards/len_reward": 0.6171875,
      "step": 100
    },
    {
      "completion_length": 248.7359375,
      "epoch": 0.9955555555555555,
      "grad_norm": 3.922514403777451,
      "kl": 0.8494384765625,
      "learning_rate": 6.066498153718734e-07,
      "loss": 0.034,
      "reward": 3.094921875,
      "reward_std": 0.3606603669002652,
      "rewards/accuracy_reward": 0.762890625,
      "rewards/format_reward": 0.972265625,
      "rewards/influence_reward": 0.740234375,
      "rewards/len_reward": 0.61953125,
      "step": 105
    },
    {
      "completion_length": 248.48076923076923,
      "epoch": 1.037925925925926,
      "grad_norm": 2.362598693464187,
      "kl": 1.0575557255244756,
      "learning_rate": 5.659460856710345e-07,
      "loss": 0.0421,
      "reward": 3.0646853146853146,
      "reward_std": 0.38184449826920785,
      "rewards/accuracy_reward": 0.7583041958041958,
      "rewards/format_reward": 0.9632867132867133,
      "rewards/influence_reward": 0.7185314685314685,
      "rewards/len_reward": 0.6245629370629371,
      "step": 110
    },
    {
      "completion_length": 242.28203125,
      "epoch": 1.0853333333333333,
      "grad_norm": 3.91822834491212,
      "kl": 1.124951171875,
      "learning_rate": 5.247918773366111e-07,
      "loss": 0.0451,
      "reward": 3.13359375,
      "reward_std": 0.3510558642446995,
      "rewards/accuracy_reward": 0.771875,
      "rewards/format_reward": 0.96875,
      "rewards/influence_reward": 0.74921875,
      "rewards/len_reward": 0.64375,
      "step": 115
    },
    {
      "completion_length": 240.628515625,
      "epoch": 1.1327407407407408,
      "grad_norm": 277.16445822272766,
      "kl": 1.0545654296875,
      "learning_rate": 4.834683153413459e-07,
      "loss": 0.0422,
      "reward": 3.1484375,
      "reward_std": 0.39152231309562924,
      "rewards/accuracy_reward": 0.78046875,
      "rewards/format_reward": 0.968359375,
      "rewards/influence_reward": 0.74609375,
      "rewards/len_reward": 0.653515625,
      "step": 120
    },
    {
      "completion_length": 241.14921875,
      "epoch": 1.1801481481481482,
      "grad_norm": 2.240406956312204,
      "kl": 0.8530029296875,
      "learning_rate": 4.4225768151520694e-07,
      "loss": 0.0341,
      "reward": 3.0328125,
      "reward_std": 0.39965488519519565,
      "rewards/accuracy_reward": 0.74921875,
      "rewards/format_reward": 0.96484375,
      "rewards/influence_reward": 0.716796875,
      "rewards/len_reward": 0.601953125,
      "step": 125
    },
    {
      "completion_length": 237.059375,
      "epoch": 1.2275555555555555,
      "grad_norm": 2.219711590955101,
      "kl": 0.8648193359375,
      "learning_rate": 4.0144148627425986e-07,
      "loss": 0.0346,
      "reward": 3.112890625,
      "reward_std": 0.3836476540192962,
      "rewards/accuracy_reward": 0.77890625,
      "rewards/format_reward": 0.965625,
      "rewards/influence_reward": 0.74765625,
      "rewards/len_reward": 0.620703125,
      "step": 130
    },
    {
      "completion_length": 234.14453125,
      "epoch": 1.274962962962963,
      "grad_norm": 2.2407662818336984,
      "kl": 0.8214111328125,
      "learning_rate": 3.612985456190778e-07,
      "loss": 0.0329,
      "reward": 3.105859375,
      "reward_std": 0.4161649979650974,
      "rewards/accuracy_reward": 0.778515625,
      "rewards/format_reward": 0.96796875,
      "rewards/influence_reward": 0.75078125,
      "rewards/len_reward": 0.60859375,
      "step": 135
    },
    {
      "completion_length": 236.843359375,
      "epoch": 1.3223703703703704,
      "grad_norm": 2.593597501974838,
      "kl": 0.8451416015625,
      "learning_rate": 3.221030765387417e-07,
      "loss": 0.0338,
      "reward": 3.133203125,
      "reward_std": 0.39431957099586723,
      "rewards/accuracy_reward": 0.776171875,
      "rewards/format_reward": 0.97109375,
      "rewards/influence_reward": 0.7515625,
      "rewards/len_reward": 0.634375,
      "step": 140
    },
    {
      "completion_length": 233.726953125,
      "epoch": 1.3697777777777778,
      "grad_norm": 3.043917358461435,
      "kl": 0.7835205078125,
      "learning_rate": 2.841228238307536e-07,
      "loss": 0.0313,
      "reward": 3.11328125,
      "reward_std": 0.3921460255980492,
      "rewards/accuracy_reward": 0.778515625,
      "rewards/format_reward": 0.96875,
      "rewards/influence_reward": 0.739453125,
      "rewards/len_reward": 0.6265625,
      "step": 145
    },
    {
      "completion_length": 234.728125,
      "epoch": 1.417185185185185,
      "grad_norm": 3.174132829179658,
      "kl": 0.8005859375,
      "learning_rate": 2.476172311325783e-07,
      "loss": 0.032,
      "reward": 3.1015625,
      "reward_std": 0.3549242248758674,
      "rewards/accuracy_reward": 0.758984375,
      "rewards/format_reward": 0.972265625,
      "rewards/influence_reward": 0.725390625,
      "rewards/len_reward": 0.644921875,
      "step": 150
    },
    {
      "completion_length": 234.570703125,
      "epoch": 1.4645925925925927,
      "grad_norm": 2.2903193614930886,
      "kl": 0.7829345703125,
      "learning_rate": 2.128356686585282e-07,
      "loss": 0.0313,
      "reward": 3.09609375,
      "reward_std": 0.359303823299706,
      "rewards/accuracy_reward": 0.7578125,
      "rewards/format_reward": 0.9703125,
      "rewards/influence_reward": 0.73125,
      "rewards/len_reward": 0.63671875,
      "step": 155
    },
    {
      "completion_length": 236.69375,
      "epoch": 1.512,
      "grad_norm": 2.3733278066870964,
      "kl": 0.7982421875,
      "learning_rate": 1.8001572974834168e-07,
      "loss": 0.0319,
      "reward": 3.117578125,
      "reward_std": 0.36292177233844997,
      "rewards/accuracy_reward": 0.768359375,
      "rewards/format_reward": 0.970703125,
      "rewards/influence_reward": 0.75,
      "rewards/len_reward": 0.628515625,
      "step": 160
    },
    {
      "completion_length": 235.683203125,
      "epoch": 1.5594074074074074,
      "grad_norm": 2.1844884573476535,
      "kl": 0.73896484375,
      "learning_rate": 1.493816078637557e-07,
      "loss": 0.0296,
      "reward": 3.10546875,
      "reward_std": 0.38580444287508725,
      "rewards/accuracy_reward": 0.76484375,
      "rewards/format_reward": 0.9703125,
      "rewards/influence_reward": 0.73515625,
      "rewards/len_reward": 0.63515625,
      "step": 165
    },
    {
      "completion_length": 238.08671875,
      "epoch": 1.6068148148148147,
      "grad_norm": 2.8344173780376734,
      "kl": 0.72591552734375,
      "learning_rate": 1.2114256511983274e-07,
      "loss": 0.029,
      "reward": 3.14921875,
      "reward_std": 0.37141652777791023,
      "rewards/accuracy_reward": 0.780078125,
      "rewards/format_reward": 0.9703125,
      "rewards/influence_reward": 0.75234375,
      "rewards/len_reward": 0.646484375,
      "step": 170
    },
    {
      "completion_length": 235.180859375,
      "epoch": 1.6542222222222223,
      "grad_norm": 3.3091191944433844,
      "kl": 0.782958984375,
      "learning_rate": 9.549150281252632e-08,
      "loss": 0.0313,
      "reward": 3.16484375,
      "reward_std": 0.36811935994774103,
      "rewards/accuracy_reward": 0.77890625,
      "rewards/format_reward": 0.971484375,
      "rewards/influence_reward": 0.75,
      "rewards/len_reward": 0.664453125,
      "step": 175
    },
    {
      "completion_length": 239.29921875,
      "epoch": 1.7016296296296296,
      "grad_norm": 1.8304678954270466,
      "kl": 0.7712158203125,
      "learning_rate": 7.260364370723043e-08,
      "loss": 0.0308,
      "reward": 3.130859375,
      "reward_std": 0.3786254977807403,
      "rewards/accuracy_reward": 0.766015625,
      "rewards/format_reward": 0.966796875,
      "rewards/influence_reward": 0.7328125,
      "rewards/len_reward": 0.665234375,
      "step": 180
    },
    {
      "completion_length": 237.4890625,
      "epoch": 1.749037037037037,
      "grad_norm": 1.7368566190889854,
      "kl": 0.78505859375,
      "learning_rate": 5.263533508961826e-08,
      "loss": 0.0314,
      "reward": 3.069140625,
      "reward_std": 0.367980245873332,
      "rewards/accuracy_reward": 0.753125,
      "rewards/format_reward": 0.96484375,
      "rewards/influence_reward": 0.726171875,
      "rewards/len_reward": 0.625,
      "step": 185
    },
    {
      "completion_length": 236.05078125,
      "epoch": 1.7964444444444445,
      "grad_norm": 2.453068026223167,
      "kl": 0.7768310546875,
      "learning_rate": 3.572298075514652e-08,
      "loss": 0.0311,
      "reward": 3.134765625,
      "reward_std": 0.3666324861347675,
      "rewards/accuracy_reward": 0.769140625,
      "rewards/format_reward": 0.96875,
      "rewards/influence_reward": 0.74765625,
      "rewards/len_reward": 0.64921875,
      "step": 190
    },
    {
      "completion_length": 235.984765625,
      "epoch": 1.8438518518518519,
      "grad_norm": 2.2804450209595295,
      "kl": 0.807861328125,
      "learning_rate": 2.1982109232821176e-08,
      "loss": 0.0323,
      "reward": 3.1125,
      "reward_std": 0.4026956457644701,
      "rewards/accuracy_reward": 0.771484375,
      "rewards/format_reward": 0.966796875,
      "rewards/influence_reward": 0.7421875,
      "rewards/len_reward": 0.63203125,
      "step": 195
    },
    {
      "completion_length": 238.119140625,
      "epoch": 1.8912592592592592,
      "grad_norm": 1.8523977240035852,
      "kl": 0.7931640625,
      "learning_rate": 1.1506584608200364e-08,
      "loss": 0.0317,
      "reward": 3.0828125,
      "reward_std": 0.357252630777657,
      "rewards/accuracy_reward": 0.765234375,
      "rewards/format_reward": 0.96875,
      "rewards/influence_reward": 0.734375,
      "rewards/len_reward": 0.614453125,
      "step": 200
    },
    {
      "completion_length": 242.7140625,
      "epoch": 1.9386666666666668,
      "grad_norm": 1.7926737038649005,
      "kl": 0.808837890625,
      "learning_rate": 4.367965336512403e-09,
      "loss": 0.0324,
      "reward": 3.124609375,
      "reward_std": 0.3813207222148776,
      "rewards/accuracy_reward": 0.777734375,
      "rewards/format_reward": 0.96328125,
      "rewards/influence_reward": 0.737109375,
      "rewards/len_reward": 0.646484375,
      "step": 205
    },
    {
      "completion_length": 238.176953125,
      "epoch": 1.986074074074074,
      "grad_norm": 2.6711903628121485,
      "kl": 0.79169921875,
      "learning_rate": 6.150154258476314e-10,
      "loss": 0.0317,
      "reward": 3.097265625,
      "reward_std": 0.36863092239946127,
      "rewards/accuracy_reward": 0.7625,
      "rewards/format_reward": 0.973046875,
      "rewards/influence_reward": 0.739453125,
      "rewards/len_reward": 0.622265625,
      "step": 210
    },
    {
      "completion_length": 240.27393617021278,
      "epoch": 2.0,
      "kl": 0.7930518617021277,
      "reward": 3.098404255319149,
      "reward_std": 0.3825969715067681,
      "rewards/accuracy_reward": 0.7726063829787234,
      "rewards/format_reward": 0.9707446808510638,
      "rewards/influence_reward": 0.7353723404255319,
      "rewards/len_reward": 0.6196808510638298,
      "step": 212,
      "total_flos": 0.0,
      "train_loss": 0.030741853463762212,
      "train_runtime": 151347.1984,
      "train_samples_per_second": 0.178,
      "train_steps_per_second": 0.001
    }
  ],
  "logging_steps": 5,
  "max_steps": 212,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 2,
  "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
}