{
  "best_global_step": null,
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 2.0,
  "eval_steps": 500,
  "global_step": 422,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "completion_length": 343.505859375,
      "epoch": 0.0047484790028194095,
      "grad_norm": 10.941497736666665,
      "kl": 1.436471939086914e-05,
      "learning_rate": 0.0,
      "loss": 0.0,
      "reward": 2.490234375,
      "reward_std": 0.557172168046236,
      "rewards/accuracy_reward": 0.66796875,
      "rewards/format_reward": 0.806640625,
      "rewards/influence_reward": 0.53515625,
      "rewards/len_reward": 0.48046875,
      "step": 1
    },
    {
      "completion_length": 341.50390625,
      "epoch": 0.023742395014097046,
      "grad_norm": 6.812195416096856,
      "kl": 0.0044515132904052734,
      "learning_rate": 9.302325581395349e-08,
      "loss": 0.0002,
      "reward": 2.46826171875,
      "reward_std": 0.5800532991997898,
      "rewards/accuracy_reward": 0.66259765625,
      "rewards/format_reward": 0.8154296875,
      "rewards/influence_reward": 0.537109375,
      "rewards/len_reward": 0.453125,
      "step": 5
    },
    {
      "completion_length": 355.537890625,
      "epoch": 0.04748479002819409,
      "grad_norm": 8.69360815652138,
      "kl": 0.008914566040039063,
      "learning_rate": 2.0930232558139536e-07,
      "loss": 0.0004,
      "reward": 2.44375,
      "reward_std": 0.6262629209086299,
      "rewards/accuracy_reward": 0.653125,
      "rewards/format_reward": 0.804296875,
      "rewards/influence_reward": 0.519140625,
      "rewards/len_reward": 0.4671875,
      "step": 10
    },
    {
      "completion_length": 359.13984375,
      "epoch": 0.07122718504229114,
      "grad_norm": 9.167296678106975,
      "kl": 0.1373016357421875,
      "learning_rate": 3.2558139534883724e-07,
      "loss": 0.0055,
      "reward": 2.444140625,
      "reward_std": 0.5922891769558192,
      "rewards/accuracy_reward": 0.666015625,
      "rewards/format_reward": 0.79921875,
      "rewards/influence_reward": 0.525390625,
      "rewards/len_reward": 0.453515625,
      "step": 15
    },
    {
      "completion_length": 351.53828125,
      "epoch": 0.09496958005638818,
      "grad_norm": 20.494707498797204,
      "kl": 0.620166015625,
      "learning_rate": 4.4186046511627905e-07,
      "loss": 0.0248,
      "reward": 2.49609375,
      "reward_std": 0.6080626923590898,
      "rewards/accuracy_reward": 0.655078125,
      "rewards/format_reward": 0.81171875,
      "rewards/influence_reward": 0.533984375,
      "rewards/len_reward": 0.4953125,
      "step": 20
    },
    {
      "completion_length": 350.94140625,
      "epoch": 0.11871197507048524,
      "grad_norm": 58.39828164144229,
      "kl": 4.012548828125,
      "learning_rate": 5.581395348837209e-07,
      "loss": 0.1604,
      "reward": 2.5171875,
      "reward_std": 0.5175810419023037,
      "rewards/accuracy_reward": 0.652734375,
      "rewards/format_reward": 0.86015625,
      "rewards/influence_reward": 0.5484375,
      "rewards/len_reward": 0.455859375,
      "step": 25
    },
    {
      "completion_length": 344.183203125,
      "epoch": 0.14245437008458228,
      "grad_norm": 10.671491690481664,
      "kl": 1.959765625,
      "learning_rate": 6.744186046511627e-07,
      "loss": 0.0784,
      "reward": 2.559375,
      "reward_std": 0.49339695163071157,
      "rewards/accuracy_reward": 0.635546875,
      "rewards/format_reward": 0.8859375,
      "rewards/influence_reward": 0.553515625,
      "rewards/len_reward": 0.484375,
      "step": 30
    },
    {
      "completion_length": 312.225390625,
      "epoch": 0.16619676509867934,
      "grad_norm": 6.830335531593763,
      "kl": 1.739990234375,
      "learning_rate": 7.906976744186046e-07,
      "loss": 0.0696,
      "reward": 2.652734375,
      "reward_std": 0.4439069716259837,
      "rewards/accuracy_reward": 0.637109375,
      "rewards/format_reward": 0.9421875,
      "rewards/influence_reward": 0.56328125,
      "rewards/len_reward": 0.51015625,
      "step": 35
    },
    {
      "completion_length": 284.68359375,
      "epoch": 0.18993916011277637,
      "grad_norm": 14.080133543924038,
      "kl": 1.54609375,
      "learning_rate": 9.069767441860464e-07,
      "loss": 0.0619,
      "reward": 2.736328125,
      "reward_std": 0.48739379905164243,
      "rewards/accuracy_reward": 0.658203125,
      "rewards/format_reward": 0.94453125,
      "rewards/influence_reward": 0.5828125,
      "rewards/len_reward": 0.55078125,
      "step": 40
    },
    {
      "completion_length": 276.867578125,
      "epoch": 0.21368155512687342,
      "grad_norm": 35.20677672719882,
      "kl": 1.3846435546875,
      "learning_rate": 9.999828225438954e-07,
      "loss": 0.0554,
      "reward": 2.808203125,
      "reward_std": 0.46657639369368553,
      "rewards/accuracy_reward": 0.635546875,
      "rewards/format_reward": 0.954296875,
      "rewards/influence_reward": 0.573046875,
      "rewards/len_reward": 0.6453125,
      "step": 45
    },
    {
      "completion_length": 273.9359375,
      "epoch": 0.23742395014097048,
      "grad_norm": 3.603130834120874,
      "kl": 1.1903564453125,
      "learning_rate": 9.993817354984484e-07,
      "loss": 0.0476,
      "reward": 2.736328125,
      "reward_std": 0.4892417475581169,
      "rewards/accuracy_reward": 0.63984375,
      "rewards/format_reward": 0.940625,
      "rewards/influence_reward": 0.554296875,
      "rewards/len_reward": 0.6015625,
      "step": 50
    },
    {
      "completion_length": 271.577734375,
      "epoch": 0.26116634515506754,
      "grad_norm": 2.2725600898834406,
      "kl": 1.1795166015625,
      "learning_rate": 9.979229555432882e-07,
      "loss": 0.0472,
      "reward": 2.798828125,
      "reward_std": 0.46620402578264475,
      "rewards/accuracy_reward": 0.65,
      "rewards/format_reward": 0.951171875,
      "rewards/influence_reward": 0.5890625,
      "rewards/len_reward": 0.60859375,
      "step": 55
    },
    {
      "completion_length": 286.709375,
      "epoch": 0.28490874016916456,
      "grad_norm": 4.953405517687021,
      "kl": 1.2238525390625,
      "learning_rate": 9.95608988146948e-07,
      "loss": 0.0489,
      "reward": 2.83984375,
      "reward_std": 0.47557592410594224,
      "rewards/accuracy_reward": 0.676953125,
      "rewards/format_reward": 0.94140625,
      "rewards/influence_reward": 0.602734375,
      "rewards/len_reward": 0.61875,
      "step": 60
    },
    {
      "completion_length": 277.02578125,
      "epoch": 0.3086511351832616,
      "grad_norm": 3.0787337409013547,
      "kl": 1.3475830078125,
      "learning_rate": 9.924438075705754e-07,
      "loss": 0.0539,
      "reward": 2.82109375,
      "reward_std": 0.45666036382317543,
      "rewards/accuracy_reward": 0.678515625,
      "rewards/format_reward": 0.938671875,
      "rewards/influence_reward": 0.605859375,
      "rewards/len_reward": 0.598046875,
      "step": 65
    },
    {
      "completion_length": 285.7953125,
      "epoch": 0.3323935301973587,
      "grad_norm": 4.0730313066810995,
      "kl": 1.3648681640625,
      "learning_rate": 9.884328500421004e-07,
      "loss": 0.0546,
      "reward": 2.772265625,
      "reward_std": 0.44936057794839146,
      "rewards/accuracy_reward": 0.64375,
      "rewards/format_reward": 0.9421875,
      "rewards/influence_reward": 0.58046875,
      "rewards/len_reward": 0.605859375,
      "step": 70
    },
    {
      "completion_length": 292.43828125,
      "epoch": 0.3561359252114557,
      "grad_norm": 3.1544402674020002,
      "kl": 1.1553466796875,
      "learning_rate": 9.835830044194623e-07,
      "loss": 0.0462,
      "reward": 2.688671875,
      "reward_std": 0.4725625243037939,
      "rewards/accuracy_reward": 0.6109375,
      "rewards/format_reward": 0.9296875,
      "rewards/influence_reward": 0.532421875,
      "rewards/len_reward": 0.615625,
      "step": 75
    },
    {
      "completion_length": 286.590234375,
      "epoch": 0.37987832022555273,
      "grad_norm": 4.765364778587015,
      "kl": 1.1624267578125,
      "learning_rate": 9.779026003589303e-07,
      "loss": 0.0465,
      "reward": 2.708984375,
      "reward_std": 0.4846075477078557,
      "rewards/accuracy_reward": 0.6375,
      "rewards/format_reward": 0.930078125,
      "rewards/influence_reward": 0.556640625,
      "rewards/len_reward": 0.584765625,
      "step": 80
    },
    {
      "completion_length": 290.792578125,
      "epoch": 0.4036207152396498,
      "grad_norm": 9.253671242707203,
      "kl": 1.3268310546875,
      "learning_rate": 9.714013940088387e-07,
      "loss": 0.0531,
      "reward": 2.683203125,
      "reward_std": 0.4955078672617674,
      "rewards/accuracy_reward": 0.630859375,
      "rewards/format_reward": 0.926953125,
      "rewards/influence_reward": 0.55625,
      "rewards/len_reward": 0.569140625,
      "step": 85
    },
    {
      "completion_length": 281.421484375,
      "epoch": 0.42736311025374685,
      "grad_norm": 3.4409983083101086,
      "kl": 1.6419921875,
      "learning_rate": 9.640905512533092e-07,
      "loss": 0.0657,
      "reward": 2.748046875,
      "reward_std": 0.4972141746431589,
      "rewards/accuracy_reward": 0.632421875,
      "rewards/format_reward": 0.945703125,
      "rewards/influence_reward": 0.561328125,
      "rewards/len_reward": 0.60859375,
      "step": 90
    },
    {
      "completion_length": 267.971875,
      "epoch": 0.4511055052678439,
      "grad_norm": 2.2522797459391732,
      "kl": 1.63046875,
      "learning_rate": 9.55982628534738e-07,
      "loss": 0.0652,
      "reward": 2.778125,
      "reward_std": 0.4521504515781999,
      "rewards/accuracy_reward": 0.6265625,
      "rewards/format_reward": 0.943359375,
      "rewards/influence_reward": 0.56796875,
      "rewards/len_reward": 0.640234375,
      "step": 95
    },
    {
      "completion_length": 267.776953125,
      "epoch": 0.47484790028194096,
      "grad_norm": 2.4372708411881794,
      "kl": 1.320361328125,
      "learning_rate": 9.470915512879851e-07,
      "loss": 0.0528,
      "reward": 2.808984375,
      "reward_std": 0.4696818504482508,
      "rewards/accuracy_reward": 0.650390625,
      "rewards/format_reward": 0.941015625,
      "rewards/influence_reward": 0.58515625,
      "rewards/len_reward": 0.632421875,
      "step": 100
    },
    {
      "completion_length": 283.4515625,
      "epoch": 0.498590295296038,
      "grad_norm": 3.8496772212244457,
      "kl": 1.447314453125,
      "learning_rate": 9.374325900233086e-07,
      "loss": 0.0579,
      "reward": 2.8171875,
      "reward_std": 0.45389450453221797,
      "rewards/accuracy_reward": 0.660546875,
      "rewards/format_reward": 0.953515625,
      "rewards/influence_reward": 0.585546875,
      "rewards/len_reward": 0.617578125,
      "step": 105
    },
    {
      "completion_length": 294.027734375,
      "epoch": 0.5223326903101351,
      "grad_norm": 3.309296740246584,
      "kl": 1.67744140625,
      "learning_rate": 9.270223340991146e-07,
      "loss": 0.0671,
      "reward": 2.75546875,
      "reward_std": 0.4890925087034702,
      "rewards/accuracy_reward": 0.646875,
      "rewards/format_reward": 0.95625,
      "rewards/influence_reward": 0.56953125,
      "rewards/len_reward": 0.5828125,
      "step": 110
    },
    {
      "completion_length": 295.9484375,
      "epoch": 0.5460750853242321,
      "grad_norm": 4.3161975783345845,
      "kl": 1.809423828125,
      "learning_rate": 9.158786632295775e-07,
      "loss": 0.0724,
      "reward": 2.72734375,
      "reward_std": 0.46550923883914946,
      "rewards/accuracy_reward": 0.637109375,
      "rewards/format_reward": 0.954296875,
      "rewards/influence_reward": 0.56328125,
      "rewards/len_reward": 0.57265625,
      "step": 115
    },
    {
      "completion_length": 297.998046875,
      "epoch": 0.5698174803383291,
      "grad_norm": 5.658396439699509,
      "kl": 1.731787109375,
      "learning_rate": 9.040207167760585e-07,
      "loss": 0.0693,
      "reward": 2.66875,
      "reward_std": 0.4691320048645139,
      "rewards/accuracy_reward": 0.621875,
      "rewards/format_reward": 0.945703125,
      "rewards/influence_reward": 0.54765625,
      "rewards/len_reward": 0.553515625,
      "step": 120
    },
    {
      "completion_length": 308.9625,
      "epoch": 0.5935598753524262,
      "grad_norm": 2.884522521427685,
      "kl": 1.869189453125,
      "learning_rate": 8.9146886087507e-07,
      "loss": 0.0748,
      "reward": 2.70234375,
      "reward_std": 0.42319196127355097,
      "rewards/accuracy_reward": 0.6296875,
      "rewards/format_reward": 0.946875,
      "rewards/influence_reward": 0.546484375,
      "rewards/len_reward": 0.579296875,
      "step": 125
    },
    {
      "completion_length": 300.675390625,
      "epoch": 0.6173022703665232,
      "grad_norm": 4.353864200081107,
      "kl": 1.741796875,
      "learning_rate": 8.782446534592411e-07,
      "loss": 0.0697,
      "reward": 2.71953125,
      "reward_std": 0.46506577655673026,
      "rewards/accuracy_reward": 0.640234375,
      "rewards/format_reward": 0.943359375,
      "rewards/influence_reward": 0.56171875,
      "rewards/len_reward": 0.57421875,
      "step": 130
    },
    {
      "completion_length": 301.005859375,
      "epoch": 0.6410446653806202,
      "grad_norm": 2.418685025022163,
      "kl": 1.816259765625,
      "learning_rate": 8.643708072313617e-07,
      "loss": 0.0726,
      "reward": 2.6859375,
      "reward_std": 0.4477614754810929,
      "rewards/accuracy_reward": 0.6265625,
      "rewards/format_reward": 0.953125,
      "rewards/influence_reward": 0.546484375,
      "rewards/len_reward": 0.559765625,
      "step": 135
    },
    {
      "completion_length": 299.0046875,
      "epoch": 0.6647870603947174,
      "grad_norm": 2.3934470580786984,
      "kl": 1.63916015625,
      "learning_rate": 8.498711506550983e-07,
      "loss": 0.0656,
      "reward": 2.665234375,
      "reward_std": 0.42022476736456155,
      "rewards/accuracy_reward": 0.593359375,
      "rewards/format_reward": 0.965234375,
      "rewards/influence_reward": 0.522265625,
      "rewards/len_reward": 0.584375,
      "step": 140
    },
    {
      "completion_length": 310.663671875,
      "epoch": 0.6885294554088144,
      "grad_norm": 5.124954589573938,
      "kl": 1.480908203125,
      "learning_rate": 8.347705870293799e-07,
      "loss": 0.0592,
      "reward": 2.5953125,
      "reward_std": 0.466478811763227,
      "rewards/accuracy_reward": 0.591796875,
      "rewards/format_reward": 0.955859375,
      "rewards/influence_reward": 0.511328125,
      "rewards/len_reward": 0.536328125,
      "step": 145
    },
    {
      "completion_length": 304.512109375,
      "epoch": 0.7122718504229114,
      "grad_norm": 2.7544392716511696,
      "kl": 1.58447265625,
      "learning_rate": 8.190950517167437e-07,
      "loss": 0.0634,
      "reward": 2.71015625,
      "reward_std": 0.44989644438028337,
      "rewards/accuracy_reward": 0.643359375,
      "rewards/format_reward": 0.955859375,
      "rewards/influence_reward": 0.56171875,
      "rewards/len_reward": 0.54921875,
      "step": 150
    },
    {
      "completion_length": 307.515625,
      "epoch": 0.7360142454370084,
      "grad_norm": 3.2720338878640036,
      "kl": 1.801708984375,
      "learning_rate": 8.028714675991006e-07,
      "loss": 0.0721,
      "reward": 2.696875,
      "reward_std": 0.463063583150506,
      "rewards/accuracy_reward": 0.639453125,
      "rewards/format_reward": 0.953515625,
      "rewards/influence_reward": 0.552734375,
      "rewards/len_reward": 0.551171875,
      "step": 155
    },
    {
      "completion_length": 306.36328125,
      "epoch": 0.7597566404511055,
      "grad_norm": 4.11327264014585,
      "kl": 1.860302734375,
      "learning_rate": 7.861276988374301e-07,
      "loss": 0.0744,
      "reward": 2.70234375,
      "reward_std": 0.4653250968083739,
      "rewards/accuracy_reward": 0.632421875,
      "rewards/format_reward": 0.955078125,
      "rewards/influence_reward": 0.55078125,
      "rewards/len_reward": 0.5640625,
      "step": 160
    },
    {
      "completion_length": 299.812890625,
      "epoch": 0.7834990354652025,
      "grad_norm": 4.8236586704402855,
      "kl": 2.14326171875,
      "learning_rate": 7.688925030148148e-07,
      "loss": 0.0857,
      "reward": 2.673828125,
      "reward_std": 0.44505598582327366,
      "rewards/accuracy_reward": 0.621875,
      "rewards/format_reward": 0.959765625,
      "rewards/influence_reward": 0.556640625,
      "rewards/len_reward": 0.535546875,
      "step": 165
    },
    {
      "completion_length": 305.41484375,
      "epoch": 0.8072414304792996,
      "grad_norm": 1.849946965836104,
      "kl": 2.1720703125,
      "learning_rate": 7.51195481745018e-07,
      "loss": 0.0869,
      "reward": 2.708984375,
      "reward_std": 0.43415932413190605,
      "rewards/accuracy_reward": 0.62265625,
      "rewards/format_reward": 0.958203125,
      "rewards/influence_reward": 0.56171875,
      "rewards/len_reward": 0.56640625,
      "step": 170
    },
    {
      "completion_length": 301.258203125,
      "epoch": 0.8309838254933967,
      "grad_norm": 2.440806378569226,
      "kl": 1.9740234375,
      "learning_rate": 7.330670298314281e-07,
      "loss": 0.0789,
      "reward": 2.70859375,
      "reward_std": 0.46049848962575196,
      "rewards/accuracy_reward": 0.62734375,
      "rewards/format_reward": 0.951953125,
      "rewards/influence_reward": 0.538671875,
      "rewards/len_reward": 0.590625,
      "step": 175
    },
    {
      "completion_length": 301.55,
      "epoch": 0.8547262205074937,
      "grad_norm": 5.1360667348261915,
      "kl": 2.40302734375,
      "learning_rate": 7.145382830636924e-07,
      "loss": 0.0962,
      "reward": 2.7078125,
      "reward_std": 0.44812944661825893,
      "rewards/accuracy_reward": 0.625390625,
      "rewards/format_reward": 0.959375,
      "rewards/influence_reward": 0.553515625,
      "rewards/len_reward": 0.56953125,
      "step": 180
    },
    {
      "completion_length": 296.28515625,
      "epoch": 0.8784686155215907,
      "grad_norm": 3.0756451628776627,
      "kl": 2.0728515625,
      "learning_rate": 6.956410647417016e-07,
      "loss": 0.0829,
      "reward": 2.679296875,
      "reward_std": 0.44053732715547084,
      "rewards/accuracy_reward": 0.624609375,
      "rewards/format_reward": 0.95,
      "rewards/influence_reward": 0.552734375,
      "rewards/len_reward": 0.551953125,
      "step": 185
    },
    {
      "completion_length": 290.825,
      "epoch": 0.9022110105356878,
      "grad_norm": 2.58004505898485,
      "kl": 1.95107421875,
      "learning_rate": 6.764078310187667e-07,
      "loss": 0.078,
      "reward": 2.743359375,
      "reward_std": 0.45461403094232083,
      "rewards/accuracy_reward": 0.62734375,
      "rewards/format_reward": 0.9625,
      "rewards/influence_reward": 0.560546875,
      "rewards/len_reward": 0.59296875,
      "step": 190
    },
    {
      "completion_length": 301.6453125,
      "epoch": 0.9259534055497848,
      "grad_norm": 3.359431223961218,
      "kl": 1.895068359375,
      "learning_rate": 6.568716151578653e-07,
      "loss": 0.0758,
      "reward": 2.6140625,
      "reward_std": 0.45234644059091805,
      "rewards/accuracy_reward": 0.591015625,
      "rewards/format_reward": 0.95703125,
      "rewards/influence_reward": 0.521875,
      "rewards/len_reward": 0.544140625,
      "step": 195
    },
    {
      "completion_length": 300.809765625,
      "epoch": 0.9496958005638819,
      "grad_norm": 2.072678305669791,
      "kl": 1.892919921875,
      "learning_rate": 6.370659707966967e-07,
      "loss": 0.0757,
      "reward": 2.6984375,
      "reward_std": 0.4385804546996951,
      "rewards/accuracy_reward": 0.602734375,
      "rewards/format_reward": 0.957421875,
      "rewards/influence_reward": 0.5296875,
      "rewards/len_reward": 0.60859375,
      "step": 200
    },
    {
      "completion_length": 295.50078125,
      "epoch": 0.973438195577979,
      "grad_norm": 2.641906292114807,
      "kl": 1.882763671875,
      "learning_rate": 6.170249143189878e-07,
      "loss": 0.0753,
      "reward": 2.726953125,
      "reward_std": 0.45993594732135534,
      "rewards/accuracy_reward": 0.61875,
      "rewards/format_reward": 0.965625,
      "rewards/influence_reward": 0.553515625,
      "rewards/len_reward": 0.5890625,
      "step": 205
    },
    {
      "completion_length": 294.291796875,
      "epoch": 0.997180590592076,
      "grad_norm": 3.1118319118270263,
      "kl": 1.721240234375,
      "learning_rate": 5.967828664310282e-07,
      "loss": 0.0689,
      "reward": 2.689453125,
      "reward_std": 0.4471969947218895,
      "rewards/accuracy_reward": 0.618359375,
      "rewards/format_reward": 0.96171875,
      "rewards/influence_reward": 0.556640625,
      "rewards/len_reward": 0.552734375,
      "step": 210
    },
    {
      "completion_length": 298.9141156462585,
      "epoch": 1.0189939160112775,
      "grad_norm": 1.4510607380023717,
      "kl": 1.6323341836734695,
      "learning_rate": 5.763745930437786e-07,
      "loss": 0.0657,
      "reward": 2.7891156462585034,
      "reward_std": 0.4109960768093057,
      "rewards/accuracy_reward": 0.6547619047619048,
      "rewards/format_reward": 0.9596088435374149,
      "rewards/influence_reward": 0.5778061224489796,
      "rewards/len_reward": 0.5969387755102041,
      "step": 215
    },
    {
      "completion_length": 292.062109375,
      "epoch": 1.0427363110253747,
      "grad_norm": 2.3079311627757115,
      "kl": 1.639599609375,
      "learning_rate": 5.558351455620851e-07,
      "loss": 0.0656,
      "reward": 2.71171875,
      "reward_std": 0.40011929515749217,
      "rewards/accuracy_reward": 0.61640625,
      "rewards/format_reward": 0.9703125,
      "rewards/influence_reward": 0.537890625,
      "rewards/len_reward": 0.587109375,
      "step": 220
    },
    {
      "completion_length": 297.239453125,
      "epoch": 1.0664787060394718,
      "grad_norm": 1.8840863111335822,
      "kl": 2.28310546875,
      "learning_rate": 5.351998006835562e-07,
      "loss": 0.0914,
      "reward": 2.684375,
      "reward_std": 0.45486766248941424,
      "rewards/accuracy_reward": 0.603515625,
      "rewards/format_reward": 0.962890625,
      "rewards/influence_reward": 0.537890625,
      "rewards/len_reward": 0.580078125,
      "step": 225
    },
    {
      "completion_length": 288.18125,
      "epoch": 1.0902211010535687,
      "grad_norm": 2.0051543482646657,
      "kl": 1.657861328125,
      "learning_rate": 5.145039998104975e-07,
      "loss": 0.0663,
      "reward": 2.69765625,
      "reward_std": 0.4361386923119426,
      "rewards/accuracy_reward": 0.622265625,
      "rewards/format_reward": 0.969140625,
      "rewards/influence_reward": 0.552734375,
      "rewards/len_reward": 0.553515625,
      "step": 230
    },
    {
      "completion_length": 287.385546875,
      "epoch": 1.1139634960676659,
      "grad_norm": 1.4748593771929919,
      "kl": 1.78935546875,
      "learning_rate": 4.937832881789634e-07,
      "loss": 0.0716,
      "reward": 2.787109375,
      "reward_std": 0.4470808956772089,
      "rewards/accuracy_reward": 0.64296875,
      "rewards/format_reward": 0.969921875,
      "rewards/influence_reward": 0.569921875,
      "rewards/len_reward": 0.604296875,
      "step": 235
    },
    {
      "completion_length": 288.096484375,
      "epoch": 1.1377058910817628,
      "grad_norm": 10.954523562177771,
      "kl": 1.878369140625,
      "learning_rate": 4.730732538094749e-07,
      "loss": 0.0751,
      "reward": 2.716796875,
      "reward_std": 0.4533087441697717,
      "rewards/accuracy_reward": 0.608203125,
      "rewards/format_reward": 0.96484375,
      "rewards/influence_reward": 0.543359375,
      "rewards/len_reward": 0.600390625,
      "step": 240
    },
    {
      "completion_length": 286.948828125,
      "epoch": 1.16144828609586,
      "grad_norm": 3.7350800690768913,
      "kl": 1.802978515625,
      "learning_rate": 4.524094663842527e-07,
      "loss": 0.0721,
      "reward": 2.716015625,
      "reward_std": 0.46027831770479677,
      "rewards/accuracy_reward": 0.605078125,
      "rewards/format_reward": 0.96640625,
      "rewards/influence_reward": 0.529296875,
      "rewards/len_reward": 0.615234375,
      "step": 245
    },
    {
      "completion_length": 298.440234375,
      "epoch": 1.1851906811099568,
      "grad_norm": 1.7998199572787943,
      "kl": 1.9130859375,
      "learning_rate": 4.3182741615594863e-07,
      "loss": 0.0765,
      "reward": 2.73046875,
      "reward_std": 0.4339360993355513,
      "rewards/accuracy_reward": 0.629296875,
      "rewards/format_reward": 0.96875,
      "rewards/influence_reward": 0.55546875,
      "rewards/len_reward": 0.576953125,
      "step": 250
    },
    {
      "completion_length": 297.07421875,
      "epoch": 1.208933076124054,
      "grad_norm": 2.1433576353118755,
      "kl": 2.023828125,
      "learning_rate": 4.1136245299279625e-07,
      "loss": 0.081,
      "reward": 2.69375,
      "reward_std": 0.4361835679039359,
      "rewards/accuracy_reward": 0.6046875,
      "rewards/format_reward": 0.962109375,
      "rewards/influence_reward": 0.5265625,
      "rewards/len_reward": 0.600390625,
      "step": 255
    },
    {
      "completion_length": 283.55859375,
      "epoch": 1.2326754711381511,
      "grad_norm": 3.56858833503255,
      "kl": 1.9943359375,
      "learning_rate": 3.910497256648742e-07,
      "loss": 0.0798,
      "reward": 2.776171875,
      "reward_std": 0.41840124800801276,
      "rewards/accuracy_reward": 0.62421875,
      "rewards/format_reward": 0.975,
      "rewards/influence_reward": 0.55703125,
      "rewards/len_reward": 0.619921875,
      "step": 260
    },
    {
      "completion_length": 285.401953125,
      "epoch": 1.256417866152248,
      "grad_norm": 3.0202540131791316,
      "kl": 1.812353515625,
      "learning_rate": 3.7092412147575756e-07,
      "loss": 0.0725,
      "reward": 2.667578125,
      "reward_std": 0.4479133376851678,
      "rewards/accuracy_reward": 0.5859375,
      "rewards/format_reward": 0.97109375,
      "rewards/influence_reward": 0.51328125,
      "rewards/len_reward": 0.597265625,
      "step": 265
    },
    {
      "completion_length": 296.51953125,
      "epoch": 1.2801602611663452,
      "grad_norm": 6.685851651155463,
      "kl": 1.824169921875,
      "learning_rate": 3.510202063432397e-07,
      "loss": 0.073,
      "reward": 2.68125,
      "reward_std": 0.45598011631518603,
      "rewards/accuracy_reward": 0.59453125,
      "rewards/format_reward": 0.97109375,
      "rewards/influence_reward": 0.526171875,
      "rewards/len_reward": 0.589453125,
      "step": 270
    },
    {
      "completion_length": 297.57734375,
      "epoch": 1.3039026561804423,
      "grad_norm": 2.291528264931079,
      "kl": 1.847607421875,
      "learning_rate": 3.3137216543203747e-07,
      "loss": 0.0739,
      "reward": 2.69765625,
      "reward_std": 0.41086965166032313,
      "rewards/accuracy_reward": 0.59140625,
      "rewards/format_reward": 0.9765625,
      "rewards/influence_reward": 0.53046875,
      "rewards/len_reward": 0.59921875,
      "step": 275
    },
    {
      "completion_length": 290.746875,
      "epoch": 1.3276450511945392,
      "grad_norm": 1.8784141288314866,
      "kl": 1.92822265625,
      "learning_rate": 3.1201374444044414e-07,
      "loss": 0.0771,
      "reward": 2.68671875,
      "reward_std": 0.43726298715919254,
      "rewards/accuracy_reward": 0.606640625,
      "rewards/format_reward": 0.96875,
      "rewards/influence_reward": 0.5296875,
      "rewards/len_reward": 0.581640625,
      "step": 280
    },
    {
      "completion_length": 285.90390625,
      "epoch": 1.3513874462086364,
      "grad_norm": 3.8666864047404657,
      "kl": 2.049609375,
      "learning_rate": 2.929781916417696e-07,
      "loss": 0.082,
      "reward": 2.79140625,
      "reward_std": 0.4504884161055088,
      "rewards/accuracy_reward": 0.635546875,
      "rewards/format_reward": 0.97578125,
      "rewards/influence_reward": 0.56875,
      "rewards/len_reward": 0.611328125,
      "step": 285
    },
    {
      "completion_length": 278.126171875,
      "epoch": 1.3751298412227333,
      "grad_norm": 2.7940015666308584,
      "kl": 1.9935546875,
      "learning_rate": 2.742982007801121e-07,
      "loss": 0.0797,
      "reward": 2.714453125,
      "reward_std": 0.40985080897808074,
      "rewards/accuracy_reward": 0.605859375,
      "rewards/format_reward": 0.97578125,
      "rewards/influence_reward": 0.553125,
      "rewards/len_reward": 0.5796875,
      "step": 290
    },
    {
      "completion_length": 282.807421875,
      "epoch": 1.3988722362368304,
      "grad_norm": 2.1190543029510374,
      "kl": 2.0150390625,
      "learning_rate": 2.5600585491854113e-07,
      "loss": 0.0806,
      "reward": 2.736328125,
      "reward_std": 0.4611986864358187,
      "rewards/accuracy_reward": 0.625,
      "rewards/format_reward": 0.966015625,
      "rewards/influence_reward": 0.5453125,
      "rewards/len_reward": 0.6,
      "step": 295
    },
    {
      "completion_length": 285.337890625,
      "epoch": 1.4226146312509274,
      "grad_norm": 2.0524239671070235,
      "kl": 2.023779296875,
      "learning_rate": 2.3813257133612823e-07,
      "loss": 0.081,
      "reward": 2.7375,
      "reward_std": 0.4188694503158331,
      "rewards/accuracy_reward": 0.615625,
      "rewards/format_reward": 0.96796875,
      "rewards/influence_reward": 0.546484375,
      "rewards/len_reward": 0.607421875,
      "step": 300
    },
    {
      "completion_length": 280.583203125,
      "epoch": 1.4463570262650245,
      "grad_norm": 2.4639075460957103,
      "kl": 1.916455078125,
      "learning_rate": 2.207090475684702e-07,
      "loss": 0.0766,
      "reward": 2.778515625,
      "reward_std": 0.4049776840955019,
      "rewards/accuracy_reward": 0.621484375,
      "rewards/format_reward": 0.978125,
      "rewards/influence_reward": 0.55703125,
      "rewards/len_reward": 0.621875,
      "step": 305
    },
    {
      "completion_length": 279.21640625,
      "epoch": 1.4700994212791216,
      "grad_norm": 1.9580691543159996,
      "kl": 1.886328125,
      "learning_rate": 2.0376520868437836e-07,
      "loss": 0.0754,
      "reward": 2.69140625,
      "reward_std": 0.4295359971001744,
      "rewards/accuracy_reward": 0.584765625,
      "rewards/format_reward": 0.9734375,
      "rewards/influence_reward": 0.512109375,
      "rewards/len_reward": 0.62109375,
      "step": 310
    },
    {
      "completion_length": 287.18671875,
      "epoch": 1.4938418162932185,
      "grad_norm": 2.334280950294775,
      "kl": 1.902001953125,
      "learning_rate": 1.8733015588928547e-07,
      "loss": 0.0761,
      "reward": 2.748046875,
      "reward_std": 0.4672464972361922,
      "rewards/accuracy_reward": 0.620703125,
      "rewards/format_reward": 0.979296875,
      "rewards/influence_reward": 0.544140625,
      "rewards/len_reward": 0.60390625,
      "step": 315
    },
    {
      "completion_length": 279.612109375,
      "epoch": 1.5175842113073155,
      "grad_norm": 2.383804496413666,
      "kl": 1.881494140625,
      "learning_rate": 1.7143211654364758e-07,
      "loss": 0.0753,
      "reward": 2.704296875,
      "reward_std": 0.46042243372648956,
      "rewards/accuracy_reward": 0.609375,
      "rewards/format_reward": 0.97578125,
      "rewards/influence_reward": 0.540234375,
      "rewards/len_reward": 0.57890625,
      "step": 320
    },
    {
      "completion_length": 276.201171875,
      "epoch": 1.5413266063214128,
      "grad_norm": 1.8621036607786687,
      "kl": 2.042041015625,
      "learning_rate": 1.5609839568218247e-07,
      "loss": 0.0817,
      "reward": 2.78984375,
      "reward_std": 0.4288489479571581,
      "rewards/accuracy_reward": 0.640625,
      "rewards/format_reward": 0.97890625,
      "rewards/influence_reward": 0.5796875,
      "rewards/len_reward": 0.590625,
      "step": 325
    },
    {
      "completion_length": 285.061328125,
      "epoch": 1.5650690013355097,
      "grad_norm": 2.541259800858008,
      "kl": 2.11279296875,
      "learning_rate": 1.413553291172106e-07,
      "loss": 0.0845,
      "reward": 2.744140625,
      "reward_std": 0.43316187243908644,
      "rewards/accuracy_reward": 0.624609375,
      "rewards/format_reward": 0.961328125,
      "rewards/influence_reward": 0.56328125,
      "rewards/len_reward": 0.594921875,
      "step": 330
    },
    {
      "completion_length": 286.321875,
      "epoch": 1.5888113963496067,
      "grad_norm": 2.4383264225155825,
      "kl": 2.04287109375,
      "learning_rate": 1.2722823820664575e-07,
      "loss": 0.0817,
      "reward": 2.7609375,
      "reward_std": 0.42981729917228223,
      "rewards/accuracy_reward": 0.623828125,
      "rewards/format_reward": 0.973046875,
      "rewards/influence_reward": 0.549609375,
      "rewards/len_reward": 0.614453125,
      "step": 335
    },
    {
      "completion_length": 281.6328125,
      "epoch": 1.6125537913637038,
      "grad_norm": 2.1689228150160225,
      "kl": 2.00263671875,
      "learning_rate": 1.1374138636432051e-07,
      "loss": 0.0801,
      "reward": 2.738671875,
      "reward_std": 0.4134707026183605,
      "rewards/accuracy_reward": 0.61015625,
      "rewards/format_reward": 0.97421875,
      "rewards/influence_reward": 0.54609375,
      "rewards/len_reward": 0.608203125,
      "step": 340
    },
    {
      "completion_length": 292.502734375,
      "epoch": 1.636296186377801,
      "grad_norm": 4.449673329252311,
      "kl": 1.98564453125,
      "learning_rate": 1.0091793738734111e-07,
      "loss": 0.0794,
      "reward": 2.7078125,
      "reward_std": 0.4420934099704027,
      "rewards/accuracy_reward": 0.5953125,
      "rewards/format_reward": 0.9703125,
      "rewards/influence_reward": 0.52890625,
      "rewards/len_reward": 0.61328125,
      "step": 345
    },
    {
      "completion_length": 284.086328125,
      "epoch": 1.6600385813918979,
      "grad_norm": 2.2530249348579403,
      "kl": 1.920751953125,
      "learning_rate": 8.877991567204351e-08,
      "loss": 0.0768,
      "reward": 2.77578125,
      "reward_std": 0.40416364911943675,
      "rewards/accuracy_reward": 0.6421875,
      "rewards/format_reward": 0.97265625,
      "rewards/influence_reward": 0.559765625,
      "rewards/len_reward": 0.601171875,
      "step": 350
    },
    {
      "completion_length": 286.23359375,
      "epoch": 1.683780976405995,
      "grad_norm": 11.607212223835898,
      "kl": 1.99482421875,
      "learning_rate": 7.734816838688246e-08,
      "loss": 0.0798,
      "reward": 2.7375,
      "reward_std": 0.42118714693933723,
      "rewards/accuracy_reward": 0.609375,
      "rewards/format_reward": 0.972265625,
      "rewards/influence_reward": 0.54375,
      "rewards/len_reward": 0.612109375,
      "step": 355
    },
    {
      "completion_length": 288.31875,
      "epoch": 1.7075233714200921,
      "grad_norm": 2.560035528567494,
      "kl": 1.884912109375,
      "learning_rate": 6.664232966721995e-08,
      "loss": 0.0754,
      "reward": 2.655078125,
      "reward_std": 0.44481576029211284,
      "rewards/accuracy_reward": 0.591796875,
      "rewards/format_reward": 0.9671875,
      "rewards/influence_reward": 0.51796875,
      "rewards/len_reward": 0.578125,
      "step": 360
    },
    {
      "completion_length": 292.05546875,
      "epoch": 1.731265766434189,
      "grad_norm": 2.703658020341151,
      "kl": 1.88125,
      "learning_rate": 5.668078689351008e-08,
      "loss": 0.0753,
      "reward": 2.731640625,
      "reward_std": 0.4349971186369658,
      "rewards/accuracy_reward": 0.603515625,
      "rewards/format_reward": 0.97265625,
      "rewards/influence_reward": 0.535546875,
      "rewards/len_reward": 0.619921875,
      "step": 365
    },
    {
      "completion_length": 289.07265625,
      "epoch": 1.755008161448286,
      "grad_norm": 2.15796183633457,
      "kl": 1.920556640625,
      "learning_rate": 4.748064911079669e-08,
      "loss": 0.0768,
      "reward": 2.773046875,
      "reward_std": 0.45684912260621785,
      "rewards/accuracy_reward": 0.634375,
      "rewards/format_reward": 0.96875,
      "rewards/influence_reward": 0.569140625,
      "rewards/len_reward": 0.60078125,
      "step": 370
    },
    {
      "completion_length": 297.7921875,
      "epoch": 1.7787505564623831,
      "grad_norm": 1.9525269307848967,
      "kl": 1.950634765625,
      "learning_rate": 3.905771764376553e-08,
      "loss": 0.078,
      "reward": 2.762890625,
      "reward_std": 0.43305158857256176,
      "rewards/accuracy_reward": 0.616796875,
      "rewards/format_reward": 0.970703125,
      "rewards/influence_reward": 0.544921875,
      "rewards/len_reward": 0.63046875,
      "step": 375
    },
    {
      "completion_length": 286.23125,
      "epoch": 1.8024929514764803,
      "grad_norm": 1.8795557235019384,
      "kl": 1.94345703125,
      "learning_rate": 3.142645895781715e-08,
      "loss": 0.0777,
      "reward": 2.705859375,
      "reward_std": 0.4185318287461996,
      "rewards/accuracy_reward": 0.6078125,
      "rewards/format_reward": 0.969140625,
      "rewards/influence_reward": 0.53125,
      "rewards/len_reward": 0.59765625,
      "step": 380
    },
    {
      "completion_length": 284.5234375,
      "epoch": 1.8262353464905772,
      "grad_norm": 1.8303802044437847,
      "kl": 1.9130859375,
      "learning_rate": 2.459997981277423e-08,
      "loss": 0.0765,
      "reward": 2.743359375,
      "reward_std": 0.42376566473394633,
      "rewards/accuracy_reward": 0.6296875,
      "rewards/format_reward": 0.973828125,
      "rewards/influence_reward": 0.556640625,
      "rewards/len_reward": 0.583203125,
      "step": 385
    },
    {
      "completion_length": 287.645703125,
      "epoch": 1.8499777415046743,
      "grad_norm": 2.1486695226548314,
      "kl": 1.919140625,
      "learning_rate": 1.859000475189687e-08,
      "loss": 0.0767,
      "reward": 2.70625,
      "reward_std": 0.43876812513917685,
      "rewards/accuracy_reward": 0.62109375,
      "rewards/format_reward": 0.965625,
      "rewards/influence_reward": 0.5328125,
      "rewards/len_reward": 0.58671875,
      "step": 390
    },
    {
      "completion_length": 289.16953125,
      "epoch": 1.8737201365187715,
      "grad_norm": 2.3275579827754704,
      "kl": 1.96728515625,
      "learning_rate": 1.3406855964866404e-08,
      "loss": 0.0787,
      "reward": 2.69453125,
      "reward_std": 0.44194721039384605,
      "rewards/accuracy_reward": 0.603125,
      "rewards/format_reward": 0.971875,
      "rewards/influence_reward": 0.533984375,
      "rewards/len_reward": 0.585546875,
      "step": 395
    },
    {
      "completion_length": 285.14296875,
      "epoch": 1.8974625315328684,
      "grad_norm": 2.1178151388074538,
      "kl": 1.926513671875,
      "learning_rate": 9.059435559326256e-09,
      "loss": 0.077,
      "reward": 2.69609375,
      "reward_std": 0.4095570512115955,
      "rewards/accuracy_reward": 0.604296875,
      "rewards/format_reward": 0.9703125,
      "rewards/influence_reward": 0.529296875,
      "rewards/len_reward": 0.5921875,
      "step": 400
    },
    {
      "completion_length": 291.73046875,
      "epoch": 1.9212049265469653,
      "grad_norm": 3.0382797909309063,
      "kl": 1.9125,
      "learning_rate": 5.555210271427191e-09,
      "loss": 0.0765,
      "reward": 2.74375,
      "reward_std": 0.45224274341017007,
      "rewards/accuracy_reward": 0.627734375,
      "rewards/format_reward": 0.968359375,
      "rewards/influence_reward": 0.544140625,
      "rewards/len_reward": 0.603515625,
      "step": 405
    },
    {
      "completion_length": 285.82109375,
      "epoch": 1.9449473215610624,
      "grad_norm": 2.223879198885574,
      "kl": 1.93515625,
      "learning_rate": 2.900198641635876e-09,
      "loss": 0.0774,
      "reward": 2.766015625,
      "reward_std": 0.4194388585165143,
      "rewards/accuracy_reward": 0.62734375,
      "rewards/format_reward": 0.973046875,
      "rewards/influence_reward": 0.558203125,
      "rewards/len_reward": 0.607421875,
      "step": 410
    },
    {
      "completion_length": 284.583203125,
      "epoch": 1.9686897165751596,
      "grad_norm": 2.3711705196769834,
      "kl": 1.942138671875,
      "learning_rate": 1.0989606778344373e-09,
      "loss": 0.0777,
      "reward": 2.782421875,
      "reward_std": 0.42840481735765934,
      "rewards/accuracy_reward": 0.637109375,
      "rewards/format_reward": 0.969140625,
      "rewards/influence_reward": 0.568359375,
      "rewards/len_reward": 0.6078125,
      "step": 415
    },
    {
      "completion_length": 297.46796875,
      "epoch": 1.9924321115892565,
      "grad_norm": 18.12331360764631,
      "kl": 1.966943359375,
      "learning_rate": 1.5459002346324136e-10,
      "loss": 0.0787,
      "reward": 2.726171875,
      "reward_std": 0.4440956773236394,
      "rewards/accuracy_reward": 0.6171875,
      "rewards/format_reward": 0.96875,
      "rewards/influence_reward": 0.5265625,
      "rewards/len_reward": 0.613671875,
      "step": 420
    },
    {
      "completion_length": 282.22058823529414,
      "epoch": 2.0,
      "kl": 1.9358149509803921,
      "reward": 2.8063725490196076,
      "reward_std": 0.42265554269154865,
      "rewards/accuracy_reward": 0.6323529411764706,
      "rewards/format_reward": 0.9816176470588235,
      "rewards/influence_reward": 0.5637254901960784,
      "rewards/len_reward": 0.6286764705882353,
      "step": 422,
      "total_flos": 0.0,
      "train_loss": 0.07021697652106884,
      "train_runtime": 419033.0805,
      "train_samples_per_second": 0.129,
      "train_steps_per_second": 0.001
    }
  ],
  "logging_steps": 5,
  "max_steps": 422,
  "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
}