{
  "best_global_step": null,
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 0.6928406466512702,
  "eval_steps": 500,
  "global_step": 150,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "completion_length": 107.87500381469727,
      "epoch": 0.004618937644341801,
      "grad_norm": 10.236255645751953,
      "kl": 0.0,
      "learning_rate": 1e-06,
      "loss": -0.0,
      "reward": 1.8541667461395264,
      "reward_std": 0.1378917135298252,
      "rewards/accuracy_reward": 0.8541666865348816,
      "rewards/format_reward": 1.0,
      "step": 1
    },
    {
      "completion_length": 96.77083587646484,
      "epoch": 0.009237875288683603,
      "grad_norm": 5.244990348815918,
      "kl": 0.00017547607421875,
      "learning_rate": 1e-06,
      "loss": 0.0,
      "reward": 1.7708333730697632,
      "reward_std": 0.17311252653598785,
      "rewards/accuracy_reward": 0.7708333432674408,
      "rewards/format_reward": 1.0,
      "step": 2
    },
    {
      "completion_length": 120.25,
      "epoch": 0.013856812933025405,
      "grad_norm": 4.743775367736816,
      "kl": 0.0006608963012695312,
      "learning_rate": 1e-06,
      "loss": 0.0,
      "reward": 1.7916666865348816,
      "reward_std": 0.179558377712965,
      "rewards/accuracy_reward": 0.7916666865348816,
      "rewards/format_reward": 1.0,
      "step": 3
    },
    {
      "completion_length": 94.3125,
      "epoch": 0.018475750577367205,
      "grad_norm": 3.1872682571411133,
      "kl": 0.0009775161743164062,
      "learning_rate": 1e-06,
      "loss": 0.0,
      "reward": 1.8125000596046448,
      "reward_std": 0.1378917098045349,
      "rewards/accuracy_reward": 0.8125000298023224,
      "rewards/format_reward": 1.0,
      "step": 4
    },
    {
      "completion_length": 76.95833587646484,
      "epoch": 0.023094688221709007,
      "grad_norm": 5.366634368896484,
      "kl": 0.00531768798828125,
      "learning_rate": 1e-06,
      "loss": 0.0002,
      "reward": 1.8750000596046448,
      "reward_std": 0.1666666716337204,
      "rewards/accuracy_reward": 0.9166666865348816,
      "rewards/format_reward": 0.9583333730697632,
      "step": 5
    },
    {
      "completion_length": 100.64583587646484,
      "epoch": 0.02771362586605081,
      "grad_norm": 3.2625043392181396,
      "kl": 0.0061187744140625,
      "learning_rate": 1e-06,
      "loss": 0.0002,
      "reward": 1.8958333730697632,
      "reward_std": 0.17311251163482666,
      "rewards/accuracy_reward": 0.8958333432674408,
      "rewards/format_reward": 1.0,
      "step": 6
    },
    {
      "completion_length": 100.83333587646484,
      "epoch": 0.03233256351039261,
      "grad_norm": 2.418282985687256,
      "kl": 0.00472259521484375,
      "learning_rate": 1e-06,
      "loss": 0.0002,
      "reward": 1.7708333730697632,
      "reward_std": 0.17311252281069756,
      "rewards/accuracy_reward": 0.7708333730697632,
      "rewards/format_reward": 1.0,
      "step": 7
    },
    {
      "completion_length": 111.625,
      "epoch": 0.03695150115473441,
      "grad_norm": 8.21818733215332,
      "kl": 0.003204345703125,
      "learning_rate": 1e-06,
      "loss": 0.0001,
      "reward": 1.8750000596046448,
      "reward_std": 0.13144585862755775,
      "rewards/accuracy_reward": 0.8750000298023224,
      "rewards/format_reward": 1.0,
      "step": 8
    },
    {
      "completion_length": 93.5,
      "epoch": 0.04157043879907621,
      "grad_norm": 4.702052593231201,
      "kl": 0.008941650390625,
      "learning_rate": 1e-06,
      "loss": 0.0004,
      "reward": 1.8125000596046448,
      "reward_std": 0.25644585490226746,
      "rewards/accuracy_reward": 0.8125000298023224,
      "rewards/format_reward": 1.0,
      "step": 9
    },
    {
      "completion_length": 148.33333587646484,
      "epoch": 0.046189376443418015,
      "grad_norm": 2.0280299186706543,
      "kl": 0.002277374267578125,
      "learning_rate": 1e-06,
      "loss": 0.0001,
      "reward": 1.8125000596046448,
      "reward_std": 0.18600423261523247,
      "rewards/accuracy_reward": 0.8125000298023224,
      "rewards/format_reward": 1.0,
      "step": 10
    },
    {
      "completion_length": 76.89583587646484,
      "epoch": 0.050808314087759814,
      "grad_norm": 3.9683003425598145,
      "kl": 0.003360748291015625,
      "learning_rate": 1e-06,
      "loss": 0.0001,
      "reward": 1.8541667461395264,
      "reward_std": 0.1250000037252903,
      "rewards/accuracy_reward": 0.8541666865348816,
      "rewards/format_reward": 1.0,
      "step": 11
    },
    {
      "completion_length": 96.02083587646484,
      "epoch": 0.05542725173210162,
      "grad_norm": 10.515838623046875,
      "kl": 0.01108551025390625,
      "learning_rate": 1e-06,
      "loss": 0.0004,
      "reward": 1.7083333730697632,
      "reward_std": 0.29811252653598785,
      "rewards/accuracy_reward": 0.7083333730697632,
      "rewards/format_reward": 1.0,
      "step": 12
    },
    {
      "completion_length": 75.54166793823242,
      "epoch": 0.06004618937644342,
      "grad_norm": 4.704844951629639,
      "kl": 0.0057830810546875,
      "learning_rate": 1e-06,
      "loss": 0.0002,
      "reward": 1.8958333730697632,
      "reward_std": 0.1250000037252903,
      "rewards/accuracy_reward": 0.8958333432674408,
      "rewards/format_reward": 1.0,
      "step": 13
    },
    {
      "completion_length": 100.41667175292969,
      "epoch": 0.06466512702078522,
      "grad_norm": 7.272250175476074,
      "kl": 0.005889892578125,
      "learning_rate": 1e-06,
      "loss": 0.0002,
      "reward": 1.7916667461395264,
      "reward_std": 0.25,
      "rewards/accuracy_reward": 0.7916666865348816,
      "rewards/format_reward": 1.0,
      "step": 14
    },
    {
      "completion_length": 104.33333587646484,
      "epoch": 0.06928406466512702,
      "grad_norm": 3.6556761264801025,
      "kl": 0.0079803466796875,
      "learning_rate": 1e-06,
      "loss": 0.0003,
      "reward": 1.9375000596046448,
      "reward_std": 0.1250000037252903,
      "rewards/accuracy_reward": 0.9375000298023224,
      "rewards/format_reward": 1.0,
      "step": 15
    },
    {
      "completion_length": 94.54167175292969,
      "epoch": 0.07390300230946882,
      "grad_norm": 4.878111362457275,
      "kl": 0.00466156005859375,
      "learning_rate": 1e-06,
      "loss": 0.0002,
      "reward": 1.7291666865348816,
      "reward_std": 0.17311252281069756,
      "rewards/accuracy_reward": 0.7291666865348816,
      "rewards/format_reward": 1.0,
      "step": 16
    },
    {
      "completion_length": 69.5625,
      "epoch": 0.07852193995381063,
      "grad_norm": 11.191930770874023,
      "kl": 0.0043792724609375,
      "learning_rate": 1e-06,
      "loss": 0.0002,
      "reward": 1.8125000596046448,
      "reward_std": 0.2212250456213951,
      "rewards/accuracy_reward": 0.8125000298023224,
      "rewards/format_reward": 1.0,
      "step": 17
    },
    {
      "completion_length": 89.70833396911621,
      "epoch": 0.08314087759815242,
      "grad_norm": 3.1972787380218506,
      "kl": 0.0032033920288085938,
      "learning_rate": 1e-06,
      "loss": 0.0001,
      "reward": 1.8541666865348816,
      "reward_std": 0.2212250456213951,
      "rewards/accuracy_reward": 0.8541666865348816,
      "rewards/format_reward": 1.0,
      "step": 18
    },
    {
      "completion_length": 81.89583587646484,
      "epoch": 0.08775981524249422,
      "grad_norm": 30.53135108947754,
      "kl": 0.007843017578125,
      "learning_rate": 1e-06,
      "loss": 0.0003,
      "reward": 1.9375000596046448,
      "reward_std": 0.1250000037252903,
      "rewards/accuracy_reward": 0.9375000298023224,
      "rewards/format_reward": 1.0,
      "step": 19
    },
    {
      "completion_length": 115.5625057220459,
      "epoch": 0.09237875288683603,
      "grad_norm": 2.5540764331817627,
      "kl": 0.0017900466918945312,
      "learning_rate": 1e-06,
      "loss": 0.0001,
      "reward": 1.7916666865348816,
      "reward_std": 0.2500000111758709,
      "rewards/accuracy_reward": 0.7916666865348816,
      "rewards/format_reward": 1.0,
      "step": 20
    },
    {
      "completion_length": 107.08333587646484,
      "epoch": 0.09699769053117784,
      "grad_norm": 1.230578899383545,
      "kl": 0.0011348724365234375,
      "learning_rate": 1e-06,
      "loss": 0.0,
      "reward": 1.8750000596046448,
      "reward_std": 0.1666666716337204,
      "rewards/accuracy_reward": 0.8750000298023224,
      "rewards/format_reward": 1.0,
      "step": 21
    },
    {
      "completion_length": 46.52083396911621,
      "epoch": 0.10161662817551963,
      "grad_norm": 6.826656341552734,
      "kl": 0.02459096908569336,
      "learning_rate": 1e-06,
      "loss": 0.001,
      "reward": 1.8958333730697632,
      "reward_std": 0.2083333432674408,
      "rewards/accuracy_reward": 0.8958333432674408,
      "rewards/format_reward": 1.0,
      "step": 22
    },
    {
      "completion_length": 79.6875,
      "epoch": 0.10623556581986143,
      "grad_norm": 6.442953109741211,
      "kl": 0.004077911376953125,
      "learning_rate": 1e-06,
      "loss": 0.0002,
      "reward": 1.8125000596046448,
      "reward_std": 0.3045583665370941,
      "rewards/accuracy_reward": 0.8333333730697632,
      "rewards/format_reward": 0.9791666865348816,
      "step": 23
    },
    {
      "completion_length": 80.70833587646484,
      "epoch": 0.11085450346420324,
      "grad_norm": 9.262118339538574,
      "kl": 0.0145263671875,
      "learning_rate": 1e-06,
      "loss": 0.0006,
      "reward": 1.7708333730697632,
      "reward_std": 0.25644584745168686,
      "rewards/accuracy_reward": 0.7708333432674408,
      "rewards/format_reward": 1.0,
      "step": 24
    },
    {
      "completion_length": 72.6875,
      "epoch": 0.11547344110854503,
      "grad_norm": 2.5963733196258545,
      "kl": 0.019866943359375,
      "learning_rate": 1e-06,
      "loss": 0.0008,
      "reward": 1.8750000596046448,
      "reward_std": 0.0833333358168602,
      "rewards/accuracy_reward": 0.8750000298023224,
      "rewards/format_reward": 1.0,
      "step": 25
    },
    {
      "completion_length": 79.37500190734863,
      "epoch": 0.12009237875288684,
      "grad_norm": 8.336856842041016,
      "kl": 0.02008056640625,
      "learning_rate": 1e-06,
      "loss": 0.0008,
      "reward": 1.6875000596046448,
      "reward_std": 0.25644585117697716,
      "rewards/accuracy_reward": 0.6875000298023224,
      "rewards/format_reward": 1.0,
      "step": 26
    },
    {
      "completion_length": 54.64583396911621,
      "epoch": 0.12471131639722864,
      "grad_norm": 7.557518482208252,
      "kl": 0.01947021484375,
      "learning_rate": 1e-06,
      "loss": 0.0008,
      "reward": 1.8125,
      "reward_std": 0.2083333358168602,
      "rewards/accuracy_reward": 0.8125,
      "rewards/format_reward": 1.0,
      "step": 27
    },
    {
      "completion_length": 57.4375,
      "epoch": 0.12933025404157045,
      "grad_norm": 4.195393085479736,
      "kl": 0.004122734069824219,
      "learning_rate": 1e-06,
      "loss": 0.0002,
      "reward": 1.9583333730697632,
      "reward_std": 0.0833333358168602,
      "rewards/accuracy_reward": 0.9583333432674408,
      "rewards/format_reward": 1.0,
      "step": 28
    },
    {
      "completion_length": 31.750000953674316,
      "epoch": 0.13394919168591224,
      "grad_norm": 8.026168823242188,
      "kl": 0.0205078125,
      "learning_rate": 1e-06,
      "loss": 0.0008,
      "reward": 1.8333333730697632,
      "reward_std": 0.0833333358168602,
      "rewards/accuracy_reward": 0.8333333432674408,
      "rewards/format_reward": 1.0,
      "step": 29
    },
    {
      "completion_length": 63.0625,
      "epoch": 0.13856812933025403,
      "grad_norm": 9.261226654052734,
      "kl": 0.01236724853515625,
      "learning_rate": 1e-06,
      "loss": 0.0005,
      "reward": 1.875,
      "reward_std": 0.048112522810697556,
      "rewards/accuracy_reward": 0.875,
      "rewards/format_reward": 1.0,
      "step": 30
    },
    {
      "completion_length": 81.93750190734863,
      "epoch": 0.14318706697459585,
      "grad_norm": 8.847457885742188,
      "kl": 0.01412200927734375,
      "learning_rate": 1e-06,
      "loss": 0.0006,
      "reward": 1.8750000596046448,
      "reward_std": 0.179558377712965,
      "rewards/accuracy_reward": 0.8958333730697632,
      "rewards/format_reward": 0.9791666865348816,
      "step": 31
    },
    {
      "completion_length": 63.125,
      "epoch": 0.14780600461893764,
      "grad_norm": 9.623985290527344,
      "kl": 0.0164337158203125,
      "learning_rate": 1e-06,
      "loss": 0.0007,
      "reward": 1.8333333730697632,
      "reward_std": 0.2628917098045349,
      "rewards/accuracy_reward": 0.8333333432674408,
      "rewards/format_reward": 1.0,
      "step": 32
    },
    {
      "completion_length": 65.35416793823242,
      "epoch": 0.15242494226327943,
      "grad_norm": 4.50661563873291,
      "kl": 0.01575469970703125,
      "learning_rate": 1e-06,
      "loss": 0.0006,
      "reward": 1.6875000596046448,
      "reward_std": 0.2083333358168602,
      "rewards/accuracy_reward": 0.6875000149011612,
      "rewards/format_reward": 1.0,
      "step": 33
    },
    {
      "completion_length": 65.10416793823242,
      "epoch": 0.15704387990762125,
      "grad_norm": 8.077963829040527,
      "kl": 0.06109619140625,
      "learning_rate": 1e-06,
      "loss": 0.0024,
      "reward": 1.8750000596046448,
      "reward_std": 0.2500000074505806,
      "rewards/accuracy_reward": 0.8750000298023224,
      "rewards/format_reward": 1.0,
      "step": 34
    },
    {
      "completion_length": 74.45833587646484,
      "epoch": 0.16166281755196305,
      "grad_norm": 9.616883277893066,
      "kl": 0.00418853759765625,
      "learning_rate": 1e-06,
      "loss": 0.0002,
      "reward": 1.9166666865348816,
      "reward_std": 0.0833333358168602,
      "rewards/accuracy_reward": 0.9166666865348816,
      "rewards/format_reward": 1.0,
      "step": 35
    },
    {
      "completion_length": 56.16666793823242,
      "epoch": 0.16628175519630484,
      "grad_norm": 8.547237396240234,
      "kl": 0.014141082763671875,
      "learning_rate": 1e-06,
      "loss": 0.0006,
      "reward": 1.875,
      "reward_std": 0.1666666716337204,
      "rewards/accuracy_reward": 0.875,
      "rewards/format_reward": 1.0,
      "step": 36
    },
    {
      "completion_length": 97.93750381469727,
      "epoch": 0.17090069284064666,
      "grad_norm": 0.9761996865272522,
      "kl": 0.01053619384765625,
      "learning_rate": 1e-06,
      "loss": 0.0004,
      "reward": 1.9166666865348816,
      "reward_std": 0.1666666679084301,
      "rewards/accuracy_reward": 0.9166666865348816,
      "rewards/format_reward": 1.0,
      "step": 37
    },
    {
      "completion_length": 101.79166793823242,
      "epoch": 0.17551963048498845,
      "grad_norm": 5.636431694030762,
      "kl": 0.0035552978515625,
      "learning_rate": 1e-06,
      "loss": 0.0001,
      "reward": 1.8541667461395264,
      "reward_std": 0.2212250456213951,
      "rewards/accuracy_reward": 0.8541666865348816,
      "rewards/format_reward": 1.0,
      "step": 38
    },
    {
      "completion_length": 50.29166793823242,
      "epoch": 0.18013856812933027,
      "grad_norm": 5.020724773406982,
      "kl": 0.006103515625,
      "learning_rate": 1e-06,
      "loss": 0.0002,
      "reward": 1.8750000596046448,
      "reward_std": 0.179558377712965,
      "rewards/accuracy_reward": 0.8750000298023224,
      "rewards/format_reward": 1.0,
      "step": 39
    },
    {
      "completion_length": 68.02083587646484,
      "epoch": 0.18475750577367206,
      "grad_norm": 1.0703792572021484,
      "kl": 0.0052642822265625,
      "learning_rate": 1e-06,
      "loss": 0.0002,
      "reward": 1.9166667461395264,
      "reward_std": 0.13144585862755775,
      "rewards/accuracy_reward": 0.9166666865348816,
      "rewards/format_reward": 1.0,
      "step": 40
    },
    {
      "completion_length": 75.22916984558105,
      "epoch": 0.18937644341801385,
      "grad_norm": 2.469010829925537,
      "kl": 0.0019931793212890625,
      "learning_rate": 1e-06,
      "loss": 0.0001,
      "reward": 1.8958333730697632,
      "reward_std": 0.08977919071912766,
      "rewards/accuracy_reward": 0.8958333432674408,
      "rewards/format_reward": 1.0,
      "step": 41
    },
    {
      "completion_length": 75.1875,
      "epoch": 0.19399538106235567,
      "grad_norm": 4.2055230140686035,
      "kl": 0.003253936767578125,
      "learning_rate": 1e-06,
      "loss": 0.0001,
      "reward": 1.9375000596046448,
      "reward_std": 0.08977919071912766,
      "rewards/accuracy_reward": 0.9375000298023224,
      "rewards/format_reward": 1.0,
      "step": 42
    },
    {
      "completion_length": 79.54166793823242,
      "epoch": 0.19861431870669746,
      "grad_norm": 7.644351482391357,
      "kl": 0.0106658935546875,
      "learning_rate": 1e-06,
      "loss": 0.0004,
      "reward": 1.875,
      "reward_std": 0.25,
      "rewards/accuracy_reward": 0.875,
      "rewards/format_reward": 1.0,
      "step": 43
    },
    {
      "completion_length": 81.04166793823242,
      "epoch": 0.20323325635103925,
      "grad_norm": 0.7677356600761414,
      "kl": 0.00148773193359375,
      "learning_rate": 1e-06,
      "loss": 0.0001,
      "reward": 1.8750000596046448,
      "reward_std": 0.21477919444441795,
      "rewards/accuracy_reward": 0.8750000298023224,
      "rewards/format_reward": 1.0,
      "step": 44
    },
    {
      "completion_length": 62.770835876464844,
      "epoch": 0.20785219399538107,
      "grad_norm": 4.3740692138671875,
      "kl": 0.002414703369140625,
      "learning_rate": 1e-06,
      "loss": 0.0001,
      "reward": 1.8750000596046448,
      "reward_std": 0.1666666716337204,
      "rewards/accuracy_reward": 0.8750000298023224,
      "rewards/format_reward": 1.0,
      "step": 45
    },
    {
      "completion_length": 78.35416793823242,
      "epoch": 0.21247113163972287,
      "grad_norm": 4.38349723815918,
      "kl": 0.005157470703125,
      "learning_rate": 1e-06,
      "loss": 0.0002,
      "reward": 1.9375000596046448,
      "reward_std": 0.1250000037252903,
      "rewards/accuracy_reward": 0.9375000298023224,
      "rewards/format_reward": 1.0,
      "step": 46
    },
    {
      "completion_length": 59.9375,
      "epoch": 0.21709006928406466,
      "grad_norm": 0.012896657921373844,
      "kl": 0.001483917236328125,
      "learning_rate": 1e-06,
      "loss": 0.0001,
      "reward": 2.0,
      "reward_std": 0.0,
      "rewards/accuracy_reward": 1.0,
      "rewards/format_reward": 1.0,
      "step": 47
    },
    {
      "completion_length": 63.520835876464844,
      "epoch": 0.22170900692840648,
      "grad_norm": 1.7839175462722778,
      "kl": 0.0340423583984375,
      "learning_rate": 1e-06,
      "loss": 0.0014,
      "reward": 1.9791666865348816,
      "reward_std": 0.0416666679084301,
      "rewards/accuracy_reward": 0.9791666865348816,
      "rewards/format_reward": 1.0,
      "step": 48
    },
    {
      "completion_length": 52.41666793823242,
      "epoch": 0.22632794457274827,
      "grad_norm": 15.218417167663574,
      "kl": 0.0085296630859375,
      "learning_rate": 1e-06,
      "loss": 0.0003,
      "reward": 1.875,
      "reward_std": 0.0833333358168602,
      "rewards/accuracy_reward": 0.875,
      "rewards/format_reward": 1.0,
      "step": 49
    },
    {
      "completion_length": 42.12500286102295,
      "epoch": 0.23094688221709006,
      "grad_norm": 0.7014623284339905,
      "kl": 0.001987457275390625,
      "learning_rate": 1e-06,
      "loss": 0.0001,
      "reward": 1.9583333730697632,
      "reward_std": 0.0833333358168602,
      "rewards/accuracy_reward": 0.9583333432674408,
      "rewards/format_reward": 1.0,
      "step": 50
    },
    {
      "completion_length": 48.666666984558105,
      "epoch": 0.23556581986143188,
      "grad_norm": 0.49798333644866943,
      "kl": 0.0025177001953125,
      "learning_rate": 1e-06,
      "loss": 0.0001,
      "reward": 1.9375,
      "reward_std": 0.08977919071912766,
      "rewards/accuracy_reward": 0.9375,
      "rewards/format_reward": 1.0,
      "step": 51
    },
    {
      "completion_length": 67.47916793823242,
      "epoch": 0.24018475750577367,
      "grad_norm": 0.28330135345458984,
      "kl": 0.00286102294921875,
      "learning_rate": 1e-06,
      "loss": 0.0001,
      "reward": 1.9583333730697632,
      "reward_std": 0.048112522810697556,
      "rewards/accuracy_reward": 0.9583333432674408,
      "rewards/format_reward": 1.0,
      "step": 52
    },
    {
      "completion_length": 58.020835876464844,
      "epoch": 0.24480369515011546,
      "grad_norm": 4.528405666351318,
      "kl": 0.004150390625,
      "learning_rate": 1e-06,
      "loss": 0.0002,
      "reward": 1.8958333730697632,
      "reward_std": 0.0416666679084301,
      "rewards/accuracy_reward": 0.8958333432674408,
      "rewards/format_reward": 1.0,
      "step": 53
    },
    {
      "completion_length": 47.91666793823242,
      "epoch": 0.24942263279445728,
      "grad_norm": 0.3298056125640869,
      "kl": 0.00391387939453125,
      "learning_rate": 1e-06,
      "loss": 0.0002,
      "reward": 1.9791666865348816,
      "reward_std": 0.0416666679084301,
      "rewards/accuracy_reward": 0.9791666865348816,
      "rewards/format_reward": 1.0,
      "step": 54
    },
    {
      "completion_length": 33.47916793823242,
      "epoch": 0.2540415704387991,
      "grad_norm": 0.016688158735632896,
      "kl": 0.00406646728515625,
      "learning_rate": 1e-06,
      "loss": 0.0002,
      "reward": 2.0,
      "reward_std": 0.0,
      "rewards/accuracy_reward": 1.0,
      "rewards/format_reward": 1.0,
      "step": 55
    },
    {
      "completion_length": 55.979169845581055,
      "epoch": 0.2586605080831409,
      "grad_norm": 0.3446481227874756,
      "kl": 0.004150390625,
      "learning_rate": 1e-06,
      "loss": 0.0002,
      "reward": 1.9791666865348816,
      "reward_std": 0.0416666679084301,
      "rewards/accuracy_reward": 0.9791666865348816,
      "rewards/format_reward": 1.0,
      "step": 56
    },
    {
      "completion_length": 31.604167938232422,
      "epoch": 0.2632794457274827,
      "grad_norm": 4.186825275421143,
      "kl": 0.0166015625,
      "learning_rate": 1e-06,
      "loss": 0.0007,
      "reward": 1.8750000596046448,
      "reward_std": 0.0833333358168602,
      "rewards/accuracy_reward": 0.8750000298023224,
      "rewards/format_reward": 1.0,
      "step": 57
    },
    {
      "completion_length": 54.708335876464844,
      "epoch": 0.2678983833718245,
      "grad_norm": 0.45439592003822327,
      "kl": 0.005859375,
      "learning_rate": 1e-06,
      "loss": 0.0002,
      "reward": 1.8958333730697632,
      "reward_std": 0.0416666679084301,
      "rewards/accuracy_reward": 0.8958333730697632,
      "rewards/format_reward": 1.0,
      "step": 58
    },
    {
      "completion_length": 44.14583396911621,
      "epoch": 0.27251732101616627,
      "grad_norm": 0.3110947012901306,
      "kl": 0.01617431640625,
      "learning_rate": 1e-06,
      "loss": 0.0006,
      "reward": 1.9583333730697632,
      "reward_std": 0.048112522810697556,
      "rewards/accuracy_reward": 0.9583333432674408,
      "rewards/format_reward": 1.0,
      "step": 59
    },
    {
      "completion_length": 56.56250190734863,
      "epoch": 0.27713625866050806,
      "grad_norm": 4.412627220153809,
      "kl": 0.00714111328125,
      "learning_rate": 1e-06,
      "loss": 0.0003,
      "reward": 1.8750000596046448,
      "reward_std": 0.1666666716337204,
      "rewards/accuracy_reward": 0.8750000298023224,
      "rewards/format_reward": 1.0,
      "step": 60
    },
    {
      "completion_length": 24.52083396911621,
      "epoch": 0.2817551963048499,
      "grad_norm": 12.474504470825195,
      "kl": 0.02862548828125,
      "learning_rate": 1e-06,
      "loss": 0.0011,
      "reward": 1.7500000596046448,
      "reward_std": 0.0833333358168602,
      "rewards/accuracy_reward": 0.7500000298023224,
      "rewards/format_reward": 1.0,
      "step": 61
    },
    {
      "completion_length": 42.08333396911621,
      "epoch": 0.2863741339491917,
      "grad_norm": 6.133255481719971,
      "kl": 0.02166748046875,
      "learning_rate": 1e-06,
      "loss": 0.0009,
      "reward": 1.9375,
      "reward_std": 0.0416666679084301,
      "rewards/accuracy_reward": 0.9375,
      "rewards/format_reward": 1.0,
      "step": 62
    },
    {
      "completion_length": 32.33333396911621,
      "epoch": 0.2909930715935335,
      "grad_norm": 18.329042434692383,
      "kl": 0.01953125,
      "learning_rate": 1e-06,
      "loss": 0.0008,
      "reward": 1.8958333730697632,
      "reward_std": 0.1250000037252903,
      "rewards/accuracy_reward": 0.8958333432674408,
      "rewards/format_reward": 1.0,
      "step": 63
    },
    {
      "completion_length": 41.87500190734863,
      "epoch": 0.2956120092378753,
      "grad_norm": 0.8285086750984192,
      "kl": 0.00627899169921875,
      "learning_rate": 1e-06,
      "loss": 0.0003,
      "reward": 1.8333333730697632,
      "reward_std": 0.1666666679084301,
      "rewards/accuracy_reward": 0.8333333730697632,
      "rewards/format_reward": 1.0,
      "step": 64
    },
    {
      "completion_length": 41.395835876464844,
      "epoch": 0.3002309468822171,
      "grad_norm": 12.128033638000488,
      "kl": 0.01610565185546875,
      "learning_rate": 1e-06,
      "loss": 0.0006,
      "reward": 1.9375,
      "reward_std": 0.125,
      "rewards/accuracy_reward": 0.9375,
      "rewards/format_reward": 1.0,
      "step": 65
    },
    {
      "completion_length": 51.145835876464844,
      "epoch": 0.30484988452655887,
      "grad_norm": 0.4265430271625519,
      "kl": 0.0070648193359375,
      "learning_rate": 1e-06,
      "loss": 0.0003,
      "reward": 1.9583333730697632,
      "reward_std": 0.0833333358168602,
      "rewards/accuracy_reward": 0.9583333432674408,
      "rewards/format_reward": 1.0,
      "step": 66
    },
    {
      "completion_length": 43.50000190734863,
      "epoch": 0.3094688221709007,
      "grad_norm": 5.070745468139648,
      "kl": 0.0081787109375,
      "learning_rate": 1e-06,
      "loss": 0.0003,
      "reward": 1.8958333730697632,
      "reward_std": 0.0416666679084301,
      "rewards/accuracy_reward": 0.8958333432674408,
      "rewards/format_reward": 1.0,
      "step": 67
    },
    {
      "completion_length": 42.604166984558105,
      "epoch": 0.3140877598152425,
      "grad_norm": 8.272510528564453,
      "kl": 0.01300048828125,
      "learning_rate": 1e-06,
      "loss": 0.0005,
      "reward": 1.8333333730697632,
      "reward_std": 0.1666666716337204,
      "rewards/accuracy_reward": 0.8333333432674408,
      "rewards/format_reward": 1.0,
      "step": 68
    },
    {
      "completion_length": 45.33333396911621,
      "epoch": 0.3187066974595843,
      "grad_norm": 0.4940468668937683,
      "kl": 0.00400543212890625,
      "learning_rate": 1e-06,
      "loss": 0.0002,
      "reward": 1.9583333730697632,
      "reward_std": 0.0833333358168602,
      "rewards/accuracy_reward": 0.9583333730697632,
      "rewards/format_reward": 1.0,
      "step": 69
    },
    {
      "completion_length": 34.70833492279053,
      "epoch": 0.3233256351039261,
      "grad_norm": 26.54204750061035,
      "kl": 0.02557373046875,
      "learning_rate": 1e-06,
      "loss": 0.001,
      "reward": 1.8541666865348816,
      "reward_std": 0.1250000037252903,
      "rewards/accuracy_reward": 0.875,
      "rewards/format_reward": 0.9791666865348816,
      "step": 70
    },
    {
      "completion_length": 18.750000953674316,
      "epoch": 0.3279445727482679,
      "grad_norm": 14.518416404724121,
      "kl": 0.00531005859375,
      "learning_rate": 1e-06,
      "loss": 0.0002,
      "reward": 1.9375000596046448,
      "reward_std": 0.08977919071912766,
      "rewards/accuracy_reward": 0.9375000298023224,
      "rewards/format_reward": 1.0,
      "step": 71
    },
    {
      "completion_length": 28.541666984558105,
      "epoch": 0.3325635103926097,
      "grad_norm": 9.401073455810547,
      "kl": 0.004241943359375,
      "learning_rate": 1e-06,
      "loss": 0.0002,
      "reward": 1.9375,
      "reward_std": 0.0416666679084301,
      "rewards/accuracy_reward": 0.9375,
      "rewards/format_reward": 1.0,
      "step": 72
    },
    {
      "completion_length": 33.47916793823242,
      "epoch": 0.3371824480369515,
      "grad_norm": 0.5438477396965027,
      "kl": 0.00969696044921875,
      "learning_rate": 1e-06,
      "loss": 0.0004,
      "reward": 1.9791666865348816,
      "reward_std": 0.0416666679084301,
      "rewards/accuracy_reward": 0.9791666865348816,
      "rewards/format_reward": 1.0,
      "step": 73
    },
    {
      "completion_length": 28.625000953674316,
      "epoch": 0.3418013856812933,
      "grad_norm": 12.923347473144531,
      "kl": 0.006134033203125,
      "learning_rate": 1e-06,
      "loss": 0.0002,
      "reward": 1.8541666865348816,
      "reward_std": 0.08977919071912766,
      "rewards/accuracy_reward": 0.8541666865348816,
      "rewards/format_reward": 1.0,
      "step": 74
    },
    {
      "completion_length": 38.875,
      "epoch": 0.3464203233256351,
      "grad_norm": 0.082447350025177,
      "kl": 0.0061798095703125,
      "learning_rate": 1e-06,
      "loss": 0.0002,
      "reward": 2.0,
      "reward_std": 0.0,
      "rewards/accuracy_reward": 1.0,
      "rewards/format_reward": 1.0,
      "step": 75
    },
    {
      "completion_length": 32.291666984558105,
      "epoch": 0.3510392609699769,
      "grad_norm": 6.839282989501953,
      "kl": 0.02349853515625,
      "learning_rate": 1e-06,
      "loss": 0.0009,
      "reward": 1.8750000596046448,
      "reward_std": 0.0833333358168602,
      "rewards/accuracy_reward": 0.8750000298023224,
      "rewards/format_reward": 1.0,
      "step": 76
    },
    {
      "completion_length": 29.812500953674316,
      "epoch": 0.3556581986143187,
      "grad_norm": 0.014031085185706615,
      "kl": 0.0072021484375,
      "learning_rate": 1e-06,
      "loss": 0.0003,
      "reward": 1.9166666865348816,
      "reward_std": 0.0,
      "rewards/accuracy_reward": 0.9166666865348816,
      "rewards/format_reward": 1.0,
      "step": 77
    },
    {
      "completion_length": 19.812500476837158,
      "epoch": 0.36027713625866054,
      "grad_norm": 0.3273457884788513,
      "kl": 0.0062103271484375,
      "learning_rate": 1e-06,
      "loss": 0.0002,
      "reward": 1.9791666865348816,
      "reward_std": 0.0416666679084301,
      "rewards/accuracy_reward": 0.9791666865348816,
      "rewards/format_reward": 1.0,
      "step": 78
    },
    {
      "completion_length": 25.354167461395264,
      "epoch": 0.3648960739030023,
      "grad_norm": 0.3809925615787506,
      "kl": 0.00506591796875,
      "learning_rate": 1e-06,
      "loss": 0.0002,
      "reward": 1.9791666865348816,
      "reward_std": 0.0416666679084301,
      "rewards/accuracy_reward": 0.9791666865348816,
      "rewards/format_reward": 1.0,
      "step": 79
    },
    {
      "completion_length": 18.791667461395264,
      "epoch": 0.3695150115473441,
      "grad_norm": 0.3350464105606079,
      "kl": 0.0046234130859375,
      "learning_rate": 1e-06,
      "loss": 0.0002,
      "reward": 1.9791666865348816,
      "reward_std": 0.0416666679084301,
      "rewards/accuracy_reward": 0.9791666865348816,
      "rewards/format_reward": 1.0,
      "step": 80
    },
    {
      "completion_length": 18.83333396911621,
      "epoch": 0.3741339491916859,
      "grad_norm": 0.16990438103675842,
      "kl": 0.01031494140625,
      "learning_rate": 1e-06,
      "loss": 0.0004,
      "reward": 1.9166666865348816,
      "reward_std": 0.0,
      "rewards/accuracy_reward": 0.9166666865348816,
      "rewards/format_reward": 1.0,
      "step": 81
    },
    {
      "completion_length": 26.895833492279053,
      "epoch": 0.3787528868360277,
      "grad_norm": 42.352142333984375,
      "kl": 0.0153961181640625,
      "learning_rate": 1e-06,
      "loss": 0.0006,
      "reward": 1.9166666865348816,
      "reward_std": 0.0833333358168602,
      "rewards/accuracy_reward": 0.9166666865348816,
      "rewards/format_reward": 1.0,
      "step": 82
    },
    {
      "completion_length": 22.791666984558105,
      "epoch": 0.3833718244803695,
      "grad_norm": 38.687313079833984,
      "kl": 0.023101806640625,
      "learning_rate": 1e-06,
      "loss": 0.0009,
      "reward": 1.7708333730697632,
      "reward_std": 0.17311252653598785,
      "rewards/accuracy_reward": 0.7708333432674408,
      "rewards/format_reward": 1.0,
      "step": 83
    },
    {
      "completion_length": 15.166666984558105,
      "epoch": 0.38799076212471134,
      "grad_norm": 11.10213851928711,
      "kl": 0.00698089599609375,
      "learning_rate": 1e-06,
      "loss": 0.0003,
      "reward": 1.9791666865348816,
      "reward_std": 0.0416666679084301,
      "rewards/accuracy_reward": 0.9791666865348816,
      "rewards/format_reward": 1.0,
      "step": 84
    },
    {
      "completion_length": 20.6875,
      "epoch": 0.39260969976905313,
      "grad_norm": 0.2621253430843353,
      "kl": 0.01202392578125,
      "learning_rate": 1e-06,
      "loss": 0.0005,
      "reward": 1.9791666865348816,
      "reward_std": 0.0416666679084301,
      "rewards/accuracy_reward": 0.9791666865348816,
      "rewards/format_reward": 1.0,
      "step": 85
    },
    {
      "completion_length": 33.02083396911621,
      "epoch": 0.3972286374133949,
      "grad_norm": 0.24522998929023743,
      "kl": 0.005706787109375,
      "learning_rate": 1e-06,
      "loss": 0.0002,
      "reward": 2.0,
      "reward_std": 0.0,
      "rewards/accuracy_reward": 1.0,
      "rewards/format_reward": 1.0,
      "step": 86
    },
    {
      "completion_length": 25.25,
      "epoch": 0.4018475750577367,
      "grad_norm": 0.32983526587486267,
      "kl": 0.0078277587890625,
      "learning_rate": 1e-06,
      "loss": 0.0003,
      "reward": 1.8958333730697632,
      "reward_std": 0.0416666679084301,
      "rewards/accuracy_reward": 0.8958333432674408,
      "rewards/format_reward": 1.0,
      "step": 87
    },
    {
      "completion_length": 29.416666984558105,
      "epoch": 0.4064665127020785,
      "grad_norm": 10.310896873474121,
      "kl": 0.022125244140625,
      "learning_rate": 1e-06,
      "loss": 0.0009,
      "reward": 1.9375,
      "reward_std": 0.0416666679084301,
      "rewards/accuracy_reward": 0.9375,
      "rewards/format_reward": 1.0,
      "step": 88
    },
    {
      "completion_length": 52.645835876464844,
      "epoch": 0.4110854503464203,
      "grad_norm": 0.3584893047809601,
      "kl": 0.02850341796875,
      "learning_rate": 1e-06,
      "loss": 0.0011,
      "reward": 1.8958333730697632,
      "reward_std": 0.0416666679084301,
      "rewards/accuracy_reward": 0.8958333432674408,
      "rewards/format_reward": 1.0,
      "step": 89
    },
    {
      "completion_length": 20.750000476837158,
      "epoch": 0.41570438799076215,
      "grad_norm": 1.2553701400756836,
      "kl": 0.0139312744140625,
      "learning_rate": 1e-06,
      "loss": 0.0006,
      "reward": 1.9791666865348816,
      "reward_std": 0.0416666679084301,
      "rewards/accuracy_reward": 0.9791666865348816,
      "rewards/format_reward": 1.0,
      "step": 90
    },
    {
      "completion_length": 39.9791693687439,
      "epoch": 0.42032332563510394,
      "grad_norm": 0.2528979182243347,
      "kl": 0.0077056884765625,
      "learning_rate": 1e-06,
      "loss": 0.0003,
      "reward": 1.9791666865348816,
      "reward_std": 0.0416666679084301,
      "rewards/accuracy_reward": 0.9791666865348816,
      "rewards/format_reward": 1.0,
      "step": 91
    },
    {
      "completion_length": 33.83333396911621,
      "epoch": 0.42494226327944573,
      "grad_norm": 0.5499228835105896,
      "kl": 0.011077880859375,
      "learning_rate": 1e-06,
      "loss": 0.0004,
      "reward": 1.9583333730697632,
      "reward_std": 0.0833333358168602,
      "rewards/accuracy_reward": 0.9583333730697632,
      "rewards/format_reward": 1.0,
      "step": 92
    },
    {
      "completion_length": 27.541666984558105,
      "epoch": 0.4295612009237875,
      "grad_norm": 0.012784978374838829,
      "kl": 0.0097503662109375,
      "learning_rate": 1e-06,
      "loss": 0.0004,
      "reward": 2.0,
      "reward_std": 0.0,
      "rewards/accuracy_reward": 1.0,
      "rewards/format_reward": 1.0,
      "step": 93
    },
    {
      "completion_length": 24.229167461395264,
      "epoch": 0.4341801385681293,
      "grad_norm": 13.934257507324219,
      "kl": 0.014495849609375,
      "learning_rate": 1e-06,
      "loss": 0.0006,
      "reward": 1.9583333730697632,
      "reward_std": 0.0833333358168602,
      "rewards/accuracy_reward": 0.9583333432674408,
      "rewards/format_reward": 1.0,
      "step": 94
    },
    {
      "completion_length": 33.33333492279053,
      "epoch": 0.4387990762124711,
      "grad_norm": 0.22108934819698334,
      "kl": 0.0045166015625,
      "learning_rate": 1e-06,
      "loss": 0.0002,
      "reward": 1.8958333730697632,
      "reward_std": 0.0416666679084301,
      "rewards/accuracy_reward": 0.8958333432674408,
      "rewards/format_reward": 1.0,
      "step": 95
    },
    {
      "completion_length": 20.875000953674316,
      "epoch": 0.44341801385681295,
      "grad_norm": 0.01029158290475607,
      "kl": 0.00502777099609375,
      "learning_rate": 1e-06,
      "loss": 0.0002,
      "reward": 2.0,
      "reward_std": 0.0,
      "rewards/accuracy_reward": 1.0,
      "rewards/format_reward": 1.0,
      "step": 96
    },
    {
      "completion_length": 31.583334922790527,
      "epoch": 0.44803695150115475,
      "grad_norm": 0.28462010622024536,
      "kl": 0.0048828125,
      "learning_rate": 1e-06,
      "loss": 0.0002,
      "reward": 1.9791666865348816,
      "reward_std": 0.0416666679084301,
      "rewards/accuracy_reward": 0.9791666865348816,
      "rewards/format_reward": 1.0,
      "step": 97
    },
    {
      "completion_length": 15.0,
      "epoch": 0.45265588914549654,
      "grad_norm": 0.047917552292346954,
      "kl": 0.0049591064453125,
      "learning_rate": 1e-06,
      "loss": 0.0002,
      "reward": 2.0,
      "reward_std": 0.0,
      "rewards/accuracy_reward": 1.0,
      "rewards/format_reward": 1.0,
      "step": 98
    },
    {
      "completion_length": 18.479166984558105,
      "epoch": 0.45727482678983833,
      "grad_norm": 22.465944290161133,
      "kl": 0.0059356689453125,
      "learning_rate": 1e-06,
      "loss": 0.0002,
      "reward": 1.9375,
      "reward_std": 0.08977919071912766,
      "rewards/accuracy_reward": 0.9375,
      "rewards/format_reward": 1.0,
      "step": 99
    },
    {
      "completion_length": 25.083333492279053,
      "epoch": 0.4618937644341801,
      "grad_norm": 0.008041063323616982,
      "kl": 0.003936767578125,
      "learning_rate": 1e-06,
      "loss": 0.0002,
      "reward": 2.0,
      "reward_std": 0.0,
      "rewards/accuracy_reward": 1.0,
      "rewards/format_reward": 1.0,
      "step": 100
    },
    {
      "completion_length": 37.833335876464844,
      "epoch": 0.4665127020785219,
      "grad_norm": 14.107870101928711,
      "kl": 0.007110595703125,
      "learning_rate": 1e-06,
      "loss": 0.0003,
      "reward": 1.7500000596046448,
      "reward_std": 0.1666666716337204,
      "rewards/accuracy_reward": 0.7500000298023224,
      "rewards/format_reward": 1.0,
      "step": 101
    },
    {
      "completion_length": 15.104166984558105,
      "epoch": 0.47113163972286376,
      "grad_norm": 15.883159637451172,
      "kl": 0.0056915283203125,
      "learning_rate": 1e-06,
      "loss": 0.0002,
      "reward": 1.9791666865348816,
      "reward_std": 0.0416666679084301,
      "rewards/accuracy_reward": 0.9791666865348816,
      "rewards/format_reward": 1.0,
      "step": 102
    },
    {
      "completion_length": 18.979167461395264,
      "epoch": 0.47575057736720555,
      "grad_norm": 11.000266075134277,
      "kl": 0.0062103271484375,
      "learning_rate": 1e-06,
      "loss": 0.0002,
      "reward": 1.9375,
      "reward_std": 0.0416666679084301,
      "rewards/accuracy_reward": 0.9375,
      "rewards/format_reward": 1.0,
      "step": 103
    },
    {
      "completion_length": 18.875000953674316,
      "epoch": 0.48036951501154734,
      "grad_norm": 0.06595253199338913,
      "kl": 0.0058135986328125,
      "learning_rate": 1e-06,
      "loss": 0.0002,
      "reward": 1.8333333730697632,
      "reward_std": 0.0,
      "rewards/accuracy_reward": 0.8333333730697632,
      "rewards/format_reward": 1.0,
      "step": 104
    },
    {
      "completion_length": 20.312500476837158,
      "epoch": 0.48498845265588914,
      "grad_norm": 7.73568868637085,
      "kl": 0.01102447509765625,
      "learning_rate": 1e-06,
      "loss": 0.0004,
      "reward": 1.9583333730697632,
      "reward_std": 0.048112522810697556,
      "rewards/accuracy_reward": 0.9583333432674408,
      "rewards/format_reward": 1.0,
      "step": 105
    },
    {
      "completion_length": 33.91666793823242,
      "epoch": 0.4896073903002309,
      "grad_norm": 1.0296730995178223,
      "kl": 0.0074462890625,
      "learning_rate": 1e-06,
      "loss": 0.0003,
      "reward": 1.9375,
      "reward_std": 0.0416666679084301,
      "rewards/accuracy_reward": 0.9375,
      "rewards/format_reward": 1.0,
      "step": 106
    },
    {
      "completion_length": 20.000000476837158,
      "epoch": 0.4942263279445728,
      "grad_norm": 0.011166680604219437,
      "kl": 0.0047149658203125,
      "learning_rate": 1e-06,
      "loss": 0.0002,
      "reward": 2.0,
      "reward_std": 0.0,
      "rewards/accuracy_reward": 1.0,
      "rewards/format_reward": 1.0,
      "step": 107
    },
    {
      "completion_length": 20.4375,
      "epoch": 0.49884526558891457,
      "grad_norm": 0.09496494382619858,
      "kl": 0.00506591796875,
      "learning_rate": 1e-06,
      "loss": 0.0002,
      "reward": 2.0,
      "reward_std": 0.0,
      "rewards/accuracy_reward": 1.0,
      "rewards/format_reward": 1.0,
      "step": 108
    },
    {
      "completion_length": 19.791666984558105,
      "epoch": 0.5034642032332564,
      "grad_norm": 0.013560502789914608,
      "kl": 0.00475311279296875,
      "learning_rate": 1e-06,
      "loss": 0.0002,
      "reward": 2.0,
      "reward_std": 0.0,
      "rewards/accuracy_reward": 1.0,
      "rewards/format_reward": 1.0,
      "step": 109
    },
    {
      "completion_length": 15.0,
      "epoch": 0.5080831408775982,
      "grad_norm": 0.056342754513025284,
      "kl": 0.00507354736328125,
      "learning_rate": 1e-06,
      "loss": 0.0002,
      "reward": 1.9166666865348816,
      "reward_std": 0.0,
      "rewards/accuracy_reward": 0.9166666865348816,
      "rewards/format_reward": 1.0,
      "step": 110
    },
    {
      "completion_length": 31.125001907348633,
      "epoch": 0.5127020785219399,
      "grad_norm": 12.659052848815918,
      "kl": 0.0098724365234375,
      "learning_rate": 1e-06,
      "loss": 0.0004,
      "reward": 1.9166666865348816,
      "reward_std": 0.0833333358168602,
      "rewards/accuracy_reward": 0.9166666865348816,
      "rewards/format_reward": 1.0,
      "step": 111
    },
    {
      "completion_length": 21.0,
      "epoch": 0.5173210161662818,
      "grad_norm": 2.68220591545105,
      "kl": 0.01091766357421875,
      "learning_rate": 1e-06,
      "loss": 0.0004,
      "reward": 1.9166666865348816,
      "reward_std": 0.0,
      "rewards/accuracy_reward": 0.9166666865348816,
      "rewards/format_reward": 1.0,
      "step": 112
    },
    {
      "completion_length": 29.041666984558105,
      "epoch": 0.5219399538106235,
      "grad_norm": 38.36140823364258,
      "kl": 0.0079193115234375,
      "learning_rate": 1e-06,
      "loss": 0.0003,
      "reward": 1.9375000596046448,
      "reward_std": 0.08977919071912766,
      "rewards/accuracy_reward": 0.9375000298023224,
      "rewards/format_reward": 1.0,
      "step": 113
    },
    {
      "completion_length": 26.375000953674316,
      "epoch": 0.5265588914549654,
      "grad_norm": 4.6412272453308105,
      "kl": 0.024200439453125,
      "learning_rate": 1e-06,
      "loss": 0.001,
      "reward": 1.7708333730697632,
      "reward_std": 0.08977919071912766,
      "rewards/accuracy_reward": 0.7708333432674408,
      "rewards/format_reward": 1.0,
      "step": 114
    },
    {
      "completion_length": 24.70833396911621,
      "epoch": 0.5311778290993071,
      "grad_norm": 0.01729178987443447,
      "kl": 0.014862060546875,
      "learning_rate": 1e-06,
      "loss": 0.0006,
      "reward": 2.0,
      "reward_std": 0.0,
      "rewards/accuracy_reward": 1.0,
      "rewards/format_reward": 1.0,
      "step": 115
    },
    {
      "completion_length": 22.875,
      "epoch": 0.535796766743649,
      "grad_norm": 1.4758672714233398,
      "kl": 0.011444091796875,
      "learning_rate": 1e-06,
      "loss": 0.0005,
      "reward": 1.9166666865348816,
      "reward_std": 0.0,
      "rewards/accuracy_reward": 0.9166666865348816,
      "rewards/format_reward": 1.0,
      "step": 116
    },
    {
      "completion_length": 45.125,
      "epoch": 0.5404157043879908,
      "grad_norm": 0.3602856695652008,
      "kl": 0.01806640625,
      "learning_rate": 1e-06,
      "loss": 0.0007,
      "reward": 1.9791666865348816,
      "reward_std": 0.0416666679084301,
      "rewards/accuracy_reward": 0.9791666865348816,
      "rewards/format_reward": 1.0,
      "step": 117
    },
    {
      "completion_length": 32.33333396911621,
      "epoch": 0.5450346420323325,
      "grad_norm": 11.787064552307129,
      "kl": 0.0201416015625,
      "learning_rate": 1e-06,
      "loss": 0.0008,
      "reward": 1.9791666865348816,
      "reward_std": 0.0416666679084301,
      "rewards/accuracy_reward": 0.9791666865348816,
      "rewards/format_reward": 1.0,
      "step": 118
    },
    {
      "completion_length": 33.10416793823242,
      "epoch": 0.5496535796766744,
      "grad_norm": 16.938676834106445,
      "kl": 0.0124664306640625,
      "learning_rate": 1e-06,
      "loss": 0.0005,
      "reward": 1.9375,
      "reward_std": 0.0416666679084301,
      "rewards/accuracy_reward": 0.9375,
      "rewards/format_reward": 1.0,
      "step": 119
    },
    {
      "completion_length": 21.854167461395264,
      "epoch": 0.5542725173210161,
      "grad_norm": 5.987547874450684,
      "kl": 0.017303466796875,
      "learning_rate": 1e-06,
      "loss": 0.0007,
      "reward": 1.8958333730697632,
      "reward_std": 0.0416666679084301,
      "rewards/accuracy_reward": 0.8958333730697632,
      "rewards/format_reward": 1.0,
      "step": 120
    },
    {
      "completion_length": 24.729167938232422,
      "epoch": 0.558891454965358,
      "grad_norm": 0.01790229044854641,
      "kl": 0.009857177734375,
      "learning_rate": 1e-06,
      "loss": 0.0004,
      "reward": 1.9166666865348816,
      "reward_std": 0.0,
      "rewards/accuracy_reward": 0.9166666865348816,
      "rewards/format_reward": 1.0,
      "step": 121
    },
    {
      "completion_length": 15.333333492279053,
      "epoch": 0.5635103926096998,
      "grad_norm": 0.004795704036951065,
      "kl": 0.0053558349609375,
      "learning_rate": 1e-06,
      "loss": 0.0002,
      "reward": 2.0,
      "reward_std": 0.0,
      "rewards/accuracy_reward": 1.0,
      "rewards/format_reward": 1.0,
      "step": 122
    },
    {
      "completion_length": 19.25,
      "epoch": 0.5681293302540416,
      "grad_norm": 0.021862277761101723,
      "kl": 0.0130615234375,
      "learning_rate": 1e-06,
      "loss": 0.0005,
      "reward": 2.0,
      "reward_std": 0.0,
      "rewards/accuracy_reward": 1.0,
      "rewards/format_reward": 1.0,
      "step": 123
    },
    {
      "completion_length": 18.20833396911621,
      "epoch": 0.5727482678983834,
      "grad_norm": 0.01567975804209709,
      "kl": 0.0094451904296875,
      "learning_rate": 1e-06,
      "loss": 0.0004,
      "reward": 2.0,
      "reward_std": 0.0,
      "rewards/accuracy_reward": 1.0,
      "rewards/format_reward": 1.0,
      "step": 124
    },
    {
      "completion_length": 21.479166984558105,
      "epoch": 0.5773672055427251,
      "grad_norm": 8.731499671936035,
      "kl": 0.0161590576171875,
      "learning_rate": 1e-06,
      "loss": 0.0006,
      "reward": 1.9791666865348816,
      "reward_std": 0.0416666679084301,
      "rewards/accuracy_reward": 0.9791666865348816,
      "rewards/format_reward": 1.0,
      "step": 125
    },
    {
      "completion_length": 15.083333492279053,
      "epoch": 0.581986143187067,
      "grad_norm": 25.284887313842773,
      "kl": 0.006011962890625,
      "learning_rate": 1e-06,
      "loss": 0.0002,
      "reward": 1.9583333730697632,
      "reward_std": 0.048112522810697556,
      "rewards/accuracy_reward": 0.9583333432674408,
      "rewards/format_reward": 1.0,
      "step": 126
    },
    {
      "completion_length": 24.312500953674316,
      "epoch": 0.5866050808314087,
      "grad_norm": 0.015062646940350533,
      "kl": 0.0047454833984375,
      "learning_rate": 1e-06,
      "loss": 0.0002,
      "reward": 2.0,
      "reward_std": 0.0,
      "rewards/accuracy_reward": 1.0,
      "rewards/format_reward": 1.0,
      "step": 127
    },
    {
      "completion_length": 37.66666793823242,
      "epoch": 0.5912240184757506,
      "grad_norm": 0.048227909952402115,
      "kl": 0.0377197265625,
      "learning_rate": 1e-06,
      "loss": 0.0015,
      "reward": 1.9166666865348816,
      "reward_std": 0.0,
      "rewards/accuracy_reward": 0.9166666865348816,
      "rewards/format_reward": 1.0,
      "step": 128
    },
    {
      "completion_length": 19.291667461395264,
      "epoch": 0.5958429561200924,
      "grad_norm": 8.027989387512207,
      "kl": 0.0084075927734375,
      "learning_rate": 1e-06,
      "loss": 0.0003,
      "reward": 1.8750000596046448,
      "reward_std": 0.048112522810697556,
      "rewards/accuracy_reward": 0.8750000298023224,
      "rewards/format_reward": 1.0,
      "step": 129
    },
    {
      "completion_length": 15.083333492279053,
      "epoch": 0.6004618937644342,
      "grad_norm": 0.0062526981346309185,
      "kl": 0.00661468505859375,
      "learning_rate": 1e-06,
      "loss": 0.0003,
      "reward": 2.0,
      "reward_std": 0.0,
      "rewards/accuracy_reward": 1.0,
      "rewards/format_reward": 1.0,
      "step": 130
    },
    {
      "completion_length": 22.375000476837158,
      "epoch": 0.605080831408776,
      "grad_norm": 0.026607630774378777,
      "kl": 0.0137176513671875,
      "learning_rate": 1e-06,
      "loss": 0.0005,
      "reward": 2.0,
      "reward_std": 0.0,
      "rewards/accuracy_reward": 1.0,
      "rewards/format_reward": 1.0,
      "step": 131
    },
    {
      "completion_length": 20.95833396911621,
      "epoch": 0.6096997690531177,
      "grad_norm": 0.00871030893176794,
      "kl": 0.0045928955078125,
      "learning_rate": 1e-06,
      "loss": 0.0002,
      "reward": 2.0,
      "reward_std": 0.0,
      "rewards/accuracy_reward": 1.0,
      "rewards/format_reward": 1.0,
      "step": 132
    },
    {
      "completion_length": 19.291666984558105,
      "epoch": 0.6143187066974596,
      "grad_norm": 0.0312328077852726,
      "kl": 0.0061187744140625,
      "learning_rate": 1e-06,
      "loss": 0.0002,
      "reward": 1.9166666865348816,
      "reward_std": 0.0,
      "rewards/accuracy_reward": 0.9166666865348816,
      "rewards/format_reward": 1.0,
      "step": 133
    },
    {
      "completion_length": 35.750000953674316,
      "epoch": 0.6189376443418014,
      "grad_norm": 0.07837596535682678,
      "kl": 0.01300048828125,
      "learning_rate": 1e-06,
      "loss": 0.0005,
      "reward": 2.0,
      "reward_std": 0.0,
      "rewards/accuracy_reward": 1.0,
      "rewards/format_reward": 1.0,
      "step": 134
    },
    {
      "completion_length": 15.250000476837158,
      "epoch": 0.6235565819861432,
      "grad_norm": 0.02661307342350483,
      "kl": 0.0044708251953125,
      "learning_rate": 1e-06,
      "loss": 0.0002,
      "reward": 1.8333333730697632,
      "reward_std": 0.0,
      "rewards/accuracy_reward": 0.8333333730697632,
      "rewards/format_reward": 1.0,
      "step": 135
    },
    {
      "completion_length": 47.916669845581055,
      "epoch": 0.628175519630485,
      "grad_norm": 0.025101438164711,
      "kl": 0.0069580078125,
      "learning_rate": 1e-06,
      "loss": 0.0003,
      "reward": 1.9166666865348816,
      "reward_std": 0.0,
      "rewards/accuracy_reward": 0.9166666865348816,
      "rewards/format_reward": 1.0,
      "step": 136
    },
    {
      "completion_length": 26.479167461395264,
      "epoch": 0.6327944572748267,
      "grad_norm": 22.01622200012207,
      "kl": 0.02655029296875,
      "learning_rate": 1e-06,
      "loss": 0.0011,
      "reward": 1.9791666865348816,
      "reward_std": 0.0416666679084301,
      "rewards/accuracy_reward": 0.9791666865348816,
      "rewards/format_reward": 1.0,
      "step": 137
    },
    {
      "completion_length": 31.95833396911621,
      "epoch": 0.6374133949191686,
      "grad_norm": 26.566116333007812,
      "kl": 0.012664794921875,
      "learning_rate": 1e-06,
      "loss": 0.0005,
      "reward": 1.8541666865348816,
      "reward_std": 0.08977919071912766,
      "rewards/accuracy_reward": 0.8541666865348816,
      "rewards/format_reward": 1.0,
      "step": 138
    },
    {
      "completion_length": 24.416667461395264,
      "epoch": 0.6420323325635104,
      "grad_norm": 0.47595617175102234,
      "kl": 0.006439208984375,
      "learning_rate": 1e-06,
      "loss": 0.0003,
      "reward": 1.9791666865348816,
      "reward_std": 0.0416666679084301,
      "rewards/accuracy_reward": 0.9791666865348816,
      "rewards/format_reward": 1.0,
      "step": 139
    },
    {
      "completion_length": 20.437500476837158,
      "epoch": 0.6466512702078522,
      "grad_norm": 0.010236713103950024,
      "kl": 0.005096435546875,
      "learning_rate": 1e-06,
      "loss": 0.0002,
      "reward": 1.9166666865348816,
      "reward_std": 0.0,
      "rewards/accuracy_reward": 0.9166666865348816,
      "rewards/format_reward": 1.0,
      "step": 140
    },
    {
      "completion_length": 35.16666793823242,
      "epoch": 0.651270207852194,
      "grad_norm": 0.3795807659626007,
      "kl": 0.0071563720703125,
      "learning_rate": 1e-06,
      "loss": 0.0003,
      "reward": 1.9791666865348816,
      "reward_std": 0.0416666679084301,
      "rewards/accuracy_reward": 0.9791666865348816,
      "rewards/format_reward": 1.0,
      "step": 141
    },
    {
      "completion_length": 27.229166984558105,
      "epoch": 0.6558891454965358,
      "grad_norm": 0.025517858564853668,
      "kl": 0.0072021484375,
      "learning_rate": 1e-06,
      "loss": 0.0003,
      "reward": 2.0,
      "reward_std": 0.0,
      "rewards/accuracy_reward": 1.0,
      "rewards/format_reward": 1.0,
      "step": 142
    },
    {
      "completion_length": 15.0,
      "epoch": 0.6605080831408776,
      "grad_norm": 7.9069013595581055,
      "kl": 0.00848388671875,
      "learning_rate": 1e-06,
      "loss": 0.0003,
      "reward": 1.9791666865348816,
      "reward_std": 0.0416666679084301,
      "rewards/accuracy_reward": 0.9791666865348816,
      "rewards/format_reward": 1.0,
      "step": 143
    },
    {
      "completion_length": 17.83333396911621,
      "epoch": 0.6651270207852193,
      "grad_norm": 0.014802374877035618,
      "kl": 0.00531005859375,
      "learning_rate": 1e-06,
      "loss": 0.0002,
      "reward": 1.8333333730697632,
      "reward_std": 0.0,
      "rewards/accuracy_reward": 0.8333333730697632,
      "rewards/format_reward": 1.0,
      "step": 144
    },
    {
      "completion_length": 23.291667461395264,
      "epoch": 0.6697459584295612,
      "grad_norm": 8.15732479095459,
      "kl": 0.005279541015625,
      "learning_rate": 1e-06,
      "loss": 0.0002,
      "reward": 1.9583333730697632,
      "reward_std": 0.048112522810697556,
      "rewards/accuracy_reward": 0.9583333432674408,
      "rewards/format_reward": 1.0,
      "step": 145
    },
    {
      "completion_length": 21.791667461395264,
      "epoch": 0.674364896073903,
      "grad_norm": 0.008288037963211536,
      "kl": 0.00370025634765625,
      "learning_rate": 1e-06,
      "loss": 0.0001,
      "reward": 2.0,
      "reward_std": 0.0,
      "rewards/accuracy_reward": 1.0,
      "rewards/format_reward": 1.0,
      "step": 146
    },
    {
      "completion_length": 34.47916793823242,
      "epoch": 0.6789838337182448,
      "grad_norm": 0.3849964439868927,
      "kl": 0.011077880859375,
      "learning_rate": 1e-06,
      "loss": 0.0004,
      "reward": 1.8125000596046448,
      "reward_std": 0.0416666679084301,
      "rewards/accuracy_reward": 0.8125000298023224,
      "rewards/format_reward": 1.0,
      "step": 147
    },
    {
      "completion_length": 28.166667461395264,
      "epoch": 0.6836027713625866,
      "grad_norm": 0.012812400236725807,
      "kl": 0.00555419921875,
      "learning_rate": 1e-06,
      "loss": 0.0002,
      "reward": 2.0,
      "reward_std": 0.0,
      "rewards/accuracy_reward": 1.0,
      "rewards/format_reward": 1.0,
      "step": 148
    },
    {
      "completion_length": 45.85416793823242,
      "epoch": 0.6882217090069284,
      "grad_norm": 0.016709528863430023,
      "kl": 0.009246826171875,
      "learning_rate": 1e-06,
      "loss": 0.0004,
      "reward": 2.0,
      "reward_std": 0.0,
      "rewards/accuracy_reward": 1.0,
      "rewards/format_reward": 1.0,
      "step": 149
    },
    {
      "completion_length": 51.25000286102295,
      "epoch": 0.6928406466512702,
      "grad_norm": 0.6184998154640198,
      "kl": 0.012664794921875,
      "learning_rate": 1e-06,
      "loss": 0.0005,
      "reward": 1.9583333730697632,
      "reward_std": 0.0833333358168602,
      "rewards/accuracy_reward": 0.9583333730697632,
      "rewards/format_reward": 1.0,
      "step": 150
    }
  ],
  "logging_steps": 1.0,
  "max_steps": 216,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 1,
  "save_steps": 50,
  "stateful_callbacks": {
    "TrainerControl": {
      "args": {
        "should_epoch_stop": false,
        "should_evaluate": false,
        "should_log": false,
        "should_save": true,
        "should_training_stop": false
      },
      "attributes": {}
    }
  },
  "total_flos": 0.0,
  "train_batch_size": 1,
  "trial_name": null,
  "trial_params": null
}