{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 2.992,
  "eval_steps": 10,
  "global_step": 279,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "clip_ratio": 0.0,
      "completion_length": 231.22917556762695,
      "epoch": 0.010666666666666666,
      "grad_norm": 0.1495383381843567,
      "kl": 0.0,
      "learning_rate": 1.0714285714285714e-07,
      "loss": -0.0555,
      "reward": 0.1458333358168602,
      "reward_std": 0.2350771240890026,
      "rewards/accuracy_reward": 0.0833333358168602,
      "rewards/format_reward": 0.06250000186264515,
      "step": 1
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 271.77084732055664,
      "epoch": 0.021333333333333333,
      "grad_norm": 0.11679606884717941,
      "kl": 0.0,
      "learning_rate": 2.1428571428571428e-07,
      "loss": -0.0391,
      "reward": 0.1458333358168602,
      "reward_std": 0.27258946746587753,
      "rewards/accuracy_reward": 0.06250000186264515,
      "rewards/format_reward": 0.0833333358168602,
      "step": 2
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 331.87500953674316,
      "epoch": 0.032,
      "grad_norm": 0.16674447059631348,
      "kl": 0.00020113587379455566,
      "learning_rate": 3.214285714285714e-07,
      "loss": -0.0214,
      "reward": 0.14583333767950535,
      "reward_std": 0.28219256922602654,
      "rewards/accuracy_reward": 0.10416666977107525,
      "rewards/format_reward": 0.0416666679084301,
      "step": 3
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 305.97917556762695,
      "epoch": 0.042666666666666665,
      "grad_norm": 0.16403762996196747,
      "kl": 9.453296661376953e-05,
      "learning_rate": 4.2857142857142857e-07,
      "loss": 0.0042,
      "reward": 0.1666666716337204,
      "reward_std": 0.3332235999405384,
      "rewards/accuracy_reward": 0.0416666679084301,
      "rewards/format_reward": 0.1250000037252903,
      "step": 4
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 292.6458463668823,
      "epoch": 0.05333333333333334,
      "grad_norm": 0.026218686252832413,
      "kl": 0.00013911724090576172,
      "learning_rate": 5.357142857142858e-07,
      "loss": -0.0034,
      "reward": 0.02083333395421505,
      "reward_std": 0.05103103443980217,
      "rewards/accuracy_reward": 0.02083333395421505,
      "rewards/format_reward": 0.0,
      "step": 5
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 235.87500858306885,
      "epoch": 0.064,
      "grad_norm": 0.17623676359653473,
      "kl": 0.0001900792121887207,
      "learning_rate": 6.428571428571428e-07,
      "loss": -0.082,
      "reward": 0.1875000037252903,
      "reward_std": 0.2996268533170223,
      "rewards/accuracy_reward": 0.10416666977107525,
      "rewards/format_reward": 0.0833333358168602,
      "step": 6
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 266.6666774749756,
      "epoch": 0.07466666666666667,
      "grad_norm": 0.13698340952396393,
      "kl": 0.00019168853759765625,
      "learning_rate": 7.5e-07,
      "loss": 0.1021,
      "reward": 0.2083333395421505,
      "reward_std": 0.43528567254543304,
      "rewards/accuracy_reward": 0.1250000037252903,
      "rewards/format_reward": 0.0833333358168602,
      "step": 7
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 221.31250762939453,
      "epoch": 0.08533333333333333,
      "grad_norm": 0.20153407752513885,
      "kl": 0.0002624690532684326,
      "learning_rate": 8.571428571428571e-07,
      "loss": -0.0188,
      "reward": 0.35416667349636555,
      "reward_std": 0.48367293924093246,
      "rewards/accuracy_reward": 0.1666666716337204,
      "rewards/format_reward": 0.1875000037252903,
      "step": 8
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 256.7291736602783,
      "epoch": 0.096,
      "grad_norm": 0.14511053264141083,
      "kl": 0.0002912282943725586,
      "learning_rate": 9.642857142857145e-07,
      "loss": 0.0278,
      "reward": 0.27083334140479565,
      "reward_std": 0.4177170880138874,
      "rewards/accuracy_reward": 0.14583333767950535,
      "rewards/format_reward": 0.1250000037252903,
      "step": 9
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 275.1458396911621,
      "epoch": 0.10666666666666667,
      "grad_norm": 0.10830721259117126,
      "kl": 0.0003374814987182617,
      "learning_rate": 1.0714285714285716e-06,
      "loss": -0.0591,
      "reward": 0.16666666977107525,
      "reward_std": 0.18404608592391014,
      "rewards/accuracy_reward": 0.1458333358168602,
      "rewards/format_reward": 0.02083333395421505,
      "step": 10
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 203.83333778381348,
      "epoch": 0.11733333333333333,
      "grad_norm": 0.16608387231826782,
      "kl": 0.0005745887756347656,
      "learning_rate": 1.1785714285714285e-06,
      "loss": 0.0155,
      "reward": 0.22916667349636555,
      "reward_std": 0.41129202395677567,
      "rewards/accuracy_reward": 0.10416666977107525,
      "rewards/format_reward": 0.1250000037252903,
      "step": 11
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 275.6458396911621,
      "epoch": 0.128,
      "grad_norm": 0.16276516020298004,
      "kl": 0.0009386539459228516,
      "learning_rate": 1.2857142857142856e-06,
      "loss": -0.1055,
      "reward": 0.3333333395421505,
      "reward_std": 0.46875908970832825,
      "rewards/accuracy_reward": 0.1666666716337204,
      "rewards/format_reward": 0.16666666977107525,
      "step": 12
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 209.54167366027832,
      "epoch": 0.13866666666666666,
      "grad_norm": 0.20444652438163757,
      "kl": 0.0017161369323730469,
      "learning_rate": 1.392857142857143e-06,
      "loss": 0.0046,
      "reward": 0.27083334140479565,
      "reward_std": 0.4217669703066349,
      "rewards/accuracy_reward": 0.10416666977107525,
      "rewards/format_reward": 0.1666666679084301,
      "step": 13
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 211.5000057220459,
      "epoch": 0.14933333333333335,
      "grad_norm": 0.1791943609714508,
      "kl": 0.0035800933837890625,
      "learning_rate": 1.5e-06,
      "loss": -0.0985,
      "reward": 0.3750000074505806,
      "reward_std": 0.49719163402915,
      "rewards/accuracy_reward": 0.16666666977107525,
      "rewards/format_reward": 0.2083333395421505,
      "step": 14
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 292.8750057220459,
      "epoch": 0.16,
      "grad_norm": 0.1786985695362091,
      "kl": 0.0047359466552734375,
      "learning_rate": 1.6071428571428572e-06,
      "loss": -0.0043,
      "reward": 0.37500000931322575,
      "reward_std": 0.44172170385718346,
      "rewards/accuracy_reward": 0.06250000186264515,
      "rewards/format_reward": 0.3125000074505806,
      "step": 15
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 273.43750762939453,
      "epoch": 0.17066666666666666,
      "grad_norm": 0.17925456166267395,
      "kl": 0.010840415954589844,
      "learning_rate": 1.7142857142857143e-06,
      "loss": 0.0074,
      "reward": 0.4583333432674408,
      "reward_std": 0.38552645966410637,
      "rewards/accuracy_reward": 0.10416666977107525,
      "rewards/format_reward": 0.3541666716337204,
      "step": 16
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 264.7916793823242,
      "epoch": 0.18133333333333335,
      "grad_norm": 0.1461903601884842,
      "kl": 0.02383136749267578,
      "learning_rate": 1.8214285714285714e-06,
      "loss": -0.0197,
      "reward": 0.6250000037252903,
      "reward_std": 0.46026117354631424,
      "rewards/accuracy_reward": 0.2083333395421505,
      "rewards/format_reward": 0.4166666753590107,
      "step": 17
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 319.12500762939453,
      "epoch": 0.192,
      "grad_norm": 0.17338670790195465,
      "kl": 0.036746978759765625,
      "learning_rate": 1.928571428571429e-06,
      "loss": 0.0239,
      "reward": 0.6458333395421505,
      "reward_std": 0.5965815186500549,
      "rewards/accuracy_reward": 0.1875000037252903,
      "rewards/format_reward": 0.45833334513008595,
      "step": 18
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 218.85417366027832,
      "epoch": 0.20266666666666666,
      "grad_norm": 0.15455670654773712,
      "kl": 0.028100967407226562,
      "learning_rate": 2.035714285714286e-06,
      "loss": 0.0165,
      "reward": 0.9791666939854622,
      "reward_std": 0.6326877325773239,
      "rewards/accuracy_reward": 0.2291666716337204,
      "rewards/format_reward": 0.7500000298023224,
      "step": 19
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 328.1875114440918,
      "epoch": 0.21333333333333335,
      "grad_norm": 0.3660297095775604,
      "kl": 0.2696857452392578,
      "learning_rate": 2.142857142857143e-06,
      "loss": 0.0074,
      "reward": 0.5416666828095913,
      "reward_std": 0.6121251657605171,
      "rewards/accuracy_reward": 0.06250000186264515,
      "rewards/format_reward": 0.4791666753590107,
      "step": 20
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 185.58334159851074,
      "epoch": 0.224,
      "grad_norm": 0.18908099830150604,
      "kl": 0.034381866455078125,
      "learning_rate": 2.25e-06,
      "loss": 0.0302,
      "reward": 0.9166666846722364,
      "reward_std": 0.46722716465592384,
      "rewards/accuracy_reward": 0.2708333432674408,
      "rewards/format_reward": 0.6458333488553762,
      "step": 21
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 269.4583435058594,
      "epoch": 0.23466666666666666,
      "grad_norm": 0.23737290501594543,
      "kl": 0.07430791854858398,
      "learning_rate": 2.357142857142857e-06,
      "loss": 0.0205,
      "reward": 0.6875000186264515,
      "reward_std": 0.41912321373820305,
      "rewards/accuracy_reward": 0.02083333395421505,
      "rewards/format_reward": 0.6666666846722364,
      "step": 22
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 289.9791774749756,
      "epoch": 0.24533333333333332,
      "grad_norm": 0.11483895033597946,
      "kl": 0.026653289794921875,
      "learning_rate": 2.4642857142857143e-06,
      "loss": 0.0206,
      "reward": 0.7708333469927311,
      "reward_std": 0.34674229472875595,
      "rewards/accuracy_reward": 0.06250000186264515,
      "rewards/format_reward": 0.7083333395421505,
      "step": 23
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 256.7083377838135,
      "epoch": 0.256,
      "grad_norm": 0.10966558754444122,
      "kl": 0.009334564208984375,
      "learning_rate": 2.571428571428571e-06,
      "loss": 0.0168,
      "reward": 0.9583333730697632,
      "reward_std": 0.2861081585288048,
      "rewards/accuracy_reward": 0.1875000074505806,
      "rewards/format_reward": 0.7708333395421505,
      "step": 24
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 255.9583396911621,
      "epoch": 0.26666666666666666,
      "grad_norm": 0.19421663880348206,
      "kl": 0.01947784423828125,
      "learning_rate": 2.678571428571429e-06,
      "loss": 0.0648,
      "reward": 0.8750000335276127,
      "reward_std": 0.46787626668810844,
      "rewards/accuracy_reward": 0.1250000037252903,
      "rewards/format_reward": 0.7500000186264515,
      "step": 25
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 270.50000762939453,
      "epoch": 0.2773333333333333,
      "grad_norm": 0.13444656133651733,
      "kl": 0.013866424560546875,
      "learning_rate": 2.785714285714286e-06,
      "loss": 0.132,
      "reward": 0.833333358168602,
      "reward_std": 0.43123577907681465,
      "rewards/accuracy_reward": 0.0833333358168602,
      "rewards/format_reward": 0.7500000074505806,
      "step": 26
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 292.6458377838135,
      "epoch": 0.288,
      "grad_norm": 0.10114189982414246,
      "kl": 0.01815032958984375,
      "learning_rate": 2.892857142857143e-06,
      "loss": 0.0158,
      "reward": 0.9166666939854622,
      "reward_std": 0.37223926186561584,
      "rewards/accuracy_reward": 0.20833333767950535,
      "rewards/format_reward": 0.7083333432674408,
      "step": 27
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 263.29167079925537,
      "epoch": 0.2986666666666667,
      "grad_norm": 0.14505182206630707,
      "kl": 0.022125244140625,
      "learning_rate": 3e-06,
      "loss": 0.082,
      "reward": 0.9166667126119137,
      "reward_std": 0.41380149498581886,
      "rewards/accuracy_reward": 0.20833333767950535,
      "rewards/format_reward": 0.7083333507180214,
      "step": 28
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 256.2500057220459,
      "epoch": 0.30933333333333335,
      "grad_norm": 0.07133425027132034,
      "kl": 0.026397705078125,
      "learning_rate": 2.999882508108334e-06,
      "loss": 0.0633,
      "reward": 0.8958333432674408,
      "reward_std": 0.1801304928958416,
      "rewards/accuracy_reward": 0.0,
      "rewards/format_reward": 0.8958333432674408,
      "step": 29
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 241.0000057220459,
      "epoch": 0.32,
      "grad_norm": 0.1608133763074875,
      "kl": 0.03473663330078125,
      "learning_rate": 2.9995300508391286e-06,
      "loss": 0.0251,
      "reward": 1.0416666939854622,
      "reward_std": 0.4235428161919117,
      "rewards/accuracy_reward": 0.1875000074505806,
      "rewards/format_reward": 0.854166679084301,
      "step": 30
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 242.22917366027832,
      "epoch": 0.33066666666666666,
      "grad_norm": 0.15004193782806396,
      "kl": 0.02217864990234375,
      "learning_rate": 2.998942683406879e-06,
      "loss": 0.0159,
      "reward": 1.0000000223517418,
      "reward_std": 0.5234047770500183,
      "rewards/accuracy_reward": 0.1875000037252903,
      "rewards/format_reward": 0.8125000298023224,
      "step": 31
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 195.08334064483643,
      "epoch": 0.3413333333333333,
      "grad_norm": 0.1392156332731247,
      "kl": 0.03990936279296875,
      "learning_rate": 2.998120497826133e-06,
      "loss": 0.048,
      "reward": 0.9791666939854622,
      "reward_std": 0.29669977352023125,
      "rewards/accuracy_reward": 0.1458333358168602,
      "rewards/format_reward": 0.8333333469927311,
      "step": 32
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 229.68751049041748,
      "epoch": 0.352,
      "grad_norm": 0.15871571004390717,
      "kl": 0.030364990234375,
      "learning_rate": 2.9970636228970757e-06,
      "loss": 0.0226,
      "reward": 1.0416666939854622,
      "reward_std": 0.2957112565636635,
      "rewards/accuracy_reward": 0.10416666977107525,
      "rewards/format_reward": 0.9375000074505806,
      "step": 33
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 158.83333778381348,
      "epoch": 0.3626666666666667,
      "grad_norm": 0.16293902695178986,
      "kl": 0.02352142333984375,
      "learning_rate": 2.995772224185354e-06,
      "loss": 0.0433,
      "reward": 1.2708333432674408,
      "reward_std": 0.37061261385679245,
      "rewards/accuracy_reward": 0.2916666716337204,
      "rewards/format_reward": 0.9791666716337204,
      "step": 34
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 249.41667556762695,
      "epoch": 0.37333333333333335,
      "grad_norm": 0.1561712920665741,
      "kl": 0.02712249755859375,
      "learning_rate": 2.9942465039961373e-06,
      "loss": 0.0097,
      "reward": 1.1041666716337204,
      "reward_std": 0.34607556089758873,
      "rewards/accuracy_reward": 0.1875000037252903,
      "rewards/format_reward": 0.9166666865348816,
      "step": 35
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 269.75000762939453,
      "epoch": 0.384,
      "grad_norm": 0.12326981872320175,
      "kl": 0.0439910888671875,
      "learning_rate": 2.992486701342427e-06,
      "loss": 0.0882,
      "reward": 1.0000000298023224,
      "reward_std": 0.39079636335372925,
      "rewards/accuracy_reward": 0.1250000037252903,
      "rewards/format_reward": 0.8750000223517418,
      "step": 36
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 195.60417461395264,
      "epoch": 0.39466666666666667,
      "grad_norm": 0.14010730385780334,
      "kl": 0.0265045166015625,
      "learning_rate": 2.9904930919076145e-06,
      "loss": 0.0183,
      "reward": 1.1875000298023224,
      "reward_std": 0.2996268570423126,
      "rewards/accuracy_reward": 0.2291666716337204,
      "rewards/format_reward": 0.9583333432674408,
      "step": 37
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 264.81250953674316,
      "epoch": 0.4053333333333333,
      "grad_norm": 0.16210472583770752,
      "kl": 0.0369720458984375,
      "learning_rate": 2.9882659880022905e-06,
      "loss": 0.0318,
      "reward": 1.1250000521540642,
      "reward_std": 0.5272140763700008,
      "rewards/accuracy_reward": 0.3333333432674408,
      "rewards/format_reward": 0.7916666865348816,
      "step": 38
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 180.5833396911621,
      "epoch": 0.416,
      "grad_norm": 0.13184523582458496,
      "kl": 0.0410919189453125,
      "learning_rate": 2.985805738515323e-06,
      "loss": -0.0404,
      "reward": 1.2291666865348816,
      "reward_std": 0.3411714509129524,
      "rewards/accuracy_reward": 0.25000000558793545,
      "rewards/format_reward": 0.9791666716337204,
      "step": 39
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 206.54166793823242,
      "epoch": 0.4266666666666667,
      "grad_norm": 0.13690926134586334,
      "kl": 0.039154052734375,
      "learning_rate": 2.9831127288592004e-06,
      "loss": 0.0823,
      "reward": 1.062500037252903,
      "reward_std": 0.40168892219662666,
      "rewards/accuracy_reward": 0.1666666716337204,
      "rewards/format_reward": 0.8958333507180214,
      "step": 40
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 258.16667556762695,
      "epoch": 0.43733333333333335,
      "grad_norm": 0.17689989507198334,
      "kl": 0.03472900390625,
      "learning_rate": 2.9801873809096543e-06,
      "loss": 0.0241,
      "reward": 1.145833346992731,
      "reward_std": 0.46066781133413315,
      "rewards/accuracy_reward": 0.3125000074505806,
      "rewards/format_reward": 0.8333333414047956,
      "step": 41
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 201.5000057220459,
      "epoch": 0.448,
      "grad_norm": 0.15305694937705994,
      "kl": 0.08642578125,
      "learning_rate": 2.97703015293957e-06,
      "loss": -0.0321,
      "reward": 1.1875000298023224,
      "reward_std": 0.4862990714609623,
      "rewards/accuracy_reward": 0.2916666753590107,
      "rewards/format_reward": 0.8958333507180214,
      "step": 42
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 164.06250286102295,
      "epoch": 0.45866666666666667,
      "grad_norm": 0.14044952392578125,
      "kl": 0.0565032958984375,
      "learning_rate": 2.9736415395471975e-06,
      "loss": 0.0723,
      "reward": 1.333333358168602,
      "reward_std": 0.4847872518002987,
      "rewards/accuracy_reward": 0.45833334513008595,
      "rewards/format_reward": 0.8750000149011612,
      "step": 43
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 190.35417366027832,
      "epoch": 0.4693333333333333,
      "grad_norm": 0.16754870116710663,
      "kl": 0.05095672607421875,
      "learning_rate": 2.970022071578666e-06,
      "loss": 0.0547,
      "reward": 1.4166667014360428,
      "reward_std": 0.43123577907681465,
      "rewards/accuracy_reward": 0.5416666846722364,
      "rewards/format_reward": 0.8750000074505806,
      "step": 44
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 257.3541736602783,
      "epoch": 0.48,
      "grad_norm": 0.11928242444992065,
      "kl": 0.05057525634765625,
      "learning_rate": 2.9661723160448267e-06,
      "loss": -0.0214,
      "reward": 1.2083333656191826,
      "reward_std": 0.3922024853527546,
      "rewards/accuracy_reward": 0.29166667349636555,
      "rewards/format_reward": 0.916666679084301,
      "step": 45
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 248.25000953674316,
      "epoch": 0.49066666666666664,
      "grad_norm": 0.406486839056015,
      "kl": 0.0576629638671875,
      "learning_rate": 2.962092876032427e-06,
      "loss": 0.0367,
      "reward": 1.2708333656191826,
      "reward_std": 0.36417657881975174,
      "rewards/accuracy_reward": 0.3750000111758709,
      "rewards/format_reward": 0.8958333432674408,
      "step": 46
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 240.0833387374878,
      "epoch": 0.5013333333333333,
      "grad_norm": 0.10968135297298431,
      "kl": 0.059417724609375,
      "learning_rate": 2.957784390609632e-06,
      "loss": 0.0364,
      "reward": 1.3958333656191826,
      "reward_std": 0.34674229472875595,
      "rewards/accuracy_reward": 0.4791666753590107,
      "rewards/format_reward": 0.916666679084301,
      "step": 47
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 255.45833778381348,
      "epoch": 0.512,
      "grad_norm": 0.09772133082151413,
      "kl": 0.0493011474609375,
      "learning_rate": 2.9532475347259113e-06,
      "loss": 0.0263,
      "reward": 1.5416667014360428,
      "reward_std": 0.3881702236831188,
      "rewards/accuracy_reward": 0.625000013038516,
      "rewards/format_reward": 0.916666679084301,
      "step": 48
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 230.87500762939453,
      "epoch": 0.5226666666666666,
      "grad_norm": 0.1263599842786789,
      "kl": 0.0523681640625,
      "learning_rate": 2.9484830191063057e-06,
      "loss": 0.0451,
      "reward": 1.3541666939854622,
      "reward_std": 0.5295314528048038,
      "rewards/accuracy_reward": 0.4791666753590107,
      "rewards/format_reward": 0.8750000149011612,
      "step": 49
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 312.2708377838135,
      "epoch": 0.5333333333333333,
      "grad_norm": 0.08942104130983353,
      "kl": 0.0318603515625,
      "learning_rate": 2.9434915901400857e-06,
      "loss": 0.06,
      "reward": 1.3125000298023224,
      "reward_std": 0.3947119526565075,
      "rewards/accuracy_reward": 0.41666667349636555,
      "rewards/format_reward": 0.8958333432674408,
      "step": 50
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 233.83334159851074,
      "epoch": 0.544,
      "grad_norm": 0.15756255388259888,
      "kl": 0.071136474609375,
      "learning_rate": 2.938274029763826e-06,
      "loss": 0.054,
      "reward": 1.250000037252903,
      "reward_std": 0.45534609258174896,
      "rewards/accuracy_reward": 0.37500000931322575,
      "rewards/format_reward": 0.8750000223517418,
      "step": 51
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 258.87500762939453,
      "epoch": 0.5546666666666666,
      "grad_norm": 0.11138404905796051,
      "kl": 0.033782958984375,
      "learning_rate": 2.9328311553389116e-06,
      "loss": 0.0751,
      "reward": 1.354166716337204,
      "reward_std": 0.5238114185631275,
      "rewards/accuracy_reward": 0.4583333432674408,
      "rewards/format_reward": 0.8958333507180214,
      "step": 52
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 256.56250953674316,
      "epoch": 0.5653333333333334,
      "grad_norm": 0.10758252441883087,
      "kl": 0.0422515869140625,
      "learning_rate": 2.927163819523492e-06,
      "loss": 0.0381,
      "reward": 1.437500037252903,
      "reward_std": 0.3842546418309212,
      "rewards/accuracy_reward": 0.5208333507180214,
      "rewards/format_reward": 0.916666679084301,
      "step": 53
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 309.83333587646484,
      "epoch": 0.576,
      "grad_norm": 0.10542929172515869,
      "kl": 0.0524139404296875,
      "learning_rate": 2.9212729101389084e-06,
      "loss": 0.0156,
      "reward": 1.5416666939854622,
      "reward_std": 0.5237057246267796,
      "rewards/accuracy_reward": 0.6875000149011612,
      "rewards/format_reward": 0.8541666939854622,
      "step": 54
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 369.87500762939453,
      "epoch": 0.5866666666666667,
      "grad_norm": 0.07796894013881683,
      "kl": 0.04473876953125,
      "learning_rate": 2.9151593500306103e-06,
      "loss": 0.0122,
      "reward": 1.2500000298023224,
      "reward_std": 0.3506578877568245,
      "rewards/accuracy_reward": 0.29166666977107525,
      "rewards/format_reward": 0.9583333432674408,
      "step": 55
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 243.08334159851074,
      "epoch": 0.5973333333333334,
      "grad_norm": 0.12414506822824478,
      "kl": 0.0529937744140625,
      "learning_rate": 2.9088240969235863e-06,
      "loss": 0.0219,
      "reward": 1.4583333879709244,
      "reward_std": 0.5347039774060249,
      "rewards/accuracy_reward": 0.5000000149011612,
      "rewards/format_reward": 0.9583333432674408,
      "step": 56
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 176.43750476837158,
      "epoch": 0.608,
      "grad_norm": 0.12592697143554688,
      "kl": 0.0508575439453125,
      "learning_rate": 2.902268143272333e-06,
      "loss": -0.0349,
      "reward": 1.8125000298023224,
      "reward_std": 0.33713918924331665,
      "rewards/accuracy_reward": 0.8125000223517418,
      "rewards/format_reward": 1.0,
      "step": 57
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 284.16666984558105,
      "epoch": 0.6186666666666667,
      "grad_norm": 0.08683699369430542,
      "kl": 0.0404052734375,
      "learning_rate": 2.895492516105378e-06,
      "loss": 0.0729,
      "reward": 1.4791666865348816,
      "reward_std": 0.37575671821832657,
      "rewards/accuracy_reward": 0.5208333507180214,
      "rewards/format_reward": 0.9583333432674408,
      "step": 58
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 305.87500381469727,
      "epoch": 0.6293333333333333,
      "grad_norm": 0.09318725019693375,
      "kl": 0.03826904296875,
      "learning_rate": 2.888498276864393e-06,
      "loss": 0.0262,
      "reward": 1.2708333656191826,
      "reward_std": 0.36417657881975174,
      "rewards/accuracy_reward": 0.33333333767950535,
      "rewards/format_reward": 0.9375000074505806,
      "step": 59
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 337.0416736602783,
      "epoch": 0.64,
      "grad_norm": 0.09196493029594421,
      "kl": 0.045013427734375,
      "learning_rate": 2.8812865212379087e-06,
      "loss": 0.0465,
      "reward": 1.4791667014360428,
      "reward_std": 0.33713918924331665,
      "rewards/accuracy_reward": 0.500000013038516,
      "rewards/format_reward": 0.9791666716337204,
      "step": 60
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 285.41667556762695,
      "epoch": 0.6506666666666666,
      "grad_norm": 0.12164270132780075,
      "kl": 0.0420989990234375,
      "learning_rate": 2.8738583789896743e-06,
      "loss": 0.0701,
      "reward": 1.5208333879709244,
      "reward_std": 0.5107103250920773,
      "rewards/accuracy_reward": 0.562500013038516,
      "rewards/format_reward": 0.9583333432674408,
      "step": 61
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 346.9166793823242,
      "epoch": 0.6613333333333333,
      "grad_norm": 0.08939547091722488,
      "kl": 0.0489654541015625,
      "learning_rate": 2.866215013781668e-06,
      "loss": -0.0066,
      "reward": 1.3750000447034836,
      "reward_std": 0.4743013270199299,
      "rewards/accuracy_reward": 0.45833334140479565,
      "rewards/format_reward": 0.916666679084301,
      "step": 62
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 301.2708435058594,
      "epoch": 0.672,
      "grad_norm": 0.10984641313552856,
      "kl": 0.0488128662109375,
      "learning_rate": 2.858357622991806e-06,
      "loss": -0.0121,
      "reward": 1.541666716337204,
      "reward_std": 0.46232306212186813,
      "rewards/accuracy_reward": 0.6458333563059568,
      "rewards/format_reward": 0.8958333507180214,
      "step": 63
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 269.2500066757202,
      "epoch": 0.6826666666666666,
      "grad_norm": 0.11388546973466873,
      "kl": 0.067291259765625,
      "learning_rate": 2.8502874375263646e-06,
      "loss": 0.0396,
      "reward": 1.2916666977107525,
      "reward_std": 0.3332235999405384,
      "rewards/accuracy_reward": 0.43750001303851604,
      "rewards/format_reward": 0.8541666753590107,
      "step": 64
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 227.08334159851074,
      "epoch": 0.6933333333333334,
      "grad_norm": 0.13996504247188568,
      "kl": 0.078887939453125,
      "learning_rate": 2.8420057216271533e-06,
      "loss": 0.0677,
      "reward": 1.4166667014360428,
      "reward_std": 0.3977733291685581,
      "rewards/accuracy_reward": 0.5208333432674408,
      "rewards/format_reward": 0.8958333358168602,
      "step": 65
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 279.1458396911621,
      "epoch": 0.704,
      "grad_norm": 0.061006851494312286,
      "kl": 0.0504913330078125,
      "learning_rate": 2.8335137726734607e-06,
      "loss": -0.0082,
      "reward": 1.5000000149011612,
      "reward_std": 0.18404608592391014,
      "rewards/accuracy_reward": 0.5208333358168602,
      "rewards/format_reward": 0.9791666716337204,
      "step": 66
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 227.02083587646484,
      "epoch": 0.7146666666666667,
      "grad_norm": 0.12499740719795227,
      "kl": 0.0751800537109375,
      "learning_rate": 2.8248129209788164e-06,
      "loss": -0.0358,
      "reward": 1.479166716337204,
      "reward_std": 0.47038573771715164,
      "rewards/accuracy_reward": 0.5625000186264515,
      "rewards/format_reward": 0.916666679084301,
      "step": 67
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 279.9791774749756,
      "epoch": 0.7253333333333334,
      "grad_norm": 0.08705603331327438,
      "kl": 0.0396881103515625,
      "learning_rate": 2.8159045295825857e-06,
      "loss": -0.0075,
      "reward": 1.4583333730697632,
      "reward_std": 0.45827317237854004,
      "rewards/accuracy_reward": 0.500000013038516,
      "rewards/format_reward": 0.9583333432674408,
      "step": 68
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 240.45834159851074,
      "epoch": 0.736,
      "grad_norm": 0.1296224296092987,
      "kl": 0.04766845703125,
      "learning_rate": 2.806789994036445e-06,
      "loss": 0.0998,
      "reward": 1.4583333879709244,
      "reward_std": 0.5228229016065598,
      "rewards/accuracy_reward": 0.47916668094694614,
      "rewards/format_reward": 0.9791666716337204,
      "step": 69
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 301.2916793823242,
      "epoch": 0.7466666666666667,
      "grad_norm": 0.10123682767152786,
      "kl": 0.06207275390625,
      "learning_rate": 2.7974707421857584e-06,
      "loss": 0.0305,
      "reward": 1.520833358168602,
      "reward_std": 0.2982207275927067,
      "rewards/accuracy_reward": 0.645833345130086,
      "rewards/format_reward": 0.8750000074505806,
      "step": 70
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 292.62500762939453,
      "epoch": 0.7573333333333333,
      "grad_norm": 0.10017960518598557,
      "kl": 0.0426025390625,
      "learning_rate": 2.7879482339458973e-06,
      "loss": 0.0018,
      "reward": 1.458333358168602,
      "reward_std": 0.2861081585288048,
      "rewards/accuracy_reward": 0.479166679084301,
      "rewards/format_reward": 0.9791666716337204,
      "step": 71
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 300.12500953674316,
      "epoch": 0.768,
      "grad_norm": 0.12141850590705872,
      "kl": 0.0420684814453125,
      "learning_rate": 2.778223961073537e-06,
      "loss": 0.0561,
      "reward": 1.4375000298023224,
      "reward_std": 0.41619613766670227,
      "rewards/accuracy_reward": 0.5000000186264515,
      "rewards/format_reward": 0.9375000149011612,
      "step": 72
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 232.16666984558105,
      "epoch": 0.7786666666666666,
      "grad_norm": 0.12889333069324493,
      "kl": 0.057708740234375,
      "learning_rate": 2.7682994469329646e-06,
      "loss": -0.04,
      "reward": 1.5625000596046448,
      "reward_std": 0.5363415889441967,
      "rewards/accuracy_reward": 0.6250000223517418,
      "rewards/format_reward": 0.9375000149011612,
      "step": 73
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 307.2291793823242,
      "epoch": 0.7893333333333333,
      "grad_norm": 0.13781766593456268,
      "kl": 0.0377960205078125,
      "learning_rate": 2.7581762462574334e-06,
      "loss": 0.014,
      "reward": 1.3958333730697632,
      "reward_std": 0.4997011013329029,
      "rewards/accuracy_reward": 0.47916668467223644,
      "rewards/format_reward": 0.9166666716337204,
      "step": 74
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 397.3958511352539,
      "epoch": 0.8,
      "grad_norm": 0.08684942871332169,
      "kl": 0.0388031005859375,
      "learning_rate": 2.747855944905606e-06,
      "loss": 0.0606,
      "reward": 1.229166716337204,
      "reward_std": 0.5326506160199642,
      "rewards/accuracy_reward": 0.35416666977107525,
      "rewards/format_reward": 0.8750000149011612,
      "step": 75
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 299.77084159851074,
      "epoch": 0.8106666666666666,
      "grad_norm": 0.10988432168960571,
      "kl": 0.0503997802734375,
      "learning_rate": 2.73734015961312e-06,
      "loss": -0.0247,
      "reward": 1.3333333730697632,
      "reward_std": 0.5198958218097687,
      "rewards/accuracy_reward": 0.5208333469927311,
      "rewards/format_reward": 0.8125000149011612,
      "step": 76
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 324.7916774749756,
      "epoch": 0.8213333333333334,
      "grad_norm": 0.13095711171627045,
      "kl": 0.070037841796875,
      "learning_rate": 2.7266305377393183e-06,
      "loss": 0.0494,
      "reward": 1.5833333432674408,
      "reward_std": 0.400033675134182,
      "rewards/accuracy_reward": 0.6666666734963655,
      "rewards/format_reward": 0.9166666865348816,
      "step": 77
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 356.4583435058594,
      "epoch": 0.832,
      "grad_norm": 0.10965587943792343,
      "kl": 0.04058837890625,
      "learning_rate": 2.7157287570091765e-06,
      "loss": 0.036,
      "reward": 1.354166716337204,
      "reward_std": 0.6204284392297268,
      "rewards/accuracy_reward": 0.47916668094694614,
      "rewards/format_reward": 0.8750000223517418,
      "step": 78
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 352.4166831970215,
      "epoch": 0.8426666666666667,
      "grad_norm": 0.11178199201822281,
      "kl": 0.048248291015625,
      "learning_rate": 2.7046365252504835e-06,
      "loss": 0.019,
      "reward": 1.5000000298023224,
      "reward_std": 0.5423886179924011,
      "rewards/accuracy_reward": 0.6250000186264515,
      "rewards/format_reward": 0.8750000223517418,
      "step": 79
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 318.5000171661377,
      "epoch": 0.8533333333333334,
      "grad_norm": 0.11200235784053802,
      "kl": 0.04592132568359375,
      "learning_rate": 2.693355580126295e-06,
      "loss": 0.0629,
      "reward": 1.416666716337204,
      "reward_std": 0.5645233504474163,
      "rewards/accuracy_reward": 0.4791666828095913,
      "rewards/format_reward": 0.9375000074505806,
      "step": 80
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 349.708345413208,
      "epoch": 0.864,
      "grad_norm": 0.09885663539171219,
      "kl": 0.0634307861328125,
      "learning_rate": 2.6818876888627203e-06,
      "loss": 0.0439,
      "reward": 1.5416666865348816,
      "reward_std": 0.49666832759976387,
      "rewards/accuracy_reward": 0.6875000074505806,
      "rewards/format_reward": 0.8541666865348816,
      "step": 81
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 292.77084159851074,
      "epoch": 0.8746666666666667,
      "grad_norm": 0.09445224702358246,
      "kl": 0.0454254150390625,
      "learning_rate": 2.6702346479720777e-06,
      "loss": 0.0829,
      "reward": 1.4791667088866234,
      "reward_std": 0.48782002553343773,
      "rewards/accuracy_reward": 0.6041666865348816,
      "rewards/format_reward": 0.8750000223517418,
      "step": 82
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 373.3125057220459,
      "epoch": 0.8853333333333333,
      "grad_norm": 0.6235383152961731,
      "kl": 0.069488525390625,
      "learning_rate": 2.6583982829714576e-06,
      "loss": 0.04,
      "reward": 1.3958333730697632,
      "reward_std": 0.4621887616813183,
      "rewards/accuracy_reward": 0.520833345130086,
      "rewards/format_reward": 0.8750000149011612,
      "step": 83
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 298.9583396911621,
      "epoch": 0.896,
      "grad_norm": 0.10517609119415283,
      "kl": 0.0439910888671875,
      "learning_rate": 2.6463804480967447e-06,
      "loss": 0.0363,
      "reward": 1.5000000298023224,
      "reward_std": 0.4338619112968445,
      "rewards/accuracy_reward": 0.5833333469927311,
      "rewards/format_reward": 0.916666679084301,
      "step": 84
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 347.3750057220459,
      "epoch": 0.9066666666666666,
      "grad_norm": 0.10065872222185135,
      "kl": 0.0599365234375,
      "learning_rate": 2.6341830260121445e-06,
      "loss": 0.1186,
      "reward": 1.5000000447034836,
      "reward_std": 0.5534840673208237,
      "rewards/accuracy_reward": 0.625000013038516,
      "rewards/format_reward": 0.8750000223517418,
      "step": 85
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 387.0833396911621,
      "epoch": 0.9173333333333333,
      "grad_norm": 0.12794648110866547,
      "kl": 0.090423583984375,
      "learning_rate": 2.6218079275152485e-06,
      "loss": 0.045,
      "reward": 1.2500000447034836,
      "reward_std": 0.4701542444527149,
      "rewards/accuracy_reward": 0.4583333469927311,
      "rewards/format_reward": 0.7916666828095913,
      "step": 86
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 391.37501525878906,
      "epoch": 0.928,
      "grad_norm": 0.09694866091012955,
      "kl": 0.0460968017578125,
      "learning_rate": 2.609257091237699e-06,
      "loss": 0.1368,
      "reward": 1.2916667014360428,
      "reward_std": 0.57484245672822,
      "rewards/accuracy_reward": 0.41666667722165585,
      "rewards/format_reward": 0.8750000149011612,
      "step": 87
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 362.1875104904175,
      "epoch": 0.9386666666666666,
      "grad_norm": 0.08955412358045578,
      "kl": 0.0487823486328125,
      "learning_rate": 2.5965324833414926e-06,
      "loss": 0.0046,
      "reward": 1.5416667014360428,
      "reward_std": 0.39372343569993973,
      "rewards/accuracy_reward": 0.625000013038516,
      "rewards/format_reward": 0.9166666865348816,
      "step": 88
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 412.12501525878906,
      "epoch": 0.9493333333333334,
      "grad_norm": 0.5077324509620667,
      "kl": 0.1267852783203125,
      "learning_rate": 2.5836360972109642e-06,
      "loss": 0.0258,
      "reward": 1.2500000223517418,
      "reward_std": 0.46246128901839256,
      "rewards/accuracy_reward": 0.4583333395421505,
      "rewards/format_reward": 0.7916666865348816,
      "step": 89
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 304.1666717529297,
      "epoch": 0.96,
      "grad_norm": 0.11206508427858353,
      "kl": 0.064788818359375,
      "learning_rate": 2.5705699531405174e-06,
      "loss": 0.0635,
      "reward": 1.708333358168602,
      "reward_std": 0.4028032459318638,
      "rewards/accuracy_reward": 0.7916666753590107,
      "rewards/format_reward": 0.9166666865348816,
      "step": 90
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 345.958345413208,
      "epoch": 0.9706666666666667,
      "grad_norm": 0.09778965264558792,
      "kl": 0.050811767578125,
      "learning_rate": 2.55733609801813e-06,
      "loss": -0.0167,
      "reward": 1.5416667014360428,
      "reward_std": 0.3776952736079693,
      "rewards/accuracy_reward": 0.6458333469927311,
      "rewards/format_reward": 0.8958333358168602,
      "step": 91
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 296.50000381469727,
      "epoch": 0.9813333333333333,
      "grad_norm": 0.12521831691265106,
      "kl": 0.0517120361328125,
      "learning_rate": 2.5439366050046985e-06,
      "loss": 0.0045,
      "reward": 1.5416666865348816,
      "reward_std": 0.42428741604089737,
      "rewards/accuracy_reward": 0.6041666753590107,
      "rewards/format_reward": 0.9375000149011612,
      "step": 92
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 335.70834159851074,
      "epoch": 0.992,
      "grad_norm": 0.10977984219789505,
      "kl": 0.0595550537109375,
      "learning_rate": 2.5303735732092655e-06,
      "loss": 0.0105,
      "reward": 1.4375000149011612,
      "reward_std": 0.47319112718105316,
      "rewards/accuracy_reward": 0.6458333488553762,
      "rewards/format_reward": 0.791666679084301,
      "step": 93
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 298.2500057220459,
      "epoch": 1.0106666666666666,
      "grad_norm": 0.11219625920057297,
      "kl": 0.0492706298828125,
      "learning_rate": 2.51664912736018e-06,
      "loss": -0.0016,
      "reward": 1.5625000447034836,
      "reward_std": 0.48367293924093246,
      "rewards/accuracy_reward": 0.5625000167638063,
      "rewards/format_reward": 1.0,
      "step": 94
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 291.3125114440918,
      "epoch": 1.0213333333333334,
      "grad_norm": 0.097702756524086,
      "kl": 0.0426025390625,
      "learning_rate": 2.5027654174722495e-06,
      "loss": 0.0599,
      "reward": 1.5000000298023224,
      "reward_std": 0.45132480561733246,
      "rewards/accuracy_reward": 0.6041666828095913,
      "rewards/format_reward": 0.8958333507180214,
      "step": 95
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 289.00000953674316,
      "epoch": 1.032,
      "grad_norm": 0.12104532122612,
      "kl": 0.0511474609375,
      "learning_rate": 2.4887246185099237e-06,
      "loss": 0.0013,
      "reward": 1.6875000298023224,
      "reward_std": 0.4161961302161217,
      "rewards/accuracy_reward": 0.7500000204890966,
      "rewards/format_reward": 0.9375000149011612,
      "step": 96
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 218.04167366027832,
      "epoch": 1.0426666666666666,
      "grad_norm": 0.148694708943367,
      "kl": 0.09942626953125,
      "learning_rate": 2.474528930046578e-06,
      "loss": -0.0069,
      "reward": 1.6041666865348816,
      "reward_std": 0.33713919296860695,
      "rewards/accuracy_reward": 0.6458333395421505,
      "rewards/format_reward": 0.9583333432674408,
      "step": 97
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 246.31250858306885,
      "epoch": 1.0533333333333332,
      "grad_norm": 0.08420921862125397,
      "kl": 0.0697021484375,
      "learning_rate": 2.4601805759199317e-06,
      "loss": -0.0038,
      "reward": 1.645833358168602,
      "reward_std": 0.26070838421583176,
      "rewards/accuracy_reward": 0.666666679084301,
      "rewards/format_reward": 0.9791666716337204,
      "step": 98
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 337.3333396911621,
      "epoch": 1.064,
      "grad_norm": 0.12480373680591583,
      "kl": 0.054962158203125,
      "learning_rate": 2.445681803883678e-06,
      "loss": 0.0435,
      "reward": 1.4791667014360428,
      "reward_std": 0.5356924869120121,
      "rewards/accuracy_reward": 0.5625000111758709,
      "rewards/format_reward": 0.9166666865348816,
      "step": 99
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 257.0625057220459,
      "epoch": 1.0746666666666667,
      "grad_norm": 0.1519816517829895,
      "kl": 0.0865478515625,
      "learning_rate": 2.4310348852553554e-06,
      "loss": -0.0236,
      "reward": 1.520833358168602,
      "reward_std": 0.5126873664557934,
      "rewards/accuracy_reward": 0.6250000186264515,
      "rewards/format_reward": 0.895833358168602,
      "step": 100
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 279.02084732055664,
      "epoch": 1.0853333333333333,
      "grad_norm": 0.10319739580154419,
      "kl": 0.0516510009765625,
      "learning_rate": 2.4162421145605308e-06,
      "loss": 0.0704,
      "reward": 1.5416666865348816,
      "reward_std": 0.37979811057448387,
      "rewards/accuracy_reward": 0.6458333507180214,
      "rewards/format_reward": 0.8958333432674408,
      "step": 101
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 294.45834159851074,
      "epoch": 1.096,
      "grad_norm": 0.13318555057048798,
      "kl": 0.069122314453125,
      "learning_rate": 2.4013058091733546e-06,
      "loss": 0.0196,
      "reward": 1.3958333730697632,
      "reward_std": 0.4621887616813183,
      "rewards/accuracy_reward": 0.4583333469927311,
      "rewards/format_reward": 0.9375000149011612,
      "step": 102
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 257.93750953674316,
      "epoch": 1.1066666666666667,
      "grad_norm": 0.49007534980773926,
      "kl": 0.1214599609375,
      "learning_rate": 2.3862283089535265e-06,
      "loss": 0.0464,
      "reward": 1.6041666865348816,
      "reward_std": 0.4967849962413311,
      "rewards/accuracy_reward": 0.750000013038516,
      "rewards/format_reward": 0.8541666865348816,
      "step": 103
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 215.58334159851074,
      "epoch": 1.1173333333333333,
      "grad_norm": 0.15618140995502472,
      "kl": 0.0696868896484375,
      "learning_rate": 2.371011975879742e-06,
      "loss": -0.0101,
      "reward": 1.5000000149011612,
      "reward_std": 0.43767181411385536,
      "rewards/accuracy_reward": 0.6875000111758709,
      "rewards/format_reward": 0.8125000149011612,
      "step": 104
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 220.60417366027832,
      "epoch": 1.1280000000000001,
      "grad_norm": 0.34512364864349365,
      "kl": 0.144317626953125,
      "learning_rate": 2.3556591936796777e-06,
      "loss": 0.0326,
      "reward": 1.5208333730697632,
      "reward_std": 0.44323352724313736,
      "rewards/accuracy_reward": 0.687500013038516,
      "rewards/format_reward": 0.8333333507180214,
      "step": 105
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 238.3750057220459,
      "epoch": 1.1386666666666667,
      "grad_norm": 0.1124984472990036,
      "kl": 0.0436859130859375,
      "learning_rate": 2.340172367456564e-06,
      "loss": -0.0044,
      "reward": 1.6041667014360428,
      "reward_std": 0.5307883769273758,
      "rewards/accuracy_reward": 0.729166679084301,
      "rewards/format_reward": 0.8750000149011612,
      "step": 106
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 235.91667366027832,
      "epoch": 1.1493333333333333,
      "grad_norm": 0.09784164279699326,
      "kl": 0.053192138671875,
      "learning_rate": 2.3245539233124133e-06,
      "loss": -0.0207,
      "reward": 1.5000000298023224,
      "reward_std": 0.3506578914821148,
      "rewards/accuracy_reward": 0.5625000093132257,
      "rewards/format_reward": 0.9375000074505806,
      "step": 107
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 222.02083778381348,
      "epoch": 1.16,
      "grad_norm": 0.11026975512504578,
      "kl": 0.0535888671875,
      "learning_rate": 2.308806307967955e-06,
      "loss": 0.0335,
      "reward": 1.645833358168602,
      "reward_std": 0.36124950647354126,
      "rewards/accuracy_reward": 0.7500000223517418,
      "rewards/format_reward": 0.8958333358168602,
      "step": 108
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 280.7291793823242,
      "epoch": 1.1706666666666667,
      "grad_norm": 0.13585244119167328,
      "kl": 0.0595703125,
      "learning_rate": 2.2929319883793442e-06,
      "loss": 0.026,
      "reward": 1.479166716337204,
      "reward_std": 0.5093041993677616,
      "rewards/accuracy_reward": 0.666666679084301,
      "rewards/format_reward": 0.8125000149011612,
      "step": 109
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 319.4583396911621,
      "epoch": 1.1813333333333333,
      "grad_norm": 0.11203665286302567,
      "kl": 0.056427001953125,
      "learning_rate": 2.276933451351696e-06,
      "loss": 0.0592,
      "reward": 1.2708333805203438,
      "reward_std": 0.5797660648822784,
      "rewards/accuracy_reward": 0.500000013038516,
      "rewards/format_reward": 0.7708333432674408,
      "step": 110
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 298.00000381469727,
      "epoch": 1.192,
      "grad_norm": 0.14347125589847565,
      "kl": 0.056915283203125,
      "learning_rate": 2.2608132031495184e-06,
      "loss": 0.0683,
      "reward": 1.4583333730697632,
      "reward_std": 0.6156517639756203,
      "rewards/accuracy_reward": 0.6875000149011612,
      "rewards/format_reward": 0.7708333507180214,
      "step": 111
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 287.2916774749756,
      "epoch": 1.2026666666666666,
      "grad_norm": 0.11557626724243164,
      "kl": 0.0446014404296875,
      "learning_rate": 2.244573769104084e-06,
      "loss": 0.0335,
      "reward": 1.479166716337204,
      "reward_std": 0.5093042030930519,
      "rewards/accuracy_reward": 0.6666666716337204,
      "rewards/format_reward": 0.8125000186264515,
      "step": 112
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 312.93750762939453,
      "epoch": 1.2133333333333334,
      "grad_norm": 0.10706419497728348,
      "kl": 0.045074462890625,
      "learning_rate": 2.228217693217826e-06,
      "loss": 0.0568,
      "reward": 1.354166716337204,
      "reward_std": 0.6839981600642204,
      "rewards/accuracy_reward": 0.6041666828095913,
      "rewards/format_reward": 0.7500000298023224,
      "step": 113
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 352.7083435058594,
      "epoch": 1.224,
      "grad_norm": 0.10810855031013489,
      "kl": 0.050567626953125,
      "learning_rate": 2.211747537765808e-06,
      "loss": 0.0611,
      "reward": 1.2708333805203438,
      "reward_std": 0.7179597914218903,
      "rewards/accuracy_reward": 0.5625000149011612,
      "rewards/format_reward": 0.7083333507180214,
      "step": 114
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 265.6250057220459,
      "epoch": 1.2346666666666666,
      "grad_norm": 0.13173377513885498,
      "kl": 0.217529296875,
      "learning_rate": 2.195165882894322e-06,
      "loss": -0.0123,
      "reward": 1.4583333879709244,
      "reward_std": 0.6409131288528442,
      "rewards/accuracy_reward": 0.6875000111758709,
      "rewards/format_reward": 0.770833358168602,
      "step": 115
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 386.06250762939453,
      "epoch": 1.2453333333333334,
      "grad_norm": 0.09368361532688141,
      "kl": 0.0453033447265625,
      "learning_rate": 2.1784753262166986e-06,
      "loss": 0.0297,
      "reward": 1.2083333805203438,
      "reward_std": 0.46787626668810844,
      "rewards/accuracy_reward": 0.35416667349636555,
      "rewards/format_reward": 0.854166679084301,
      "step": 116
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 281.27084159851074,
      "epoch": 1.256,
      "grad_norm": 0.10779143124818802,
      "kl": 0.054718017578125,
      "learning_rate": 2.161678482406374e-06,
      "loss": 0.0179,
      "reward": 1.5000000447034836,
      "reward_std": 0.4393179304897785,
      "rewards/accuracy_reward": 0.6666666753590107,
      "rewards/format_reward": 0.833333358168602,
      "step": 117
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 350.04167079925537,
      "epoch": 1.2666666666666666,
      "grad_norm": 0.08757103979587555,
      "kl": 0.053070068359375,
      "learning_rate": 2.144777982787286e-06,
      "loss": 0.0239,
      "reward": 1.4166666865348816,
      "reward_std": 0.5397448539733887,
      "rewards/accuracy_reward": 0.604166679084301,
      "rewards/format_reward": 0.8125000223517418,
      "step": 118
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 248.54167556762695,
      "epoch": 1.2773333333333334,
      "grad_norm": 0.0667010247707367,
      "kl": 0.058319091796875,
      "learning_rate": 2.127776474921661e-06,
      "loss": 0.004,
      "reward": 1.8125000298023224,
      "reward_std": 0.26070839166641235,
      "rewards/accuracy_reward": 0.8541666753590107,
      "rewards/format_reward": 0.9583333432674408,
      "step": 119
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 340.87500953674316,
      "epoch": 1.288,
      "grad_norm": 0.1155412346124649,
      "kl": 0.0574951171875,
      "learning_rate": 2.1106766221952607e-06,
      "loss": 0.0287,
      "reward": 1.500000037252903,
      "reward_std": 0.5172696895897388,
      "rewards/accuracy_reward": 0.6041666828095913,
      "rewards/format_reward": 0.8958333507180214,
      "step": 120
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 300.62500953674316,
      "epoch": 1.2986666666666666,
      "grad_norm": 0.10186120867729187,
      "kl": 0.053436279296875,
      "learning_rate": 2.0934811034001427e-06,
      "loss": 0.055,
      "reward": 1.5000000149011612,
      "reward_std": 0.400033675134182,
      "rewards/accuracy_reward": 0.6250000111758709,
      "rewards/format_reward": 0.8750000074505806,
      "step": 121
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 292.72917556762695,
      "epoch": 1.3093333333333335,
      "grad_norm": 0.1855934113264084,
      "kl": 0.0568084716796875,
      "learning_rate": 2.0761926123150166e-06,
      "loss": 0.0365,
      "reward": 1.5625000298023224,
      "reward_std": 0.2446802221238613,
      "rewards/accuracy_reward": 0.5833333395421505,
      "rewards/format_reward": 0.9791666716337204,
      "step": 122
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 348.6666793823242,
      "epoch": 1.32,
      "grad_norm": 0.10667165368795395,
      "kl": 0.076904296875,
      "learning_rate": 2.058813857283244e-06,
      "loss": 0.0521,
      "reward": 1.5208333656191826,
      "reward_std": 0.4552404023706913,
      "rewards/accuracy_reward": 0.6458333544433117,
      "rewards/format_reward": 0.8750000074505806,
      "step": 123
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 328.22918224334717,
      "epoch": 1.3306666666666667,
      "grad_norm": 27.496614456176758,
      "kl": 1.2711334228515625,
      "learning_rate": 2.041347560788562e-06,
      "loss": 0.1228,
      "reward": 1.6041667014360428,
      "reward_std": 0.4337562173604965,
      "rewards/accuracy_reward": 0.6666666809469461,
      "rewards/format_reward": 0.9375000149011612,
      "step": 124
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 326.8333435058594,
      "epoch": 1.3413333333333333,
      "grad_norm": 0.13310205936431885,
      "kl": 0.0502166748046875,
      "learning_rate": 2.0237964590285924e-06,
      "loss": 0.0235,
      "reward": 1.5625000447034836,
      "reward_std": 0.5452955961227417,
      "rewards/accuracy_reward": 0.6458333488553762,
      "rewards/format_reward": 0.9166666865348816,
      "step": 125
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 217.8958396911621,
      "epoch": 1.3519999999999999,
      "grad_norm": 0.1421211063861847,
      "kl": 0.07293701171875,
      "learning_rate": 2.0061633014861975e-06,
      "loss": 0.0336,
      "reward": 1.5208333879709244,
      "reward_std": 0.4217669740319252,
      "rewards/accuracy_reward": 0.5625000186264515,
      "rewards/format_reward": 0.9583333432674408,
      "step": 126
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 287.79168128967285,
      "epoch": 1.3626666666666667,
      "grad_norm": 0.11143215000629425,
      "kl": 0.078277587890625,
      "learning_rate": 1.988450850498758e-06,
      "loss": 0.0335,
      "reward": 1.5416667014360428,
      "reward_std": 0.3881702274084091,
      "rewards/accuracy_reward": 0.6458333432674408,
      "rewards/format_reward": 0.8958333507180214,
      "step": 127
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 357.0833511352539,
      "epoch": 1.3733333333333333,
      "grad_norm": 0.08851885795593262,
      "kl": 0.05126953125,
      "learning_rate": 1.970661880825437e-06,
      "loss": 0.0769,
      "reward": 1.3125000223517418,
      "reward_std": 0.4403064474463463,
      "rewards/accuracy_reward": 0.4375000074505806,
      "rewards/format_reward": 0.8750000223517418,
      "step": 128
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 245.2708396911621,
      "epoch": 1.384,
      "grad_norm": 0.11802337318658829,
      "kl": 0.0472564697265625,
      "learning_rate": 1.952799179212498e-06,
      "loss": 0.0265,
      "reward": 1.583333358168602,
      "reward_std": 0.40823065117001534,
      "rewards/accuracy_reward": 0.625000013038516,
      "rewards/format_reward": 0.9583333432674408,
      "step": 129
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 185.97917079925537,
      "epoch": 1.3946666666666667,
      "grad_norm": 0.15199995040893555,
      "kl": 0.080413818359375,
      "learning_rate": 1.934865543956745e-06,
      "loss": 0.0216,
      "reward": 1.7083333730697632,
      "reward_std": 0.4326419048011303,
      "rewards/accuracy_reward": 0.7083333507180214,
      "rewards/format_reward": 1.0,
      "step": 130
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 291.2916793823242,
      "epoch": 1.4053333333333333,
      "grad_norm": 0.11653240770101547,
      "kl": 0.0476531982421875,
      "learning_rate": 1.916863784467152e-06,
      "loss": 0.0277,
      "reward": 1.5416667088866234,
      "reward_std": 0.4870755970478058,
      "rewards/accuracy_reward": 0.6250000149011612,
      "rewards/format_reward": 0.9166666716337204,
      "step": 131
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 310.37501335144043,
      "epoch": 1.416,
      "grad_norm": 0.11371561139822006,
      "kl": 0.049407958984375,
      "learning_rate": 1.8987967208247542e-06,
      "loss": 0.0587,
      "reward": 1.5833333730697632,
      "reward_std": 0.39372343569993973,
      "rewards/accuracy_reward": 0.6250000111758709,
      "rewards/format_reward": 0.9583333432674408,
      "step": 132
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 217.06250762939453,
      "epoch": 1.4266666666666667,
      "grad_norm": 0.05566725507378578,
      "kl": 0.06402587890625,
      "learning_rate": 1.8806671833408633e-06,
      "loss": 0.0115,
      "reward": 1.75,
      "reward_std": 0.17304783314466476,
      "rewards/accuracy_reward": 0.8333333358168602,
      "rewards/format_reward": 0.9166666716337204,
      "step": 133
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 233.18750762939453,
      "epoch": 1.4373333333333334,
      "grad_norm": 0.09656023234128952,
      "kl": 0.0499114990234375,
      "learning_rate": 1.8624780121136837e-06,
      "loss": 0.0088,
      "reward": 1.833333358168602,
      "reward_std": 0.2861081622540951,
      "rewards/accuracy_reward": 0.854166679084301,
      "rewards/format_reward": 0.9791666716337204,
      "step": 134
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 265.62500953674316,
      "epoch": 1.448,
      "grad_norm": 0.12826627492904663,
      "kl": 0.062347412109375,
      "learning_rate": 1.8442320565833972e-06,
      "loss": 0.0249,
      "reward": 1.5000000447034836,
      "reward_std": 0.5420222245156765,
      "rewards/accuracy_reward": 0.5416666809469461,
      "rewards/format_reward": 0.9583333358168602,
      "step": 135
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 239.37500953674316,
      "epoch": 1.4586666666666668,
      "grad_norm": 0.10006551444530487,
      "kl": 0.0582275390625,
      "learning_rate": 1.8259321750857772e-06,
      "loss": -0.0374,
      "reward": 1.7291667014360428,
      "reward_std": 0.4090071730315685,
      "rewards/accuracy_reward": 0.7708333507180214,
      "rewards/format_reward": 0.9583333358168602,
      "step": 136
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 259.45834159851074,
      "epoch": 1.4693333333333334,
      "grad_norm": 0.1015496775507927,
      "kl": 0.0584716796875,
      "learning_rate": 1.8075812344044163e-06,
      "loss": 0.0308,
      "reward": 1.7083333730697632,
      "reward_std": 0.39372343569993973,
      "rewards/accuracy_reward": 0.7708333469927311,
      "rewards/format_reward": 0.9375000149011612,
      "step": 137
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 241.56250381469727,
      "epoch": 1.48,
      "grad_norm": 0.0967569649219513,
      "kl": 0.085906982421875,
      "learning_rate": 1.7891821093216275e-06,
      "loss": 0.0161,
      "reward": 1.6875000447034836,
      "reward_std": 0.41129202023148537,
      "rewards/accuracy_reward": 0.7916666716337204,
      "rewards/format_reward": 0.8958333507180214,
      "step": 138
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 341.37500953674316,
      "epoch": 1.4906666666666666,
      "grad_norm": 0.9276979565620422,
      "kl": 0.562530517578125,
      "learning_rate": 1.770737682168091e-06,
      "loss": 0.1031,
      "reward": 1.5833333730697632,
      "reward_std": 0.4673529714345932,
      "rewards/accuracy_reward": 0.6458333544433117,
      "rewards/format_reward": 0.9375000149011612,
      "step": 139
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 204.06250381469727,
      "epoch": 1.5013333333333332,
      "grad_norm": 0.09736520051956177,
      "kl": 0.105438232421875,
      "learning_rate": 1.7522508423713243e-06,
      "loss": 0.0251,
      "reward": 1.8541667014360428,
      "reward_std": 0.2446802221238613,
      "rewards/accuracy_reward": 0.854166679084301,
      "rewards/format_reward": 1.0,
      "step": 140
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 347.8541679382324,
      "epoch": 1.512,
      "grad_norm": 0.12234628945589066,
      "kl": 0.0500640869140625,
      "learning_rate": 1.7337244860030312e-06,
      "loss": 0.0863,
      "reward": 1.3333333656191826,
      "reward_std": 0.47278037667274475,
      "rewards/accuracy_reward": 0.47916667722165585,
      "rewards/format_reward": 0.854166679084301,
      "step": 141
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 237.06250858306885,
      "epoch": 1.5226666666666666,
      "grad_norm": 0.15673072636127472,
      "kl": 0.0669403076171875,
      "learning_rate": 1.7151615153254192e-06,
      "loss": 0.0404,
      "reward": 1.4791667014360428,
      "reward_std": 0.5729752071201801,
      "rewards/accuracy_reward": 0.5833333469927311,
      "rewards/format_reward": 0.8958333395421505,
      "step": 142
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 256.10417652130127,
      "epoch": 1.5333333333333332,
      "grad_norm": 0.16718435287475586,
      "kl": 0.0815582275390625,
      "learning_rate": 1.6965648383365416e-06,
      "loss": 0.1272,
      "reward": 1.4583333730697632,
      "reward_std": 0.5867235325276852,
      "rewards/accuracy_reward": 0.5625000186264515,
      "rewards/format_reward": 0.895833358168602,
      "step": 143
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 282.1458377838135,
      "epoch": 1.544,
      "grad_norm": 0.11697231978178024,
      "kl": 0.060821533203125,
      "learning_rate": 1.6779373683147423e-06,
      "loss": 0.1063,
      "reward": 1.3541667014360428,
      "reward_std": 0.5518574230372906,
      "rewards/accuracy_reward": 0.47916668094694614,
      "rewards/format_reward": 0.8750000223517418,
      "step": 144
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 275.29167556762695,
      "epoch": 1.5546666666666666,
      "grad_norm": 0.13972164690494537,
      "kl": 0.0799560546875,
      "learning_rate": 1.6592820233622743e-06,
      "loss": -0.0019,
      "reward": 1.2500000298023224,
      "reward_std": 0.5609171241521835,
      "rewards/accuracy_reward": 0.43750001303851604,
      "rewards/format_reward": 0.8125000298023224,
      "step": 145
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 235.35417461395264,
      "epoch": 1.5653333333333332,
      "grad_norm": 0.1573922336101532,
      "kl": 0.084381103515625,
      "learning_rate": 1.6406017259481624e-06,
      "loss": 0.1167,
      "reward": 1.3125000298023224,
      "reward_std": 0.46066780760884285,
      "rewards/accuracy_reward": 0.41666666977107525,
      "rewards/format_reward": 0.8958333358168602,
      "step": 146
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 315.25001335144043,
      "epoch": 1.576,
      "grad_norm": 0.15172944962978363,
      "kl": 0.080963134765625,
      "learning_rate": 1.621899402450379e-06,
      "loss": 0.1219,
      "reward": 1.2708333656191826,
      "reward_std": 0.6084391921758652,
      "rewards/accuracy_reward": 0.4583333358168602,
      "rewards/format_reward": 0.8125000298023224,
      "step": 147
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 263.4375066757202,
      "epoch": 1.5866666666666667,
      "grad_norm": 0.12377157062292099,
      "kl": 0.07403564453125,
      "learning_rate": 1.6031779826974138e-06,
      "loss": 0.0255,
      "reward": 1.395833358168602,
      "reward_std": 0.4362565539777279,
      "rewards/accuracy_reward": 0.5416666716337204,
      "rewards/format_reward": 0.8541666865348816,
      "step": 148
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 272.3750114440918,
      "epoch": 1.5973333333333333,
      "grad_norm": 0.46126094460487366,
      "kl": 0.209686279296875,
      "learning_rate": 1.5844403995092944e-06,
      "loss": 0.0734,
      "reward": 1.145833358168602,
      "reward_std": 0.5279995314776897,
      "rewards/accuracy_reward": 0.31250000931322575,
      "rewards/format_reward": 0.8333333507180214,
      "step": 149
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 219.7291717529297,
      "epoch": 1.608,
      "grad_norm": 0.17344419658184052,
      "kl": 0.077972412109375,
      "learning_rate": 1.5656895882381465e-06,
      "loss": 0.0446,
      "reward": 1.104166716337204,
      "reward_std": 0.5293822586536407,
      "rewards/accuracy_reward": 0.2500000037252903,
      "rewards/format_reward": 0.8541666939854622,
      "step": 150
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 295.9166717529297,
      "epoch": 1.6186666666666667,
      "grad_norm": 0.13225041329860687,
      "kl": 0.078277587890625,
      "learning_rate": 1.5469284863083523e-06,
      "loss": 0.1267,
      "reward": 1.2083333805203438,
      "reward_std": 0.44083888083696365,
      "rewards/accuracy_reward": 0.33333334140479565,
      "rewards/format_reward": 0.8750000223517418,
      "step": 151
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 253.854172706604,
      "epoch": 1.6293333333333333,
      "grad_norm": 0.14670467376708984,
      "kl": 0.06512451171875,
      "learning_rate": 1.5281600327563853e-06,
      "loss": 0.0528,
      "reward": 1.2916667014360428,
      "reward_std": 0.46722717583179474,
      "rewards/accuracy_reward": 0.354166679084301,
      "rewards/format_reward": 0.9375000149011612,
      "step": 152
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 305.4166774749756,
      "epoch": 1.6400000000000001,
      "grad_norm": 0.14602994918823242,
      "kl": 0.1068115234375,
      "learning_rate": 1.5093871677703947e-06,
      "loss": 0.0707,
      "reward": 1.1875000149011612,
      "reward_std": 0.6286378428339958,
      "rewards/accuracy_reward": 0.41666667349636555,
      "rewards/format_reward": 0.7708333507180214,
      "step": 153
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 284.2083396911621,
      "epoch": 1.6506666666666665,
      "grad_norm": 0.15345576405525208,
      "kl": 0.06597900390625,
      "learning_rate": 1.4906128322296056e-06,
      "loss": 0.0552,
      "reward": 1.2291666865348816,
      "reward_std": 0.5342973358929157,
      "rewards/accuracy_reward": 0.31250000558793545,
      "rewards/format_reward": 0.9166666865348816,
      "step": 154
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 309.1666736602783,
      "epoch": 1.6613333333333333,
      "grad_norm": 0.17641578614711761,
      "kl": 0.1270904541015625,
      "learning_rate": 1.471839967243615e-06,
      "loss": 0.0082,
      "reward": 1.1041667088866234,
      "reward_std": 0.5523697510361671,
      "rewards/accuracy_reward": 0.3125000111758709,
      "rewards/format_reward": 0.791666679084301,
      "step": 155
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 322.75000762939453,
      "epoch": 1.6720000000000002,
      "grad_norm": 0.15972626209259033,
      "kl": 0.0926513671875,
      "learning_rate": 1.453071513691648e-06,
      "loss": 0.1647,
      "reward": 1.2291667014360428,
      "reward_std": 0.5694059133529663,
      "rewards/accuracy_reward": 0.3958333469927311,
      "rewards/format_reward": 0.833333358168602,
      "step": 156
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 258.6875114440918,
      "epoch": 1.6826666666666665,
      "grad_norm": 0.1186479851603508,
      "kl": 0.081817626953125,
      "learning_rate": 1.4343104117618536e-06,
      "loss": 0.0049,
      "reward": 1.2500000223517418,
      "reward_std": 0.4595342054963112,
      "rewards/accuracy_reward": 0.3750000074505806,
      "rewards/format_reward": 0.8750000149011612,
      "step": 157
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 193.104172706604,
      "epoch": 1.6933333333333334,
      "grad_norm": 0.08673811703920364,
      "kl": 0.0623321533203125,
      "learning_rate": 1.415559600490706e-06,
      "loss": 0.0657,
      "reward": 1.6041666716337204,
      "reward_std": 0.18796167895197868,
      "rewards/accuracy_reward": 0.6458333358168602,
      "rewards/format_reward": 0.9583333432674408,
      "step": 158
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 218.0416717529297,
      "epoch": 1.704,
      "grad_norm": 0.1484222710132599,
      "kl": 0.088836669921875,
      "learning_rate": 1.3968220173025865e-06,
      "loss": 0.0912,
      "reward": 1.5208333879709244,
      "reward_std": 0.572458777576685,
      "rewards/accuracy_reward": 0.5833333544433117,
      "rewards/format_reward": 0.9375000149011612,
      "step": 159
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 303.95834159851074,
      "epoch": 1.7146666666666666,
      "grad_norm": 0.2447938323020935,
      "kl": 0.1371002197265625,
      "learning_rate": 1.3781005975496212e-06,
      "loss": 0.0618,
      "reward": 1.2083333730697632,
      "reward_std": 0.34925176575779915,
      "rewards/accuracy_reward": 0.31250001303851604,
      "rewards/format_reward": 0.8958333507180214,
      "step": 160
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 263.3125057220459,
      "epoch": 1.7253333333333334,
      "grad_norm": 0.14777325093746185,
      "kl": 0.159271240234375,
      "learning_rate": 1.359398274051838e-06,
      "loss": 0.0329,
      "reward": 1.3125000149011612,
      "reward_std": 0.5235714204609394,
      "rewards/accuracy_reward": 0.4375000037252903,
      "rewards/format_reward": 0.8750000149011612,
      "step": 161
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 279.91667556762695,
      "epoch": 1.736,
      "grad_norm": 0.11606959253549576,
      "kl": 0.12255859375,
      "learning_rate": 1.340717976637726e-06,
      "loss": 0.0324,
      "reward": 1.4375000298023224,
      "reward_std": 0.45774073526263237,
      "rewards/accuracy_reward": 0.5416666753590107,
      "rewards/format_reward": 0.8958333507180214,
      "step": 162
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 252.3125057220459,
      "epoch": 1.7466666666666666,
      "grad_norm": 0.1397114396095276,
      "kl": 0.0972442626953125,
      "learning_rate": 1.3220626316852584e-06,
      "loss": 0.0687,
      "reward": 1.4791666865348816,
      "reward_std": 0.4081249535083771,
      "rewards/accuracy_reward": 0.6666666772216558,
      "rewards/format_reward": 0.8125000223517418,
      "step": 163
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 315.2708435058594,
      "epoch": 1.7573333333333334,
      "grad_norm": 0.1347351223230362,
      "kl": 0.113433837890625,
      "learning_rate": 1.303435161663459e-06,
      "loss": 0.1106,
      "reward": 1.1666667088866234,
      "reward_std": 0.45271995291113853,
      "rewards/accuracy_reward": 0.29166666977107525,
      "rewards/format_reward": 0.8750000223517418,
      "step": 164
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 216.47917556762695,
      "epoch": 1.768,
      "grad_norm": 0.15313582122325897,
      "kl": 0.0865936279296875,
      "learning_rate": 1.2848384846745813e-06,
      "loss": 0.0198,
      "reward": 1.5416667014360428,
      "reward_std": 0.41228053718805313,
      "rewards/accuracy_reward": 0.604166679084301,
      "rewards/format_reward": 0.9375000074505806,
      "step": 165
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 250.6875114440918,
      "epoch": 1.7786666666666666,
      "grad_norm": 0.1456070989370346,
      "kl": 0.09686279296875,
      "learning_rate": 1.2662755139969693e-06,
      "loss": -0.0239,
      "reward": 1.333333358168602,
      "reward_std": 0.5820958688855171,
      "rewards/accuracy_reward": 0.4166666753590107,
      "rewards/format_reward": 0.9166666716337204,
      "step": 166
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 318.395845413208,
      "epoch": 1.7893333333333334,
      "grad_norm": 0.1404711902141571,
      "kl": 0.091278076171875,
      "learning_rate": 1.2477491576286764e-06,
      "loss": 0.0999,
      "reward": 1.3125000149011612,
      "reward_std": 0.4405704736709595,
      "rewards/accuracy_reward": 0.41666667722165585,
      "rewards/format_reward": 0.8958333507180214,
      "step": 167
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 231.89583778381348,
      "epoch": 1.8,
      "grad_norm": 0.14985869824886322,
      "kl": 0.104583740234375,
      "learning_rate": 1.2292623178319094e-06,
      "loss": 0.0291,
      "reward": 1.333333358168602,
      "reward_std": 0.5864746198058128,
      "rewards/accuracy_reward": 0.5000000204890966,
      "rewards/format_reward": 0.8333333432674408,
      "step": 168
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 268.00000762939453,
      "epoch": 1.8106666666666666,
      "grad_norm": 0.16692346334457397,
      "kl": 0.071533203125,
      "learning_rate": 1.2108178906783732e-06,
      "loss": 0.0409,
      "reward": 1.354166716337204,
      "reward_std": 0.5293822661042213,
      "rewards/accuracy_reward": 0.45833334140479565,
      "rewards/format_reward": 0.895833358168602,
      "step": 169
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 246.33333826065063,
      "epoch": 1.8213333333333335,
      "grad_norm": 0.17591020464897156,
      "kl": 0.09210205078125,
      "learning_rate": 1.1924187655955838e-06,
      "loss": 0.0802,
      "reward": 1.4583333805203438,
      "reward_std": 0.44083888456225395,
      "rewards/accuracy_reward": 0.5416666772216558,
      "rewards/format_reward": 0.916666679084301,
      "step": 170
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 283.4166774749756,
      "epoch": 1.8319999999999999,
      "grad_norm": 0.1486297994852066,
      "kl": 0.1509857177734375,
      "learning_rate": 1.1740678249142229e-06,
      "loss": 0.0796,
      "reward": 1.3958333730697632,
      "reward_std": 0.532146617770195,
      "rewards/accuracy_reward": 0.5000000149011612,
      "rewards/format_reward": 0.8958333507180214,
      "step": 171
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 198.7083396911621,
      "epoch": 1.8426666666666667,
      "grad_norm": 0.23767401278018951,
      "kl": 0.145599365234375,
      "learning_rate": 1.1557679434166027e-06,
      "loss": 0.0035,
      "reward": 1.4791666865348816,
      "reward_std": 0.47406983748078346,
      "rewards/accuracy_reward": 0.541666679084301,
      "rewards/format_reward": 0.9375000149011612,
      "step": 172
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 224.75000476837158,
      "epoch": 1.8533333333333335,
      "grad_norm": 0.13122382760047913,
      "kl": 0.11077880859375,
      "learning_rate": 1.1375219878863162e-06,
      "loss": 0.0779,
      "reward": 1.5416667014360428,
      "reward_std": 0.38677507638931274,
      "rewards/accuracy_reward": 0.6041666809469461,
      "rewards/format_reward": 0.9375000149011612,
      "step": 173
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 246.08333587646484,
      "epoch": 1.8639999999999999,
      "grad_norm": 0.13901762664318085,
      "kl": 0.099822998046875,
      "learning_rate": 1.1193328166591372e-06,
      "loss": 0.0029,
      "reward": 1.458333358168602,
      "reward_std": 0.45446737110614777,
      "rewards/accuracy_reward": 0.5625000055879354,
      "rewards/format_reward": 0.8958333432674408,
      "step": 174
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 279.9166736602783,
      "epoch": 1.8746666666666667,
      "grad_norm": 0.304503470659256,
      "kl": 0.228118896484375,
      "learning_rate": 1.1012032791752457e-06,
      "loss": 0.0385,
      "reward": 1.2291667088866234,
      "reward_std": 0.4801270626485348,
      "rewards/accuracy_reward": 0.3750000074505806,
      "rewards/format_reward": 0.8541666865348816,
      "step": 175
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 244.0208396911621,
      "epoch": 1.8853333333333333,
      "grad_norm": 0.2561553120613098,
      "kl": 0.191009521484375,
      "learning_rate": 1.0831362155328477e-06,
      "loss": 0.0531,
      "reward": 1.6875000149011612,
      "reward_std": 0.4051978886127472,
      "rewards/accuracy_reward": 0.750000013038516,
      "rewards/format_reward": 0.9375000149011612,
      "step": 176
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 344.43750953674316,
      "epoch": 1.896,
      "grad_norm": 0.24141816794872284,
      "kl": 0.21771240234375,
      "learning_rate": 1.0651344560432548e-06,
      "loss": 0.1034,
      "reward": 1.4375000223517418,
      "reward_std": 0.5041032508015633,
      "rewards/accuracy_reward": 0.6041666734963655,
      "rewards/format_reward": 0.8333333469927311,
      "step": 177
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 237.37500762939453,
      "epoch": 1.9066666666666667,
      "grad_norm": 1.193719744682312,
      "kl": 0.32879638671875,
      "learning_rate": 1.0472008207875023e-06,
      "loss": 0.1244,
      "reward": 1.479166716337204,
      "reward_std": 0.5614495500922203,
      "rewards/accuracy_reward": 0.6250000186264515,
      "rewards/format_reward": 0.854166679084301,
      "step": 178
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 261.7083435058594,
      "epoch": 1.9173333333333333,
      "grad_norm": 0.46435582637786865,
      "kl": 0.276611328125,
      "learning_rate": 1.0293381191745637e-06,
      "loss": 0.0264,
      "reward": 1.4375000447034836,
      "reward_std": 0.4688647873699665,
      "rewards/accuracy_reward": 0.5000000204890966,
      "rewards/format_reward": 0.9375000149011612,
      "step": 179
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 292.27084255218506,
      "epoch": 1.928,
      "grad_norm": 0.19408948719501495,
      "kl": 0.187713623046875,
      "learning_rate": 1.0115491495012424e-06,
      "loss": -0.002,
      "reward": 1.5000000149011612,
      "reward_std": 0.20815639942884445,
      "rewards/accuracy_reward": 0.5208333376795053,
      "rewards/format_reward": 0.9791666716337204,
      "step": 180
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 326.85417079925537,
      "epoch": 1.9386666666666668,
      "grad_norm": 0.34896135330200195,
      "kl": 0.253265380859375,
      "learning_rate": 9.938366985138028e-07,
      "loss": 0.1119,
      "reward": 1.3750000186264515,
      "reward_std": 0.4847872592508793,
      "rewards/accuracy_reward": 0.5625000093132257,
      "rewards/format_reward": 0.8125000111758709,
      "step": 181
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 246.3333396911621,
      "epoch": 1.9493333333333334,
      "grad_norm": 0.242198184132576,
      "kl": 0.17230224609375,
      "learning_rate": 9.762035409714077e-07,
      "loss": 0.0225,
      "reward": 1.2500000298023224,
      "reward_std": 0.45622891560196877,
      "rewards/accuracy_reward": 0.37500000931322575,
      "rewards/format_reward": 0.8750000149011612,
      "step": 182
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 252.4375057220459,
      "epoch": 1.96,
      "grad_norm": 1.2867988348007202,
      "kl": 0.237274169921875,
      "learning_rate": 9.586524392114385e-07,
      "loss": 0.1312,
      "reward": 1.4166667014360428,
      "reward_std": 0.5692892409861088,
      "rewards/accuracy_reward": 0.5416666753590107,
      "rewards/format_reward": 0.8750000223517418,
      "step": 183
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 252.2500114440918,
      "epoch": 1.9706666666666668,
      "grad_norm": 0.4133116602897644,
      "kl": 0.166107177734375,
      "learning_rate": 9.411861427167567e-07,
      "loss": 0.1611,
      "reward": 1.5208333507180214,
      "reward_std": 0.508791871368885,
      "rewards/accuracy_reward": 0.6250000093132257,
      "rewards/format_reward": 0.8958333432674408,
      "step": 184
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 297.16666984558105,
      "epoch": 1.9813333333333332,
      "grad_norm": 2.6324448585510254,
      "kl": 0.236083984375,
      "learning_rate": 9.238073876849837e-07,
      "loss": 0.0799,
      "reward": 1.4166666865348816,
      "reward_std": 0.6997495777904987,
      "rewards/accuracy_reward": 0.5833333525806665,
      "rewards/format_reward": 0.8333333507180214,
      "step": 185
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 350.0833435058594,
      "epoch": 1.992,
      "grad_norm": 1.6513264179229736,
      "kl": 0.191741943359375,
      "learning_rate": 9.065188965998574e-07,
      "loss": 0.1732,
      "reward": 1.4375000447034836,
      "reward_std": 0.6652876362204552,
      "rewards/accuracy_reward": 0.6041666828095913,
      "rewards/format_reward": 0.8333333656191826,
      "step": 186
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 263.8333387374878,
      "epoch": 2.010666666666667,
      "grad_norm": 0.46523988246917725,
      "kl": 0.34124755859375,
      "learning_rate": 8.893233778047394e-07,
      "loss": 0.0944,
      "reward": 1.4166666865348816,
      "reward_std": 0.4923461005091667,
      "rewards/accuracy_reward": 0.5833333469927311,
      "rewards/format_reward": 0.8333333544433117,
      "step": 187
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 241.3750057220459,
      "epoch": 2.021333333333333,
      "grad_norm": 5.950314521789551,
      "kl": 0.441497802734375,
      "learning_rate": 8.722235250783389e-07,
      "loss": 0.0085,
      "reward": 1.5625000298023224,
      "reward_std": 0.5480775870382786,
      "rewards/accuracy_reward": 0.6875000167638063,
      "rewards/format_reward": 0.8750000223517418,
      "step": 188
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 239.95834255218506,
      "epoch": 2.032,
      "grad_norm": 2.313253879547119,
      "kl": 0.481903076171875,
      "learning_rate": 8.552220172127148e-07,
      "loss": 0.1195,
      "reward": 1.6250000223517418,
      "reward_std": 0.4617997705936432,
      "rewards/accuracy_reward": 0.750000013038516,
      "rewards/format_reward": 0.8750000149011612,
      "step": 189
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 333.2291774749756,
      "epoch": 2.042666666666667,
      "grad_norm": 393.3891296386719,
      "kl": 54.324493408203125,
      "learning_rate": 8.383215175936263e-07,
      "loss": 3.1915,
      "reward": 1.3750000223517418,
      "reward_std": 0.5343292653560638,
      "rewards/accuracy_reward": 0.5625000037252903,
      "rewards/format_reward": 0.8125000149011612,
      "step": 190
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 288.06250381469727,
      "epoch": 2.0533333333333332,
      "grad_norm": 17.025423049926758,
      "kl": 5.16571044921875,
      "learning_rate": 8.215246737833017e-07,
      "loss": 0.3571,
      "reward": 1.3333333730697632,
      "reward_std": 0.5607051253318787,
      "rewards/accuracy_reward": 0.4791666753590107,
      "rewards/format_reward": 0.8541666865348816,
      "step": 191
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 242.31250762939453,
      "epoch": 2.064,
      "grad_norm": 1.5437623262405396,
      "kl": 1.153076171875,
      "learning_rate": 8.048341171056786e-07,
      "loss": 0.0747,
      "reward": 1.5625000447034836,
      "reward_std": 0.49742312356829643,
      "rewards/accuracy_reward": 0.666666679084301,
      "rewards/format_reward": 0.8958333432674408,
      "step": 192
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 250.291672706604,
      "epoch": 2.074666666666667,
      "grad_norm": 2.553690195083618,
      "kl": 0.81292724609375,
      "learning_rate": 7.882524622341923e-07,
      "loss": 0.1897,
      "reward": 1.5208333656191826,
      "reward_std": 0.6309319958090782,
      "rewards/accuracy_reward": 0.6666666939854622,
      "rewards/format_reward": 0.8541666865348816,
      "step": 193
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 276.79167556762695,
      "epoch": 2.0853333333333333,
      "grad_norm": 0.9575795531272888,
      "kl": 0.219451904296875,
      "learning_rate": 7.717823067821738e-07,
      "loss": 0.0771,
      "reward": 1.4166667014360428,
      "reward_std": 0.3477308116853237,
      "rewards/accuracy_reward": 0.5000000111758709,
      "rewards/format_reward": 0.9166666865348816,
      "step": 194
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 236.22917366027832,
      "epoch": 2.096,
      "grad_norm": 1.3622400760650635,
      "kl": 0.867919921875,
      "learning_rate": 7.554262308959165e-07,
      "loss": 0.0914,
      "reward": 1.1458333805203438,
      "reward_std": 0.5349354632198811,
      "rewards/accuracy_reward": 0.29166667349636555,
      "rewards/format_reward": 0.854166679084301,
      "step": 195
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 277.3958396911621,
      "epoch": 2.1066666666666665,
      "grad_norm": 4.180922031402588,
      "kl": 0.69647216796875,
      "learning_rate": 7.39186796850482e-07,
      "loss": 0.1392,
      "reward": 1.416666679084301,
      "reward_std": 0.5008263885974884,
      "rewards/accuracy_reward": 0.5416666734963655,
      "rewards/format_reward": 0.8750000149011612,
      "step": 196
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 233.37500190734863,
      "epoch": 2.1173333333333333,
      "grad_norm": 0.28783828020095825,
      "kl": 0.64825439453125,
      "learning_rate": 7.23066548648304e-07,
      "loss": 0.0323,
      "reward": 1.458333358168602,
      "reward_std": 0.5146066322922707,
      "rewards/accuracy_reward": 0.5833333432674408,
      "rewards/format_reward": 0.8750000223517418,
      "step": 197
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 400.9791831970215,
      "epoch": 2.128,
      "grad_norm": 10.193907737731934,
      "kl": 5.00872802734375,
      "learning_rate": 7.070680116206564e-07,
      "loss": 0.3473,
      "reward": 1.0625000335276127,
      "reward_std": 0.5005942918360233,
      "rewards/accuracy_reward": 0.3125000037252903,
      "rewards/format_reward": 0.7500000111758709,
      "step": 198
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 292.7916736602783,
      "epoch": 2.1386666666666665,
      "grad_norm": 4.077642440795898,
      "kl": 2.90966796875,
      "learning_rate": 6.911936920320457e-07,
      "loss": 0.1194,
      "reward": 1.4166667014360428,
      "reward_std": 0.5656285360455513,
      "rewards/accuracy_reward": 0.5000000186264515,
      "rewards/format_reward": 0.916666679084301,
      "step": 199
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 225.8958396911621,
      "epoch": 2.1493333333333333,
      "grad_norm": 0.7050092816352844,
      "kl": 0.34698486328125,
      "learning_rate": 6.754460766875871e-07,
      "loss": 0.0508,
      "reward": 1.5208333730697632,
      "reward_std": 0.40168892219662666,
      "rewards/accuracy_reward": 0.583333345130086,
      "rewards/format_reward": 0.9375000149011612,
      "step": 200
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 173.95833778381348,
      "epoch": 2.16,
      "grad_norm": 0.23802195489406586,
      "kl": 0.095611572265625,
      "learning_rate": 6.598276325434357e-07,
      "loss": 0.0267,
      "reward": 1.7916666865348816,
      "reward_std": 0.3410547971725464,
      "rewards/accuracy_reward": 0.8333333432674408,
      "rewards/format_reward": 0.9583333432674408,
      "step": 201
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 266.3541736602783,
      "epoch": 2.1706666666666665,
      "grad_norm": 0.40206822752952576,
      "kl": 0.666473388671875,
      "learning_rate": 6.443408063203226e-07,
      "loss": 0.0029,
      "reward": 1.4375000447034836,
      "reward_std": 0.5253323614597321,
      "rewards/accuracy_reward": 0.5625000018626451,
      "rewards/format_reward": 0.8750000223517418,
      "step": 202
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 228.9583396911621,
      "epoch": 2.1813333333333333,
      "grad_norm": 1.0147886276245117,
      "kl": 0.338043212890625,
      "learning_rate": 6.289880241202581e-07,
      "loss": 0.043,
      "reward": 1.2708333730697632,
      "reward_std": 0.5363415889441967,
      "rewards/accuracy_reward": 0.3541666753590107,
      "rewards/format_reward": 0.916666679084301,
      "step": 203
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 221.7916774749756,
      "epoch": 2.192,
      "grad_norm": 0.427720844745636,
      "kl": 0.18035888671875,
      "learning_rate": 6.137716910464735e-07,
      "loss": 0.0187,
      "reward": 1.5625000447034836,
      "reward_std": 0.5079090520739555,
      "rewards/accuracy_reward": 0.6250000111758709,
      "rewards/format_reward": 0.9375000149011612,
      "step": 204
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 235.62500762939453,
      "epoch": 2.2026666666666666,
      "grad_norm": 0.5082700848579407,
      "kl": 1.269378662109375,
      "learning_rate": 5.986941908266454e-07,
      "loss": 0.1469,
      "reward": 1.4375000447034836,
      "reward_std": 0.5468165427446365,
      "rewards/accuracy_reward": 0.5833333488553762,
      "rewards/format_reward": 0.854166679084301,
      "step": 205
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 228.1875057220459,
      "epoch": 2.2133333333333334,
      "grad_norm": 7.290508270263672,
      "kl": 4.46014404296875,
      "learning_rate": 5.837578854394692e-07,
      "loss": 0.2142,
      "reward": 1.3125000298023224,
      "reward_std": 0.44334835186600685,
      "rewards/accuracy_reward": 0.45833333767950535,
      "rewards/format_reward": 0.8541666865348816,
      "step": 206
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 248.1875057220459,
      "epoch": 2.224,
      "grad_norm": 5.754692077636719,
      "kl": 6.70330810546875,
      "learning_rate": 5.689651147446448e-07,
      "loss": 0.4078,
      "reward": 1.2500000223517418,
      "reward_std": 0.8125958703458309,
      "rewards/accuracy_reward": 0.520833345130086,
      "rewards/format_reward": 0.7291666865348816,
      "step": 207
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 276.5000057220459,
      "epoch": 2.2346666666666666,
      "grad_norm": 0.4427819550037384,
      "kl": 0.77191162109375,
      "learning_rate": 5.543181961163219e-07,
      "loss": 0.1116,
      "reward": 1.3958333730697632,
      "reward_std": 0.4217669740319252,
      "rewards/accuracy_reward": 0.479166679084301,
      "rewards/format_reward": 0.9166666865348816,
      "step": 208
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 191.64583778381348,
      "epoch": 2.2453333333333334,
      "grad_norm": 0.782500684261322,
      "kl": 1.3341064453125,
      "learning_rate": 5.398194240800684e-07,
      "loss": 0.0145,
      "reward": 1.5833333730697632,
      "reward_std": 0.5104185156524181,
      "rewards/accuracy_reward": 0.6875000149011612,
      "rewards/format_reward": 0.895833358168602,
      "step": 209
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 267.18750762939453,
      "epoch": 2.2560000000000002,
      "grad_norm": 1.0296471118927002,
      "kl": 1.434326171875,
      "learning_rate": 5.254710699534226e-07,
      "loss": 0.1032,
      "reward": 1.3750000298023224,
      "reward_std": 0.4010416902601719,
      "rewards/accuracy_reward": 0.5416666697710752,
      "rewards/format_reward": 0.8333333432674408,
      "step": 210
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 308.89584159851074,
      "epoch": 2.2666666666666666,
      "grad_norm": 2.316876173019409,
      "kl": 1.29791259765625,
      "learning_rate": 5.112753814900761e-07,
      "loss": 0.1656,
      "reward": 1.312500037252903,
      "reward_std": 0.5606077648699284,
      "rewards/accuracy_reward": 0.4583333395421505,
      "rewards/format_reward": 0.8541666865348816,
      "step": 211
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 198.3125057220459,
      "epoch": 2.2773333333333334,
      "grad_norm": 0.34542033076286316,
      "kl": 1.111846923828125,
      "learning_rate": 4.97234582527751e-07,
      "loss": 0.018,
      "reward": 1.3333333805203438,
      "reward_std": 0.6683094389736652,
      "rewards/accuracy_reward": 0.5416666846722364,
      "rewards/format_reward": 0.7916666828095913,
      "step": 212
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 316.75001335144043,
      "epoch": 2.288,
      "grad_norm": 1.203397274017334,
      "kl": 1.38311767578125,
      "learning_rate": 4.8335087263982e-07,
      "loss": 0.2258,
      "reward": 1.3333333730697632,
      "reward_std": 0.607686098664999,
      "rewards/accuracy_reward": 0.5000000055879354,
      "rewards/format_reward": 0.833333358168602,
      "step": 213
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 296.8125114440918,
      "epoch": 2.2986666666666666,
      "grad_norm": 0.3318883180618286,
      "kl": 0.48272705078125,
      "learning_rate": 4.6962642679073465e-07,
      "loss": 0.034,
      "reward": 1.2083333730697632,
      "reward_std": 0.5359239801764488,
      "rewards/accuracy_reward": 0.31250000931322575,
      "rewards/format_reward": 0.8958333507180214,
      "step": 214
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 286.9166736602783,
      "epoch": 2.3093333333333335,
      "grad_norm": 0.5079717636108398,
      "kl": 0.31884765625,
      "learning_rate": 4.560633949953017e-07,
      "loss": 0.051,
      "reward": 1.333333358168602,
      "reward_std": 0.5560254342854023,
      "rewards/accuracy_reward": 0.5000000149011612,
      "rewards/format_reward": 0.8333333507180214,
      "step": 215
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 211.1250057220459,
      "epoch": 2.32,
      "grad_norm": 0.673991858959198,
      "kl": 0.325927734375,
      "learning_rate": 4.426639019818704e-07,
      "loss": -0.0239,
      "reward": 1.4375000298023224,
      "reward_std": 0.5759567655622959,
      "rewards/accuracy_reward": 0.5625000149011612,
      "rewards/format_reward": 0.8750000223517418,
      "step": 216
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 249.66667556762695,
      "epoch": 2.3306666666666667,
      "grad_norm": 0.620853841304779,
      "kl": 0.3583984375,
      "learning_rate": 4.2943004685948286e-07,
      "loss": 0.0766,
      "reward": 1.2708333432674408,
      "reward_std": 0.5561920776963234,
      "rewards/accuracy_reward": 0.4166666753590107,
      "rewards/format_reward": 0.854166679084301,
      "step": 217
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 291.3333396911621,
      "epoch": 2.3413333333333335,
      "grad_norm": 10.582584381103516,
      "kl": 2.179534912109375,
      "learning_rate": 4.163639027890363e-07,
      "loss": 0.1387,
      "reward": 1.3125000409781933,
      "reward_std": 0.5307883769273758,
      "rewards/accuracy_reward": 0.4375000111758709,
      "rewards/format_reward": 0.8750000111758709,
      "step": 218
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 249.1250057220459,
      "epoch": 2.352,
      "grad_norm": 1.4043989181518555,
      "kl": 0.674163818359375,
      "learning_rate": 4.034675166585079e-07,
      "loss": 0.1281,
      "reward": 1.4166667312383652,
      "reward_std": 0.5429009459912777,
      "rewards/accuracy_reward": 0.5000000186264515,
      "rewards/format_reward": 0.9166666865348816,
      "step": 219
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 297.16667461395264,
      "epoch": 2.3626666666666667,
      "grad_norm": 3.6459124088287354,
      "kl": 2.717041015625,
      "learning_rate": 3.907429087623007e-07,
      "loss": 0.1155,
      "reward": 1.2291666865348816,
      "reward_std": 0.41912320628762245,
      "rewards/accuracy_reward": 0.3958333432674408,
      "rewards/format_reward": 0.8333333432674408,
      "step": 220
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 152.5416717529297,
      "epoch": 2.3733333333333335,
      "grad_norm": 0.8172355890274048,
      "kl": 0.51953125,
      "learning_rate": 3.7819207248475174e-07,
      "loss": 0.0473,
      "reward": 1.6666666865348816,
      "reward_std": 0.40420936793088913,
      "rewards/accuracy_reward": 0.7500000111758709,
      "rewards/format_reward": 0.916666679084301,
      "step": 221
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 234.00000953674316,
      "epoch": 2.384,
      "grad_norm": 0.4453510046005249,
      "kl": 0.760650634765625,
      "learning_rate": 3.6581697398785615e-07,
      "loss": -0.0164,
      "reward": 1.4791666865348816,
      "reward_std": 0.3892955183982849,
      "rewards/accuracy_reward": 0.5208333414047956,
      "rewards/format_reward": 0.9583333432674408,
      "step": 222
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 223.3958396911621,
      "epoch": 2.3946666666666667,
      "grad_norm": 0.7755252122879028,
      "kl": 1.18450927734375,
      "learning_rate": 3.5361955190325563e-07,
      "loss": 0.0217,
      "reward": 1.5833333730697632,
      "reward_std": 0.5788923390209675,
      "rewards/accuracy_reward": 0.7291666846722364,
      "rewards/format_reward": 0.8541666939854622,
      "step": 223
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 200.2916717529297,
      "epoch": 2.405333333333333,
      "grad_norm": 0.29831400513648987,
      "kl": 0.40399169921875,
      "learning_rate": 3.4160171702854286e-07,
      "loss": -0.0345,
      "reward": 1.4583333730697632,
      "reward_std": 0.5722163133323193,
      "rewards/accuracy_reward": 0.6041666753590107,
      "rewards/format_reward": 0.8541666865348816,
      "step": 224
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 232.39583778381348,
      "epoch": 2.416,
      "grad_norm": 0.2781633138656616,
      "kl": 0.917236328125,
      "learning_rate": 3.297653520279226e-07,
      "loss": 0.0222,
      "reward": 1.3958333656191826,
      "reward_std": 0.6155543886125088,
      "rewards/accuracy_reward": 0.5625000186264515,
      "rewards/format_reward": 0.833333358168602,
      "step": 225
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 236.77083778381348,
      "epoch": 2.4266666666666667,
      "grad_norm": 0.24500270187854767,
      "kl": 0.32452392578125,
      "learning_rate": 3.1811231113727974e-07,
      "loss": 0.043,
      "reward": 1.5208333730697632,
      "reward_std": 0.5257022529840469,
      "rewards/accuracy_reward": 0.6250000186264515,
      "rewards/format_reward": 0.8958333507180214,
      "step": 226
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 210.14583778381348,
      "epoch": 2.437333333333333,
      "grad_norm": 0.28082796931266785,
      "kl": 0.30218505859375,
      "learning_rate": 3.066444198737053e-07,
      "loss": 0.0462,
      "reward": 1.5208333730697632,
      "reward_std": 0.5423685237765312,
      "rewards/accuracy_reward": 0.6250000186264515,
      "rewards/format_reward": 0.895833358168602,
      "step": 227
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 236.43750762939453,
      "epoch": 2.448,
      "grad_norm": 0.32010531425476074,
      "kl": 0.2791748046875,
      "learning_rate": 2.9536347474951645e-07,
      "loss": 0.0316,
      "reward": 1.7291667014360428,
      "reward_std": 0.4177280552685261,
      "rewards/accuracy_reward": 0.791666679084301,
      "rewards/format_reward": 0.9375000149011612,
      "step": 228
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 243.35417366027832,
      "epoch": 2.458666666666667,
      "grad_norm": 0.4656471014022827,
      "kl": 0.6329345703125,
      "learning_rate": 2.842712429908237e-07,
      "loss": 0.0578,
      "reward": 1.187500037252903,
      "reward_std": 0.566876970231533,
      "rewards/accuracy_reward": 0.3125000074505806,
      "rewards/format_reward": 0.8750000149011612,
      "step": 229
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 266.68750381469727,
      "epoch": 2.469333333333333,
      "grad_norm": 5.0469970703125,
      "kl": 2.81072998046875,
      "learning_rate": 2.733694622606824e-07,
      "loss": 0.109,
      "reward": 1.2708333730697632,
      "reward_std": 0.7010384388267994,
      "rewards/accuracy_reward": 0.4583333432674408,
      "rewards/format_reward": 0.8125000149011612,
      "step": 230
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 239.02083683013916,
      "epoch": 2.48,
      "grad_norm": 0.692191481590271,
      "kl": 0.800323486328125,
      "learning_rate": 2.626598403868801e-07,
      "loss": 0.0852,
      "reward": 1.5000000298023224,
      "reward_std": 0.5464099086821079,
      "rewards/accuracy_reward": 0.6041666828095913,
      "rewards/format_reward": 0.8958333507180214,
      "step": 231
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 204.97917556762695,
      "epoch": 2.490666666666667,
      "grad_norm": 0.18402956426143646,
      "kl": 0.36541748046875,
      "learning_rate": 2.521440550943943e-07,
      "loss": 0.0105,
      "reward": 1.7500000149011612,
      "reward_std": 0.3694790154695511,
      "rewards/accuracy_reward": 0.8333333507180214,
      "rewards/format_reward": 0.9166666716337204,
      "step": 232
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 261.1666774749756,
      "epoch": 2.501333333333333,
      "grad_norm": 0.18328428268432617,
      "kl": 0.095855712890625,
      "learning_rate": 2.418237537425666e-07,
      "loss": 0.0137,
      "reward": 1.7291667014360428,
      "reward_std": 0.40168891847133636,
      "rewards/accuracy_reward": 0.7708333432674408,
      "rewards/format_reward": 0.9583333432674408,
      "step": 233
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 215.5416717529297,
      "epoch": 2.512,
      "grad_norm": 0.19842679798603058,
      "kl": 0.108673095703125,
      "learning_rate": 2.3170055306703512e-07,
      "loss": 0.028,
      "reward": 1.7500000447034836,
      "reward_std": 0.45132480934262276,
      "rewards/accuracy_reward": 0.770833358168602,
      "rewards/format_reward": 0.9791666716337204,
      "step": 234
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 258.0208396911621,
      "epoch": 2.522666666666667,
      "grad_norm": 0.9317381381988525,
      "kl": 0.91571044921875,
      "learning_rate": 2.2177603892646277e-07,
      "loss": 0.0325,
      "reward": 1.4791667014360428,
      "reward_std": 0.49830594658851624,
      "rewards/accuracy_reward": 0.604166679084301,
      "rewards/format_reward": 0.8750000149011612,
      "step": 235
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 216.5208396911621,
      "epoch": 2.533333333333333,
      "grad_norm": 2.2101762294769287,
      "kl": 1.422576904296875,
      "learning_rate": 2.1205176605410259e-07,
      "loss": 0.0724,
      "reward": 1.6875000447034836,
      "reward_std": 0.3842546343803406,
      "rewards/accuracy_reward": 0.7708333414047956,
      "rewards/format_reward": 0.916666679084301,
      "step": 236
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 243.12500667572021,
      "epoch": 2.544,
      "grad_norm": 1.1581038236618042,
      "kl": 1.95135498046875,
      "learning_rate": 2.0252925781424153e-07,
      "loss": 0.1607,
      "reward": 1.270833358168602,
      "reward_std": 0.6249947585165501,
      "rewards/accuracy_reward": 0.47916667722165585,
      "rewards/format_reward": 0.7916666939854622,
      "step": 237
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 242.5208444595337,
      "epoch": 2.554666666666667,
      "grad_norm": 0.6918954849243164,
      "kl": 1.07720947265625,
      "learning_rate": 1.932100059635551e-07,
      "loss": 0.119,
      "reward": 1.3541667014360428,
      "reward_std": 0.51335409283638,
      "rewards/accuracy_reward": 0.4791666865348816,
      "rewards/format_reward": 0.8750000149011612,
      "step": 238
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 281.1458377838135,
      "epoch": 2.5653333333333332,
      "grad_norm": 0.5743387341499329,
      "kl": 0.401519775390625,
      "learning_rate": 1.8409547041741436e-07,
      "loss": 0.0128,
      "reward": 1.3125000298023224,
      "reward_std": 0.4759799800813198,
      "rewards/accuracy_reward": 0.4375000111758709,
      "rewards/format_reward": 0.8750000074505806,
      "step": 239
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 338.0000057220459,
      "epoch": 2.576,
      "grad_norm": 0.6467074155807495,
      "kl": 1.7817535400390625,
      "learning_rate": 1.751870790211838e-07,
      "loss": 0.1999,
      "reward": 1.3750000447034836,
      "reward_std": 0.5198958218097687,
      "rewards/accuracy_reward": 0.479166679084301,
      "rewards/format_reward": 0.8958333507180214,
      "step": 240
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 275.14584732055664,
      "epoch": 2.586666666666667,
      "grad_norm": 0.6359933614730835,
      "kl": 0.8695068359375,
      "learning_rate": 1.6648622732653946e-07,
      "loss": 0.0497,
      "reward": 1.4375000521540642,
      "reward_std": 0.547709733247757,
      "rewards/accuracy_reward": 0.5416666753590107,
      "rewards/format_reward": 0.8958333432674408,
      "step": 241
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 241.33334159851074,
      "epoch": 2.5973333333333333,
      "grad_norm": 0.8483588695526123,
      "kl": 0.48760986328125,
      "learning_rate": 1.5799427837284686e-07,
      "loss": 0.1047,
      "reward": 1.6666667014360428,
      "reward_std": 0.37073594331741333,
      "rewards/accuracy_reward": 0.7500000149011612,
      "rewards/format_reward": 0.9166666865348816,
      "step": 242
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 235.08333587646484,
      "epoch": 2.608,
      "grad_norm": 0.25336307287216187,
      "kl": 0.22003173828125,
      "learning_rate": 1.4971256247363534e-07,
      "loss": 0.0624,
      "reward": 1.5833333730697632,
      "reward_std": 0.5708101950585842,
      "rewards/accuracy_reward": 0.6875000279396772,
      "rewards/format_reward": 0.8958333432674408,
      "step": 243
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 291.1041793823242,
      "epoch": 2.618666666666667,
      "grad_norm": 0.47303256392478943,
      "kl": 0.993865966796875,
      "learning_rate": 1.416423770081941e-07,
      "loss": 0.1299,
      "reward": 1.4583333544433117,
      "reward_std": 0.5008263885974884,
      "rewards/accuracy_reward": 0.6666666828095913,
      "rewards/format_reward": 0.7916666902601719,
      "step": 244
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 282.0416774749756,
      "epoch": 2.6293333333333333,
      "grad_norm": 0.39533504843711853,
      "kl": 0.520782470703125,
      "learning_rate": 1.3378498621833236e-07,
      "loss": -0.0644,
      "reward": 1.3958333730697632,
      "reward_std": 0.6059102527797222,
      "rewards/accuracy_reward": 0.5833333507180214,
      "rewards/format_reward": 0.8125000223517418,
      "step": 245
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 315.14584732055664,
      "epoch": 2.64,
      "grad_norm": 0.33665895462036133,
      "kl": 0.689300537109375,
      "learning_rate": 1.2614162101032601e-07,
      "loss": 0.0385,
      "reward": 1.4166666939854622,
      "reward_std": 0.4743013381958008,
      "rewards/accuracy_reward": 0.5625000093132257,
      "rewards/format_reward": 0.8541666716337204,
      "step": 246
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 213.68750762939453,
      "epoch": 2.6506666666666665,
      "grad_norm": 0.25907960534095764,
      "kl": 0.36181640625,
      "learning_rate": 1.1871347876209121e-07,
      "loss": 0.0062,
      "reward": 1.520833358168602,
      "reward_std": 0.5440386645495892,
      "rewards/accuracy_reward": 0.6458333525806665,
      "rewards/format_reward": 0.8750000149011612,
      "step": 247
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 305.4166793823242,
      "epoch": 2.6613333333333333,
      "grad_norm": 0.44137367606163025,
      "kl": 0.75616455078125,
      "learning_rate": 1.1150172313560741e-07,
      "loss": 0.0573,
      "reward": 1.3750000223517418,
      "reward_std": 0.5896422937512398,
      "rewards/accuracy_reward": 0.5208333544433117,
      "rewards/format_reward": 0.8541666865348816,
      "step": 248
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 230.31250667572021,
      "epoch": 2.672,
      "grad_norm": 0.3653818666934967,
      "kl": 0.227935791015625,
      "learning_rate": 1.0450748389462184e-07,
      "loss": 0.0375,
      "reward": 1.3958333805203438,
      "reward_std": 0.48087166249752045,
      "rewards/accuracy_reward": 0.5000000149011612,
      "rewards/format_reward": 0.8958333432674408,
      "step": 249
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 240.56250762939453,
      "epoch": 2.6826666666666665,
      "grad_norm": 0.27294793725013733,
      "kl": 0.223388671875,
      "learning_rate": 9.773185672766705e-08,
      "loss": -0.0183,
      "reward": 1.6041666865348816,
      "reward_std": 0.4456373043358326,
      "rewards/accuracy_reward": 0.666666679084301,
      "rewards/format_reward": 0.9375000149011612,
      "step": 250
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 414.2916774749756,
      "epoch": 2.6933333333333334,
      "grad_norm": 0.2690182328224182,
      "kl": 0.7999267578125,
      "learning_rate": 9.117590307641361e-08,
      "loss": 0.1233,
      "reward": 1.000000026077032,
      "reward_std": 0.6379684209823608,
      "rewards/accuracy_reward": 0.2708333358168602,
      "rewards/format_reward": 0.7291666902601719,
      "step": 251
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 290.645845413208,
      "epoch": 2.7039999999999997,
      "grad_norm": 0.19759991765022278,
      "kl": 0.17645263671875,
      "learning_rate": 8.484064996939023e-08,
      "loss": 0.0121,
      "reward": 1.437500037252903,
      "reward_std": 0.5738539323210716,
      "rewards/accuracy_reward": 0.5833333469927311,
      "rewards/format_reward": 0.8541666828095913,
      "step": 252
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 327.3541736602783,
      "epoch": 2.7146666666666666,
      "grad_norm": 0.2967400848865509,
      "kl": 0.84234619140625,
      "learning_rate": 7.872708986109183e-08,
      "loss": 0.112,
      "reward": 1.2916667014360428,
      "reward_std": 0.519895825535059,
      "rewards/accuracy_reward": 0.4791666716337204,
      "rewards/format_reward": 0.8125000149011612,
      "step": 253
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 320.6458396911621,
      "epoch": 2.7253333333333334,
      "grad_norm": 1.6768146753311157,
      "kl": 1.749481201171875,
      "learning_rate": 7.283618047650808e-08,
      "loss": 0.1938,
      "reward": 1.2291666865348816,
      "reward_std": 0.5054984018206596,
      "rewards/accuracy_reward": 0.41666667349636555,
      "rewards/format_reward": 0.8125000149011612,
      "step": 254
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 190.7916717529297,
      "epoch": 2.7359999999999998,
      "grad_norm": 8.16866397857666,
      "kl": 2.960968017578125,
      "learning_rate": 6.716884466108869e-08,
      "loss": 0.283,
      "reward": 1.6875000298023224,
      "reward_std": 0.3987618423998356,
      "rewards/accuracy_reward": 0.750000013038516,
      "rewards/format_reward": 0.9375000149011612,
      "step": 255
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 282.4166736602783,
      "epoch": 2.7466666666666666,
      "grad_norm": 0.19895876944065094,
      "kl": 0.157257080078125,
      "learning_rate": 6.17259702361741e-08,
      "loss": 0.0076,
      "reward": 1.645833358168602,
      "reward_std": 0.4148009791970253,
      "rewards/accuracy_reward": 0.6875000167638063,
      "rewards/format_reward": 0.9583333432674408,
      "step": 256
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 289.1875057220459,
      "epoch": 2.7573333333333334,
      "grad_norm": 0.549140214920044,
      "kl": 0.666778564453125,
      "learning_rate": 5.65084098599144e-08,
      "loss": 0.0692,
      "reward": 1.2500000447034836,
      "reward_std": 0.6597342528402805,
      "rewards/accuracy_reward": 0.4583333507180214,
      "rewards/format_reward": 0.7916666865348816,
      "step": 257
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 229.06250381469727,
      "epoch": 2.768,
      "grad_norm": 0.37792423367500305,
      "kl": 0.859893798828125,
      "learning_rate": 5.151698089369433e-08,
      "loss": 0.0291,
      "reward": 1.4791667088866234,
      "reward_std": 0.5829372256994247,
      "rewards/accuracy_reward": 0.6875000055879354,
      "rewards/format_reward": 0.791666679084301,
      "step": 258
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 240.12500762939453,
      "epoch": 2.7786666666666666,
      "grad_norm": 0.340668261051178,
      "kl": 0.816009521484375,
      "learning_rate": 4.675246527408872e-08,
      "loss": 0.009,
      "reward": 1.5000000447034836,
      "reward_std": 0.6165339834988117,
      "rewards/accuracy_reward": 0.7083333395421505,
      "rewards/format_reward": 0.7916666865348816,
      "step": 259
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 246.7708396911621,
      "epoch": 2.7893333333333334,
      "grad_norm": 0.18053409457206726,
      "kl": 0.27801513671875,
      "learning_rate": 4.221560939036828e-08,
      "loss": -0.0014,
      "reward": 1.3125000298023224,
      "reward_std": 0.5489303506910801,
      "rewards/accuracy_reward": 0.4166666753590107,
      "rewards/format_reward": 0.895833358168602,
      "step": 260
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 306.04167556762695,
      "epoch": 2.8,
      "grad_norm": 0.19975829124450684,
      "kl": 0.544586181640625,
      "learning_rate": 3.790712396757317e-08,
      "loss": 0.0101,
      "reward": 1.416666716337204,
      "reward_std": 0.604806911200285,
      "rewards/accuracy_reward": 0.5208333488553762,
      "rewards/format_reward": 0.8958333358168602,
      "step": 261
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 227.0000057220459,
      "epoch": 2.8106666666666666,
      "grad_norm": 0.27892568707466125,
      "kl": 0.60784912109375,
      "learning_rate": 3.382768395517355e-08,
      "loss": 0.0042,
      "reward": 1.3750000298023224,
      "reward_std": 0.5251008756458759,
      "rewards/accuracy_reward": 0.479166679084301,
      "rewards/format_reward": 0.8958333432674408,
      "step": 262
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 317.79167652130127,
      "epoch": 2.8213333333333335,
      "grad_norm": 0.2454572319984436,
      "kl": 0.6148834228515625,
      "learning_rate": 2.997792842133445e-08,
      "loss": 0.0885,
      "reward": 1.5208333507180214,
      "reward_std": 0.3212082237005234,
      "rewards/accuracy_reward": 0.6250000018626451,
      "rewards/format_reward": 0.8958333432674408,
      "step": 263
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 240.93750762939453,
      "epoch": 2.832,
      "grad_norm": 3.458434820175171,
      "kl": 1.1170654296875,
      "learning_rate": 2.6358460452802768e-08,
      "loss": 0.1461,
      "reward": 1.458333358168602,
      "reward_std": 0.5896422863006592,
      "rewards/accuracy_reward": 0.5833333469927311,
      "rewards/format_reward": 0.8750000223517418,
      "step": 264
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 245.12500762939453,
      "epoch": 2.8426666666666667,
      "grad_norm": 0.4039785861968994,
      "kl": 1.179351806640625,
      "learning_rate": 2.296984706042993e-08,
      "loss": 0.0867,
      "reward": 1.5625000298023224,
      "reward_std": 0.47457385063171387,
      "rewards/accuracy_reward": 0.6875000074505806,
      "rewards/format_reward": 0.8750000149011612,
      "step": 265
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 293.22916984558105,
      "epoch": 2.8533333333333335,
      "grad_norm": 0.689044713973999,
      "kl": 0.975433349609375,
      "learning_rate": 1.9812619090345984e-08,
      "loss": 0.1793,
      "reward": 1.6250000298023224,
      "reward_std": 0.4975505545735359,
      "rewards/accuracy_reward": 0.7083333432674408,
      "rewards/format_reward": 0.916666679084301,
      "step": 266
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 348.16668128967285,
      "epoch": 2.864,
      "grad_norm": 0.7113028764724731,
      "kl": 2.130126953125,
      "learning_rate": 1.688727114079963e-08,
      "loss": 0.1079,
      "reward": 0.9583333507180214,
      "reward_std": 0.6428273692727089,
      "rewards/accuracy_reward": 0.33333334513008595,
      "rewards/format_reward": 0.6250000037252903,
      "step": 267
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 289.5208435058594,
      "epoch": 2.8746666666666667,
      "grad_norm": 0.8639241456985474,
      "kl": 1.56182861328125,
      "learning_rate": 1.4194261484677007e-08,
      "loss": 0.1785,
      "reward": 1.5000000298023224,
      "reward_std": 0.514606636017561,
      "rewards/accuracy_reward": 0.6458333488553762,
      "rewards/format_reward": 0.8541666865348816,
      "step": 268
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 196.2291717529297,
      "epoch": 2.8853333333333335,
      "grad_norm": 0.28236812353134155,
      "kl": 0.495849609375,
      "learning_rate": 1.1734011997709626e-08,
      "loss": 0.0362,
      "reward": 1.6041667014360428,
      "reward_std": 0.5063771232962608,
      "rewards/accuracy_reward": 0.7291666809469461,
      "rewards/format_reward": 0.8750000223517418,
      "step": 269
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 254.4791717529297,
      "epoch": 2.896,
      "grad_norm": 0.4359928071498871,
      "kl": 0.602508544921875,
      "learning_rate": 9.50690809238558e-09,
      "loss": 0.0818,
      "reward": 1.6458333730697632,
      "reward_std": 0.5747367590665817,
      "rewards/accuracy_reward": 0.7291666865348816,
      "rewards/format_reward": 0.9166666865348816,
      "step": 270
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 300.5208492279053,
      "epoch": 2.9066666666666667,
      "grad_norm": 0.33899879455566406,
      "kl": 0.670623779296875,
      "learning_rate": 7.513298657572831e-09,
      "loss": 0.0838,
      "reward": 1.3958333432674408,
      "reward_std": 0.36908068880438805,
      "rewards/accuracy_reward": 0.47916666977107525,
      "rewards/format_reward": 0.9166666865348816,
      "step": 271
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 225.29167079925537,
      "epoch": 2.9173333333333336,
      "grad_norm": 0.22239045798778534,
      "kl": 0.837432861328125,
      "learning_rate": 5.753496003862935e-09,
      "loss": 0.0877,
      "reward": 1.416666690260172,
      "reward_std": 0.46875908970832825,
      "rewards/accuracy_reward": 0.6041666772216558,
      "rewards/format_reward": 0.8125000037252903,
      "step": 272
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 289.27084159851074,
      "epoch": 2.928,
      "grad_norm": 0.4770262837409973,
      "kl": 0.46295166015625,
      "learning_rate": 4.227775814646173e-09,
      "loss": 0.0406,
      "reward": 1.1458333805203438,
      "reward_std": 0.4217669777572155,
      "rewards/accuracy_reward": 0.25000000931322575,
      "rewards/format_reward": 0.8958333432674408,
      "step": 273
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 249.45834159851074,
      "epoch": 2.9386666666666668,
      "grad_norm": 0.523398220539093,
      "kl": 0.4814453125,
      "learning_rate": 2.9363771029242705e-09,
      "loss": 0.0887,
      "reward": 1.5625000596046448,
      "reward_std": 0.5939319953322411,
      "rewards/accuracy_reward": 0.7083333544433117,
      "rewards/format_reward": 0.8541667014360428,
      "step": 274
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 246.04167366027832,
      "epoch": 2.9493333333333336,
      "grad_norm": 0.3436894416809082,
      "kl": 0.813873291015625,
      "learning_rate": 1.8795021738670093e-09,
      "loss": 0.0461,
      "reward": 1.5208333507180214,
      "reward_std": 0.5120971687138081,
      "rewards/accuracy_reward": 0.6666666828095913,
      "rewards/format_reward": 0.854166679084301,
      "step": 275
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 274.6666774749756,
      "epoch": 2.96,
      "grad_norm": 0.20680680871009827,
      "kl": 0.593109130859375,
      "learning_rate": 1.0573165931209162e-09,
      "loss": 0.0129,
      "reward": 1.5416667014360428,
      "reward_std": 0.43123577535152435,
      "rewards/accuracy_reward": 0.6666666846722364,
      "rewards/format_reward": 0.8750000111758709,
      "step": 276
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 222.81250762939453,
      "epoch": 2.970666666666667,
      "grad_norm": 0.2603120803833008,
      "kl": 0.629364013671875,
      "learning_rate": 4.699491608713413e-10,
      "loss": -0.0279,
      "reward": 1.479166716337204,
      "reward_std": 0.5843288898468018,
      "rewards/accuracy_reward": 0.604166692122817,
      "rewards/format_reward": 0.8750000223517418,
      "step": 277
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 321.8958435058594,
      "epoch": 2.981333333333333,
      "grad_norm": 1.4922990798950195,
      "kl": 1.21771240234375,
      "learning_rate": 1.1749189166609943e-10,
      "loss": 0.1116,
      "reward": 1.2708333656191826,
      "reward_std": 0.677177045494318,
      "rewards/accuracy_reward": 0.45833333767950535,
      "rewards/format_reward": 0.8125000298023224,
      "step": 278
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 225.45833778381348,
      "epoch": 2.992,
      "grad_norm": 0.13138599693775177,
      "kl": 0.090789794921875,
      "learning_rate": 0.0,
      "loss": -0.0398,
      "reward": 1.770833358168602,
      "reward_std": 0.26070838794112206,
      "rewards/accuracy_reward": 0.791666679084301,
      "rewards/format_reward": 0.9791666716337204,
      "step": 279
    },
    {
      "epoch": 2.992,
      "step": 279,
      "total_flos": 0.0,
      "train_loss": 0.061770447607879864,
      "train_runtime": 25801.9425,
      "train_samples_per_second": 0.087,
      "train_steps_per_second": 0.011
    }
  ],
  "logging_steps": 1,
  "max_steps": 279,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 3,
  "save_steps": 500,
  "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": 2,
  "trial_name": null,
  "trial_params": null
}