{
  "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": 574.645845413208,
      "epoch": 0.010666666666666666,
      "grad_norm": 0.08630427718162537,
      "kl": 0.0,
      "learning_rate": 1.0714285714285714e-07,
      "loss": -0.0341,
      "reward": 0.06250000186264515,
      "reward_std": 0.11558076366782188,
      "rewards/accuracy_reward": 0.06250000186264515,
      "rewards/format_reward": 0.0,
      "step": 1
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 611.5833511352539,
      "epoch": 0.021333333333333333,
      "grad_norm": 0.06199095398187637,
      "kl": 0.0,
      "learning_rate": 2.1428571428571428e-07,
      "loss": -0.0023,
      "reward": 0.02083333395421505,
      "reward_std": 0.05103103443980217,
      "rewards/accuracy_reward": 0.02083333395421505,
      "rewards/format_reward": 0.0,
      "step": 2
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 514.1250114440918,
      "epoch": 0.032,
      "grad_norm": 0.04586375877261162,
      "kl": 0.000579833984375,
      "learning_rate": 3.214285714285714e-07,
      "loss": 0.0497,
      "reward": 0.02083333395421505,
      "reward_std": 0.05103103443980217,
      "rewards/accuracy_reward": 0.02083333395421505,
      "rewards/format_reward": 0.0,
      "step": 3
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 554.4375152587891,
      "epoch": 0.042666666666666665,
      "grad_norm": 0.0028707189485430717,
      "kl": 0.0006541013717651367,
      "learning_rate": 4.2857142857142857e-07,
      "loss": 0.0,
      "reward": 0.0,
      "reward_std": 0.0,
      "rewards/accuracy_reward": 0.0,
      "rewards/format_reward": 0.0,
      "step": 4
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 508.04168701171875,
      "epoch": 0.05333333333333334,
      "grad_norm": 0.1056743636727333,
      "kl": 0.0003535747528076172,
      "learning_rate": 5.357142857142858e-07,
      "loss": 0.0614,
      "reward": 0.06250000186264515,
      "reward_std": 0.11558076739311218,
      "rewards/accuracy_reward": 0.06250000186264515,
      "rewards/format_reward": 0.0,
      "step": 5
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 543.0833511352539,
      "epoch": 0.064,
      "grad_norm": 0.04604552313685417,
      "kl": 0.00043952465057373047,
      "learning_rate": 6.428571428571428e-07,
      "loss": -0.0038,
      "reward": 0.02083333395421505,
      "reward_std": 0.05103103816509247,
      "rewards/accuracy_reward": 0.02083333395421505,
      "rewards/format_reward": 0.0,
      "step": 6
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 585.8958511352539,
      "epoch": 0.07466666666666667,
      "grad_norm": 0.06841025501489639,
      "kl": 0.00035643577575683594,
      "learning_rate": 7.5e-07,
      "loss": 0.0413,
      "reward": 0.0416666679084301,
      "reward_std": 0.06454972922801971,
      "rewards/accuracy_reward": 0.0416666679084301,
      "rewards/format_reward": 0.0,
      "step": 7
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 563.3541870117188,
      "epoch": 0.08533333333333333,
      "grad_norm": 0.0019720818381756544,
      "kl": 0.0005422830581665039,
      "learning_rate": 8.571428571428571e-07,
      "loss": 0.0,
      "reward": 0.0,
      "reward_std": 0.0,
      "rewards/accuracy_reward": 0.0,
      "rewards/format_reward": 0.0,
      "step": 8
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 643.0625152587891,
      "epoch": 0.096,
      "grad_norm": 0.0895163044333458,
      "kl": 0.00038993358612060547,
      "learning_rate": 9.642857142857145e-07,
      "loss": -0.044,
      "reward": 0.02083333395421505,
      "reward_std": 0.05103103443980217,
      "rewards/accuracy_reward": 0.02083333395421505,
      "rewards/format_reward": 0.0,
      "step": 9
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 607.1458511352539,
      "epoch": 0.10666666666666667,
      "grad_norm": 0.03434349223971367,
      "kl": 0.00034165382385253906,
      "learning_rate": 1.0714285714285716e-06,
      "loss": 0.0228,
      "reward": 0.02083333395421505,
      "reward_std": 0.05103103443980217,
      "rewards/accuracy_reward": 0.02083333395421505,
      "rewards/format_reward": 0.0,
      "step": 10
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 528.0000152587891,
      "epoch": 0.11733333333333333,
      "grad_norm": 0.2434433549642563,
      "kl": 0.0005402565002441406,
      "learning_rate": 1.1785714285714285e-06,
      "loss": -0.0169,
      "reward": 0.0416666679084301,
      "reward_std": 0.10206207633018494,
      "rewards/accuracy_reward": 0.0416666679084301,
      "rewards/format_reward": 0.0,
      "step": 11
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 483.83335876464844,
      "epoch": 0.128,
      "grad_norm": 0.002469642786309123,
      "kl": 0.0004532337188720703,
      "learning_rate": 1.2857142857142856e-06,
      "loss": 0.0,
      "reward": 0.0,
      "reward_std": 0.0,
      "rewards/accuracy_reward": 0.0,
      "rewards/format_reward": 0.0,
      "step": 12
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 513.5625038146973,
      "epoch": 0.13866666666666666,
      "grad_norm": 0.14749591052532196,
      "kl": 0.000500798225402832,
      "learning_rate": 1.392857142857143e-06,
      "loss": -0.0378,
      "reward": 0.0416666679084301,
      "reward_std": 0.10206207633018494,
      "rewards/accuracy_reward": 0.0416666679084301,
      "rewards/format_reward": 0.0,
      "step": 13
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 538.7916831970215,
      "epoch": 0.14933333333333335,
      "grad_norm": 0.161040261387825,
      "kl": 0.0008957386016845703,
      "learning_rate": 1.5e-06,
      "loss": 0.0062,
      "reward": 0.0416666679084301,
      "reward_std": 0.10206206887960434,
      "rewards/accuracy_reward": 0.0416666679084301,
      "rewards/format_reward": 0.0,
      "step": 14
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 596.1875114440918,
      "epoch": 0.16,
      "grad_norm": 0.0028237502556294203,
      "kl": 0.0005574226379394531,
      "learning_rate": 1.6071428571428572e-06,
      "loss": 0.0,
      "reward": 0.0,
      "reward_std": 0.0,
      "rewards/accuracy_reward": 0.0,
      "rewards/format_reward": 0.0,
      "step": 15
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 534.1458587646484,
      "epoch": 0.17066666666666666,
      "grad_norm": 0.25467991828918457,
      "kl": 0.0010123252868652344,
      "learning_rate": 1.7142857142857143e-06,
      "loss": -0.0508,
      "reward": 0.0416666679084301,
      "reward_std": 0.10206207260489464,
      "rewards/accuracy_reward": 0.0416666679084301,
      "rewards/format_reward": 0.0,
      "step": 16
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 484.0416851043701,
      "epoch": 0.18133333333333335,
      "grad_norm": 0.011162322014570236,
      "kl": 0.0026879310607910156,
      "learning_rate": 1.8214285714285714e-06,
      "loss": 0.0,
      "reward": 0.0,
      "reward_std": 0.0,
      "rewards/accuracy_reward": 0.0,
      "rewards/format_reward": 0.0,
      "step": 17
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 616.6875114440918,
      "epoch": 0.192,
      "grad_norm": 0.10696219652891159,
      "kl": 0.0013022422790527344,
      "learning_rate": 1.928571428571429e-06,
      "loss": 0.0673,
      "reward": 0.06250000186264515,
      "reward_std": 0.1530931033194065,
      "rewards/accuracy_reward": 0.06250000186264515,
      "rewards/format_reward": 0.0,
      "step": 18
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 542.3750152587891,
      "epoch": 0.20266666666666666,
      "grad_norm": 0.10040425509214401,
      "kl": 0.001857757568359375,
      "learning_rate": 2.035714285714286e-06,
      "loss": 0.0068,
      "reward": 0.02083333395421505,
      "reward_std": 0.05103103816509247,
      "rewards/accuracy_reward": 0.02083333395421505,
      "rewards/format_reward": 0.0,
      "step": 19
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 631.1041831970215,
      "epoch": 0.21333333333333335,
      "grad_norm": 0.07833883166313171,
      "kl": 0.002624988555908203,
      "learning_rate": 2.142857142857143e-06,
      "loss": 0.0135,
      "reward": 0.0416666679084301,
      "reward_std": 0.10206207260489464,
      "rewards/accuracy_reward": 0.0416666679084301,
      "rewards/format_reward": 0.0,
      "step": 20
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 422.77084159851074,
      "epoch": 0.224,
      "grad_norm": 0.11672168970108032,
      "kl": 0.0066432952880859375,
      "learning_rate": 2.25e-06,
      "loss": -0.0038,
      "reward": 0.0625,
      "reward_std": 0.06846532225608826,
      "rewards/accuracy_reward": 0.0625,
      "rewards/format_reward": 0.0,
      "step": 21
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 511.95834732055664,
      "epoch": 0.23466666666666666,
      "grad_norm": 0.05832681804895401,
      "kl": 0.004792213439941406,
      "learning_rate": 2.357142857142857e-06,
      "loss": 0.0088,
      "reward": 0.0625,
      "reward_std": 0.06846532225608826,
      "rewards/accuracy_reward": 0.0625,
      "rewards/format_reward": 0.0,
      "step": 22
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 454.6666793823242,
      "epoch": 0.24533333333333332,
      "grad_norm": 0.05373668298125267,
      "kl": 0.006567955017089844,
      "learning_rate": 2.4642857142857143e-06,
      "loss": 0.0188,
      "reward": 0.02083333395421505,
      "reward_std": 0.05103103443980217,
      "rewards/accuracy_reward": 0.02083333395421505,
      "rewards/format_reward": 0.0,
      "step": 23
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 457.27084732055664,
      "epoch": 0.256,
      "grad_norm": 0.18046098947525024,
      "kl": 0.0054378509521484375,
      "learning_rate": 2.571428571428571e-06,
      "loss": 0.011,
      "reward": 0.06250000186264515,
      "reward_std": 0.11558076366782188,
      "rewards/accuracy_reward": 0.06250000186264515,
      "rewards/format_reward": 0.0,
      "step": 24
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 514.3333473205566,
      "epoch": 0.26666666666666666,
      "grad_norm": 0.1545768529176712,
      "kl": 0.006916046142578125,
      "learning_rate": 2.678571428571429e-06,
      "loss": 0.0313,
      "reward": 0.0416666679084301,
      "reward_std": 0.10206207260489464,
      "rewards/accuracy_reward": 0.0416666679084301,
      "rewards/format_reward": 0.0,
      "step": 25
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 543.0833511352539,
      "epoch": 0.2773333333333333,
      "grad_norm": 0.08341816812753677,
      "kl": 0.0072002410888671875,
      "learning_rate": 2.785714285714286e-06,
      "loss": -0.0054,
      "reward": 0.02083333395421505,
      "reward_std": 0.05103103443980217,
      "rewards/accuracy_reward": 0.02083333395421505,
      "rewards/format_reward": 0.0,
      "step": 26
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 500.02084732055664,
      "epoch": 0.288,
      "grad_norm": 0.1588888019323349,
      "kl": 0.018650054931640625,
      "learning_rate": 2.892857142857143e-06,
      "loss": -0.0055,
      "reward": 0.02083333395421505,
      "reward_std": 0.05103103443980217,
      "rewards/accuracy_reward": 0.02083333395421505,
      "rewards/format_reward": 0.0,
      "step": 27
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 517.6041889190674,
      "epoch": 0.2986666666666667,
      "grad_norm": 0.1424867957830429,
      "kl": 0.007252693176269531,
      "learning_rate": 3e-06,
      "loss": -0.003,
      "reward": 0.0416666679084301,
      "reward_std": 0.10206206887960434,
      "rewards/accuracy_reward": 0.0416666679084301,
      "rewards/format_reward": 0.0,
      "step": 28
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 477.33334732055664,
      "epoch": 0.30933333333333335,
      "grad_norm": 0.11833926290273666,
      "kl": 0.01897430419921875,
      "learning_rate": 2.999882508108334e-06,
      "loss": 0.0413,
      "reward": 0.08333333395421505,
      "reward_std": 0.11949636042118073,
      "rewards/accuracy_reward": 0.08333333395421505,
      "rewards/format_reward": 0.0,
      "step": 29
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 529.1041851043701,
      "epoch": 0.32,
      "grad_norm": 0.21591709554195404,
      "kl": 0.010650634765625,
      "learning_rate": 2.9995300508391286e-06,
      "loss": -0.0722,
      "reward": 0.16666667349636555,
      "reward_std": 0.20412414520978928,
      "rewards/accuracy_reward": 0.16666667349636555,
      "rewards/format_reward": 0.0,
      "step": 30
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 487.5416831970215,
      "epoch": 0.33066666666666666,
      "grad_norm": 0.12658198177814484,
      "kl": 0.011867523193359375,
      "learning_rate": 2.998942683406879e-06,
      "loss": 0.0104,
      "reward": 0.18750000558793545,
      "reward_std": 0.2446802221238613,
      "rewards/accuracy_reward": 0.18750000558793545,
      "rewards/format_reward": 0.0,
      "step": 31
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 460.0625114440918,
      "epoch": 0.3413333333333333,
      "grad_norm": 0.21254189312458038,
      "kl": 0.02234649658203125,
      "learning_rate": 2.998120497826133e-06,
      "loss": 0.0341,
      "reward": 0.08333333395421505,
      "reward_std": 0.11949635669589043,
      "rewards/accuracy_reward": 0.08333333395421505,
      "rewards/format_reward": 0.0,
      "step": 32
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 501.5625114440918,
      "epoch": 0.352,
      "grad_norm": 0.18026530742645264,
      "kl": 0.013641357421875,
      "learning_rate": 2.9970636228970757e-06,
      "loss": -0.0473,
      "reward": 0.12500000186264515,
      "reward_std": 0.22155842930078506,
      "rewards/accuracy_reward": 0.12500000186264515,
      "rewards/format_reward": 0.0,
      "step": 33
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 513.2916774749756,
      "epoch": 0.3626666666666667,
      "grad_norm": 0.1767341047525406,
      "kl": 0.021457672119140625,
      "learning_rate": 2.995772224185354e-06,
      "loss": 0.0094,
      "reward": 0.0833333358168602,
      "reward_std": 0.16661180183291435,
      "rewards/accuracy_reward": 0.0833333358168602,
      "rewards/format_reward": 0.0,
      "step": 34
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 660.7083511352539,
      "epoch": 0.37333333333333335,
      "grad_norm": 0.13715602457523346,
      "kl": 0.022609710693359375,
      "learning_rate": 2.9942465039961373e-06,
      "loss": -0.0303,
      "reward": 0.06250000186264515,
      "reward_std": 0.1530931070446968,
      "rewards/accuracy_reward": 0.06250000186264515,
      "rewards/format_reward": 0.0,
      "step": 35
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 580.8125152587891,
      "epoch": 0.384,
      "grad_norm": 0.11364670842885971,
      "kl": 0.011260986328125,
      "learning_rate": 2.992486701342427e-06,
      "loss": 0.0458,
      "reward": 0.1458333358168602,
      "reward_std": 0.2350771278142929,
      "rewards/accuracy_reward": 0.1458333358168602,
      "rewards/format_reward": 0.0,
      "step": 36
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 500.08334732055664,
      "epoch": 0.39466666666666667,
      "grad_norm": 0.09874270111322403,
      "kl": 0.015888214111328125,
      "learning_rate": 2.9904930919076145e-06,
      "loss": -0.0039,
      "reward": 0.0833333358168602,
      "reward_std": 0.06454972922801971,
      "rewards/accuracy_reward": 0.0833333358168602,
      "rewards/format_reward": 0.0,
      "step": 37
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 628.7708511352539,
      "epoch": 0.4053333333333333,
      "grad_norm": 0.089125856757164,
      "kl": 0.011135101318359375,
      "learning_rate": 2.9882659880022905e-06,
      "loss": -0.0547,
      "reward": 0.12500000186264515,
      "reward_std": 0.18404608964920044,
      "rewards/accuracy_reward": 0.12500000186264515,
      "rewards/format_reward": 0.0,
      "step": 38
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 614.1666889190674,
      "epoch": 0.416,
      "grad_norm": 0.2361649125814438,
      "kl": 0.016826629638671875,
      "learning_rate": 2.985805738515323e-06,
      "loss": 0.0092,
      "reward": 0.18750000558793545,
      "reward_std": 0.1801304928958416,
      "rewards/accuracy_reward": 0.18750000558793545,
      "rewards/format_reward": 0.0,
      "step": 39
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 500.7708492279053,
      "epoch": 0.4266666666666667,
      "grad_norm": 0.2772132158279419,
      "kl": 0.01801300048828125,
      "learning_rate": 2.9831127288592004e-06,
      "loss": -0.0077,
      "reward": 0.2291666753590107,
      "reward_std": 0.28219256550073624,
      "rewards/accuracy_reward": 0.2291666753590107,
      "rewards/format_reward": 0.0,
      "step": 40
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 615.9375114440918,
      "epoch": 0.43733333333333335,
      "grad_norm": 0.11632812768220901,
      "kl": 0.013370513916015625,
      "learning_rate": 2.9801873809096543e-06,
      "loss": 0.0736,
      "reward": 0.18750000558793545,
      "reward_std": 0.2446802258491516,
      "rewards/accuracy_reward": 0.18750000558793545,
      "rewards/format_reward": 0.0,
      "step": 41
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 481.37500762939453,
      "epoch": 0.448,
      "grad_norm": 0.16239772737026215,
      "kl": 0.02106475830078125,
      "learning_rate": 2.97703015293957e-06,
      "loss": 0.018,
      "reward": 0.20833333767950535,
      "reward_std": 0.3506578877568245,
      "rewards/accuracy_reward": 0.20833333767950535,
      "rewards/format_reward": 0.0,
      "step": 42
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 447.4166831970215,
      "epoch": 0.45866666666666667,
      "grad_norm": 0.17852915823459625,
      "kl": 0.027374267578125,
      "learning_rate": 2.9736415395471975e-06,
      "loss": 0.0367,
      "reward": 0.2916666753590107,
      "reward_std": 0.24859581887722015,
      "rewards/accuracy_reward": 0.2916666753590107,
      "rewards/format_reward": 0.0,
      "step": 43
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 424.91667556762695,
      "epoch": 0.4693333333333333,
      "grad_norm": 0.17440026998519897,
      "kl": 0.025390625,
      "learning_rate": 2.970022071578666e-06,
      "loss": 0.0344,
      "reward": 0.5833333432674408,
      "reward_std": 0.38552646338939667,
      "rewards/accuracy_reward": 0.5833333432674408,
      "rewards/format_reward": 0.0,
      "step": 44
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 595.6041793823242,
      "epoch": 0.48,
      "grad_norm": 0.691380500793457,
      "kl": 0.03966522216796875,
      "learning_rate": 2.9661723160448267e-06,
      "loss": 0.0369,
      "reward": 0.18750000186264515,
      "reward_std": 0.11558076366782188,
      "rewards/accuracy_reward": 0.18750000186264515,
      "rewards/format_reward": 0.0,
      "step": 45
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 583.2291870117188,
      "epoch": 0.49066666666666664,
      "grad_norm": 0.10043969005346298,
      "kl": 0.016597747802734375,
      "learning_rate": 2.962092876032427e-06,
      "loss": 0.0466,
      "reward": 0.3958333358168602,
      "reward_std": 0.2350771240890026,
      "rewards/accuracy_reward": 0.3958333358168602,
      "rewards/format_reward": 0.0,
      "step": 46
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 535.3750152587891,
      "epoch": 0.5013333333333333,
      "grad_norm": 0.2007218450307846,
      "kl": 0.025909423828125,
      "learning_rate": 2.957784390609632e-06,
      "loss": -0.069,
      "reward": 0.4375000111758709,
      "reward_std": 0.3816108666360378,
      "rewards/accuracy_reward": 0.4375000111758709,
      "rewards/format_reward": 0.0,
      "step": 47
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 489.0833435058594,
      "epoch": 0.512,
      "grad_norm": 0.1545257568359375,
      "kl": 0.03205108642578125,
      "learning_rate": 2.9532475347259113e-06,
      "loss": 0.0074,
      "reward": 0.4791666753590107,
      "reward_std": 0.3170611374080181,
      "rewards/accuracy_reward": 0.4791666753590107,
      "rewards/format_reward": 0.0,
      "step": 48
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 518.416675567627,
      "epoch": 0.5226666666666666,
      "grad_norm": 0.15209315717220306,
      "kl": 0.0297698974609375,
      "learning_rate": 2.9484830191063057e-06,
      "loss": 0.0626,
      "reward": 0.47916667722165585,
      "reward_std": 0.299626849591732,
      "rewards/accuracy_reward": 0.47916667722165585,
      "rewards/format_reward": 0.0,
      "step": 49
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 587.9375152587891,
      "epoch": 0.5333333333333333,
      "grad_norm": 0.14053413271903992,
      "kl": 0.023651123046875,
      "learning_rate": 2.9434915901400857e-06,
      "loss": 0.0446,
      "reward": 0.3125000074505806,
      "reward_std": 0.36417657881975174,
      "rewards/accuracy_reward": 0.3125000074505806,
      "rewards/format_reward": 0.0,
      "step": 50
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 533.2083415985107,
      "epoch": 0.544,
      "grad_norm": 0.14134930074214935,
      "kl": 0.024688720703125,
      "learning_rate": 2.938274029763826e-06,
      "loss": 0.0683,
      "reward": 0.4166666753590107,
      "reward_std": 0.2861081659793854,
      "rewards/accuracy_reward": 0.4166666753590107,
      "rewards/format_reward": 0.0,
      "step": 51
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 601.0625152587891,
      "epoch": 0.5546666666666666,
      "grad_norm": 0.14649143815040588,
      "kl": 0.02349853515625,
      "learning_rate": 2.9328311553389116e-06,
      "loss": 0.0642,
      "reward": 0.22916667349636555,
      "reward_std": 0.3092299550771713,
      "rewards/accuracy_reward": 0.22916667349636555,
      "rewards/format_reward": 0.0,
      "step": 52
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 562.3541793823242,
      "epoch": 0.5653333333333334,
      "grad_norm": 0.1704944372177124,
      "kl": 0.0292816162109375,
      "learning_rate": 2.927163819523492e-06,
      "loss": -0.0439,
      "reward": 0.2916666753590107,
      "reward_std": 0.2861081585288048,
      "rewards/accuracy_reward": 0.2916666753590107,
      "rewards/format_reward": 0.0,
      "step": 53
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 487.8541831970215,
      "epoch": 0.576,
      "grad_norm": 0.20558209717273712,
      "kl": 0.04265594482421875,
      "learning_rate": 2.9212729101389084e-06,
      "loss": 0.0393,
      "reward": 0.4166666828095913,
      "reward_std": 0.3332235999405384,
      "rewards/accuracy_reward": 0.4166666828095913,
      "rewards/format_reward": 0.0,
      "step": 54
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 638.2500228881836,
      "epoch": 0.5866666666666667,
      "grad_norm": 0.2073916643857956,
      "kl": 0.02713775634765625,
      "learning_rate": 2.9151593500306103e-06,
      "loss": 0.043,
      "reward": 0.2708333395421505,
      "reward_std": 0.2621145099401474,
      "rewards/accuracy_reward": 0.2708333395421505,
      "rewards/format_reward": 0.0,
      "step": 55
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 432.4791717529297,
      "epoch": 0.5973333333333334,
      "grad_norm": 0.14717544615268707,
      "kl": 0.030670166015625,
      "learning_rate": 2.9088240969235863e-06,
      "loss": 0.012,
      "reward": 0.4791666753590107,
      "reward_std": 0.42872630804777145,
      "rewards/accuracy_reward": 0.4791666753590107,
      "rewards/format_reward": 0.0,
      "step": 56
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 434.2291793823242,
      "epoch": 0.608,
      "grad_norm": 0.1894180178642273,
      "kl": 0.069061279296875,
      "learning_rate": 2.902268143272333e-06,
      "loss": 0.0308,
      "reward": 0.6666666697710752,
      "reward_std": 0.24859581515192986,
      "rewards/accuracy_reward": 0.6666666697710752,
      "rewards/format_reward": 0.0,
      "step": 57
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 467.66667556762695,
      "epoch": 0.6186666666666667,
      "grad_norm": 0.11211913079023361,
      "kl": 0.0320892333984375,
      "learning_rate": 2.895492516105378e-06,
      "loss": 0.0359,
      "reward": 0.4583333395421505,
      "reward_std": 0.22155842557549477,
      "rewards/accuracy_reward": 0.4583333395421505,
      "rewards/format_reward": 0.0,
      "step": 58
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 611.6875076293945,
      "epoch": 0.6293333333333333,
      "grad_norm": 0.12812590599060059,
      "kl": 0.03234100341796875,
      "learning_rate": 2.888498276864393e-06,
      "loss": 0.0159,
      "reward": 0.3125000074505806,
      "reward_std": 0.2621145099401474,
      "rewards/accuracy_reward": 0.3125000074505806,
      "rewards/format_reward": 0.0,
      "step": 59
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 699.4583473205566,
      "epoch": 0.64,
      "grad_norm": 0.0962519571185112,
      "kl": 0.02121734619140625,
      "learning_rate": 2.8812865212379087e-06,
      "loss": -0.016,
      "reward": 0.22916666977107525,
      "reward_std": 0.1801304928958416,
      "rewards/accuracy_reward": 0.22916666977107525,
      "rewards/format_reward": 0.0,
      "step": 60
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 537.1458435058594,
      "epoch": 0.6506666666666666,
      "grad_norm": 0.4079749286174774,
      "kl": 0.03366851806640625,
      "learning_rate": 2.8738583789896743e-06,
      "loss": -0.0003,
      "reward": 0.3333333395421505,
      "reward_std": 0.3680921792984009,
      "rewards/accuracy_reward": 0.3333333395421505,
      "rewards/format_reward": 0.0,
      "step": 61
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 590.4583435058594,
      "epoch": 0.6613333333333333,
      "grad_norm": 0.1474791318178177,
      "kl": 0.0286712646484375,
      "learning_rate": 2.866215013781668e-06,
      "loss": 0.0456,
      "reward": 0.20833334140479565,
      "reward_std": 0.268673874437809,
      "rewards/accuracy_reward": 0.20833334140479565,
      "rewards/format_reward": 0.0,
      "step": 62
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 505.1250190734863,
      "epoch": 0.672,
      "grad_norm": 0.20406241714954376,
      "kl": 0.0316925048828125,
      "learning_rate": 2.858357622991806e-06,
      "loss": 0.0455,
      "reward": 0.4375000111758709,
      "reward_std": 0.41912321001291275,
      "rewards/accuracy_reward": 0.4375000111758709,
      "rewards/format_reward": 0.0,
      "step": 63
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 592.6250228881836,
      "epoch": 0.6826666666666666,
      "grad_norm": 0.17144007980823517,
      "kl": 0.04378509521484375,
      "learning_rate": 2.8502874375263646e-06,
      "loss": 0.0515,
      "reward": 0.25000000186264515,
      "reward_std": 0.22155842557549477,
      "rewards/accuracy_reward": 0.25000000186264515,
      "rewards/format_reward": 0.0,
      "step": 64
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 491.75001430511475,
      "epoch": 0.6933333333333334,
      "grad_norm": 0.21236185729503632,
      "kl": 0.0411224365234375,
      "learning_rate": 2.8420057216271533e-06,
      "loss": 0.07,
      "reward": 0.4166666716337204,
      "reward_std": 0.26603010296821594,
      "rewards/accuracy_reward": 0.4166666716337204,
      "rewards/format_reward": 0.0,
      "step": 65
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 524.0000114440918,
      "epoch": 0.704,
      "grad_norm": 0.15796327590942383,
      "kl": 0.03389739990234375,
      "learning_rate": 2.8335137726734607e-06,
      "loss": 0.0943,
      "reward": 0.39583334140479565,
      "reward_std": 0.3720077648758888,
      "rewards/accuracy_reward": 0.39583334140479565,
      "rewards/format_reward": 0.0,
      "step": 66
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 500.3125114440918,
      "epoch": 0.7146666666666667,
      "grad_norm": 0.27333250641822815,
      "kl": 0.04351806640625,
      "learning_rate": 2.8248129209788164e-06,
      "loss": 0.0156,
      "reward": 0.4375000149011612,
      "reward_std": 0.36417657881975174,
      "rewards/accuracy_reward": 0.4375000149011612,
      "rewards/format_reward": 0.0,
      "step": 67
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 581.8750228881836,
      "epoch": 0.7253333333333334,
      "grad_norm": 4.9572272300720215,
      "kl": 0.10577392578125,
      "learning_rate": 2.8159045295825857e-06,
      "loss": 0.075,
      "reward": 0.2291666679084301,
      "reward_std": 0.2699456959962845,
      "rewards/accuracy_reward": 0.2291666679084301,
      "rewards/format_reward": 0.0,
      "step": 68
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 510.7916793823242,
      "epoch": 0.736,
      "grad_norm": 0.2626616060733795,
      "kl": 0.047515869140625,
      "learning_rate": 2.806789994036445e-06,
      "loss": -0.0249,
      "reward": 0.479166679084301,
      "reward_std": 0.48367294296622276,
      "rewards/accuracy_reward": 0.479166679084301,
      "rewards/format_reward": 0.0,
      "step": 69
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 478.4166793823242,
      "epoch": 0.7466666666666667,
      "grad_norm": 0.12211908400058746,
      "kl": 0.03977203369140625,
      "learning_rate": 2.7974707421857584e-06,
      "loss": 0.0285,
      "reward": 0.4166666679084301,
      "reward_std": 0.23899272084236145,
      "rewards/accuracy_reward": 0.4166666679084301,
      "rewards/format_reward": 0.0,
      "step": 70
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 528.9583492279053,
      "epoch": 0.7573333333333333,
      "grad_norm": 0.1832711100578308,
      "kl": 0.056304931640625,
      "learning_rate": 2.7879482339458973e-06,
      "loss": -0.011,
      "reward": 0.3333333432674408,
      "reward_std": 0.31314554437994957,
      "rewards/accuracy_reward": 0.3333333432674408,
      "rewards/format_reward": 0.0,
      "step": 71
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 501.89584732055664,
      "epoch": 0.768,
      "grad_norm": 0.19136710464954376,
      "kl": 0.0486297607421875,
      "learning_rate": 2.778223961073537e-06,
      "loss": 0.1018,
      "reward": 0.3541666753590107,
      "reward_std": 0.3170611411333084,
      "rewards/accuracy_reward": 0.3541666753590107,
      "rewards/format_reward": 0.0,
      "step": 72
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 417.6458435058594,
      "epoch": 0.7786666666666666,
      "grad_norm": 0.17803442478179932,
      "kl": 0.04766845703125,
      "learning_rate": 2.7682994469329646e-06,
      "loss": 0.0162,
      "reward": 0.4375000074505806,
      "reward_std": 0.36417658254504204,
      "rewards/accuracy_reward": 0.4375000074505806,
      "rewards/format_reward": 0.0,
      "step": 73
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 545.5625190734863,
      "epoch": 0.7893333333333333,
      "grad_norm": 0.2556670606136322,
      "kl": 0.0534820556640625,
      "learning_rate": 2.7581762462574334e-06,
      "loss": 0.0717,
      "reward": 0.43750000931322575,
      "reward_std": 0.2996268533170223,
      "rewards/accuracy_reward": 0.43750000931322575,
      "rewards/format_reward": 0.0,
      "step": 74
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 727.8125152587891,
      "epoch": 0.8,
      "grad_norm": 2.5011374950408936,
      "kl": 0.15460205078125,
      "learning_rate": 2.747855944905606e-06,
      "loss": 0.0504,
      "reward": 0.25000000558793545,
      "reward_std": 0.3506578877568245,
      "rewards/accuracy_reward": 0.25000000558793545,
      "rewards/format_reward": 0.0,
      "step": 75
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 559.6875152587891,
      "epoch": 0.8106666666666666,
      "grad_norm": 0.18483959138393402,
      "kl": 0.0756072998046875,
      "learning_rate": 2.73734015961312e-06,
      "loss": 0.0674,
      "reward": 0.33333334140479565,
      "reward_std": 0.3776952773332596,
      "rewards/accuracy_reward": 0.33333334140479565,
      "rewards/format_reward": 0.0,
      "step": 76
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 569.7708511352539,
      "epoch": 0.8213333333333334,
      "grad_norm": 0.23977096378803253,
      "kl": 0.077392578125,
      "learning_rate": 2.7266305377393183e-06,
      "loss": -0.0037,
      "reward": 0.37500001303851604,
      "reward_std": 0.3977733291685581,
      "rewards/accuracy_reward": 0.37500001303851604,
      "rewards/format_reward": 0.0,
      "step": 77
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 673.3541870117188,
      "epoch": 0.832,
      "grad_norm": 0.15160395205020905,
      "kl": 0.087005615234375,
      "learning_rate": 2.7157287570091765e-06,
      "loss": 0.0068,
      "reward": 0.20833333767950535,
      "reward_std": 0.3506578840315342,
      "rewards/accuracy_reward": 0.20833333767950535,
      "rewards/format_reward": 0.0,
      "step": 78
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 665.2916870117188,
      "epoch": 0.8426666666666667,
      "grad_norm": 0.18723702430725098,
      "kl": 0.086883544921875,
      "learning_rate": 2.7046365252504835e-06,
      "loss": 0.1126,
      "reward": 0.2916666716337204,
      "reward_std": 0.26603010296821594,
      "rewards/accuracy_reward": 0.2916666716337204,
      "rewards/format_reward": 0.0,
      "step": 79
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 570.3750228881836,
      "epoch": 0.8533333333333334,
      "grad_norm": 0.20665200054645538,
      "kl": 0.130096435546875,
      "learning_rate": 2.693355580126295e-06,
      "loss": 0.043,
      "reward": 0.31250000931322575,
      "reward_std": 0.2996268533170223,
      "rewards/accuracy_reward": 0.31250000931322575,
      "rewards/format_reward": 0.0,
      "step": 80
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 474.5625114440918,
      "epoch": 0.864,
      "grad_norm": 0.17555448412895203,
      "kl": 0.089813232421875,
      "learning_rate": 2.6818876888627203e-06,
      "loss": -0.0371,
      "reward": 0.3750000037252903,
      "reward_std": 0.16661180183291435,
      "rewards/accuracy_reward": 0.3750000037252903,
      "rewards/format_reward": 0.0,
      "step": 81
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 517.9583530426025,
      "epoch": 0.8746666666666667,
      "grad_norm": 0.24799272418022156,
      "kl": 0.057342529296875,
      "learning_rate": 2.6702346479720777e-06,
      "loss": 0.1192,
      "reward": 0.41666667722165585,
      "reward_std": 0.22155843675136566,
      "rewards/accuracy_reward": 0.41666667722165585,
      "rewards/format_reward": 0.0,
      "step": 82
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 539.0000114440918,
      "epoch": 0.8853333333333333,
      "grad_norm": 0.18812210857868195,
      "kl": 0.085723876953125,
      "learning_rate": 2.6583982829714576e-06,
      "loss": 0.1357,
      "reward": 0.47916667722165585,
      "reward_std": 0.36417657881975174,
      "rewards/accuracy_reward": 0.47916667722165585,
      "rewards/format_reward": 0.0,
      "step": 83
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 599.6875114440918,
      "epoch": 0.896,
      "grad_norm": 0.15599775314331055,
      "kl": 0.13250732421875,
      "learning_rate": 2.6463804480967447e-06,
      "loss": 0.1373,
      "reward": 0.20833334140479565,
      "reward_std": 0.268673874437809,
      "rewards/accuracy_reward": 0.20833334140479565,
      "rewards/format_reward": 0.0,
      "step": 84
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 623.2916831970215,
      "epoch": 0.9066666666666666,
      "grad_norm": 0.1542806178331375,
      "kl": 0.14483642578125,
      "learning_rate": 2.6341830260121445e-06,
      "loss": 0.0449,
      "reward": 0.2291666679084301,
      "reward_std": 0.1705273911356926,
      "rewards/accuracy_reward": 0.2291666679084301,
      "rewards/format_reward": 0.0,
      "step": 85
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 635.5000171661377,
      "epoch": 0.9173333333333333,
      "grad_norm": 0.12707923352718353,
      "kl": 0.112548828125,
      "learning_rate": 2.6218079275152485e-06,
      "loss": 0.0072,
      "reward": 0.22916667349636555,
      "reward_std": 0.2350771278142929,
      "rewards/accuracy_reward": 0.22916667349636555,
      "rewards/format_reward": 0.0,
      "step": 86
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 601.979190826416,
      "epoch": 0.928,
      "grad_norm": 0.15272405743598938,
      "kl": 0.16046142578125,
      "learning_rate": 2.609257091237699e-06,
      "loss": 0.0329,
      "reward": 0.2291666716337204,
      "reward_std": 0.33713919296860695,
      "rewards/accuracy_reward": 0.2291666716337204,
      "rewards/format_reward": 0.0,
      "step": 87
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 661.3958587646484,
      "epoch": 0.9386666666666666,
      "grad_norm": 0.2064783275127411,
      "kl": 0.2079620361328125,
      "learning_rate": 2.5965324833414926e-06,
      "loss": 0.0359,
      "reward": 0.22916666977107525,
      "reward_std": 0.35457348078489304,
      "rewards/accuracy_reward": 0.22916666977107525,
      "rewards/format_reward": 0.0,
      "step": 88
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 695.0416870117188,
      "epoch": 0.9493333333333334,
      "grad_norm": 0.12128940969705582,
      "kl": 0.16503143310546875,
      "learning_rate": 2.5836360972109642e-06,
      "loss": 0.0464,
      "reward": 0.2708333358168602,
      "reward_std": 0.13301505148410797,
      "rewards/accuracy_reward": 0.2708333358168602,
      "rewards/format_reward": 0.0,
      "step": 89
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 495.1458435058594,
      "epoch": 0.96,
      "grad_norm": 0.21274319291114807,
      "kl": 0.119293212890625,
      "learning_rate": 2.5705699531405174e-06,
      "loss": 0.0933,
      "reward": 0.3750000149011612,
      "reward_std": 0.3332235999405384,
      "rewards/accuracy_reward": 0.3750000149011612,
      "rewards/format_reward": 0.0,
      "step": 90
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 629.7291717529297,
      "epoch": 0.9706666666666667,
      "grad_norm": 7.209469795227051,
      "kl": 0.511688232421875,
      "learning_rate": 2.55733609801813e-06,
      "loss": 0.0816,
      "reward": 0.31250000558793545,
      "reward_std": 0.1801304928958416,
      "rewards/accuracy_reward": 0.31250000558793545,
      "rewards/format_reward": 0.0,
      "step": 91
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 537.3750190734863,
      "epoch": 0.9813333333333333,
      "grad_norm": 0.14357629418373108,
      "kl": 0.11968994140625,
      "learning_rate": 2.5439366050046985e-06,
      "loss": 0.0476,
      "reward": 0.3125000111758709,
      "reward_std": 0.2446802221238613,
      "rewards/accuracy_reward": 0.3125000111758709,
      "rewards/format_reward": 0.0,
      "step": 92
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 514.6875190734863,
      "epoch": 0.992,
      "grad_norm": 0.16206593811511993,
      "kl": 0.130645751953125,
      "learning_rate": 2.5303735732092655e-06,
      "loss": 0.0713,
      "reward": 0.37500000931322575,
      "reward_std": 0.23116153478622437,
      "rewards/accuracy_reward": 0.37500000931322575,
      "rewards/format_reward": 0.0,
      "step": 93
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 444.0000190734863,
      "epoch": 1.0106666666666666,
      "grad_norm": 0.22481516003608704,
      "kl": 0.115203857421875,
      "learning_rate": 2.51664912736018e-06,
      "loss": 0.0358,
      "reward": 0.39583334140479565,
      "reward_std": 0.309229951351881,
      "rewards/accuracy_reward": 0.39583334140479565,
      "rewards/format_reward": 0.0,
      "step": 94
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 410.1666831970215,
      "epoch": 1.0213333333333334,
      "grad_norm": 0.19855806231498718,
      "kl": 0.0841064453125,
      "learning_rate": 2.5027654174722495e-06,
      "loss": 0.0227,
      "reward": 0.43750001676380634,
      "reward_std": 0.34674229472875595,
      "rewards/accuracy_reward": 0.43750001676380634,
      "rewards/format_reward": 0.0,
      "step": 95
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 517.1666793823242,
      "epoch": 1.032,
      "grad_norm": 0.1521887183189392,
      "kl": 0.10870361328125,
      "learning_rate": 2.4887246185099237e-06,
      "loss": -0.0222,
      "reward": 0.4166666828095913,
      "reward_std": 0.3977733328938484,
      "rewards/accuracy_reward": 0.4166666828095913,
      "rewards/format_reward": 0.0,
      "step": 96
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 495.270845413208,
      "epoch": 1.0426666666666666,
      "grad_norm": 0.14659333229064941,
      "kl": 0.119964599609375,
      "learning_rate": 2.474528930046578e-06,
      "loss": 0.0107,
      "reward": 0.4375000037252903,
      "reward_std": 0.2350771203637123,
      "rewards/accuracy_reward": 0.4375000037252903,
      "rewards/format_reward": 0.0,
      "step": 97
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 525.2291870117188,
      "epoch": 1.0533333333333332,
      "grad_norm": 0.28117021918296814,
      "kl": 0.1961669921875,
      "learning_rate": 2.4601805759199317e-06,
      "loss": 0.1342,
      "reward": 0.500000013038516,
      "reward_std": 0.3881702236831188,
      "rewards/accuracy_reward": 0.500000013038516,
      "rewards/format_reward": 0.0,
      "step": 98
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 639.7916946411133,
      "epoch": 1.064,
      "grad_norm": 1.6004377603530884,
      "kl": 0.3116912841796875,
      "learning_rate": 2.445681803883678e-06,
      "loss": 0.0713,
      "reward": 0.29166666977107525,
      "reward_std": 0.18404608592391014,
      "rewards/accuracy_reward": 0.29166666977107525,
      "rewards/format_reward": 0.0,
      "step": 99
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 453.0416736602783,
      "epoch": 1.0746666666666667,
      "grad_norm": 0.13738404214382172,
      "kl": 0.066802978515625,
      "learning_rate": 2.4310348852553554e-06,
      "loss": -0.013,
      "reward": 0.45833333767950535,
      "reward_std": 0.2861081585288048,
      "rewards/accuracy_reward": 0.45833333767950535,
      "rewards/format_reward": 0.0,
      "step": 100
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 473.52084732055664,
      "epoch": 1.0853333333333333,
      "grad_norm": 0.12310273200273514,
      "kl": 0.0965118408203125,
      "learning_rate": 2.4162421145605308e-06,
      "loss": 0.047,
      "reward": 0.3958333395421505,
      "reward_std": 0.2525114081799984,
      "rewards/accuracy_reward": 0.3958333395421505,
      "rewards/format_reward": 0.0,
      "step": 101
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 612.1041831970215,
      "epoch": 1.096,
      "grad_norm": 0.12621328234672546,
      "kl": 0.1176605224609375,
      "learning_rate": 2.4013058091733546e-06,
      "loss": 0.0047,
      "reward": 0.2708333395421505,
      "reward_std": 0.2621145099401474,
      "rewards/accuracy_reward": 0.2708333395421505,
      "rewards/format_reward": 0.0,
      "step": 102
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 529.6250152587891,
      "epoch": 1.1066666666666667,
      "grad_norm": 0.1293606013059616,
      "kl": 0.1231842041015625,
      "learning_rate": 2.3862283089535265e-06,
      "loss": 0.0036,
      "reward": 0.1666666716337204,
      "reward_std": 0.2957112565636635,
      "rewards/accuracy_reward": 0.1666666716337204,
      "rewards/format_reward": 0.0,
      "step": 103
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 467.3541793823242,
      "epoch": 1.1173333333333333,
      "grad_norm": 0.16234131157398224,
      "kl": 0.1039886474609375,
      "learning_rate": 2.371011975879742e-06,
      "loss": 0.0903,
      "reward": 0.41666668094694614,
      "reward_std": 0.3602609895169735,
      "rewards/accuracy_reward": 0.41666668094694614,
      "rewards/format_reward": 0.0,
      "step": 104
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 547.9583492279053,
      "epoch": 1.1280000000000001,
      "grad_norm": 0.15858705341815948,
      "kl": 0.1283111572265625,
      "learning_rate": 2.3556591936796777e-06,
      "loss": 0.0329,
      "reward": 0.3958333432674408,
      "reward_std": 0.3816108666360378,
      "rewards/accuracy_reward": 0.3958333432674408,
      "rewards/format_reward": 0.0,
      "step": 105
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 543.8750076293945,
      "epoch": 1.1386666666666667,
      "grad_norm": 0.14797671139240265,
      "kl": 0.1473388671875,
      "learning_rate": 2.340172367456564e-06,
      "loss": 0.0572,
      "reward": 0.39583333767950535,
      "reward_std": 0.3170611374080181,
      "rewards/accuracy_reward": 0.39583333767950535,
      "rewards/format_reward": 0.0,
      "step": 106
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 619.354190826416,
      "epoch": 1.1493333333333333,
      "grad_norm": 0.1709398776292801,
      "kl": 0.2138671875,
      "learning_rate": 2.3245539233124133e-06,
      "loss": 0.1243,
      "reward": 0.37500000931322575,
      "reward_std": 0.3680921755731106,
      "rewards/accuracy_reward": 0.37500000931322575,
      "rewards/format_reward": 0.0,
      "step": 107
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 536.5416812896729,
      "epoch": 1.16,
      "grad_norm": 0.16565905511379242,
      "kl": 0.166259765625,
      "learning_rate": 2.308806307967955e-06,
      "loss": 0.0667,
      "reward": 0.39583334513008595,
      "reward_std": 0.35457348078489304,
      "rewards/accuracy_reward": 0.39583334513008595,
      "rewards/format_reward": 0.0,
      "step": 108
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 745.0833511352539,
      "epoch": 1.1706666666666667,
      "grad_norm": 0.13067631423473358,
      "kl": 0.26544189453125,
      "learning_rate": 2.2929319883793442e-06,
      "loss": 0.106,
      "reward": 0.18750000558793545,
      "reward_std": 0.28219256922602654,
      "rewards/accuracy_reward": 0.18750000558793545,
      "rewards/format_reward": 0.0,
      "step": 109
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 684.4166793823242,
      "epoch": 1.1813333333333333,
      "grad_norm": 0.15161383152008057,
      "kl": 0.28167724609375,
      "learning_rate": 2.276933451351696e-06,
      "loss": 0.0019,
      "reward": 0.20833333767950535,
      "reward_std": 0.24859581515192986,
      "rewards/accuracy_reward": 0.20833333767950535,
      "rewards/format_reward": 0.0,
      "step": 110
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 686.8125228881836,
      "epoch": 1.192,
      "grad_norm": 0.1324198693037033,
      "kl": 0.2686004638671875,
      "learning_rate": 2.2608132031495184e-06,
      "loss": 0.0964,
      "reward": 0.20833333767950535,
      "reward_std": 0.2861081622540951,
      "rewards/accuracy_reward": 0.20833333767950535,
      "rewards/format_reward": 0.0,
      "step": 111
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 717.1458473205566,
      "epoch": 1.2026666666666666,
      "grad_norm": 0.13981826603412628,
      "kl": 0.28814697265625,
      "learning_rate": 2.244573769104084e-06,
      "loss": 0.0751,
      "reward": 0.2708333395421505,
      "reward_std": 0.40168892592191696,
      "rewards/accuracy_reward": 0.2708333395421505,
      "rewards/format_reward": 0.0,
      "step": 112
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 775.5416946411133,
      "epoch": 1.2133333333333334,
      "grad_norm": 0.11597945541143417,
      "kl": 0.33941650390625,
      "learning_rate": 2.228217693217826e-06,
      "loss": 0.0816,
      "reward": 0.1250000037252903,
      "reward_std": 0.23116153106093407,
      "rewards/accuracy_reward": 0.1250000037252903,
      "rewards/format_reward": 0.0,
      "step": 113
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 657.5625152587891,
      "epoch": 1.224,
      "grad_norm": 0.12906955182552338,
      "kl": 0.248687744140625,
      "learning_rate": 2.211747537765808e-06,
      "loss": 0.0391,
      "reward": 0.22916666977107525,
      "reward_std": 0.21764283254742622,
      "rewards/accuracy_reward": 0.22916666977107525,
      "rewards/format_reward": 0.0,
      "step": 114
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 568.9791831970215,
      "epoch": 1.2346666666666666,
      "grad_norm": 0.2203153669834137,
      "kl": 0.222137451171875,
      "learning_rate": 2.195165882894322e-06,
      "loss": 0.0288,
      "reward": 0.45833335258066654,
      "reward_std": 0.3332236036658287,
      "rewards/accuracy_reward": 0.45833335258066654,
      "rewards/format_reward": 0.0,
      "step": 115
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 749.1041870117188,
      "epoch": 1.2453333333333334,
      "grad_norm": 0.10187610238790512,
      "kl": 0.3118896484375,
      "learning_rate": 2.1784753262166986e-06,
      "loss": 0.0395,
      "reward": 0.02083333395421505,
      "reward_std": 0.05103103443980217,
      "rewards/accuracy_reward": 0.02083333395421505,
      "rewards/format_reward": 0.0,
      "step": 116
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 595.2083396911621,
      "epoch": 1.256,
      "grad_norm": 0.2225155383348465,
      "kl": 0.19293212890625,
      "learning_rate": 2.161678482406374e-06,
      "loss": 0.0777,
      "reward": 0.4166666828095913,
      "reward_std": 0.2957112602889538,
      "rewards/accuracy_reward": 0.4166666828095913,
      "rewards/format_reward": 0.0,
      "step": 117
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 708.0416793823242,
      "epoch": 1.2666666666666666,
      "grad_norm": 0.2065069079399109,
      "kl": 0.260498046875,
      "learning_rate": 2.144777982787286e-06,
      "loss": 0.083,
      "reward": 0.27083334140479565,
      "reward_std": 0.2996268533170223,
      "rewards/accuracy_reward": 0.27083334140479565,
      "rewards/format_reward": 0.0,
      "step": 118
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 598.5833435058594,
      "epoch": 1.2773333333333334,
      "grad_norm": 0.19487521052360535,
      "kl": 0.270263671875,
      "learning_rate": 2.127776474921661e-06,
      "loss": 0.057,
      "reward": 0.3125,
      "reward_std": 0.20539596676826477,
      "rewards/accuracy_reward": 0.3125,
      "rewards/format_reward": 0.0,
      "step": 119
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 691.0000228881836,
      "epoch": 1.288,
      "grad_norm": 0.22089028358459473,
      "kl": 0.263824462890625,
      "learning_rate": 2.1106766221952607e-06,
      "loss": 0.0792,
      "reward": 0.2291666716337204,
      "reward_std": 0.299626849591732,
      "rewards/accuracy_reward": 0.2291666716337204,
      "rewards/format_reward": 0.0,
      "step": 120
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 560.0000133514404,
      "epoch": 1.2986666666666666,
      "grad_norm": 0.29344892501831055,
      "kl": 0.244903564453125,
      "learning_rate": 2.0934811034001427e-06,
      "loss": 0.0699,
      "reward": 0.3958333395421505,
      "reward_std": 0.46359488368034363,
      "rewards/accuracy_reward": 0.3958333395421505,
      "rewards/format_reward": 0.0,
      "step": 121
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 611.0833358764648,
      "epoch": 1.3093333333333335,
      "grad_norm": 0.22854970395565033,
      "kl": 0.3225555419921875,
      "learning_rate": 2.0761926123150166e-06,
      "loss": -0.0103,
      "reward": 0.18750000186264515,
      "reward_std": 0.2525114081799984,
      "rewards/accuracy_reward": 0.18750000186264515,
      "rewards/format_reward": 0.0,
      "step": 122
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 706.0833473205566,
      "epoch": 1.32,
      "grad_norm": 0.3734970986843109,
      "kl": 0.319671630859375,
      "learning_rate": 2.058813857283244e-06,
      "loss": 0.1124,
      "reward": 0.22916666977107525,
      "reward_std": 0.2525114119052887,
      "rewards/accuracy_reward": 0.22916666977107525,
      "rewards/format_reward": 0.0,
      "step": 123
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 622.3541831970215,
      "epoch": 1.3306666666666667,
      "grad_norm": 0.5326780676841736,
      "kl": 0.245574951171875,
      "learning_rate": 2.041347560788562e-06,
      "loss": 0.0965,
      "reward": 0.2500000037252903,
      "reward_std": 0.3680921718478203,
      "rewards/accuracy_reward": 0.2500000037252903,
      "rewards/format_reward": 0.0,
      "step": 124
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 674.9583530426025,
      "epoch": 1.3413333333333333,
      "grad_norm": 0.5754895806312561,
      "kl": 0.567291259765625,
      "learning_rate": 2.0237964590285924e-06,
      "loss": 0.037,
      "reward": 0.25000000558793545,
      "reward_std": 0.10206207260489464,
      "rewards/accuracy_reward": 0.25000000558793545,
      "rewards/format_reward": 0.0,
      "step": 125
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 567.7500152587891,
      "epoch": 1.3519999999999999,
      "grad_norm": 1.544613242149353,
      "kl": 0.3824310302734375,
      "learning_rate": 2.0061633014861975e-06,
      "loss": 0.0941,
      "reward": 0.2500000074505806,
      "reward_std": 0.3332235962152481,
      "rewards/accuracy_reward": 0.2500000074505806,
      "rewards/format_reward": 0.0,
      "step": 126
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 600.2083549499512,
      "epoch": 1.3626666666666667,
      "grad_norm": 1.894257664680481,
      "kl": 0.47564697265625,
      "learning_rate": 1.988450850498758e-06,
      "loss": 0.1021,
      "reward": 0.43750001303851604,
      "reward_std": 0.21764283627271652,
      "rewards/accuracy_reward": 0.43750001303851604,
      "rewards/format_reward": 0.0,
      "step": 127
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 746.2708587646484,
      "epoch": 1.3733333333333333,
      "grad_norm": 0.8493513464927673,
      "kl": 0.794921875,
      "learning_rate": 1.970661880825437e-06,
      "loss": 0.0542,
      "reward": 0.12500000186264515,
      "reward_std": 0.18404608592391014,
      "rewards/accuracy_reward": 0.12500000186264515,
      "rewards/format_reward": 0.0,
      "step": 128
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 564.4375190734863,
      "epoch": 1.384,
      "grad_norm": 2.1978530883789062,
      "kl": 0.4825286865234375,
      "learning_rate": 1.952799179212498e-06,
      "loss": 0.123,
      "reward": 0.37500000931322575,
      "reward_std": 0.3776952736079693,
      "rewards/accuracy_reward": 0.37500000931322575,
      "rewards/format_reward": 0.0,
      "step": 129
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 487.1041793823242,
      "epoch": 1.3946666666666667,
      "grad_norm": 2.844587802886963,
      "kl": 0.571441650390625,
      "learning_rate": 1.934865543956745e-06,
      "loss": 0.1218,
      "reward": 0.43750001303851604,
      "reward_std": 0.36417658254504204,
      "rewards/accuracy_reward": 0.43750001303851604,
      "rewards/format_reward": 0.0,
      "step": 130
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 646.0000152587891,
      "epoch": 1.4053333333333333,
      "grad_norm": 2.2201156616210938,
      "kl": 0.89276123046875,
      "learning_rate": 1.916863784467152e-06,
      "loss": 0.1161,
      "reward": 0.20833334140479565,
      "reward_std": 0.3061862140893936,
      "rewards/accuracy_reward": 0.20833334140479565,
      "rewards/format_reward": 0.0,
      "step": 131
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 684.2083473205566,
      "epoch": 1.416,
      "grad_norm": 2.4630818367004395,
      "kl": 1.4970703125,
      "learning_rate": 1.8987967208247542e-06,
      "loss": 0.1526,
      "reward": 0.2500000074505806,
      "reward_std": 0.2957112602889538,
      "rewards/accuracy_reward": 0.2500000074505806,
      "rewards/format_reward": 0.0,
      "step": 132
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 452.8541793823242,
      "epoch": 1.4266666666666667,
      "grad_norm": 1.1379987001419067,
      "kl": 0.559539794921875,
      "learning_rate": 1.8806671833408633e-06,
      "loss": 0.0492,
      "reward": 0.4583333395421505,
      "reward_std": 0.39512956142425537,
      "rewards/accuracy_reward": 0.4583333395421505,
      "rewards/format_reward": 0.0,
      "step": 133
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 531.5000190734863,
      "epoch": 1.4373333333333334,
      "grad_norm": 2.348485231399536,
      "kl": 1.101715087890625,
      "learning_rate": 1.8624780121136837e-06,
      "loss": 0.1375,
      "reward": 0.27083334140479565,
      "reward_std": 0.37465154007077217,
      "rewards/accuracy_reward": 0.27083334140479565,
      "rewards/format_reward": 0.0,
      "step": 134
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 645.3750114440918,
      "epoch": 1.448,
      "grad_norm": 3.0426082611083984,
      "kl": 1.2880859375,
      "learning_rate": 1.8442320565833972e-06,
      "loss": 0.1097,
      "reward": 0.1666666716337204,
      "reward_std": 0.19364918768405914,
      "rewards/accuracy_reward": 0.1666666716337204,
      "rewards/format_reward": 0.0,
      "step": 135
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 598.8333473205566,
      "epoch": 1.4586666666666668,
      "grad_norm": 1.8209415674209595,
      "kl": 0.7379150390625,
      "learning_rate": 1.8259321750857772e-06,
      "loss": 0.1796,
      "reward": 0.3333333395421505,
      "reward_std": 0.3680921755731106,
      "rewards/accuracy_reward": 0.3333333395421505,
      "rewards/format_reward": 0.0,
      "step": 136
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 603.8125190734863,
      "epoch": 1.4693333333333334,
      "grad_norm": 1.2022289037704468,
      "kl": 0.38909912109375,
      "learning_rate": 1.8075812344044163e-06,
      "loss": 0.0532,
      "reward": 0.3333333432674408,
      "reward_std": 0.31314554437994957,
      "rewards/accuracy_reward": 0.3333333432674408,
      "rewards/format_reward": 0.0,
      "step": 137
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 565.7500152587891,
      "epoch": 1.48,
      "grad_norm": 3.5259082317352295,
      "kl": 0.617431640625,
      "learning_rate": 1.7891821093216275e-06,
      "loss": 0.13,
      "reward": 0.3958333432674408,
      "reward_std": 0.39121396839618683,
      "rewards/accuracy_reward": 0.3958333432674408,
      "rewards/format_reward": 0.0,
      "step": 138
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 626.6250114440918,
      "epoch": 1.4906666666666666,
      "grad_norm": 2.2854840755462646,
      "kl": 0.4901123046875,
      "learning_rate": 1.770737682168091e-06,
      "loss": 0.0564,
      "reward": 0.29166667349636555,
      "reward_std": 0.31314554437994957,
      "rewards/accuracy_reward": 0.29166667349636555,
      "rewards/format_reward": 0.0,
      "step": 139
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 400.1250057220459,
      "epoch": 1.5013333333333332,
      "grad_norm": 3.279991388320923,
      "kl": 0.47369384765625,
      "learning_rate": 1.7522508423713243e-06,
      "loss": 0.1032,
      "reward": 0.3750000111758709,
      "reward_std": 0.3776952736079693,
      "rewards/accuracy_reward": 0.3750000111758709,
      "rewards/format_reward": 0.0,
      "step": 140
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 652.3958549499512,
      "epoch": 1.512,
      "grad_norm": 2.080812692642212,
      "kl": 0.7769775390625,
      "learning_rate": 1.7337244860030312e-06,
      "loss": 0.0917,
      "reward": 0.2291666716337204,
      "reward_std": 0.299626849591732,
      "rewards/accuracy_reward": 0.2291666716337204,
      "rewards/format_reward": 0.0,
      "step": 141
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 576.8958511352539,
      "epoch": 1.5226666666666666,
      "grad_norm": 3.379082441329956,
      "kl": 1.463653564453125,
      "learning_rate": 1.7151615153254192e-06,
      "loss": 0.165,
      "reward": 0.2916666679084301,
      "reward_std": 0.37592336162924767,
      "rewards/accuracy_reward": 0.2916666679084301,
      "rewards/format_reward": 0.0,
      "step": 142
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 458.1666831970215,
      "epoch": 1.5333333333333332,
      "grad_norm": 2.190816640853882,
      "kl": 0.494293212890625,
      "learning_rate": 1.6965648383365416e-06,
      "loss": -0.0006,
      "reward": 0.41666668094694614,
      "reward_std": 0.3506578877568245,
      "rewards/accuracy_reward": 0.41666668094694614,
      "rewards/format_reward": 0.0,
      "step": 143
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 517.3125152587891,
      "epoch": 1.544,
      "grad_norm": 1.580851674079895,
      "kl": 0.733123779296875,
      "learning_rate": 1.6779373683147423e-06,
      "loss": -0.0037,
      "reward": 0.43750000186264515,
      "reward_std": 0.2525114081799984,
      "rewards/accuracy_reward": 0.43750000186264515,
      "rewards/format_reward": 0.0,
      "step": 144
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 612.4375114440918,
      "epoch": 1.5546666666666666,
      "grad_norm": 3.22583270072937,
      "kl": 0.9237518310546875,
      "learning_rate": 1.6592820233622743e-06,
      "loss": 0.1054,
      "reward": 0.20833334140479565,
      "reward_std": 0.23116152733564377,
      "rewards/accuracy_reward": 0.20833334140479565,
      "rewards/format_reward": 0.0,
      "step": 145
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 553.2291870117188,
      "epoch": 1.5653333333333332,
      "grad_norm": 5.700227737426758,
      "kl": 1.231536865234375,
      "learning_rate": 1.6406017259481624e-06,
      "loss": 0.1547,
      "reward": 0.33333333767950535,
      "reward_std": 0.38552645966410637,
      "rewards/accuracy_reward": 0.33333333767950535,
      "rewards/format_reward": 0.0,
      "step": 146
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 585.9583511352539,
      "epoch": 1.576,
      "grad_norm": 2.94277024269104,
      "kl": 0.98828125,
      "learning_rate": 1.621899402450379e-06,
      "loss": 0.0967,
      "reward": 0.4375000111758709,
      "reward_std": 0.28219256922602654,
      "rewards/accuracy_reward": 0.4375000111758709,
      "rewards/format_reward": 0.0,
      "step": 147
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 504.02084732055664,
      "epoch": 1.5866666666666667,
      "grad_norm": 1.7243810892105103,
      "kl": 1.04620361328125,
      "learning_rate": 1.6031779826974138e-06,
      "loss": 0.1353,
      "reward": 0.5000000093132257,
      "reward_std": 0.2686738707125187,
      "rewards/accuracy_reward": 0.5000000093132257,
      "rewards/format_reward": 0.0,
      "step": 148
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 602.3333568572998,
      "epoch": 1.5973333333333333,
      "grad_norm": 17.420921325683594,
      "kl": 1.4764404296875,
      "learning_rate": 1.5844403995092944e-06,
      "loss": 0.1324,
      "reward": 0.18750000186264515,
      "reward_std": 0.2525114081799984,
      "rewards/accuracy_reward": 0.18750000186264515,
      "rewards/format_reward": 0.0,
      "step": 149
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 536.5208473205566,
      "epoch": 1.608,
      "grad_norm": 20.595264434814453,
      "kl": 2.333740234375,
      "learning_rate": 1.5656895882381465e-06,
      "loss": 0.1055,
      "reward": 0.31250001303851604,
      "reward_std": 0.21764283627271652,
      "rewards/accuracy_reward": 0.31250001303851604,
      "rewards/format_reward": 0.0,
      "step": 150
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 556.3125190734863,
      "epoch": 1.6186666666666667,
      "grad_norm": 4.533661842346191,
      "kl": 1.85205078125,
      "learning_rate": 1.5469284863083523e-06,
      "loss": 0.1692,
      "reward": 0.3958333469927311,
      "reward_std": 0.40168891847133636,
      "rewards/accuracy_reward": 0.3958333469927311,
      "rewards/format_reward": 0.0,
      "step": 151
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 456.81251335144043,
      "epoch": 1.6293333333333333,
      "grad_norm": 17.720441818237305,
      "kl": 1.3587646484375,
      "learning_rate": 1.5281600327563853e-06,
      "loss": 0.1272,
      "reward": 0.3333333395421505,
      "reward_std": 0.23116152733564377,
      "rewards/accuracy_reward": 0.3333333395421505,
      "rewards/format_reward": 0.0,
      "step": 152
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 626.6666831970215,
      "epoch": 1.6400000000000001,
      "grad_norm": 4.506388187408447,
      "kl": 0.3983154296875,
      "learning_rate": 1.5093871677703947e-06,
      "loss": 0.0347,
      "reward": 0.2083333358168602,
      "reward_std": 0.30354244634509087,
      "rewards/accuracy_reward": 0.2083333358168602,
      "rewards/format_reward": 0.0,
      "step": 153
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 639.0208435058594,
      "epoch": 1.6506666666666665,
      "grad_norm": 3.304605484008789,
      "kl": 1.2666015625,
      "learning_rate": 1.4906128322296056e-06,
      "loss": 0.0427,
      "reward": 0.22916667349636555,
      "reward_std": 0.27258946001529694,
      "rewards/accuracy_reward": 0.22916667349636555,
      "rewards/format_reward": 0.0,
      "step": 154
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 661.4166870117188,
      "epoch": 1.6613333333333333,
      "grad_norm": 8.943105697631836,
      "kl": 0.469970703125,
      "learning_rate": 1.471839967243615e-06,
      "loss": 0.0405,
      "reward": 0.2500000074505806,
      "reward_std": 0.19364918768405914,
      "rewards/accuracy_reward": 0.2500000074505806,
      "rewards/format_reward": 0.0,
      "step": 155
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 711.1666870117188,
      "epoch": 1.6720000000000002,
      "grad_norm": 4.7917866706848145,
      "kl": 0.383453369140625,
      "learning_rate": 1.453071513691648e-06,
      "loss": 0.0355,
      "reward": 0.18750000558793545,
      "reward_std": 0.1705273948609829,
      "rewards/accuracy_reward": 0.18750000558793545,
      "rewards/format_reward": 0.0,
      "step": 156
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 583.0833473205566,
      "epoch": 1.6826666666666665,
      "grad_norm": 11.259808540344238,
      "kl": 0.30938720703125,
      "learning_rate": 1.4343104117618536e-06,
      "loss": 0.0203,
      "reward": 0.29166667722165585,
      "reward_std": 0.2957112565636635,
      "rewards/accuracy_reward": 0.29166667722165585,
      "rewards/format_reward": 0.0,
      "step": 157
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 487.87500762939453,
      "epoch": 1.6933333333333334,
      "grad_norm": 3.858649253845215,
      "kl": 0.7235565185546875,
      "learning_rate": 1.415559600490706e-06,
      "loss": 0.0899,
      "reward": 0.5000000093132257,
      "reward_std": 0.23116153106093407,
      "rewards/accuracy_reward": 0.5000000093132257,
      "rewards/format_reward": 0.0,
      "step": 158
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 580.7916793823242,
      "epoch": 1.704,
      "grad_norm": 19.872974395751953,
      "kl": 0.8516845703125,
      "learning_rate": 1.3968220173025865e-06,
      "loss": 0.1322,
      "reward": 0.416666679084301,
      "reward_std": 0.4152076132595539,
      "rewards/accuracy_reward": 0.416666679084301,
      "rewards/format_reward": 0.0,
      "step": 159
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 588.3125152587891,
      "epoch": 1.7146666666666666,
      "grad_norm": 0.839247465133667,
      "kl": 0.3216552734375,
      "learning_rate": 1.3781005975496212e-06,
      "loss": 0.0073,
      "reward": 0.16666666977107525,
      "reward_std": 0.2861081659793854,
      "rewards/accuracy_reward": 0.16666666977107525,
      "rewards/format_reward": 0.0,
      "step": 160
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 518.6875114440918,
      "epoch": 1.7253333333333334,
      "grad_norm": 2.5837020874023438,
      "kl": 1.117645263671875,
      "learning_rate": 1.359398274051838e-06,
      "loss": 0.09,
      "reward": 0.2916666753590107,
      "reward_std": 0.2957112640142441,
      "rewards/accuracy_reward": 0.2916666753590107,
      "rewards/format_reward": 0.0,
      "step": 161
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 623.3750190734863,
      "epoch": 1.736,
      "grad_norm": 1.8010332584381104,
      "kl": 1.14996337890625,
      "learning_rate": 1.340717976637726e-06,
      "loss": 0.0947,
      "reward": 0.25000000186264515,
      "reward_std": 0.32097673043608665,
      "rewards/accuracy_reward": 0.25000000186264515,
      "rewards/format_reward": 0.0,
      "step": 162
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 545.6875076293945,
      "epoch": 1.7466666666666666,
      "grad_norm": 0.9145823121070862,
      "kl": 0.93804931640625,
      "learning_rate": 1.3220626316852584e-06,
      "loss": 0.0202,
      "reward": 0.33333334140479565,
      "reward_std": 0.16661179810762405,
      "rewards/accuracy_reward": 0.33333334140479565,
      "rewards/format_reward": 0.0,
      "step": 163
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 639.3750114440918,
      "epoch": 1.7573333333333334,
      "grad_norm": 1.514998435974121,
      "kl": 0.569366455078125,
      "learning_rate": 1.303435161663459e-06,
      "loss": 0.0524,
      "reward": 0.16666666977107525,
      "reward_std": 0.24859581515192986,
      "rewards/accuracy_reward": 0.16666666977107525,
      "rewards/format_reward": 0.0,
      "step": 164
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 558.8750152587891,
      "epoch": 1.768,
      "grad_norm": 9.262928009033203,
      "kl": 0.46624755859375,
      "learning_rate": 1.2848384846745813e-06,
      "loss": 0.105,
      "reward": 0.3958333395421505,
      "reward_std": 0.2900237552821636,
      "rewards/accuracy_reward": 0.3958333395421505,
      "rewards/format_reward": 0.0,
      "step": 165
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 718.2291793823242,
      "epoch": 1.7786666666666666,
      "grad_norm": 4.827789306640625,
      "kl": 0.4215087890625,
      "learning_rate": 1.2662755139969693e-06,
      "loss": 0.0837,
      "reward": 0.25000000558793545,
      "reward_std": 0.3506578840315342,
      "rewards/accuracy_reward": 0.25000000558793545,
      "rewards/format_reward": 0.0,
      "step": 166
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 629.2500152587891,
      "epoch": 1.7893333333333334,
      "grad_norm": 2.5728464126586914,
      "kl": 0.748077392578125,
      "learning_rate": 1.2477491576286764e-06,
      "loss": 0.0796,
      "reward": 0.3333333395421505,
      "reward_std": 0.22155842557549477,
      "rewards/accuracy_reward": 0.3333333395421505,
      "rewards/format_reward": 0.0,
      "step": 167
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 592.3541831970215,
      "epoch": 1.8,
      "grad_norm": 11.94718074798584,
      "kl": 0.812744140625,
      "learning_rate": 1.2292623178319094e-06,
      "loss": 0.0748,
      "reward": 0.27083333767950535,
      "reward_std": 0.3170611411333084,
      "rewards/accuracy_reward": 0.27083333767950535,
      "rewards/format_reward": 0.0,
      "step": 168
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 644.7500114440918,
      "epoch": 1.8106666666666666,
      "grad_norm": 0.9922601580619812,
      "kl": 1.2894439697265625,
      "learning_rate": 1.2108178906783732e-06,
      "loss": 0.0885,
      "reward": 0.25000000186264515,
      "reward_std": 0.32097673043608665,
      "rewards/accuracy_reward": 0.25000000186264515,
      "rewards/format_reward": 0.0,
      "step": 169
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 602.8125171661377,
      "epoch": 1.8213333333333335,
      "grad_norm": 5.8193278312683105,
      "kl": 2.0225830078125,
      "learning_rate": 1.1924187655955838e-06,
      "loss": 0.186,
      "reward": 0.291666679084301,
      "reward_std": 0.23116153106093407,
      "rewards/accuracy_reward": 0.291666679084301,
      "rewards/format_reward": 0.0,
      "step": 170
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 675.3125190734863,
      "epoch": 1.8319999999999999,
      "grad_norm": 4.4365949630737305,
      "kl": 2.0120849609375,
      "learning_rate": 1.1740678249142229e-06,
      "loss": 0.1596,
      "reward": 0.2500000074505806,
      "reward_std": 0.24859581515192986,
      "rewards/accuracy_reward": 0.2500000074505806,
      "rewards/format_reward": 0.0,
      "step": 171
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 549.979175567627,
      "epoch": 1.8426666666666667,
      "grad_norm": 11.282766342163086,
      "kl": 1.157684326171875,
      "learning_rate": 1.1557679434166027e-06,
      "loss": 0.1544,
      "reward": 0.4166666828095913,
      "reward_std": 0.2861081622540951,
      "rewards/accuracy_reward": 0.4166666828095913,
      "rewards/format_reward": 0.0,
      "step": 172
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 417.0416831970215,
      "epoch": 1.8533333333333335,
      "grad_norm": 3.242908000946045,
      "kl": 1.05181884765625,
      "learning_rate": 1.1375219878863162e-06,
      "loss": 0.13,
      "reward": 0.6250000093132257,
      "reward_std": 0.23116153106093407,
      "rewards/accuracy_reward": 0.6250000093132257,
      "rewards/format_reward": 0.0,
      "step": 173
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 628.7083473205566,
      "epoch": 1.8639999999999999,
      "grad_norm": 25.157915115356445,
      "kl": 4.201385498046875,
      "learning_rate": 1.1193328166591372e-06,
      "loss": 0.2222,
      "reward": 0.1666666716337204,
      "reward_std": 0.2957112565636635,
      "rewards/accuracy_reward": 0.1666666716337204,
      "rewards/format_reward": 0.0,
      "step": 174
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 685.0416793823242,
      "epoch": 1.8746666666666667,
      "grad_norm": 8.827438354492188,
      "kl": 2.48046875,
      "learning_rate": 1.1012032791752457e-06,
      "loss": 0.1483,
      "reward": 0.2291666716337204,
      "reward_std": 0.33713919296860695,
      "rewards/accuracy_reward": 0.2291666716337204,
      "rewards/format_reward": 0.0,
      "step": 175
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 575.5416793823242,
      "epoch": 1.8853333333333333,
      "grad_norm": 7.852333068847656,
      "kl": 2.4658203125,
      "learning_rate": 1.0831362155328477e-06,
      "loss": 0.1938,
      "reward": 0.41666667722165585,
      "reward_std": 0.43264190107584,
      "rewards/accuracy_reward": 0.41666667722165585,
      "rewards/format_reward": 0.0,
      "step": 176
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 657.1458473205566,
      "epoch": 1.896,
      "grad_norm": 3.781799793243408,
      "kl": 1.44403076171875,
      "learning_rate": 1.0651344560432548e-06,
      "loss": 0.0971,
      "reward": 0.2500000074505806,
      "reward_std": 0.32274864614009857,
      "rewards/accuracy_reward": 0.2500000074505806,
      "rewards/format_reward": 0.0,
      "step": 177
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 507.4375171661377,
      "epoch": 1.9066666666666667,
      "grad_norm": 15.98084545135498,
      "kl": 1.1881103515625,
      "learning_rate": 1.0472008207875023e-06,
      "loss": 0.158,
      "reward": 0.4583333469927311,
      "reward_std": 0.3776952773332596,
      "rewards/accuracy_reward": 0.4583333469927311,
      "rewards/format_reward": 0.0,
      "step": 178
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 656.3333549499512,
      "epoch": 1.9173333333333333,
      "grad_norm": 148.17950439453125,
      "kl": 6.6319580078125,
      "learning_rate": 1.0293381191745637e-06,
      "loss": 0.4726,
      "reward": 0.31250000186264515,
      "reward_std": 0.2525114081799984,
      "rewards/accuracy_reward": 0.31250000186264515,
      "rewards/format_reward": 0.0,
      "step": 179
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 678.5416774749756,
      "epoch": 1.928,
      "grad_norm": 1.2220107316970825,
      "kl": 1.0302886962890625,
      "learning_rate": 1.0115491495012424e-06,
      "loss": 0.1113,
      "reward": 0.3750000037252903,
      "reward_std": 0.19364918768405914,
      "rewards/accuracy_reward": 0.3750000037252903,
      "rewards/format_reward": 0.0,
      "step": 180
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 572.2083435058594,
      "epoch": 1.9386666666666668,
      "grad_norm": 8.053202629089355,
      "kl": 0.740478515625,
      "learning_rate": 9.938366985138028e-07,
      "loss": 0.149,
      "reward": 0.3125000074505806,
      "reward_std": 0.2525114081799984,
      "rewards/accuracy_reward": 0.3125000074505806,
      "rewards/format_reward": 0.0,
      "step": 181
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 535.9583511352539,
      "epoch": 1.9493333333333334,
      "grad_norm": 2.4490530490875244,
      "kl": 0.72894287109375,
      "learning_rate": 9.762035409714077e-07,
      "loss": 0.0826,
      "reward": 0.4166666828095913,
      "reward_std": 0.3977733328938484,
      "rewards/accuracy_reward": 0.4166666828095913,
      "rewards/format_reward": 0.0,
      "step": 182
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 535.4166851043701,
      "epoch": 1.96,
      "grad_norm": 0.8531109690666199,
      "kl": 1.231597900390625,
      "learning_rate": 9.586524392114385e-07,
      "loss": 0.1191,
      "reward": 0.3750000037252903,
      "reward_std": 0.23116153106093407,
      "rewards/accuracy_reward": 0.3750000037252903,
      "rewards/format_reward": 0.0,
      "step": 183
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 506.0208435058594,
      "epoch": 1.9706666666666668,
      "grad_norm": 7.356703281402588,
      "kl": 0.62359619140625,
      "learning_rate": 9.411861427167567e-07,
      "loss": 0.0718,
      "reward": 0.47916668094694614,
      "reward_std": 0.28219256177544594,
      "rewards/accuracy_reward": 0.47916668094694614,
      "rewards/format_reward": 0.0,
      "step": 184
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 636.7500152587891,
      "epoch": 1.9813333333333332,
      "grad_norm": 3.2374136447906494,
      "kl": 0.815521240234375,
      "learning_rate": 9.238073876849837e-07,
      "loss": 0.0789,
      "reward": 0.25000000186264515,
      "reward_std": 0.18404608592391014,
      "rewards/accuracy_reward": 0.25000000186264515,
      "rewards/format_reward": 0.0,
      "step": 185
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 617.7083435058594,
      "epoch": 1.992,
      "grad_norm": 1.628970742225647,
      "kl": 0.70062255859375,
      "learning_rate": 9.065188965998574e-07,
      "loss": 0.1044,
      "reward": 0.25000000558793545,
      "reward_std": 0.31314554810523987,
      "rewards/accuracy_reward": 0.25000000558793545,
      "rewards/format_reward": 0.0,
      "step": 186
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 629.5416831970215,
      "epoch": 2.010666666666667,
      "grad_norm": 8.64794635772705,
      "kl": 0.5065460205078125,
      "learning_rate": 8.893233778047394e-07,
      "loss": 0.1611,
      "reward": 0.3750000111758709,
      "reward_std": 0.31314554810523987,
      "rewards/accuracy_reward": 0.3750000111758709,
      "rewards/format_reward": 0.0,
      "step": 187
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 566.5625114440918,
      "epoch": 2.021333333333333,
      "grad_norm": 7.554544448852539,
      "kl": 0.850250244140625,
      "learning_rate": 8.722235250783389e-07,
      "loss": 0.1748,
      "reward": 0.4166666828095913,
      "reward_std": 0.3332235999405384,
      "rewards/accuracy_reward": 0.4166666828095913,
      "rewards/format_reward": 0.0,
      "step": 188
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 506.54167556762695,
      "epoch": 2.032,
      "grad_norm": 3.377150774002075,
      "kl": 1.17388916015625,
      "learning_rate": 8.552220172127148e-07,
      "loss": 0.1613,
      "reward": 0.5416666846722364,
      "reward_std": 0.4893604479730129,
      "rewards/accuracy_reward": 0.5416666846722364,
      "rewards/format_reward": 0.0,
      "step": 189
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 669.0625190734863,
      "epoch": 2.042666666666667,
      "grad_norm": 3.2369844913482666,
      "kl": 1.787841796875,
      "learning_rate": 8.383215175936263e-07,
      "loss": 0.2299,
      "reward": 0.2916666753590107,
      "reward_std": 0.2957112602889538,
      "rewards/accuracy_reward": 0.2916666753590107,
      "rewards/format_reward": 0.0,
      "step": 190
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 589.0833568572998,
      "epoch": 2.0533333333333332,
      "grad_norm": 1.6646206378936768,
      "kl": 1.053955078125,
      "learning_rate": 8.215246737833017e-07,
      "loss": 0.152,
      "reward": 0.2708333395421505,
      "reward_std": 0.2996268533170223,
      "rewards/accuracy_reward": 0.2708333395421505,
      "rewards/format_reward": 0.0,
      "step": 191
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 538.7291831970215,
      "epoch": 2.064,
      "grad_norm": 4.677793979644775,
      "kl": 1.664642333984375,
      "learning_rate": 8.048341171056786e-07,
      "loss": 0.1378,
      "reward": 0.37500000558793545,
      "reward_std": 0.24859581515192986,
      "rewards/accuracy_reward": 0.37500000558793545,
      "rewards/format_reward": 0.0,
      "step": 192
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 464.04168128967285,
      "epoch": 2.074666666666667,
      "grad_norm": 2.5755910873413086,
      "kl": 1.07476806640625,
      "learning_rate": 7.882524622341923e-07,
      "loss": 0.1292,
      "reward": 0.4583333507180214,
      "reward_std": 0.2861081548035145,
      "rewards/accuracy_reward": 0.4583333507180214,
      "rewards/format_reward": 0.0,
      "step": 193
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 673.6875152587891,
      "epoch": 2.0853333333333333,
      "grad_norm": 1.9688490629196167,
      "kl": 0.8976593017578125,
      "learning_rate": 7.717823067821738e-07,
      "loss": 0.1349,
      "reward": 0.43750001303851604,
      "reward_std": 0.21764283999800682,
      "rewards/accuracy_reward": 0.43750001303851604,
      "rewards/format_reward": 0.0,
      "step": 194
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 621.8958587646484,
      "epoch": 2.096,
      "grad_norm": 3.078874349594116,
      "kl": 0.706787109375,
      "learning_rate": 7.554262308959165e-07,
      "loss": 0.0235,
      "reward": 0.1875000037252903,
      "reward_std": 0.2996268533170223,
      "rewards/accuracy_reward": 0.1875000037252903,
      "rewards/format_reward": 0.0,
      "step": 195
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 702.0416946411133,
      "epoch": 2.1066666666666665,
      "grad_norm": 2.9152376651763916,
      "kl": 0.9154052734375,
      "learning_rate": 7.39186796850482e-07,
      "loss": 0.1356,
      "reward": 0.2916666679084301,
      "reward_std": 0.33841101825237274,
      "rewards/accuracy_reward": 0.2916666679084301,
      "rewards/format_reward": 0.0,
      "step": 196
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 484.37500953674316,
      "epoch": 2.1173333333333333,
      "grad_norm": 4.060208320617676,
      "kl": 0.632110595703125,
      "learning_rate": 7.23066548648304e-07,
      "loss": 0.0204,
      "reward": 0.3333333358168602,
      "reward_std": 0.12909945845603943,
      "rewards/accuracy_reward": 0.3333333358168602,
      "rewards/format_reward": 0.0,
      "step": 197
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 789.3125190734863,
      "epoch": 2.128,
      "grad_norm": 3.2686266899108887,
      "kl": 0.858154296875,
      "learning_rate": 7.070680116206564e-07,
      "loss": 0.1265,
      "reward": 0.2708333395421505,
      "reward_std": 0.2525114081799984,
      "rewards/accuracy_reward": 0.2708333395421505,
      "rewards/format_reward": 0.0,
      "step": 198
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 626.7708473205566,
      "epoch": 2.1386666666666665,
      "grad_norm": 4.074109077453613,
      "kl": 0.810546875,
      "learning_rate": 6.911936920320457e-07,
      "loss": 0.1303,
      "reward": 0.35416667349636555,
      "reward_std": 0.3816108666360378,
      "rewards/accuracy_reward": 0.35416667349636555,
      "rewards/format_reward": 0.0,
      "step": 199
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 557.5833415985107,
      "epoch": 2.1493333333333333,
      "grad_norm": 0.744455099105835,
      "kl": 0.35540771484375,
      "learning_rate": 6.754460766875871e-07,
      "loss": 0.0411,
      "reward": 0.37500000931322575,
      "reward_std": 0.2686738707125187,
      "rewards/accuracy_reward": 0.37500000931322575,
      "rewards/format_reward": 0.0,
      "step": 200
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 470.77084732055664,
      "epoch": 2.16,
      "grad_norm": 4.264385223388672,
      "kl": 0.3687896728515625,
      "learning_rate": 6.598276325434357e-07,
      "loss": 0.1808,
      "reward": 0.5833333395421505,
      "reward_std": 0.33057983219623566,
      "rewards/accuracy_reward": 0.5833333395421505,
      "rewards/format_reward": 0.0,
      "step": 201
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 569.1875076293945,
      "epoch": 2.1706666666666665,
      "grad_norm": 1.42915678024292,
      "kl": 0.556396484375,
      "learning_rate": 6.443408063203226e-07,
      "loss": 0.0941,
      "reward": 0.3541666753590107,
      "reward_std": 0.41912321001291275,
      "rewards/accuracy_reward": 0.3541666753590107,
      "rewards/format_reward": 0.0,
      "step": 202
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 511.9791793823242,
      "epoch": 2.1813333333333333,
      "grad_norm": 2.0874643325805664,
      "kl": 0.400665283203125,
      "learning_rate": 6.289880241202581e-07,
      "loss": 0.0436,
      "reward": 0.3125000074505806,
      "reward_std": 0.3170611411333084,
      "rewards/accuracy_reward": 0.3125000074505806,
      "rewards/format_reward": 0.0,
      "step": 203
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 510.20835876464844,
      "epoch": 2.192,
      "grad_norm": 1.6422734260559082,
      "kl": 0.51300048828125,
      "learning_rate": 6.137716910464735e-07,
      "loss": 0.136,
      "reward": 0.5416666772216558,
      "reward_std": 0.43264190107584,
      "rewards/accuracy_reward": 0.5416666772216558,
      "rewards/format_reward": 0.0,
      "step": 204
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 529.7083473205566,
      "epoch": 2.2026666666666666,
      "grad_norm": 1.4476531744003296,
      "kl": 0.6066131591796875,
      "learning_rate": 5.986941908266454e-07,
      "loss": 0.1223,
      "reward": 0.3958333432674408,
      "reward_std": 0.28219256550073624,
      "rewards/accuracy_reward": 0.3958333432674408,
      "rewards/format_reward": 0.0,
      "step": 205
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 576.1666793823242,
      "epoch": 2.2133333333333334,
      "grad_norm": 3.435512065887451,
      "kl": 0.8050079345703125,
      "learning_rate": 5.837578854394692e-07,
      "loss": 0.0088,
      "reward": 0.22916667349636555,
      "reward_std": 0.2446802258491516,
      "rewards/accuracy_reward": 0.22916667349636555,
      "rewards/format_reward": 0.0,
      "step": 206
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 540.6666793823242,
      "epoch": 2.224,
      "grad_norm": 3.2813377380371094,
      "kl": 0.63885498046875,
      "learning_rate": 5.689651147446448e-07,
      "loss": 0.0949,
      "reward": 0.27083334140479565,
      "reward_std": 0.34674229472875595,
      "rewards/accuracy_reward": 0.27083334140479565,
      "rewards/format_reward": 0.0,
      "step": 207
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 605.6250190734863,
      "epoch": 2.2346666666666666,
      "grad_norm": 8.23859977722168,
      "kl": 1.21954345703125,
      "learning_rate": 5.543181961163219e-07,
      "loss": 0.0795,
      "reward": 0.1666666716337204,
      "reward_std": 0.11949636042118073,
      "rewards/accuracy_reward": 0.1666666716337204,
      "rewards/format_reward": 0.0,
      "step": 208
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 451.6666793823242,
      "epoch": 2.2453333333333334,
      "grad_norm": 1.515325903892517,
      "kl": 0.95635986328125,
      "learning_rate": 5.398194240800684e-07,
      "loss": 0.1437,
      "reward": 0.5416666753590107,
      "reward_std": 0.423038799315691,
      "rewards/accuracy_reward": 0.5416666753590107,
      "rewards/format_reward": 0.0,
      "step": 209
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 598.8750152587891,
      "epoch": 2.2560000000000002,
      "grad_norm": 2.8135106563568115,
      "kl": 1.22515869140625,
      "learning_rate": 5.254710699534226e-07,
      "loss": 0.0269,
      "reward": 0.35416667722165585,
      "reward_std": 0.299626849591732,
      "rewards/accuracy_reward": 0.35416667722165585,
      "rewards/format_reward": 0.0,
      "step": 210
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 580.6041793823242,
      "epoch": 2.2666666666666666,
      "grad_norm": 1.5423381328582764,
      "kl": 0.624755859375,
      "learning_rate": 5.112753814900761e-07,
      "loss": 0.0605,
      "reward": 0.37500000931322575,
      "reward_std": 0.3776952773332596,
      "rewards/accuracy_reward": 0.37500000931322575,
      "rewards/format_reward": 0.0,
      "step": 211
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 551.0833396911621,
      "epoch": 2.2773333333333334,
      "grad_norm": 1.7441309690475464,
      "kl": 0.79541015625,
      "learning_rate": 4.97234582527751e-07,
      "loss": 0.0963,
      "reward": 0.33333333767950535,
      "reward_std": 0.38552645966410637,
      "rewards/accuracy_reward": 0.33333333767950535,
      "rewards/format_reward": 0.0,
      "step": 212
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 686.3750190734863,
      "epoch": 2.288,
      "grad_norm": 0.869959831237793,
      "kl": 0.630462646484375,
      "learning_rate": 4.8335087263982e-07,
      "loss": 0.1008,
      "reward": 0.3541666716337204,
      "reward_std": 0.299626849591732,
      "rewards/accuracy_reward": 0.3541666716337204,
      "rewards/format_reward": 0.0,
      "step": 213
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 743.2083511352539,
      "epoch": 2.2986666666666666,
      "grad_norm": 5.663978576660156,
      "kl": 0.819091796875,
      "learning_rate": 4.6962642679073465e-07,
      "loss": 0.0238,
      "reward": 0.12500000186264515,
      "reward_std": 0.18404608592391014,
      "rewards/accuracy_reward": 0.12500000186264515,
      "rewards/format_reward": 0.0,
      "step": 214
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 595.791675567627,
      "epoch": 2.3093333333333335,
      "grad_norm": 0.7046431303024292,
      "kl": 0.5963134765625,
      "learning_rate": 4.560633949953017e-07,
      "loss": 0.1039,
      "reward": 0.31250000931322575,
      "reward_std": 0.33713918924331665,
      "rewards/accuracy_reward": 0.31250000931322575,
      "rewards/format_reward": 0.0,
      "step": 215
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 514.9791793823242,
      "epoch": 2.32,
      "grad_norm": 3.9048852920532227,
      "kl": 0.498199462890625,
      "learning_rate": 4.426639019818704e-07,
      "loss": 0.0968,
      "reward": 0.37500001303851604,
      "reward_std": 0.2861081622540951,
      "rewards/accuracy_reward": 0.37500001303851604,
      "rewards/format_reward": 0.0,
      "step": 216
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 633.708345413208,
      "epoch": 2.3306666666666667,
      "grad_norm": 2.1279549598693848,
      "kl": 0.5631103515625,
      "learning_rate": 4.2943004685948286e-07,
      "loss": 0.1652,
      "reward": 0.2916666753590107,
      "reward_std": 0.3506578877568245,
      "rewards/accuracy_reward": 0.2916666753590107,
      "rewards/format_reward": 0.0,
      "step": 217
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 695.3125152587891,
      "epoch": 2.3413333333333335,
      "grad_norm": 1.3892327547073364,
      "kl": 0.384521484375,
      "learning_rate": 4.163639027890363e-07,
      "loss": 0.1241,
      "reward": 0.20833333395421505,
      "reward_std": 0.25642700120806694,
      "rewards/accuracy_reward": 0.20833333395421505,
      "rewards/format_reward": 0.0,
      "step": 218
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 664.2916831970215,
      "epoch": 2.352,
      "grad_norm": 2.659468650817871,
      "kl": 0.380615234375,
      "learning_rate": 4.034675166585079e-07,
      "loss": 0.0474,
      "reward": 0.1875000074505806,
      "reward_std": 0.11558076739311218,
      "rewards/accuracy_reward": 0.1875000074505806,
      "rewards/format_reward": 0.0,
      "step": 219
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 692.4791831970215,
      "epoch": 2.3626666666666667,
      "grad_norm": 1.3437598943710327,
      "kl": 0.438720703125,
      "learning_rate": 3.907429087623007e-07,
      "loss": 0.0757,
      "reward": 0.27083333767950535,
      "reward_std": 0.1801304928958416,
      "rewards/accuracy_reward": 0.27083333767950535,
      "rewards/format_reward": 0.0,
      "step": 220
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 468.7916831970215,
      "epoch": 2.3733333333333335,
      "grad_norm": 4.506848335266113,
      "kl": 0.224761962890625,
      "learning_rate": 3.7819207248475174e-07,
      "loss": 0.1226,
      "reward": 0.625000013038516,
      "reward_std": 0.2957112565636635,
      "rewards/accuracy_reward": 0.625000013038516,
      "rewards/format_reward": 0.0,
      "step": 221
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 620.5625228881836,
      "epoch": 2.384,
      "grad_norm": 8.756277084350586,
      "kl": 0.5113525390625,
      "learning_rate": 3.6581697398785615e-07,
      "loss": 0.1596,
      "reward": 0.29166667349636555,
      "reward_std": 0.30354244634509087,
      "rewards/accuracy_reward": 0.29166667349636555,
      "rewards/format_reward": 0.0,
      "step": 222
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 428.8125171661377,
      "epoch": 2.3946666666666667,
      "grad_norm": 3.07792067527771,
      "kl": 0.2974853515625,
      "learning_rate": 3.5361955190325563e-07,
      "loss": 0.0776,
      "reward": 0.4791666828095913,
      "reward_std": 0.36417657881975174,
      "rewards/accuracy_reward": 0.4791666828095913,
      "rewards/format_reward": 0.0,
      "step": 223
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 533.0000114440918,
      "epoch": 2.405333333333333,
      "grad_norm": 3.352797031402588,
      "kl": 0.482757568359375,
      "learning_rate": 3.4160171702854286e-07,
      "loss": 0.128,
      "reward": 0.5000000037252903,
      "reward_std": 0.30354244261980057,
      "rewards/accuracy_reward": 0.5000000037252903,
      "rewards/format_reward": 0.0,
      "step": 224
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 574.3125190734863,
      "epoch": 2.416,
      "grad_norm": 7.5084147453308105,
      "kl": 0.31280517578125,
      "learning_rate": 3.297653520279226e-07,
      "loss": 0.0974,
      "reward": 0.416666679084301,
      "reward_std": 0.4152076244354248,
      "rewards/accuracy_reward": 0.416666679084301,
      "rewards/format_reward": 0.0,
      "step": 225
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 553.7083435058594,
      "epoch": 2.4266666666666667,
      "grad_norm": 6.016626358032227,
      "kl": 0.4796142578125,
      "learning_rate": 3.1811231113727974e-07,
      "loss": 0.1104,
      "reward": 0.43750001676380634,
      "reward_std": 0.34674229100346565,
      "rewards/accuracy_reward": 0.43750001676380634,
      "rewards/format_reward": 0.0,
      "step": 226
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 495.6666793823242,
      "epoch": 2.437333333333333,
      "grad_norm": 0.3561098873615265,
      "kl": 0.18603515625,
      "learning_rate": 3.066444198737053e-07,
      "loss": 0.0663,
      "reward": 0.41666667722165585,
      "reward_std": 0.22155842930078506,
      "rewards/accuracy_reward": 0.41666667722165585,
      "rewards/format_reward": 0.0,
      "step": 227
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 497.8541793823242,
      "epoch": 2.448,
      "grad_norm": 1.9640473127365112,
      "kl": 0.3065185546875,
      "learning_rate": 2.9536347474951645e-07,
      "loss": 0.0849,
      "reward": 0.5833333563059568,
      "reward_std": 0.46232306212186813,
      "rewards/accuracy_reward": 0.5833333563059568,
      "rewards/format_reward": 0.0,
      "step": 228
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 683.7916717529297,
      "epoch": 2.458666666666667,
      "grad_norm": 1.3735461235046387,
      "kl": 0.5143280029296875,
      "learning_rate": 2.842712429908237e-07,
      "loss": 0.0995,
      "reward": 0.2291666716337204,
      "reward_std": 0.2996268533170223,
      "rewards/accuracy_reward": 0.2291666716337204,
      "rewards/format_reward": 0.0,
      "step": 229
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 600.8333511352539,
      "epoch": 2.469333333333333,
      "grad_norm": 0.5821128487586975,
      "kl": 0.4196929931640625,
      "learning_rate": 2.733694622606824e-07,
      "loss": 0.0433,
      "reward": 0.3125000074505806,
      "reward_std": 0.2525114081799984,
      "rewards/accuracy_reward": 0.3125000074505806,
      "rewards/format_reward": 0.0,
      "step": 230
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 549.1666870117188,
      "epoch": 2.48,
      "grad_norm": 3.470980167388916,
      "kl": 0.444915771484375,
      "learning_rate": 2.626598403868801e-07,
      "loss": 0.1265,
      "reward": 0.4166666828095913,
      "reward_std": 0.3332236111164093,
      "rewards/accuracy_reward": 0.4166666828095913,
      "rewards/format_reward": 0.0,
      "step": 231
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 424.87501525878906,
      "epoch": 2.490666666666667,
      "grad_norm": 3.7582547664642334,
      "kl": 0.19146728515625,
      "learning_rate": 2.521440550943943e-07,
      "loss": 0.0467,
      "reward": 0.6458333469927311,
      "reward_std": 0.43655749410390854,
      "rewards/accuracy_reward": 0.6458333469927311,
      "rewards/format_reward": 0.0,
      "step": 232
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 492.93751525878906,
      "epoch": 2.501333333333333,
      "grad_norm": 24.160253524780273,
      "kl": 2.023895263671875,
      "learning_rate": 2.418237537425666e-07,
      "loss": 0.1537,
      "reward": 0.6458333469927311,
      "reward_std": 0.309229951351881,
      "rewards/accuracy_reward": 0.6458333469927311,
      "rewards/format_reward": 0.0,
      "step": 233
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 601.8750228881836,
      "epoch": 2.512,
      "grad_norm": 5.170097827911377,
      "kl": 0.60491943359375,
      "learning_rate": 2.3170055306703512e-07,
      "loss": 0.1884,
      "reward": 0.4791666828095913,
      "reward_std": 0.4758417531847954,
      "rewards/accuracy_reward": 0.4791666828095913,
      "rewards/format_reward": 0.0,
      "step": 234
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 603.7291984558105,
      "epoch": 2.522666666666667,
      "grad_norm": 1.1363794803619385,
      "kl": 0.7254638671875,
      "learning_rate": 2.2177603892646277e-07,
      "loss": 0.0472,
      "reward": 0.4375000111758709,
      "reward_std": 0.2446802258491516,
      "rewards/accuracy_reward": 0.4375000111758709,
      "rewards/format_reward": 0.0,
      "step": 235
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 491.5833435058594,
      "epoch": 2.533333333333333,
      "grad_norm": 0.9671913385391235,
      "kl": 0.7505035400390625,
      "learning_rate": 2.1205176605410259e-07,
      "loss": 0.1173,
      "reward": 0.541666679084301,
      "reward_std": 0.442245002835989,
      "rewards/accuracy_reward": 0.541666679084301,
      "rewards/format_reward": 0.0,
      "step": 236
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 562.5208473205566,
      "epoch": 2.544,
      "grad_norm": 1.6127721071243286,
      "kl": 0.5107421875,
      "learning_rate": 2.0252925781424153e-07,
      "loss": 0.062,
      "reward": 0.2500000074505806,
      "reward_std": 0.3602609857916832,
      "rewards/accuracy_reward": 0.2500000074505806,
      "rewards/format_reward": 0.0,
      "step": 237
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 615.0833511352539,
      "epoch": 2.554666666666667,
      "grad_norm": 2.2816848754882812,
      "kl": 0.560333251953125,
      "learning_rate": 1.932100059635551e-07,
      "loss": 0.0871,
      "reward": 0.33333334140479565,
      "reward_std": 0.23116153106093407,
      "rewards/accuracy_reward": 0.33333334140479565,
      "rewards/format_reward": 0.0,
      "step": 238
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 603.1458473205566,
      "epoch": 2.5653333333333332,
      "grad_norm": 0.8450276255607605,
      "kl": 0.521514892578125,
      "learning_rate": 1.8409547041741436e-07,
      "loss": 0.1009,
      "reward": 0.2916666753590107,
      "reward_std": 0.3506578877568245,
      "rewards/accuracy_reward": 0.2916666753590107,
      "rewards/format_reward": 0.0,
      "step": 239
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 599.4583396911621,
      "epoch": 2.576,
      "grad_norm": 2.095287799835205,
      "kl": 0.6512451171875,
      "learning_rate": 1.751870790211838e-07,
      "loss": 0.0938,
      "reward": 0.22916667349636555,
      "reward_std": 0.3092299550771713,
      "rewards/accuracy_reward": 0.22916667349636555,
      "rewards/format_reward": 0.0,
      "step": 240
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 631.1875152587891,
      "epoch": 2.586666666666667,
      "grad_norm": 1.0065251588821411,
      "kl": 0.6225738525390625,
      "learning_rate": 1.6648622732653946e-07,
      "loss": 0.0584,
      "reward": 0.4166666753590107,
      "reward_std": 0.24859581887722015,
      "rewards/accuracy_reward": 0.4166666753590107,
      "rewards/format_reward": 0.0,
      "step": 241
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 526.0833511352539,
      "epoch": 2.5973333333333333,
      "grad_norm": 1.0737875699996948,
      "kl": 0.5967559814453125,
      "learning_rate": 1.5799427837284686e-07,
      "loss": 0.0785,
      "reward": 0.4166666716337204,
      "reward_std": 0.46751047670841217,
      "rewards/accuracy_reward": 0.4166666716337204,
      "rewards/format_reward": 0.0,
      "step": 242
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 599.9166793823242,
      "epoch": 2.608,
      "grad_norm": 1.0244035720825195,
      "kl": 0.69384765625,
      "learning_rate": 1.4971256247363534e-07,
      "loss": 0.184,
      "reward": 0.43750000931322575,
      "reward_std": 0.40168892219662666,
      "rewards/accuracy_reward": 0.43750000931322575,
      "rewards/format_reward": 0.0,
      "step": 243
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 645.1250228881836,
      "epoch": 2.618666666666667,
      "grad_norm": 0.9711450338363647,
      "kl": 0.89501953125,
      "learning_rate": 1.416423770081941e-07,
      "loss": 0.1233,
      "reward": 0.3541666716337204,
      "reward_std": 0.2621145099401474,
      "rewards/accuracy_reward": 0.3541666716337204,
      "rewards/format_reward": 0.0,
      "step": 244
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 638.0416793823242,
      "epoch": 2.6293333333333333,
      "grad_norm": 2.9626693725585938,
      "kl": 0.8355712890625,
      "learning_rate": 1.3378498621833236e-07,
      "loss": 0.0753,
      "reward": 0.3125000111758709,
      "reward_std": 0.28219257295131683,
      "rewards/accuracy_reward": 0.3125000111758709,
      "rewards/format_reward": 0.0,
      "step": 245
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 552.9166717529297,
      "epoch": 2.64,
      "grad_norm": 1.3115434646606445,
      "kl": 0.63812255859375,
      "learning_rate": 1.2614162101032601e-07,
      "loss": 0.0858,
      "reward": 0.2916666753590107,
      "reward_std": 0.2957112602889538,
      "rewards/accuracy_reward": 0.2916666753590107,
      "rewards/format_reward": 0.0,
      "step": 246
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 507.1250114440918,
      "epoch": 2.6506666666666665,
      "grad_norm": 0.8592305779457092,
      "kl": 0.461334228515625,
      "learning_rate": 1.1871347876209121e-07,
      "loss": 0.0866,
      "reward": 0.2708333395421505,
      "reward_std": 0.36417657881975174,
      "rewards/accuracy_reward": 0.2708333395421505,
      "rewards/format_reward": 0.0,
      "step": 247
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 598.5625171661377,
      "epoch": 2.6613333333333333,
      "grad_norm": 2.392556667327881,
      "kl": 0.491851806640625,
      "learning_rate": 1.1150172313560741e-07,
      "loss": 0.0422,
      "reward": 0.2291666753590107,
      "reward_std": 0.21764283627271652,
      "rewards/accuracy_reward": 0.2291666753590107,
      "rewards/format_reward": 0.0,
      "step": 248
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 587.854175567627,
      "epoch": 2.672,
      "grad_norm": 0.9977788925170898,
      "kl": 0.42950439453125,
      "learning_rate": 1.0450748389462184e-07,
      "loss": 0.0844,
      "reward": 0.3333333395421505,
      "reward_std": 0.18404608592391014,
      "rewards/accuracy_reward": 0.3333333395421505,
      "rewards/format_reward": 0.0,
      "step": 249
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 451.6875190734863,
      "epoch": 2.6826666666666665,
      "grad_norm": 0.8374356031417847,
      "kl": 0.724639892578125,
      "learning_rate": 9.773185672766705e-08,
      "loss": 0.0788,
      "reward": 0.3958333395421505,
      "reward_std": 0.2525114081799984,
      "rewards/accuracy_reward": 0.3958333395421505,
      "rewards/format_reward": 0.0,
      "step": 250
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 827.9166870117188,
      "epoch": 2.6933333333333334,
      "grad_norm": 4.675540924072266,
      "kl": 0.73406982421875,
      "learning_rate": 9.117590307641361e-08,
      "loss": 0.049,
      "reward": 0.0833333358168602,
      "reward_std": 0.16661180183291435,
      "rewards/accuracy_reward": 0.0833333358168602,
      "rewards/format_reward": 0.0,
      "step": 251
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 615.1041812896729,
      "epoch": 2.7039999999999997,
      "grad_norm": 2.4541940689086914,
      "kl": 0.596710205078125,
      "learning_rate": 8.484064996939023e-08,
      "loss": 0.05,
      "reward": 0.3541666716337204,
      "reward_std": 0.05103103443980217,
      "rewards/accuracy_reward": 0.3541666716337204,
      "rewards/format_reward": 0.0,
      "step": 252
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 525.6041870117188,
      "epoch": 2.7146666666666666,
      "grad_norm": 3.322172164916992,
      "kl": 0.70263671875,
      "learning_rate": 7.872708986109183e-08,
      "loss": 0.0799,
      "reward": 0.33333334140479565,
      "reward_std": 0.2686738707125187,
      "rewards/accuracy_reward": 0.33333334140479565,
      "rewards/format_reward": 0.0,
      "step": 253
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 621.5416870117188,
      "epoch": 2.7253333333333334,
      "grad_norm": 0.5797271132469177,
      "kl": 0.3603363037109375,
      "learning_rate": 7.283618047650808e-08,
      "loss": 0.116,
      "reward": 0.3333333395421505,
      "reward_std": 0.3680921755731106,
      "rewards/accuracy_reward": 0.3333333395421505,
      "rewards/format_reward": 0.0,
      "step": 254
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 501.9166831970215,
      "epoch": 2.7359999999999998,
      "grad_norm": 1.9217878580093384,
      "kl": 0.55096435546875,
      "learning_rate": 6.716884466108869e-08,
      "loss": 0.0918,
      "reward": 0.4375000111758709,
      "reward_std": 0.3816108703613281,
      "rewards/accuracy_reward": 0.4375000111758709,
      "rewards/format_reward": 0.0,
      "step": 255
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 580.4375190734863,
      "epoch": 2.7466666666666666,
      "grad_norm": 0.9500013589859009,
      "kl": 0.46240234375,
      "learning_rate": 6.17259702361741e-08,
      "loss": 0.0906,
      "reward": 0.47916667722165585,
      "reward_std": 0.2900237515568733,
      "rewards/accuracy_reward": 0.47916667722165585,
      "rewards/format_reward": 0.0,
      "step": 256
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 669.8541793823242,
      "epoch": 2.7573333333333334,
      "grad_norm": 1.2487847805023193,
      "kl": 0.4754638671875,
      "learning_rate": 5.65084098599144e-08,
      "loss": 0.0906,
      "reward": 0.20833333767950535,
      "reward_std": 0.2861081622540951,
      "rewards/accuracy_reward": 0.20833333767950535,
      "rewards/format_reward": 0.0,
      "step": 257
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 589.7708396911621,
      "epoch": 2.768,
      "grad_norm": 0.8670617938041687,
      "kl": 0.6224365234375,
      "learning_rate": 5.151698089369433e-08,
      "loss": 0.0838,
      "reward": 0.41666667349636555,
      "reward_std": 0.3506578840315342,
      "rewards/accuracy_reward": 0.41666667349636555,
      "rewards/format_reward": 0.0,
      "step": 258
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 561.979175567627,
      "epoch": 2.7786666666666666,
      "grad_norm": 3.5815389156341553,
      "kl": 0.50885009765625,
      "learning_rate": 4.675246527408872e-08,
      "loss": 0.2336,
      "reward": 0.47916668467223644,
      "reward_std": 0.41129202023148537,
      "rewards/accuracy_reward": 0.47916668467223644,
      "rewards/format_reward": 0.0,
      "step": 259
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 650.8333587646484,
      "epoch": 2.7893333333333334,
      "grad_norm": 3.793149471282959,
      "kl": 0.675048828125,
      "learning_rate": 4.221560939036828e-08,
      "loss": 0.0084,
      "reward": 0.12500000186264515,
      "reward_std": 0.22155842930078506,
      "rewards/accuracy_reward": 0.12500000186264515,
      "rewards/format_reward": 0.0,
      "step": 260
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 551.4791870117188,
      "epoch": 2.8,
      "grad_norm": 1.1272730827331543,
      "kl": 0.35919189453125,
      "learning_rate": 3.790712396757317e-08,
      "loss": 0.0984,
      "reward": 0.4583333395421505,
      "reward_std": 0.39512956142425537,
      "rewards/accuracy_reward": 0.4583333395421505,
      "rewards/format_reward": 0.0,
      "step": 261
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 555.9583549499512,
      "epoch": 2.8106666666666666,
      "grad_norm": 1.0931496620178223,
      "kl": 0.4694671630859375,
      "learning_rate": 3.382768395517355e-08,
      "loss": 0.111,
      "reward": 0.3125000074505806,
      "reward_std": 0.2525114081799984,
      "rewards/accuracy_reward": 0.3125000074505806,
      "rewards/format_reward": 0.0,
      "step": 262
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 656.5208473205566,
      "epoch": 2.8213333333333335,
      "grad_norm": 1.204640507698059,
      "kl": 0.58917236328125,
      "learning_rate": 2.997792842133445e-08,
      "loss": 0.0888,
      "reward": 0.25000000558793545,
      "reward_std": 0.31314554810523987,
      "rewards/accuracy_reward": 0.25000000558793545,
      "rewards/format_reward": 0.0,
      "step": 263
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 554.854175567627,
      "epoch": 2.832,
      "grad_norm": 0.7328877449035645,
      "kl": 0.6326904296875,
      "learning_rate": 2.6358460452802768e-08,
      "loss": 0.068,
      "reward": 0.31250000558793545,
      "reward_std": 0.41912321001291275,
      "rewards/accuracy_reward": 0.31250000558793545,
      "rewards/format_reward": 0.0,
      "step": 264
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 550.3750228881836,
      "epoch": 2.8426666666666667,
      "grad_norm": 0.8261541724205017,
      "kl": 1.0120849609375,
      "learning_rate": 2.296984706042993e-08,
      "loss": 0.0608,
      "reward": 0.29166667722165585,
      "reward_std": 0.3977733254432678,
      "rewards/accuracy_reward": 0.29166667722165585,
      "rewards/format_reward": 0.0,
      "step": 265
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 597.9583396911621,
      "epoch": 2.8533333333333335,
      "grad_norm": 1.9346776008605957,
      "kl": 0.7374420166015625,
      "learning_rate": 1.9812619090345984e-08,
      "loss": 0.1606,
      "reward": 0.3125000111758709,
      "reward_std": 0.3842546343803406,
      "rewards/accuracy_reward": 0.3125000111758709,
      "rewards/format_reward": 0.0,
      "step": 266
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 701.4375190734863,
      "epoch": 2.864,
      "grad_norm": 1.795505404472351,
      "kl": 0.527099609375,
      "learning_rate": 1.688727114079963e-08,
      "loss": 0.0511,
      "reward": 0.1875,
      "reward_std": 0.20539596676826477,
      "rewards/accuracy_reward": 0.1875,
      "rewards/format_reward": 0.0,
      "step": 267
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 631.4791870117188,
      "epoch": 2.8746666666666667,
      "grad_norm": 0.9314957857131958,
      "kl": 0.873046875,
      "learning_rate": 1.4194261484677007e-08,
      "loss": 0.1001,
      "reward": 0.27083334140479565,
      "reward_std": 0.37377968057990074,
      "rewards/accuracy_reward": 0.27083334140479565,
      "rewards/format_reward": 0.0,
      "step": 268
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 456.4166851043701,
      "epoch": 2.8853333333333335,
      "grad_norm": 2.4740889072418213,
      "kl": 0.4954833984375,
      "learning_rate": 1.1734011997709626e-08,
      "loss": 0.1033,
      "reward": 0.5416666772216558,
      "reward_std": 0.2957112602889538,
      "rewards/accuracy_reward": 0.5416666772216558,
      "rewards/format_reward": 0.0,
      "step": 269
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 538.1666793823242,
      "epoch": 2.896,
      "grad_norm": 3.4201440811157227,
      "kl": 0.3656005859375,
      "learning_rate": 9.50690809238558e-09,
      "loss": 0.1591,
      "reward": 0.43750001676380634,
      "reward_std": 0.34674229100346565,
      "rewards/accuracy_reward": 0.43750001676380634,
      "rewards/format_reward": 0.0,
      "step": 270
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 610.0000114440918,
      "epoch": 2.9066666666666667,
      "grad_norm": 1.0745947360992432,
      "kl": 0.619842529296875,
      "learning_rate": 7.513298657572831e-09,
      "loss": 0.0763,
      "reward": 0.27083333767950535,
      "reward_std": 0.28219256550073624,
      "rewards/accuracy_reward": 0.27083333767950535,
      "rewards/format_reward": 0.0,
      "step": 271
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 518.0416831970215,
      "epoch": 2.9173333333333336,
      "grad_norm": 1.7398251295089722,
      "kl": 0.53472900390625,
      "learning_rate": 5.753496003862935e-09,
      "loss": 0.0838,
      "reward": 0.4583333469927311,
      "reward_std": 0.3680921792984009,
      "rewards/accuracy_reward": 0.4583333469927311,
      "rewards/format_reward": 0.0,
      "step": 272
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 672.3333587646484,
      "epoch": 2.928,
      "grad_norm": 2.2082276344299316,
      "kl": 0.4908447265625,
      "learning_rate": 4.227775814646173e-09,
      "loss": 0.0481,
      "reward": 0.08333333395421505,
      "reward_std": 0.11949635669589043,
      "rewards/accuracy_reward": 0.08333333395421505,
      "rewards/format_reward": 0.0,
      "step": 273
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 546.2291793823242,
      "epoch": 2.9386666666666668,
      "grad_norm": 1.0781402587890625,
      "kl": 0.394317626953125,
      "learning_rate": 2.9363771029242705e-09,
      "loss": 0.0481,
      "reward": 0.3750000037252903,
      "reward_std": 0.30354244261980057,
      "rewards/accuracy_reward": 0.3750000037252903,
      "rewards/format_reward": 0.0,
      "step": 274
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 515.7291831970215,
      "epoch": 2.9493333333333336,
      "grad_norm": 0.831046462059021,
      "kl": 0.3851318359375,
      "learning_rate": 1.8795021738670093e-09,
      "loss": 0.0434,
      "reward": 0.41666667722165585,
      "reward_std": 0.23116152733564377,
      "rewards/accuracy_reward": 0.41666667722165585,
      "rewards/format_reward": 0.0,
      "step": 275
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 514.3958511352539,
      "epoch": 2.96,
      "grad_norm": 2.1620874404907227,
      "kl": 0.4932861328125,
      "learning_rate": 1.0573165931209162e-09,
      "loss": 0.0746,
      "reward": 0.45833334140479565,
      "reward_std": 0.30354245007038116,
      "rewards/accuracy_reward": 0.45833334140479565,
      "rewards/format_reward": 0.0,
      "step": 276
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 548.0000152587891,
      "epoch": 2.970666666666667,
      "grad_norm": 1.67172372341156,
      "kl": 0.3961181640625,
      "learning_rate": 4.699491608713413e-10,
      "loss": 0.1205,
      "reward": 0.35416668094694614,
      "reward_std": 0.28219256177544594,
      "rewards/accuracy_reward": 0.35416668094694614,
      "rewards/format_reward": 0.0,
      "step": 277
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 619.2291870117188,
      "epoch": 2.981333333333333,
      "grad_norm": 2.9558866024017334,
      "kl": 0.8973388671875,
      "learning_rate": 1.1749189166609943e-10,
      "loss": 0.0112,
      "reward": 0.1666666716337204,
      "reward_std": 0.19364918768405914,
      "rewards/accuracy_reward": 0.1666666716337204,
      "rewards/format_reward": 0.0,
      "step": 278
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 608.5416831970215,
      "epoch": 2.992,
      "grad_norm": 0.7390713095664978,
      "kl": 0.800933837890625,
      "learning_rate": 0.0,
      "loss": 0.1264,
      "reward": 0.41666666977107525,
      "reward_std": 0.32097673416137695,
      "rewards/accuracy_reward": 0.41666666977107525,
      "rewards/format_reward": 0.0,
      "step": 279
    },
    {
      "epoch": 2.992,
      "step": 279,
      "total_flos": 0.0,
      "train_loss": 0.070359125745952,
      "train_runtime": 12186.8413,
      "train_samples_per_second": 0.185,
      "train_steps_per_second": 0.023
    }
  ],
  "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
}