{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 10.294117647058824,
  "eval_steps": 500,
  "global_step": 700,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "clip_ratio": 0.0,
      "completion_length": 107.74038696289062,
      "epoch": 0.014705882352941176,
      "grad_norm": 4.535776925340435,
      "kl": 0.0,
      "learning_rate": 9.992647058823528e-07,
      "loss": -0.0,
      "reward": 1.7768200635910034,
      "reward_std": 0.06235586851835251,
      "rewards/accuracy_reward": 0.7828295826911926,
      "rewards/format_reward": 0.993990421295166,
      "step": 1
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 109.58173370361328,
      "epoch": 0.029411764705882353,
      "grad_norm": 4.950891299350819,
      "kl": 0.000789642333984375,
      "learning_rate": 9.98529411764706e-07,
      "loss": 0.0,
      "reward": 1.797240972518921,
      "reward_std": 0.06516173481941223,
      "rewards/accuracy_reward": 0.8056544661521912,
      "rewards/format_reward": 0.9915865659713745,
      "step": 2
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 111.5252456665039,
      "epoch": 0.04411764705882353,
      "grad_norm": 1.8505526818416387,
      "kl": 0.00093841552734375,
      "learning_rate": 9.977941176470588e-07,
      "loss": 0.0,
      "reward": 1.7827500104904175,
      "reward_std": 0.06552266329526901,
      "rewards/accuracy_reward": 0.7887594699859619,
      "rewards/format_reward": 0.993990421295166,
      "step": 3
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 110.68870544433594,
      "epoch": 0.058823529411764705,
      "grad_norm": 1.3899081066234504,
      "kl": 0.00102996826171875,
      "learning_rate": 9.970588235294117e-07,
      "loss": 0.0,
      "reward": 1.7779340744018555,
      "reward_std": 0.06615827977657318,
      "rewards/accuracy_reward": 0.7851455807685852,
      "rewards/format_reward": 0.9927884936332703,
      "step": 4
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 111.76563262939453,
      "epoch": 0.07352941176470588,
      "grad_norm": 2.31071643885251,
      "kl": 0.00112152099609375,
      "learning_rate": 9.963235294117647e-07,
      "loss": 0.0,
      "reward": 1.790160894393921,
      "reward_std": 0.06119891628623009,
      "rewards/accuracy_reward": 0.7949684262275696,
      "rewards/format_reward": 0.9951923489570618,
      "step": 5
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 109.93389892578125,
      "epoch": 0.08823529411764706,
      "grad_norm": 5.892350008855451,
      "kl": 0.001983642578125,
      "learning_rate": 9.955882352941176e-07,
      "loss": 0.0001,
      "reward": 1.8021262884140015,
      "reward_std": 0.06035131961107254,
      "rewards/accuracy_reward": 0.8033280372619629,
      "rewards/format_reward": 0.998798131942749,
      "step": 6
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 107.10938262939453,
      "epoch": 0.10294117647058823,
      "grad_norm": 1.4454822561506098,
      "kl": 0.002685546875,
      "learning_rate": 9.948529411764705e-07,
      "loss": 0.0001,
      "reward": 1.7747721672058105,
      "reward_std": 0.06199384108185768,
      "rewards/accuracy_reward": 0.7795796990394592,
      "rewards/format_reward": 0.9951923489570618,
      "step": 7
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 106.11418914794922,
      "epoch": 0.11764705882352941,
      "grad_norm": 2.5592127837818732,
      "kl": 0.0029449462890625,
      "learning_rate": 9.941176470588236e-07,
      "loss": 0.0001,
      "reward": 1.7927086353302002,
      "reward_std": 0.05324796214699745,
      "rewards/accuracy_reward": 0.793910562992096,
      "rewards/format_reward": 0.998798131942749,
      "step": 8
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 104.32933044433594,
      "epoch": 0.1323529411764706,
      "grad_norm": 1.2070688702059713,
      "kl": 0.003509521484375,
      "learning_rate": 9.933823529411765e-07,
      "loss": 0.0001,
      "reward": 1.807510495185852,
      "reward_std": 0.05738683044910431,
      "rewards/accuracy_reward": 0.8099143505096436,
      "rewards/format_reward": 0.9975962042808533,
      "step": 9
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 103.88702392578125,
      "epoch": 0.14705882352941177,
      "grad_norm": 1.1236907795600153,
      "kl": 0.004150390625,
      "learning_rate": 9.926470588235293e-07,
      "loss": 0.0002,
      "reward": 1.7909963130950928,
      "reward_std": 0.05266621708869934,
      "rewards/accuracy_reward": 0.7921981811523438,
      "rewards/format_reward": 0.998798131942749,
      "step": 10
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 103.37981414794922,
      "epoch": 0.16176470588235295,
      "grad_norm": 1.6831920052248819,
      "kl": 0.005645751953125,
      "learning_rate": 9.919117647058824e-07,
      "loss": 0.0002,
      "reward": 1.805780053138733,
      "reward_std": 0.05035817250609398,
      "rewards/accuracy_reward": 0.8069819808006287,
      "rewards/format_reward": 0.998798131942749,
      "step": 11
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 100.015625,
      "epoch": 0.17647058823529413,
      "grad_norm": 1.7093381711239366,
      "kl": 0.0076904296875,
      "learning_rate": 9.911764705882353e-07,
      "loss": 0.0003,
      "reward": 1.829386830329895,
      "reward_std": 0.04737670719623566,
      "rewards/accuracy_reward": 0.8305887579917908,
      "rewards/format_reward": 0.998798131942749,
      "step": 12
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 99.73798370361328,
      "epoch": 0.19117647058823528,
      "grad_norm": 1.5253862652760264,
      "kl": 0.00909423828125,
      "learning_rate": 9.904411764705882e-07,
      "loss": 0.0004,
      "reward": 1.7849040031433105,
      "reward_std": 0.05029381439089775,
      "rewards/accuracy_reward": 0.7849039435386658,
      "rewards/format_reward": 1.0,
      "step": 13
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 98.33534240722656,
      "epoch": 0.20588235294117646,
      "grad_norm": 1.5658868221984554,
      "kl": 0.01055908203125,
      "learning_rate": 9.897058823529412e-07,
      "loss": 0.0004,
      "reward": 1.803476095199585,
      "reward_std": 0.04388910159468651,
      "rewards/accuracy_reward": 0.803476095199585,
      "rewards/format_reward": 1.0,
      "step": 14
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 99.25601196289062,
      "epoch": 0.22058823529411764,
      "grad_norm": 9.7316953003379,
      "kl": 0.01141357421875,
      "learning_rate": 9.889705882352941e-07,
      "loss": 0.0005,
      "reward": 1.791574239730835,
      "reward_std": 0.05360911041498184,
      "rewards/accuracy_reward": 0.7927761077880859,
      "rewards/format_reward": 0.998798131942749,
      "step": 15
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 98.83293914794922,
      "epoch": 0.23529411764705882,
      "grad_norm": 2.8478472800650585,
      "kl": 0.01220703125,
      "learning_rate": 9.88235294117647e-07,
      "loss": 0.0005,
      "reward": 1.7902849912643433,
      "reward_std": 0.051616914570331573,
      "rewards/accuracy_reward": 0.7914868593215942,
      "rewards/format_reward": 0.998798131942749,
      "step": 16
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 98.93509674072266,
      "epoch": 0.25,
      "grad_norm": 2.544057708424019,
      "kl": 0.013427734375,
      "learning_rate": 9.875e-07,
      "loss": 0.0005,
      "reward": 1.8025610446929932,
      "reward_std": 0.04462622106075287,
      "rewards/accuracy_reward": 0.8025608658790588,
      "rewards/format_reward": 1.0,
      "step": 17
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 100.12860870361328,
      "epoch": 0.2647058823529412,
      "grad_norm": 2.256336954402879,
      "kl": 0.0137939453125,
      "learning_rate": 9.86764705882353e-07,
      "loss": 0.0005,
      "reward": 1.8160377740859985,
      "reward_std": 0.049329712986946106,
      "rewards/accuracy_reward": 0.81844162940979,
      "rewards/format_reward": 0.9975962042808533,
      "step": 18
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 100.83413696289062,
      "epoch": 0.27941176470588236,
      "grad_norm": 1.341985441593011,
      "kl": 0.01318359375,
      "learning_rate": 9.860294117647058e-07,
      "loss": 0.0005,
      "reward": 1.7914906740188599,
      "reward_std": 0.04698958992958069,
      "rewards/accuracy_reward": 0.7914904952049255,
      "rewards/format_reward": 1.0,
      "step": 19
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 102.64183044433594,
      "epoch": 0.29411764705882354,
      "grad_norm": 1.1417020647628722,
      "kl": 0.0126953125,
      "learning_rate": 9.85294117647059e-07,
      "loss": 0.0005,
      "reward": 1.785807728767395,
      "reward_std": 0.05590025335550308,
      "rewards/accuracy_reward": 0.7906152606010437,
      "rewards/format_reward": 0.9951923489570618,
      "step": 20
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 103.41106414794922,
      "epoch": 0.3088235294117647,
      "grad_norm": 1.3996656626391477,
      "kl": 0.0118408203125,
      "learning_rate": 9.845588235294118e-07,
      "loss": 0.0005,
      "reward": 1.8126260042190552,
      "reward_std": 0.04200880602002144,
      "rewards/accuracy_reward": 0.8126258254051208,
      "rewards/format_reward": 1.0,
      "step": 21
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 103.72956848144531,
      "epoch": 0.3235294117647059,
      "grad_norm": 1.472735800879064,
      "kl": 0.01214599609375,
      "learning_rate": 9.838235294117647e-07,
      "loss": 0.0005,
      "reward": 1.819865107536316,
      "reward_std": 0.046603430062532425,
      "rewards/accuracy_reward": 0.8210669755935669,
      "rewards/format_reward": 0.998798131942749,
      "step": 22
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 104.89904022216797,
      "epoch": 0.3382352941176471,
      "grad_norm": 1.5355589885672927,
      "kl": 0.011962890625,
      "learning_rate": 9.830882352941175e-07,
      "loss": 0.0005,
      "reward": 1.795883059501648,
      "reward_std": 0.047029443085193634,
      "rewards/accuracy_reward": 0.795883059501648,
      "rewards/format_reward": 1.0,
      "step": 23
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 105.52764892578125,
      "epoch": 0.35294117647058826,
      "grad_norm": 1.7659139234832795,
      "kl": 0.01116943359375,
      "learning_rate": 9.823529411764704e-07,
      "loss": 0.0005,
      "reward": 1.819062352180481,
      "reward_std": 0.045326538383960724,
      "rewards/accuracy_reward": 0.819062352180481,
      "rewards/format_reward": 1.0,
      "step": 24
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 105.76202392578125,
      "epoch": 0.36764705882352944,
      "grad_norm": 1.8387756083505316,
      "kl": 0.0118408203125,
      "learning_rate": 9.816176470588235e-07,
      "loss": 0.0005,
      "reward": 1.8244253396987915,
      "reward_std": 0.04212662950158119,
      "rewards/accuracy_reward": 0.8256272673606873,
      "rewards/format_reward": 0.998798131942749,
      "step": 25
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 104.07572174072266,
      "epoch": 0.38235294117647056,
      "grad_norm": 2.5672205894145668,
      "kl": 0.01123046875,
      "learning_rate": 9.808823529411764e-07,
      "loss": 0.0004,
      "reward": 1.8216837644577026,
      "reward_std": 0.03997746482491493,
      "rewards/accuracy_reward": 0.8216836452484131,
      "rewards/format_reward": 1.0,
      "step": 26
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 106.86659240722656,
      "epoch": 0.39705882352941174,
      "grad_norm": 1.9492202384769068,
      "kl": 0.0111083984375,
      "learning_rate": 9.801470588235292e-07,
      "loss": 0.0004,
      "reward": 1.8016602993011475,
      "reward_std": 0.041357386857271194,
      "rewards/accuracy_reward": 0.8028621077537537,
      "rewards/format_reward": 0.998798131942749,
      "step": 27
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 108.33173370361328,
      "epoch": 0.4117647058823529,
      "grad_norm": 1.5368419770860244,
      "kl": 0.01300048828125,
      "learning_rate": 9.794117647058823e-07,
      "loss": 0.0005,
      "reward": 1.8133355379104614,
      "reward_std": 0.040824707597494125,
      "rewards/accuracy_reward": 0.8145372271537781,
      "rewards/format_reward": 0.998798131942749,
      "step": 28
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 107.08413696289062,
      "epoch": 0.4264705882352941,
      "grad_norm": 1.922839715769838,
      "kl": 0.01397705078125,
      "learning_rate": 9.786764705882352e-07,
      "loss": 0.0006,
      "reward": 1.828223705291748,
      "reward_std": 0.038493864238262177,
      "rewards/accuracy_reward": 0.8294253945350647,
      "rewards/format_reward": 0.998798131942749,
      "step": 29
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 110.17668914794922,
      "epoch": 0.4411764705882353,
      "grad_norm": 1.136750055086542,
      "kl": 0.0137939453125,
      "learning_rate": 9.77941176470588e-07,
      "loss": 0.0006,
      "reward": 1.7888784408569336,
      "reward_std": 0.03950018435716629,
      "rewards/accuracy_reward": 0.7888783812522888,
      "rewards/format_reward": 1.0,
      "step": 30
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 108.23678588867188,
      "epoch": 0.45588235294117646,
      "grad_norm": 1.5207542293833838,
      "kl": 0.013916015625,
      "learning_rate": 9.772058823529412e-07,
      "loss": 0.0006,
      "reward": 1.8419325351715088,
      "reward_std": 0.04379018768668175,
      "rewards/accuracy_reward": 0.8443362712860107,
      "rewards/format_reward": 0.9975962042808533,
      "step": 31
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 108.47956848144531,
      "epoch": 0.47058823529411764,
      "grad_norm": 1.7773456348293046,
      "kl": 0.0145263671875,
      "learning_rate": 9.76470588235294e-07,
      "loss": 0.0006,
      "reward": 1.8314036130905151,
      "reward_std": 0.03341993689537048,
      "rewards/accuracy_reward": 0.8314037919044495,
      "rewards/format_reward": 1.0,
      "step": 32
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 107.88702392578125,
      "epoch": 0.4852941176470588,
      "grad_norm": 1.4109993423092686,
      "kl": 0.01544189453125,
      "learning_rate": 9.75735294117647e-07,
      "loss": 0.0006,
      "reward": 1.8262065649032593,
      "reward_std": 0.039565280079841614,
      "rewards/accuracy_reward": 0.8286103010177612,
      "rewards/format_reward": 0.9975962042808533,
      "step": 33
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 109.64063262939453,
      "epoch": 0.5,
      "grad_norm": 1.8369759357942195,
      "kl": 0.0150146484375,
      "learning_rate": 9.75e-07,
      "loss": 0.0006,
      "reward": 1.7878358364105225,
      "reward_std": 0.03487839177250862,
      "rewards/accuracy_reward": 0.7890377044677734,
      "rewards/format_reward": 0.998798131942749,
      "step": 34
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 108.1658706665039,
      "epoch": 0.5147058823529411,
      "grad_norm": 1.4792457830014905,
      "kl": 0.0169677734375,
      "learning_rate": 9.742647058823529e-07,
      "loss": 0.0007,
      "reward": 1.8230891227722168,
      "reward_std": 0.03256754204630852,
      "rewards/accuracy_reward": 0.823089063167572,
      "rewards/format_reward": 1.0,
      "step": 35
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 111.7596206665039,
      "epoch": 0.5294117647058824,
      "grad_norm": 1.2939561711920475,
      "kl": 0.0167236328125,
      "learning_rate": 9.735294117647057e-07,
      "loss": 0.0007,
      "reward": 1.8332778215408325,
      "reward_std": 0.03907008469104767,
      "rewards/accuracy_reward": 0.835681676864624,
      "rewards/format_reward": 0.9975962042808533,
      "step": 36
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 112.78366088867188,
      "epoch": 0.5441176470588235,
      "grad_norm": 1.1548157711618783,
      "kl": 0.0172119140625,
      "learning_rate": 9.727941176470588e-07,
      "loss": 0.0007,
      "reward": 1.8284510374069214,
      "reward_std": 0.03708866238594055,
      "rewards/accuracy_reward": 0.8296528458595276,
      "rewards/format_reward": 0.998798131942749,
      "step": 37
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 114.41346740722656,
      "epoch": 0.5588235294117647,
      "grad_norm": 1.811998457382516,
      "kl": 0.0174560546875,
      "learning_rate": 9.720588235294117e-07,
      "loss": 0.0007,
      "reward": 1.8531993627548218,
      "reward_std": 0.03520160913467407,
      "rewards/accuracy_reward": 0.8544012308120728,
      "rewards/format_reward": 0.998798131942749,
      "step": 38
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 114.73798370361328,
      "epoch": 0.5735294117647058,
      "grad_norm": 5.782853992835495,
      "kl": 0.0179443359375,
      "learning_rate": 9.713235294117646e-07,
      "loss": 0.0007,
      "reward": 1.820478081703186,
      "reward_std": 0.0349760465323925,
      "rewards/accuracy_reward": 0.8204782009124756,
      "rewards/format_reward": 1.0,
      "step": 39
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 117.25360870361328,
      "epoch": 0.5882352941176471,
      "grad_norm": 2.095875673750828,
      "kl": 0.02001953125,
      "learning_rate": 9.705882352941176e-07,
      "loss": 0.0008,
      "reward": 1.818333625793457,
      "reward_std": 0.036586832255125046,
      "rewards/accuracy_reward": 0.8195353746414185,
      "rewards/format_reward": 0.998798131942749,
      "step": 40
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 119.5564956665039,
      "epoch": 0.6029411764705882,
      "grad_norm": 1.5695506518790454,
      "kl": 0.0205078125,
      "learning_rate": 9.698529411764705e-07,
      "loss": 0.0008,
      "reward": 1.7950727939605713,
      "reward_std": 0.043350886553525925,
      "rewards/accuracy_reward": 0.797476589679718,
      "rewards/format_reward": 0.9975962042808533,
      "step": 41
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 115.74159240722656,
      "epoch": 0.6176470588235294,
      "grad_norm": 4.732453320301059,
      "kl": 0.0220947265625,
      "learning_rate": 9.691176470588234e-07,
      "loss": 0.0009,
      "reward": 1.8049029111862183,
      "reward_std": 0.03959096223115921,
      "rewards/accuracy_reward": 0.8061047792434692,
      "rewards/format_reward": 0.998798131942749,
      "step": 42
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 116.66706848144531,
      "epoch": 0.6323529411764706,
      "grad_norm": 2.4880576589131533,
      "kl": 0.021484375,
      "learning_rate": 9.683823529411765e-07,
      "loss": 0.0009,
      "reward": 1.7938587665557861,
      "reward_std": 0.03622143343091011,
      "rewards/accuracy_reward": 0.7938587665557861,
      "rewards/format_reward": 1.0,
      "step": 43
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 115.97356414794922,
      "epoch": 0.6470588235294118,
      "grad_norm": 2.4237436345937122,
      "kl": 0.022705078125,
      "learning_rate": 9.676470588235294e-07,
      "loss": 0.0009,
      "reward": 1.8250298500061035,
      "reward_std": 0.03410777822136879,
      "rewards/accuracy_reward": 0.8262317776679993,
      "rewards/format_reward": 0.998798131942749,
      "step": 44
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 113.92788696289062,
      "epoch": 0.6617647058823529,
      "grad_norm": 2.2338566414155463,
      "kl": 0.022705078125,
      "learning_rate": 9.669117647058822e-07,
      "loss": 0.0009,
      "reward": 1.8153414726257324,
      "reward_std": 0.043387699872255325,
      "rewards/accuracy_reward": 0.8189471364021301,
      "rewards/format_reward": 0.9963942766189575,
      "step": 45
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 113.89543914794922,
      "epoch": 0.6764705882352942,
      "grad_norm": 11.225902629577423,
      "kl": 0.0225830078125,
      "learning_rate": 9.661764705882353e-07,
      "loss": 0.0009,
      "reward": 1.8124586343765259,
      "reward_std": 0.033101461827754974,
      "rewards/accuracy_reward": 0.8124586343765259,
      "rewards/format_reward": 1.0,
      "step": 46
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 113.8738021850586,
      "epoch": 0.6911764705882353,
      "grad_norm": 2.4337206941159564,
      "kl": 0.0216064453125,
      "learning_rate": 9.654411764705882e-07,
      "loss": 0.0009,
      "reward": 1.82453191280365,
      "reward_std": 0.033501800149679184,
      "rewards/accuracy_reward": 0.8257339596748352,
      "rewards/format_reward": 0.998798131942749,
      "step": 47
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 112.66466522216797,
      "epoch": 0.7058823529411765,
      "grad_norm": 1.6698840088478868,
      "kl": 0.02587890625,
      "learning_rate": 9.64705882352941e-07,
      "loss": 0.001,
      "reward": 1.8177635669708252,
      "reward_std": 0.03327854722738266,
      "rewards/accuracy_reward": 0.8177635073661804,
      "rewards/format_reward": 1.0,
      "step": 48
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 113.1189956665039,
      "epoch": 0.7205882352941176,
      "grad_norm": 2.838491633259207,
      "kl": 0.02294921875,
      "learning_rate": 9.639705882352941e-07,
      "loss": 0.0009,
      "reward": 1.8294517993927002,
      "reward_std": 0.03862498328089714,
      "rewards/accuracy_reward": 0.8318556547164917,
      "rewards/format_reward": 0.9975962042808533,
      "step": 49
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 112.31250762939453,
      "epoch": 0.7352941176470589,
      "grad_norm": 1.6937318420428593,
      "kl": 0.0234375,
      "learning_rate": 9.63235294117647e-07,
      "loss": 0.0009,
      "reward": 1.8327486515045166,
      "reward_std": 0.03481902554631233,
      "rewards/accuracy_reward": 0.8339505195617676,
      "rewards/format_reward": 0.998798131942749,
      "step": 50
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 109.50241088867188,
      "epoch": 0.75,
      "grad_norm": 1.9451567930831182,
      "kl": 0.0245361328125,
      "learning_rate": 9.624999999999999e-07,
      "loss": 0.001,
      "reward": 1.839182734489441,
      "reward_std": 0.029496613889932632,
      "rewards/accuracy_reward": 0.8391827940940857,
      "rewards/format_reward": 1.0,
      "step": 51
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 111.64904022216797,
      "epoch": 0.7647058823529411,
      "grad_norm": 1.9537307288218644,
      "kl": 0.023681640625,
      "learning_rate": 9.61764705882353e-07,
      "loss": 0.0009,
      "reward": 1.8222689628601074,
      "reward_std": 0.038959018886089325,
      "rewards/accuracy_reward": 0.8234708309173584,
      "rewards/format_reward": 0.998798131942749,
      "step": 52
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 109.05529022216797,
      "epoch": 0.7794117647058824,
      "grad_norm": 2.0869715773084634,
      "kl": 0.022705078125,
      "learning_rate": 9.610294117647058e-07,
      "loss": 0.0009,
      "reward": 1.8396120071411133,
      "reward_std": 0.030058706179261208,
      "rewards/accuracy_reward": 0.8396119475364685,
      "rewards/format_reward": 1.0,
      "step": 53
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 107.25241088867188,
      "epoch": 0.7941176470588235,
      "grad_norm": 1.399697409291221,
      "kl": 0.0230712890625,
      "learning_rate": 9.602941176470587e-07,
      "loss": 0.0009,
      "reward": 1.8265293836593628,
      "reward_std": 0.02903612144291401,
      "rewards/accuracy_reward": 0.8265291452407837,
      "rewards/format_reward": 1.0,
      "step": 54
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 107.35697174072266,
      "epoch": 0.8088235294117647,
      "grad_norm": 1.9480357994696749,
      "kl": 0.022705078125,
      "learning_rate": 9.595588235294118e-07,
      "loss": 0.0009,
      "reward": 1.8046581745147705,
      "reward_std": 0.030282124876976013,
      "rewards/accuracy_reward": 0.8046582341194153,
      "rewards/format_reward": 1.0,
      "step": 55
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 107.97356414794922,
      "epoch": 0.8235294117647058,
      "grad_norm": 1.8194034842124802,
      "kl": 0.0218505859375,
      "learning_rate": 9.588235294117647e-07,
      "loss": 0.0009,
      "reward": 1.8263245820999146,
      "reward_std": 0.029453594237565994,
      "rewards/accuracy_reward": 0.8275262713432312,
      "rewards/format_reward": 0.998798131942749,
      "step": 56
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 108.06731414794922,
      "epoch": 0.8382352941176471,
      "grad_norm": 1.747459486428414,
      "kl": 0.02001953125,
      "learning_rate": 9.580882352941176e-07,
      "loss": 0.0008,
      "reward": 1.8315790891647339,
      "reward_std": 0.02826782502233982,
      "rewards/accuracy_reward": 0.8315791487693787,
      "rewards/format_reward": 1.0,
      "step": 57
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 108.09616088867188,
      "epoch": 0.8529411764705882,
      "grad_norm": 2.2648825774011856,
      "kl": 0.019775390625,
      "learning_rate": 9.573529411764706e-07,
      "loss": 0.0008,
      "reward": 1.8353791236877441,
      "reward_std": 0.028635086491703987,
      "rewards/accuracy_reward": 0.8353790640830994,
      "rewards/format_reward": 1.0,
      "step": 58
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 108.56851196289062,
      "epoch": 0.8676470588235294,
      "grad_norm": 1.3353893541096462,
      "kl": 0.02001953125,
      "learning_rate": 9.566176470588235e-07,
      "loss": 0.0008,
      "reward": 1.8366105556488037,
      "reward_std": 0.029847299680113792,
      "rewards/accuracy_reward": 0.8378124833106995,
      "rewards/format_reward": 0.998798131942749,
      "step": 59
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 107.87741088867188,
      "epoch": 0.8823529411764706,
      "grad_norm": 1.2231786582956035,
      "kl": 0.021484375,
      "learning_rate": 9.558823529411764e-07,
      "loss": 0.0009,
      "reward": 1.8441990613937378,
      "reward_std": 0.029375040903687477,
      "rewards/accuracy_reward": 0.8441989421844482,
      "rewards/format_reward": 1.0,
      "step": 60
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 109.85697174072266,
      "epoch": 0.8970588235294118,
      "grad_norm": 1.6690678560710506,
      "kl": 0.01904296875,
      "learning_rate": 9.551470588235295e-07,
      "loss": 0.0008,
      "reward": 1.8236721754074097,
      "reward_std": 0.02699529379606247,
      "rewards/accuracy_reward": 0.8236721158027649,
      "rewards/format_reward": 1.0,
      "step": 61
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 110.2908706665039,
      "epoch": 0.9117647058823529,
      "grad_norm": 6.089899029100757,
      "kl": 0.019287109375,
      "learning_rate": 9.544117647058823e-07,
      "loss": 0.0008,
      "reward": 1.8286242485046387,
      "reward_std": 0.02875707298517227,
      "rewards/accuracy_reward": 0.8286241888999939,
      "rewards/format_reward": 1.0,
      "step": 62
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 111.53846740722656,
      "epoch": 0.9264705882352942,
      "grad_norm": 1.8276415184151562,
      "kl": 0.0201416015625,
      "learning_rate": 9.536764705882352e-07,
      "loss": 0.0008,
      "reward": 1.8171989917755127,
      "reward_std": 0.033383674919605255,
      "rewards/accuracy_reward": 0.8184008002281189,
      "rewards/format_reward": 0.998798131942749,
      "step": 63
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 112.20553588867188,
      "epoch": 0.9411764705882353,
      "grad_norm": 2.253394799784533,
      "kl": 0.0206298828125,
      "learning_rate": 9.529411764705881e-07,
      "loss": 0.0008,
      "reward": 1.8275461196899414,
      "reward_std": 0.033603277057409286,
      "rewards/accuracy_reward": 0.8287478685379028,
      "rewards/format_reward": 0.998798131942749,
      "step": 64
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 110.28366088867188,
      "epoch": 0.9558823529411765,
      "grad_norm": 1.9640142035092505,
      "kl": 0.0203857421875,
      "learning_rate": 9.522058823529411e-07,
      "loss": 0.0008,
      "reward": 1.8377821445465088,
      "reward_std": 0.027394887059926987,
      "rewards/accuracy_reward": 0.8389840126037598,
      "rewards/format_reward": 0.998798131942749,
      "step": 65
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 109.56971740722656,
      "epoch": 0.9705882352941176,
      "grad_norm": 1.7539526199640756,
      "kl": 0.021484375,
      "learning_rate": 9.51470588235294e-07,
      "loss": 0.0009,
      "reward": 1.8167920112609863,
      "reward_std": 0.02794465236365795,
      "rewards/accuracy_reward": 0.8167920112609863,
      "rewards/format_reward": 1.0,
      "step": 66
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 110.77284240722656,
      "epoch": 0.9852941176470589,
      "grad_norm": 1.2641268462386457,
      "kl": 0.0228271484375,
      "learning_rate": 9.507352941176469e-07,
      "loss": 0.0009,
      "reward": 1.8030400276184082,
      "reward_std": 0.03198782354593277,
      "rewards/accuracy_reward": 0.8030399680137634,
      "rewards/format_reward": 1.0,
      "step": 67
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 108.8310546875,
      "epoch": 1.0,
      "grad_norm": 1.63346443563275,
      "kl": 0.022216796875,
      "learning_rate": 9.499999999999999e-07,
      "loss": 0.0009,
      "reward": 1.8402751684188843,
      "reward_std": 0.028587892651557922,
      "rewards/accuracy_reward": 0.8402751684188843,
      "rewards/format_reward": 1.0,
      "step": 68
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 110.18991088867188,
      "epoch": 1.0147058823529411,
      "grad_norm": 2.0248373653875955,
      "kl": 0.0225830078125,
      "learning_rate": 9.492647058823529e-07,
      "loss": 0.0009,
      "reward": 1.8379939794540405,
      "reward_std": 0.02731536142528057,
      "rewards/accuracy_reward": 0.8379938006401062,
      "rewards/format_reward": 1.0,
      "step": 69
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 108.76923370361328,
      "epoch": 1.0294117647058822,
      "grad_norm": 1.2467348089074899,
      "kl": 0.0224609375,
      "learning_rate": 9.485294117647058e-07,
      "loss": 0.0009,
      "reward": 1.839119553565979,
      "reward_std": 0.027092870324850082,
      "rewards/accuracy_reward": 0.8391194939613342,
      "rewards/format_reward": 1.0,
      "step": 70
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 107.04447174072266,
      "epoch": 1.0441176470588236,
      "grad_norm": 1.7733203946188572,
      "kl": 0.024169921875,
      "learning_rate": 9.477941176470587e-07,
      "loss": 0.001,
      "reward": 1.835349678993225,
      "reward_std": 0.028652215376496315,
      "rewards/accuracy_reward": 0.8365515470504761,
      "rewards/format_reward": 0.998798131942749,
      "step": 71
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 105.54206848144531,
      "epoch": 1.0588235294117647,
      "grad_norm": 1.417496550727975,
      "kl": 0.0255126953125,
      "learning_rate": 9.470588235294117e-07,
      "loss": 0.001,
      "reward": 1.8174885511398315,
      "reward_std": 0.03328551724553108,
      "rewards/accuracy_reward": 0.8186904191970825,
      "rewards/format_reward": 0.998798131942749,
      "step": 72
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 105.22956848144531,
      "epoch": 1.0735294117647058,
      "grad_norm": 2.347300550273842,
      "kl": 0.022705078125,
      "learning_rate": 9.463235294117646e-07,
      "loss": 0.0009,
      "reward": 1.8195103406906128,
      "reward_std": 0.028913108631968498,
      "rewards/accuracy_reward": 0.8195103406906128,
      "rewards/format_reward": 1.0,
      "step": 73
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 108.82572174072266,
      "epoch": 1.088235294117647,
      "grad_norm": 2.5948956191316306,
      "kl": 0.024169921875,
      "learning_rate": 9.455882352941176e-07,
      "loss": 0.001,
      "reward": 1.8160063028335571,
      "reward_std": 0.03166431561112404,
      "rewards/accuracy_reward": 0.8160063028335571,
      "rewards/format_reward": 1.0,
      "step": 74
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 106.91106414794922,
      "epoch": 1.1029411764705883,
      "grad_norm": 2.2984611372489847,
      "kl": 0.0245361328125,
      "learning_rate": 9.448529411764705e-07,
      "loss": 0.001,
      "reward": 1.841678261756897,
      "reward_std": 0.028328273445367813,
      "rewards/accuracy_reward": 0.8416782021522522,
      "rewards/format_reward": 1.0,
      "step": 75
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 110.30769348144531,
      "epoch": 1.1176470588235294,
      "grad_norm": 1.2558402021274238,
      "kl": 0.025146484375,
      "learning_rate": 9.441176470588234e-07,
      "loss": 0.001,
      "reward": 1.8383431434631348,
      "reward_std": 0.02794811502099037,
      "rewards/accuracy_reward": 0.8395450711250305,
      "rewards/format_reward": 0.998798131942749,
      "step": 76
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 109.67428588867188,
      "epoch": 1.1323529411764706,
      "grad_norm": 1.4150029392598835,
      "kl": 0.02392578125,
      "learning_rate": 9.433823529411764e-07,
      "loss": 0.001,
      "reward": 1.8399676084518433,
      "reward_std": 0.025521790608763695,
      "rewards/accuracy_reward": 0.8399675488471985,
      "rewards/format_reward": 1.0,
      "step": 77
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 112.96034240722656,
      "epoch": 1.1470588235294117,
      "grad_norm": 1.1950823033585802,
      "kl": 0.0264892578125,
      "learning_rate": 9.426470588235294e-07,
      "loss": 0.0011,
      "reward": 1.83869206905365,
      "reward_std": 0.030266188085079193,
      "rewards/accuracy_reward": 0.8398938775062561,
      "rewards/format_reward": 0.998798131942749,
      "step": 78
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 112.77284240722656,
      "epoch": 1.161764705882353,
      "grad_norm": 3.5214271570942355,
      "kl": 0.0272216796875,
      "learning_rate": 9.419117647058823e-07,
      "loss": 0.0011,
      "reward": 1.8469470739364624,
      "reward_std": 0.025917939841747284,
      "rewards/accuracy_reward": 0.8469470143318176,
      "rewards/format_reward": 1.0,
      "step": 79
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 114.43991088867188,
      "epoch": 1.1764705882352942,
      "grad_norm": 1.5192920476546525,
      "kl": 0.027099609375,
      "learning_rate": 9.411764705882352e-07,
      "loss": 0.0011,
      "reward": 1.8514484167099,
      "reward_std": 0.02656027302145958,
      "rewards/accuracy_reward": 0.8514483571052551,
      "rewards/format_reward": 1.0,
      "step": 80
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 114.13581848144531,
      "epoch": 1.1911764705882353,
      "grad_norm": 1.4307870600496726,
      "kl": 0.027099609375,
      "learning_rate": 9.404411764705882e-07,
      "loss": 0.0011,
      "reward": 1.848158359527588,
      "reward_std": 0.03505507484078407,
      "rewards/accuracy_reward": 0.8505621552467346,
      "rewards/format_reward": 0.9975962042808533,
      "step": 81
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 114.22356414794922,
      "epoch": 1.2058823529411764,
      "grad_norm": 1.9122780286883485,
      "kl": 0.042724609375,
      "learning_rate": 9.397058823529411e-07,
      "loss": 0.0017,
      "reward": 1.8507599830627441,
      "reward_std": 0.027777796611189842,
      "rewards/accuracy_reward": 0.8519617319107056,
      "rewards/format_reward": 0.998798131942749,
      "step": 82
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 114.24159240722656,
      "epoch": 1.2205882352941178,
      "grad_norm": 1.2485868300090033,
      "kl": 0.0281982421875,
      "learning_rate": 9.389705882352941e-07,
      "loss": 0.0011,
      "reward": 1.8573682308197021,
      "reward_std": 0.028743639588356018,
      "rewards/accuracy_reward": 0.8585702180862427,
      "rewards/format_reward": 0.998798131942749,
      "step": 83
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 112.99038696289062,
      "epoch": 1.2352941176470589,
      "grad_norm": 1.6395721460640882,
      "kl": 0.02587890625,
      "learning_rate": 9.38235294117647e-07,
      "loss": 0.001,
      "reward": 1.849989414215088,
      "reward_std": 0.03292028233408928,
      "rewards/accuracy_reward": 0.8523932695388794,
      "rewards/format_reward": 0.9975962042808533,
      "step": 84
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 112.15144348144531,
      "epoch": 1.25,
      "grad_norm": 1.8788708892556627,
      "kl": 0.029296875,
      "learning_rate": 9.374999999999999e-07,
      "loss": 0.0012,
      "reward": 1.8299332857131958,
      "reward_std": 0.030824122950434685,
      "rewards/accuracy_reward": 0.832336962223053,
      "rewards/format_reward": 0.9975962042808533,
      "step": 85
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 110.70072174072266,
      "epoch": 1.2647058823529411,
      "grad_norm": 1.284626382800343,
      "kl": 0.0257568359375,
      "learning_rate": 9.367647058823529e-07,
      "loss": 0.001,
      "reward": 1.8247534036636353,
      "reward_std": 0.02943887747824192,
      "rewards/accuracy_reward": 0.825955331325531,
      "rewards/format_reward": 0.998798131942749,
      "step": 86
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 111.10938262939453,
      "epoch": 1.2794117647058822,
      "grad_norm": 1.800940971026441,
      "kl": 0.0291748046875,
      "learning_rate": 9.360294117647059e-07,
      "loss": 0.0012,
      "reward": 1.829021692276001,
      "reward_std": 0.03014710173010826,
      "rewards/accuracy_reward": 0.8290215730667114,
      "rewards/format_reward": 1.0,
      "step": 87
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 110.72356414794922,
      "epoch": 1.2941176470588236,
      "grad_norm": 1.6669503142403084,
      "kl": 0.027587890625,
      "learning_rate": 9.352941176470588e-07,
      "loss": 0.0011,
      "reward": 1.8365869522094727,
      "reward_std": 0.035319458693265915,
      "rewards/accuracy_reward": 0.8389906287193298,
      "rewards/format_reward": 0.9975962042808533,
      "step": 88
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 109.84495544433594,
      "epoch": 1.3088235294117647,
      "grad_norm": 1.9827979032477325,
      "kl": 0.0267333984375,
      "learning_rate": 9.345588235294117e-07,
      "loss": 0.0011,
      "reward": 1.8515337705612183,
      "reward_std": 0.032718025147914886,
      "rewards/accuracy_reward": 0.8527354598045349,
      "rewards/format_reward": 0.998798131942749,
      "step": 89
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 107.54688262939453,
      "epoch": 1.3235294117647058,
      "grad_norm": 1.855181016525942,
      "kl": 0.029541015625,
      "learning_rate": 9.338235294117647e-07,
      "loss": 0.0012,
      "reward": 1.8531250953674316,
      "reward_std": 0.03333236649632454,
      "rewards/accuracy_reward": 0.8555287718772888,
      "rewards/format_reward": 0.9975962042808533,
      "step": 90
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 106.51683044433594,
      "epoch": 1.3382352941176472,
      "grad_norm": 1.542405666476553,
      "kl": 0.0299072265625,
      "learning_rate": 9.330882352941176e-07,
      "loss": 0.0012,
      "reward": 1.8587943315505981,
      "reward_std": 0.02777322567999363,
      "rewards/accuracy_reward": 0.8587943315505981,
      "rewards/format_reward": 1.0,
      "step": 91
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 106.65866088867188,
      "epoch": 1.3529411764705883,
      "grad_norm": 6.044720279013163,
      "kl": 0.027099609375,
      "learning_rate": 9.323529411764706e-07,
      "loss": 0.0011,
      "reward": 1.8246790170669556,
      "reward_std": 0.03338155895471573,
      "rewards/accuracy_reward": 0.8258809447288513,
      "rewards/format_reward": 0.998798131942749,
      "step": 92
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 107.25481414794922,
      "epoch": 1.3676470588235294,
      "grad_norm": 3.698518846068984,
      "kl": 0.026611328125,
      "learning_rate": 9.316176470588235e-07,
      "loss": 0.0011,
      "reward": 1.8631399869918823,
      "reward_std": 0.027002353221178055,
      "rewards/accuracy_reward": 0.8631399869918823,
      "rewards/format_reward": 1.0,
      "step": 93
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 104.18870544433594,
      "epoch": 1.3823529411764706,
      "grad_norm": 7.435870249889465,
      "kl": 0.027587890625,
      "learning_rate": 9.308823529411764e-07,
      "loss": 0.0011,
      "reward": 1.8605844974517822,
      "reward_std": 0.02926478162407875,
      "rewards/accuracy_reward": 0.8605843782424927,
      "rewards/format_reward": 1.0,
      "step": 94
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 106.17188262939453,
      "epoch": 1.3970588235294117,
      "grad_norm": 3.53776006589639,
      "kl": 0.038818359375,
      "learning_rate": 9.301470588235294e-07,
      "loss": 0.0016,
      "reward": 1.8662550449371338,
      "reward_std": 0.028098618611693382,
      "rewards/accuracy_reward": 0.8662549257278442,
      "rewards/format_reward": 1.0,
      "step": 95
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 105.53606414794922,
      "epoch": 1.4117647058823528,
      "grad_norm": 2.008765777352944,
      "kl": 0.0277099609375,
      "learning_rate": 9.294117647058824e-07,
      "loss": 0.0011,
      "reward": 1.8323391675949097,
      "reward_std": 0.03244079276919365,
      "rewards/accuracy_reward": 0.8335410356521606,
      "rewards/format_reward": 0.998798131942749,
      "step": 96
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 106.32813262939453,
      "epoch": 1.4264705882352942,
      "grad_norm": 1.6277308554524441,
      "kl": 0.026123046875,
      "learning_rate": 9.286764705882352e-07,
      "loss": 0.001,
      "reward": 1.8551498651504517,
      "reward_std": 0.027026960626244545,
      "rewards/accuracy_reward": 0.8551498055458069,
      "rewards/format_reward": 1.0,
      "step": 97
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 104.96514892578125,
      "epoch": 1.4411764705882353,
      "grad_norm": 1.8674445586310056,
      "kl": 0.027587890625,
      "learning_rate": 9.279411764705882e-07,
      "loss": 0.0011,
      "reward": 1.8556426763534546,
      "reward_std": 0.02985842153429985,
      "rewards/accuracy_reward": 0.855642557144165,
      "rewards/format_reward": 1.0,
      "step": 98
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 107.22116088867188,
      "epoch": 1.4558823529411764,
      "grad_norm": 2.564978355227823,
      "kl": 0.031494140625,
      "learning_rate": 9.272058823529412e-07,
      "loss": 0.0013,
      "reward": 1.8616751432418823,
      "reward_std": 0.028857456520199776,
      "rewards/accuracy_reward": 0.8628771305084229,
      "rewards/format_reward": 0.998798131942749,
      "step": 99
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 106.96034240722656,
      "epoch": 1.4705882352941178,
      "grad_norm": 1.5062130671090712,
      "kl": 0.029052734375,
      "learning_rate": 9.264705882352941e-07,
      "loss": 0.0012,
      "reward": 1.8399274349212646,
      "reward_std": 0.033529430627822876,
      "rewards/accuracy_reward": 0.8411291837692261,
      "rewards/format_reward": 0.998798131942749,
      "step": 100
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 108.31731414794922,
      "epoch": 1.4852941176470589,
      "grad_norm": 1.7672687152261588,
      "kl": 0.0301513671875,
      "learning_rate": 9.257352941176471e-07,
      "loss": 0.0012,
      "reward": 1.8533381223678589,
      "reward_std": 0.03230126574635506,
      "rewards/accuracy_reward": 0.8545399308204651,
      "rewards/format_reward": 0.998798131942749,
      "step": 101
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 109.50601196289062,
      "epoch": 1.5,
      "grad_norm": 1.2513961138527014,
      "kl": 0.0294189453125,
      "learning_rate": 9.25e-07,
      "loss": 0.0012,
      "reward": 1.8355013132095337,
      "reward_std": 0.03523796796798706,
      "rewards/accuracy_reward": 0.8367031812667847,
      "rewards/format_reward": 0.998798131942749,
      "step": 102
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 107.85096740722656,
      "epoch": 1.5147058823529411,
      "grad_norm": 1.7612446743928787,
      "kl": 0.0286865234375,
      "learning_rate": 9.242647058823529e-07,
      "loss": 0.0011,
      "reward": 1.835523247718811,
      "reward_std": 0.028739692643284798,
      "rewards/accuracy_reward": 0.8355231285095215,
      "rewards/format_reward": 1.0,
      "step": 103
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 109.06971740722656,
      "epoch": 1.5294117647058822,
      "grad_norm": 1.2684013897529154,
      "kl": 0.0296630859375,
      "learning_rate": 9.235294117647059e-07,
      "loss": 0.0012,
      "reward": 1.8417116403579712,
      "reward_std": 0.03294159471988678,
      "rewards/accuracy_reward": 0.8429136872291565,
      "rewards/format_reward": 0.998798131942749,
      "step": 104
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 110.39063262939453,
      "epoch": 1.5441176470588234,
      "grad_norm": 1.5110927030527903,
      "kl": 0.0263671875,
      "learning_rate": 9.227941176470589e-07,
      "loss": 0.0011,
      "reward": 1.8465179204940796,
      "reward_std": 0.035068195313215256,
      "rewards/accuracy_reward": 0.8489217162132263,
      "rewards/format_reward": 0.9975962042808533,
      "step": 105
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 108.76202392578125,
      "epoch": 1.5588235294117647,
      "grad_norm": 4.89157193030413,
      "kl": 0.0299072265625,
      "learning_rate": 9.220588235294117e-07,
      "loss": 0.0012,
      "reward": 1.8534250259399414,
      "reward_std": 0.029864592477679253,
      "rewards/accuracy_reward": 0.8534249663352966,
      "rewards/format_reward": 1.0,
      "step": 106
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 110.63822174072266,
      "epoch": 1.5735294117647058,
      "grad_norm": 1.397310457995284,
      "kl": 0.0264892578125,
      "learning_rate": 9.213235294117646e-07,
      "loss": 0.0011,
      "reward": 1.8602780103683472,
      "reward_std": 0.0308562982827425,
      "rewards/accuracy_reward": 0.8614799380302429,
      "rewards/format_reward": 0.998798131942749,
      "step": 107
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 112.93870544433594,
      "epoch": 1.5882352941176472,
      "grad_norm": 2.2153563584401867,
      "kl": 0.027587890625,
      "learning_rate": 9.205882352941176e-07,
      "loss": 0.0011,
      "reward": 1.8766840696334839,
      "reward_std": 0.032767944037914276,
      "rewards/accuracy_reward": 0.8778861165046692,
      "rewards/format_reward": 0.998798131942749,
      "step": 108
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 110.41706848144531,
      "epoch": 1.6029411764705883,
      "grad_norm": 1.3441095236077332,
      "kl": 0.0277099609375,
      "learning_rate": 9.198529411764705e-07,
      "loss": 0.0011,
      "reward": 1.8476648330688477,
      "reward_std": 0.03170119971036911,
      "rewards/accuracy_reward": 0.8476646542549133,
      "rewards/format_reward": 1.0,
      "step": 109
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 114.37500762939453,
      "epoch": 1.6176470588235294,
      "grad_norm": 2.087440395995137,
      "kl": 0.0302734375,
      "learning_rate": 9.191176470588234e-07,
      "loss": 0.0012,
      "reward": 1.8472872972488403,
      "reward_std": 0.032663196325302124,
      "rewards/accuracy_reward": 0.8484892249107361,
      "rewards/format_reward": 0.998798131942749,
      "step": 110
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 112.37259674072266,
      "epoch": 1.6323529411764706,
      "grad_norm": 1.540582934098744,
      "kl": 0.031494140625,
      "learning_rate": 9.183823529411764e-07,
      "loss": 0.0013,
      "reward": 1.8425360918045044,
      "reward_std": 0.0323411226272583,
      "rewards/accuracy_reward": 0.8425360918045044,
      "rewards/format_reward": 1.0,
      "step": 111
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 114.81851196289062,
      "epoch": 1.6470588235294117,
      "grad_norm": 2.4116409582345297,
      "kl": 0.02978515625,
      "learning_rate": 9.176470588235293e-07,
      "loss": 0.0012,
      "reward": 1.8361573219299316,
      "reward_std": 0.0333629846572876,
      "rewards/accuracy_reward": 0.8373592495918274,
      "rewards/format_reward": 0.998798131942749,
      "step": 112
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 114.86659240722656,
      "epoch": 1.6617647058823528,
      "grad_norm": 2.020278808245713,
      "kl": 0.031005859375,
      "learning_rate": 9.169117647058823e-07,
      "loss": 0.0012,
      "reward": 1.8621249198913574,
      "reward_std": 0.030686957761645317,
      "rewards/accuracy_reward": 0.8621248602867126,
      "rewards/format_reward": 1.0,
      "step": 113
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 117.34134674072266,
      "epoch": 1.6764705882352942,
      "grad_norm": 2.599149661763795,
      "kl": 0.0272216796875,
      "learning_rate": 9.161764705882353e-07,
      "loss": 0.0011,
      "reward": 1.8338128328323364,
      "reward_std": 0.03273686766624451,
      "rewards/accuracy_reward": 0.8350145220756531,
      "rewards/format_reward": 0.998798131942749,
      "step": 114
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 117.04206848144531,
      "epoch": 1.6911764705882353,
      "grad_norm": 1.696964274521337,
      "kl": 0.02783203125,
      "learning_rate": 9.154411764705881e-07,
      "loss": 0.0011,
      "reward": 1.8655701875686646,
      "reward_std": 0.026736797764897346,
      "rewards/accuracy_reward": 0.8655699491500854,
      "rewards/format_reward": 1.0,
      "step": 115
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 117.17788696289062,
      "epoch": 1.7058823529411766,
      "grad_norm": 1.6668443133949866,
      "kl": 0.02880859375,
      "learning_rate": 9.147058823529411e-07,
      "loss": 0.0012,
      "reward": 1.859294056892395,
      "reward_std": 0.02911749854683876,
      "rewards/accuracy_reward": 0.8604959845542908,
      "rewards/format_reward": 0.998798131942749,
      "step": 116
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 116.41706848144531,
      "epoch": 1.7205882352941178,
      "grad_norm": 1.8917048442273852,
      "kl": 0.0289306640625,
      "learning_rate": 9.139705882352941e-07,
      "loss": 0.0012,
      "reward": 1.8411531448364258,
      "reward_std": 0.03658343106508255,
      "rewards/accuracy_reward": 0.8471627831459045,
      "rewards/format_reward": 0.993990421295166,
      "step": 117
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 115.32813262939453,
      "epoch": 1.7352941176470589,
      "grad_norm": 1.6450031082088983,
      "kl": 0.0279541015625,
      "learning_rate": 9.13235294117647e-07,
      "loss": 0.0011,
      "reward": 1.8670862913131714,
      "reward_std": 0.026890011504292488,
      "rewards/accuracy_reward": 0.8670862913131714,
      "rewards/format_reward": 1.0,
      "step": 118
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 110.88341522216797,
      "epoch": 1.75,
      "grad_norm": 3.248672120006505,
      "kl": 0.029052734375,
      "learning_rate": 9.124999999999999e-07,
      "loss": 0.0012,
      "reward": 1.8559081554412842,
      "reward_std": 0.02590222656726837,
      "rewards/accuracy_reward": 0.8559079170227051,
      "rewards/format_reward": 1.0,
      "step": 119
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 112.36779022216797,
      "epoch": 1.7647058823529411,
      "grad_norm": 20.773145033253595,
      "kl": 0.0279541015625,
      "learning_rate": 9.117647058823529e-07,
      "loss": 0.0011,
      "reward": 1.8575422763824463,
      "reward_std": 0.024342134594917297,
      "rewards/accuracy_reward": 0.8575422763824463,
      "rewards/format_reward": 1.0,
      "step": 120
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 110.59375762939453,
      "epoch": 1.7794117647058822,
      "grad_norm": 1.4987523214554013,
      "kl": 0.0283203125,
      "learning_rate": 9.110294117647058e-07,
      "loss": 0.0011,
      "reward": 1.8683563470840454,
      "reward_std": 0.023474786430597305,
      "rewards/accuracy_reward": 0.8683561682701111,
      "rewards/format_reward": 1.0,
      "step": 121
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 110.4363021850586,
      "epoch": 1.7941176470588234,
      "grad_norm": 2.899436959677707,
      "kl": 0.02783203125,
      "learning_rate": 9.102941176470588e-07,
      "loss": 0.0011,
      "reward": 1.82746160030365,
      "reward_std": 0.027238713577389717,
      "rewards/accuracy_reward": 0.8286634683609009,
      "rewards/format_reward": 0.998798131942749,
      "step": 122
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 109.24279022216797,
      "epoch": 1.8088235294117647,
      "grad_norm": 1.4877031960129163,
      "kl": 0.0286865234375,
      "learning_rate": 9.095588235294118e-07,
      "loss": 0.0011,
      "reward": 1.8483201265335083,
      "reward_std": 0.024529799818992615,
      "rewards/accuracy_reward": 0.8483200669288635,
      "rewards/format_reward": 1.0,
      "step": 123
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 108.35456848144531,
      "epoch": 1.8235294117647058,
      "grad_norm": 2.1149935830252473,
      "kl": 0.0308837890625,
      "learning_rate": 9.088235294117646e-07,
      "loss": 0.0012,
      "reward": 1.8580820560455322,
      "reward_std": 0.026496434584259987,
      "rewards/accuracy_reward": 0.8580819964408875,
      "rewards/format_reward": 1.0,
      "step": 124
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 107.50721740722656,
      "epoch": 1.8382352941176472,
      "grad_norm": 5.027371264695678,
      "kl": 0.031005859375,
      "learning_rate": 9.080882352941176e-07,
      "loss": 0.0012,
      "reward": 1.867140293121338,
      "reward_std": 0.02153668738901615,
      "rewards/accuracy_reward": 0.8671402931213379,
      "rewards/format_reward": 1.0,
      "step": 125
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 105.4675521850586,
      "epoch": 1.8529411764705883,
      "grad_norm": 1.2899741793420323,
      "kl": 0.028076171875,
      "learning_rate": 9.073529411764706e-07,
      "loss": 0.0011,
      "reward": 1.8652453422546387,
      "reward_std": 0.02116318792104721,
      "rewards/accuracy_reward": 0.8652453422546387,
      "rewards/format_reward": 1.0,
      "step": 126
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 106.6502456665039,
      "epoch": 1.8676470588235294,
      "grad_norm": 2.179252488296008,
      "kl": 0.031494140625,
      "learning_rate": 9.066176470588235e-07,
      "loss": 0.0013,
      "reward": 1.8535821437835693,
      "reward_std": 0.02199378050863743,
      "rewards/accuracy_reward": 0.8535822033882141,
      "rewards/format_reward": 1.0,
      "step": 127
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 104.88581848144531,
      "epoch": 1.8823529411764706,
      "grad_norm": 1.4612044453819617,
      "kl": 0.033447265625,
      "learning_rate": 9.058823529411764e-07,
      "loss": 0.0013,
      "reward": 1.8556383848190308,
      "reward_std": 0.018579188734292984,
      "rewards/accuracy_reward": 0.8556383848190308,
      "rewards/format_reward": 1.0,
      "step": 128
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 106.41346740722656,
      "epoch": 1.8970588235294117,
      "grad_norm": 1.5286682380949737,
      "kl": 0.0296630859375,
      "learning_rate": 9.051470588235294e-07,
      "loss": 0.0012,
      "reward": 1.83783757686615,
      "reward_std": 0.025251705199480057,
      "rewards/accuracy_reward": 0.8378376364707947,
      "rewards/format_reward": 1.0,
      "step": 129
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 104.98438262939453,
      "epoch": 1.9117647058823528,
      "grad_norm": 2.4536921352293892,
      "kl": 0.029296875,
      "learning_rate": 9.044117647058823e-07,
      "loss": 0.0012,
      "reward": 1.866936445236206,
      "reward_std": 0.022518165409564972,
      "rewards/accuracy_reward": 0.8669363260269165,
      "rewards/format_reward": 1.0,
      "step": 130
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 105.33534240722656,
      "epoch": 1.9264705882352942,
      "grad_norm": 1.5117857063832663,
      "kl": 0.0284423828125,
      "learning_rate": 9.036764705882353e-07,
      "loss": 0.0011,
      "reward": 1.8531559705734253,
      "reward_std": 0.024603500962257385,
      "rewards/accuracy_reward": 0.8543578386306763,
      "rewards/format_reward": 0.998798131942749,
      "step": 131
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 107.3533706665039,
      "epoch": 1.9411764705882353,
      "grad_norm": 2.703283581674193,
      "kl": 0.0308837890625,
      "learning_rate": 9.029411764705883e-07,
      "loss": 0.0012,
      "reward": 1.8581082820892334,
      "reward_std": 0.02527976781129837,
      "rewards/accuracy_reward": 0.8581082820892334,
      "rewards/format_reward": 1.0,
      "step": 132
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 108.80048370361328,
      "epoch": 1.9558823529411766,
      "grad_norm": 2.173287786867425,
      "kl": 0.029052734375,
      "learning_rate": 9.022058823529411e-07,
      "loss": 0.0012,
      "reward": 1.8408198356628418,
      "reward_std": 0.02469984069466591,
      "rewards/accuracy_reward": 0.8408196568489075,
      "rewards/format_reward": 1.0,
      "step": 133
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 107.5769271850586,
      "epoch": 1.9705882352941178,
      "grad_norm": 1.4534120952263125,
      "kl": 0.0306396484375,
      "learning_rate": 9.014705882352941e-07,
      "loss": 0.0012,
      "reward": 1.828810214996338,
      "reward_std": 0.023025646805763245,
      "rewards/accuracy_reward": 0.8288100361824036,
      "rewards/format_reward": 1.0,
      "step": 134
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 108.14904022216797,
      "epoch": 1.9852941176470589,
      "grad_norm": 2.390365626608761,
      "kl": 0.031494140625,
      "learning_rate": 9.007352941176471e-07,
      "loss": 0.0013,
      "reward": 1.8384226560592651,
      "reward_std": 0.023323819041252136,
      "rewards/accuracy_reward": 0.8384228348731995,
      "rewards/format_reward": 1.0,
      "step": 135
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 109.38813018798828,
      "epoch": 2.0,
      "grad_norm": 1.3167695891772675,
      "kl": 0.033203125,
      "learning_rate": 9e-07,
      "loss": 0.0013,
      "reward": 1.8518694639205933,
      "reward_std": 0.021665828302502632,
      "rewards/accuracy_reward": 0.8518694639205933,
      "rewards/format_reward": 1.0,
      "step": 136
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 108.4158706665039,
      "epoch": 2.014705882352941,
      "grad_norm": 1.1129043277785302,
      "kl": 0.0296630859375,
      "learning_rate": 8.992647058823529e-07,
      "loss": 0.0012,
      "reward": 1.8364787101745605,
      "reward_std": 0.02524709887802601,
      "rewards/accuracy_reward": 0.8364787697792053,
      "rewards/format_reward": 1.0,
      "step": 137
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 110.6189956665039,
      "epoch": 2.0294117647058822,
      "grad_norm": 11.77858401808986,
      "kl": 0.0341796875,
      "learning_rate": 8.985294117647059e-07,
      "loss": 0.0014,
      "reward": 1.8593671321868896,
      "reward_std": 0.02559875324368477,
      "rewards/accuracy_reward": 0.8605691194534302,
      "rewards/format_reward": 0.998798131942749,
      "step": 138
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 108.89063262939453,
      "epoch": 2.0441176470588234,
      "grad_norm": 1.4515857339540483,
      "kl": 0.035400390625,
      "learning_rate": 8.977941176470588e-07,
      "loss": 0.0014,
      "reward": 1.8674319982528687,
      "reward_std": 0.02322627604007721,
      "rewards/accuracy_reward": 0.8674321174621582,
      "rewards/format_reward": 1.0,
      "step": 139
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 111.36058044433594,
      "epoch": 2.0588235294117645,
      "grad_norm": 1.7247871087577205,
      "kl": 0.030517578125,
      "learning_rate": 8.970588235294118e-07,
      "loss": 0.0012,
      "reward": 1.8335715532302856,
      "reward_std": 0.028056330978870392,
      "rewards/accuracy_reward": 0.8347733020782471,
      "rewards/format_reward": 0.998798131942749,
      "step": 140
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 112.83894348144531,
      "epoch": 2.073529411764706,
      "grad_norm": 1.700568612592954,
      "kl": 0.029296875,
      "learning_rate": 8.963235294117648e-07,
      "loss": 0.0012,
      "reward": 1.841809868812561,
      "reward_std": 0.02446519024670124,
      "rewards/accuracy_reward": 0.8418098092079163,
      "rewards/format_reward": 1.0,
      "step": 141
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 113.85577392578125,
      "epoch": 2.088235294117647,
      "grad_norm": 1.9489963115328834,
      "kl": 0.0296630859375,
      "learning_rate": 8.955882352941176e-07,
      "loss": 0.0012,
      "reward": 1.8537096977233887,
      "reward_std": 0.025433821603655815,
      "rewards/accuracy_reward": 0.8537096381187439,
      "rewards/format_reward": 1.0,
      "step": 142
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 115.17788696289062,
      "epoch": 2.1029411764705883,
      "grad_norm": 1.930319438615514,
      "kl": 0.0303955078125,
      "learning_rate": 8.948529411764706e-07,
      "loss": 0.0012,
      "reward": 1.8577303886413574,
      "reward_std": 0.025013966485857964,
      "rewards/accuracy_reward": 0.858932375907898,
      "rewards/format_reward": 0.998798131942749,
      "step": 143
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 115.77404022216797,
      "epoch": 2.1176470588235294,
      "grad_norm": 2.741052616431375,
      "kl": 0.031005859375,
      "learning_rate": 8.941176470588236e-07,
      "loss": 0.0012,
      "reward": 1.8680880069732666,
      "reward_std": 0.025277523323893547,
      "rewards/accuracy_reward": 0.8692898750305176,
      "rewards/format_reward": 0.998798131942749,
      "step": 144
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 116.88581848144531,
      "epoch": 2.1323529411764706,
      "grad_norm": 2.1856417966833543,
      "kl": 0.030029296875,
      "learning_rate": 8.933823529411765e-07,
      "loss": 0.0012,
      "reward": 1.851528286933899,
      "reward_std": 0.024104848504066467,
      "rewards/accuracy_reward": 0.8515281081199646,
      "rewards/format_reward": 1.0,
      "step": 145
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 116.79808044433594,
      "epoch": 2.1470588235294117,
      "grad_norm": 1.7095764168354908,
      "kl": 0.03076171875,
      "learning_rate": 8.926470588235294e-07,
      "loss": 0.0012,
      "reward": 1.8471791744232178,
      "reward_std": 0.028594160452485085,
      "rewards/accuracy_reward": 0.8483811020851135,
      "rewards/format_reward": 0.998798131942749,
      "step": 146
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 115.95673370361328,
      "epoch": 2.161764705882353,
      "grad_norm": 1.8204981844802233,
      "kl": 0.03125,
      "learning_rate": 8.919117647058824e-07,
      "loss": 0.0013,
      "reward": 1.8616198301315308,
      "reward_std": 0.0273142047226429,
      "rewards/accuracy_reward": 0.8628216981887817,
      "rewards/format_reward": 0.998798131942749,
      "step": 147
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 115.73438262939453,
      "epoch": 2.176470588235294,
      "grad_norm": 2.6286179471239906,
      "kl": 0.0299072265625,
      "learning_rate": 8.911764705882353e-07,
      "loss": 0.0012,
      "reward": 1.864734172821045,
      "reward_std": 0.027529988437891006,
      "rewards/accuracy_reward": 0.8647341132164001,
      "rewards/format_reward": 1.0,
      "step": 148
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 116.94471740722656,
      "epoch": 2.1911764705882355,
      "grad_norm": 1.2094663657381695,
      "kl": 0.033447265625,
      "learning_rate": 8.904411764705882e-07,
      "loss": 0.0013,
      "reward": 1.8421967029571533,
      "reward_std": 0.030433619394898415,
      "rewards/accuracy_reward": 0.8446005582809448,
      "rewards/format_reward": 0.9975962042808533,
      "step": 149
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 116.39904022216797,
      "epoch": 2.2058823529411766,
      "grad_norm": 1.9663042039131189,
      "kl": 0.032470703125,
      "learning_rate": 8.89705882352941e-07,
      "loss": 0.0013,
      "reward": 1.8600176572799683,
      "reward_std": 0.024169858545064926,
      "rewards/accuracy_reward": 0.860017716884613,
      "rewards/format_reward": 1.0,
      "step": 150
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 113.9627456665039,
      "epoch": 2.2205882352941178,
      "grad_norm": 2.5776707309950115,
      "kl": 0.02978515625,
      "learning_rate": 8.88970588235294e-07,
      "loss": 0.0012,
      "reward": 1.871897578239441,
      "reward_std": 0.023594854399561882,
      "rewards/accuracy_reward": 0.8718973398208618,
      "rewards/format_reward": 1.0,
      "step": 151
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 113.98678588867188,
      "epoch": 2.235294117647059,
      "grad_norm": 1.5441369185409464,
      "kl": 0.0306396484375,
      "learning_rate": 8.88235294117647e-07,
      "loss": 0.0012,
      "reward": 1.8490618467330933,
      "reward_std": 0.027410874143242836,
      "rewards/accuracy_reward": 0.8502638339996338,
      "rewards/format_reward": 0.998798131942749,
      "step": 152
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 112.43029022216797,
      "epoch": 2.25,
      "grad_norm": 1.8027917380235519,
      "kl": 0.031494140625,
      "learning_rate": 8.874999999999999e-07,
      "loss": 0.0013,
      "reward": 1.8541759252548218,
      "reward_std": 0.024694515392184258,
      "rewards/accuracy_reward": 0.8553777933120728,
      "rewards/format_reward": 0.998798131942749,
      "step": 153
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 111.06491088867188,
      "epoch": 2.264705882352941,
      "grad_norm": 2.271542926940373,
      "kl": 0.0308837890625,
      "learning_rate": 8.867647058823528e-07,
      "loss": 0.0012,
      "reward": 1.8564496040344238,
      "reward_std": 0.023326044902205467,
      "rewards/accuracy_reward": 0.856449544429779,
      "rewards/format_reward": 1.0,
      "step": 154
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 110.58053588867188,
      "epoch": 2.2794117647058822,
      "grad_norm": 1.5375482020791835,
      "kl": 0.032470703125,
      "learning_rate": 8.860294117647058e-07,
      "loss": 0.0013,
      "reward": 1.8531345129013062,
      "reward_std": 0.025648273527622223,
      "rewards/accuracy_reward": 0.8543363809585571,
      "rewards/format_reward": 0.998798131942749,
      "step": 155
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 110.21514892578125,
      "epoch": 2.2941176470588234,
      "grad_norm": 1.6774011174439099,
      "kl": 0.041259765625,
      "learning_rate": 8.852941176470587e-07,
      "loss": 0.0017,
      "reward": 1.8623310327529907,
      "reward_std": 0.024891231209039688,
      "rewards/accuracy_reward": 0.862330973148346,
      "rewards/format_reward": 1.0,
      "step": 156
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 109.63702392578125,
      "epoch": 2.3088235294117645,
      "grad_norm": 1.5805094517168279,
      "kl": 0.03515625,
      "learning_rate": 8.845588235294117e-07,
      "loss": 0.0014,
      "reward": 1.8760714530944824,
      "reward_std": 0.028571614995598793,
      "rewards/accuracy_reward": 0.8784750699996948,
      "rewards/format_reward": 0.9975962042808533,
      "step": 157
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 107.48798370361328,
      "epoch": 2.323529411764706,
      "grad_norm": 2.903562674226702,
      "kl": 0.03466796875,
      "learning_rate": 8.838235294117647e-07,
      "loss": 0.0014,
      "reward": 1.8514492511749268,
      "reward_std": 0.023621495813131332,
      "rewards/accuracy_reward": 0.8514493107795715,
      "rewards/format_reward": 1.0,
      "step": 158
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 106.39543914794922,
      "epoch": 2.338235294117647,
      "grad_norm": 2.257173366564063,
      "kl": 0.035888671875,
      "learning_rate": 8.830882352941175e-07,
      "loss": 0.0014,
      "reward": 1.8729958534240723,
      "reward_std": 0.026356099173426628,
      "rewards/accuracy_reward": 0.8729958534240723,
      "rewards/format_reward": 1.0,
      "step": 159
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 105.02764892578125,
      "epoch": 2.3529411764705883,
      "grad_norm": 1.6931345428419227,
      "kl": 0.047607421875,
      "learning_rate": 8.823529411764705e-07,
      "loss": 0.0019,
      "reward": 1.871155023574829,
      "reward_std": 0.02590741030871868,
      "rewards/accuracy_reward": 0.8723568916320801,
      "rewards/format_reward": 0.998798131942749,
      "step": 160
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 104.74519348144531,
      "epoch": 2.3676470588235294,
      "grad_norm": 2.3920729679008232,
      "kl": 0.03955078125,
      "learning_rate": 8.816176470588235e-07,
      "loss": 0.0016,
      "reward": 1.870141863822937,
      "reward_std": 0.025313733145594597,
      "rewards/accuracy_reward": 0.870141863822937,
      "rewards/format_reward": 1.0,
      "step": 161
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 101.9831771850586,
      "epoch": 2.3823529411764706,
      "grad_norm": 1.7968109427756567,
      "kl": 0.03955078125,
      "learning_rate": 8.808823529411764e-07,
      "loss": 0.0016,
      "reward": 1.8542630672454834,
      "reward_std": 0.025294849649071693,
      "rewards/accuracy_reward": 0.8542630672454834,
      "rewards/format_reward": 1.0,
      "step": 162
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 100.81731414794922,
      "epoch": 2.3970588235294117,
      "grad_norm": 3.2616152503315536,
      "kl": 0.040771484375,
      "learning_rate": 8.801470588235293e-07,
      "loss": 0.0016,
      "reward": 1.8570067882537842,
      "reward_std": 0.02526533231139183,
      "rewards/accuracy_reward": 0.8570066094398499,
      "rewards/format_reward": 1.0,
      "step": 163
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 100.97716522216797,
      "epoch": 2.411764705882353,
      "grad_norm": 1.8975512236370329,
      "kl": 0.03955078125,
      "learning_rate": 8.794117647058823e-07,
      "loss": 0.0016,
      "reward": 1.8464734554290771,
      "reward_std": 0.02490406110882759,
      "rewards/accuracy_reward": 0.8464734554290771,
      "rewards/format_reward": 1.0,
      "step": 164
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 101.0769271850586,
      "epoch": 2.426470588235294,
      "grad_norm": 3.1230913633145096,
      "kl": 0.04931640625,
      "learning_rate": 8.786764705882352e-07,
      "loss": 0.002,
      "reward": 1.8685623407363892,
      "reward_std": 0.024073276668787003,
      "rewards/accuracy_reward": 0.8685624599456787,
      "rewards/format_reward": 1.0,
      "step": 165
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 101.92668914794922,
      "epoch": 2.4411764705882355,
      "grad_norm": 2.501654602692275,
      "kl": 0.04296875,
      "learning_rate": 8.779411764705882e-07,
      "loss": 0.0017,
      "reward": 1.8564651012420654,
      "reward_std": 0.024265745654702187,
      "rewards/accuracy_reward": 0.8564649820327759,
      "rewards/format_reward": 1.0,
      "step": 166
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 101.86538696289062,
      "epoch": 2.4558823529411766,
      "grad_norm": 6.016419703798391,
      "kl": 0.0439453125,
      "learning_rate": 8.772058823529412e-07,
      "loss": 0.0018,
      "reward": 1.8798558712005615,
      "reward_std": 0.021158261224627495,
      "rewards/accuracy_reward": 0.8798558712005615,
      "rewards/format_reward": 1.0,
      "step": 167
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 105.84014892578125,
      "epoch": 2.4705882352941178,
      "grad_norm": 1.9371757598853314,
      "kl": 0.04052734375,
      "learning_rate": 8.76470588235294e-07,
      "loss": 0.0016,
      "reward": 1.882270336151123,
      "reward_std": 0.023024968802928925,
      "rewards/accuracy_reward": 0.882270336151123,
      "rewards/format_reward": 1.0,
      "step": 168
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 103.30529022216797,
      "epoch": 2.485294117647059,
      "grad_norm": 1.4411287692419261,
      "kl": 0.036376953125,
      "learning_rate": 8.75735294117647e-07,
      "loss": 0.0015,
      "reward": 1.862653374671936,
      "reward_std": 0.021055346354842186,
      "rewards/accuracy_reward": 0.8626533150672913,
      "rewards/format_reward": 1.0,
      "step": 169
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 105.5877456665039,
      "epoch": 2.5,
      "grad_norm": 1.4740075609659171,
      "kl": 0.03857421875,
      "learning_rate": 8.75e-07,
      "loss": 0.0015,
      "reward": 1.8584967851638794,
      "reward_std": 0.02370167151093483,
      "rewards/accuracy_reward": 0.8584967255592346,
      "rewards/format_reward": 1.0,
      "step": 170
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 104.28245544433594,
      "epoch": 2.514705882352941,
      "grad_norm": 1.3370741346003325,
      "kl": 0.03662109375,
      "learning_rate": 8.742647058823529e-07,
      "loss": 0.0015,
      "reward": 1.8700354099273682,
      "reward_std": 0.023715250194072723,
      "rewards/accuracy_reward": 0.8700354099273682,
      "rewards/format_reward": 1.0,
      "step": 171
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 103.4050521850586,
      "epoch": 2.5294117647058822,
      "grad_norm": 2.6769462130534345,
      "kl": 0.03369140625,
      "learning_rate": 8.735294117647058e-07,
      "loss": 0.0013,
      "reward": 1.8732657432556152,
      "reward_std": 0.01899472437798977,
      "rewards/accuracy_reward": 0.8732655644416809,
      "rewards/format_reward": 1.0,
      "step": 172
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 104.87741088867188,
      "epoch": 2.5441176470588234,
      "grad_norm": 2.5769454874734152,
      "kl": 0.0322265625,
      "learning_rate": 8.727941176470588e-07,
      "loss": 0.0013,
      "reward": 1.8616483211517334,
      "reward_std": 0.02438032068312168,
      "rewards/accuracy_reward": 0.8616481423377991,
      "rewards/format_reward": 1.0,
      "step": 173
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 102.12259674072266,
      "epoch": 2.5588235294117645,
      "grad_norm": 1.8562667369690105,
      "kl": 0.033935546875,
      "learning_rate": 8.720588235294117e-07,
      "loss": 0.0013,
      "reward": 1.855645775794983,
      "reward_std": 0.02895743027329445,
      "rewards/accuracy_reward": 0.8580495715141296,
      "rewards/format_reward": 0.9975962042808533,
      "step": 174
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 104.21154022216797,
      "epoch": 2.5735294117647056,
      "grad_norm": 2.558018085056357,
      "kl": 0.034423828125,
      "learning_rate": 8.713235294117647e-07,
      "loss": 0.0014,
      "reward": 1.884694218635559,
      "reward_std": 0.026144152507185936,
      "rewards/accuracy_reward": 0.8858961462974548,
      "rewards/format_reward": 0.998798131942749,
      "step": 175
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 104.8894271850586,
      "epoch": 2.588235294117647,
      "grad_norm": 4.093759859461043,
      "kl": 0.0322265625,
      "learning_rate": 8.705882352941177e-07,
      "loss": 0.0013,
      "reward": 1.859119176864624,
      "reward_std": 0.023388409987092018,
      "rewards/accuracy_reward": 0.859119176864624,
      "rewards/format_reward": 1.0,
      "step": 176
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 105.76923370361328,
      "epoch": 2.6029411764705883,
      "grad_norm": 2.565372217350874,
      "kl": 0.039794921875,
      "learning_rate": 8.698529411764705e-07,
      "loss": 0.0016,
      "reward": 1.8814976215362549,
      "reward_std": 0.024835962802171707,
      "rewards/accuracy_reward": 0.8826994895935059,
      "rewards/format_reward": 0.998798131942749,
      "step": 177
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 106.4158706665039,
      "epoch": 2.6176470588235294,
      "grad_norm": 1.546586759316415,
      "kl": 0.03662109375,
      "learning_rate": 8.691176470588235e-07,
      "loss": 0.0015,
      "reward": 1.8796207904815674,
      "reward_std": 0.022139329463243484,
      "rewards/accuracy_reward": 0.8796206712722778,
      "rewards/format_reward": 1.0,
      "step": 178
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 105.92668914794922,
      "epoch": 2.6323529411764706,
      "grad_norm": 1.5303977681102032,
      "kl": 0.03369140625,
      "learning_rate": 8.683823529411765e-07,
      "loss": 0.0014,
      "reward": 1.8701624870300293,
      "reward_std": 0.025447282940149307,
      "rewards/accuracy_reward": 0.8701622486114502,
      "rewards/format_reward": 1.0,
      "step": 179
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 107.36659240722656,
      "epoch": 2.6470588235294117,
      "grad_norm": 1.518878529368709,
      "kl": 0.033203125,
      "learning_rate": 8.676470588235294e-07,
      "loss": 0.0013,
      "reward": 1.8654332160949707,
      "reward_std": 0.026392469182610512,
      "rewards/accuracy_reward": 0.8654330372810364,
      "rewards/format_reward": 1.0,
      "step": 180
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 110.12620544433594,
      "epoch": 2.661764705882353,
      "grad_norm": 1.403191410442849,
      "kl": 0.03173828125,
      "learning_rate": 8.669117647058823e-07,
      "loss": 0.0013,
      "reward": 1.86957585811615,
      "reward_std": 0.025720389559864998,
      "rewards/accuracy_reward": 0.8695757985115051,
      "rewards/format_reward": 1.0,
      "step": 181
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 109.7283706665039,
      "epoch": 2.6764705882352944,
      "grad_norm": 2.022045986077511,
      "kl": 0.031494140625,
      "learning_rate": 8.661764705882353e-07,
      "loss": 0.0013,
      "reward": 1.8482249975204468,
      "reward_std": 0.03184087574481964,
      "rewards/accuracy_reward": 0.8494267463684082,
      "rewards/format_reward": 0.998798131942749,
      "step": 182
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 111.31491088867188,
      "epoch": 2.6911764705882355,
      "grad_norm": 1.4526871175791842,
      "kl": 0.043701171875,
      "learning_rate": 8.654411764705882e-07,
      "loss": 0.0017,
      "reward": 1.8396512269973755,
      "reward_std": 0.02649850584566593,
      "rewards/accuracy_reward": 0.8396511673927307,
      "rewards/format_reward": 1.0,
      "step": 183
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 113.32933044433594,
      "epoch": 2.7058823529411766,
      "grad_norm": 1.962569471145186,
      "kl": 0.034423828125,
      "learning_rate": 8.647058823529412e-07,
      "loss": 0.0014,
      "reward": 1.864058017730713,
      "reward_std": 0.026860656216740608,
      "rewards/accuracy_reward": 0.8640578389167786,
      "rewards/format_reward": 1.0,
      "step": 184
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 113.7331771850586,
      "epoch": 2.7205882352941178,
      "grad_norm": 2.2484467422992944,
      "kl": 0.03369140625,
      "learning_rate": 8.639705882352941e-07,
      "loss": 0.0013,
      "reward": 1.8754035234451294,
      "reward_std": 0.024205412715673447,
      "rewards/accuracy_reward": 0.8754035234451294,
      "rewards/format_reward": 1.0,
      "step": 185
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 113.37860870361328,
      "epoch": 2.735294117647059,
      "grad_norm": 1.997975820505161,
      "kl": 0.033935546875,
      "learning_rate": 8.63235294117647e-07,
      "loss": 0.0014,
      "reward": 1.860775351524353,
      "reward_std": 0.02340296283364296,
      "rewards/accuracy_reward": 0.8607752919197083,
      "rewards/format_reward": 1.0,
      "step": 186
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 114.65866088867188,
      "epoch": 2.75,
      "grad_norm": 1.7044143261510771,
      "kl": 0.0341796875,
      "learning_rate": 8.625e-07,
      "loss": 0.0014,
      "reward": 1.8710038661956787,
      "reward_std": 0.026455307379364967,
      "rewards/accuracy_reward": 0.8722056746482849,
      "rewards/format_reward": 0.998798131942749,
      "step": 187
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 113.46514892578125,
      "epoch": 2.764705882352941,
      "grad_norm": 1.8000725865597647,
      "kl": 0.038330078125,
      "learning_rate": 8.61764705882353e-07,
      "loss": 0.0015,
      "reward": 1.8618617057800293,
      "reward_std": 0.03181543946266174,
      "rewards/accuracy_reward": 0.8642653822898865,
      "rewards/format_reward": 0.9975962042808533,
      "step": 188
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 109.53125762939453,
      "epoch": 2.7794117647058822,
      "grad_norm": 2.339206486710046,
      "kl": 0.0322265625,
      "learning_rate": 8.610294117647059e-07,
      "loss": 0.0013,
      "reward": 1.875985860824585,
      "reward_std": 0.02267822064459324,
      "rewards/accuracy_reward": 0.875985860824585,
      "rewards/format_reward": 1.0,
      "step": 189
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 109.21634674072266,
      "epoch": 2.7941176470588234,
      "grad_norm": 1.4965339815672982,
      "kl": 0.033447265625,
      "learning_rate": 8.602941176470588e-07,
      "loss": 0.0013,
      "reward": 1.8847665786743164,
      "reward_std": 0.023552460595965385,
      "rewards/accuracy_reward": 0.8847665190696716,
      "rewards/format_reward": 1.0,
      "step": 190
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 106.6238021850586,
      "epoch": 2.8088235294117645,
      "grad_norm": 2.3107985557542774,
      "kl": 0.03369140625,
      "learning_rate": 8.595588235294118e-07,
      "loss": 0.0014,
      "reward": 1.8373372554779053,
      "reward_std": 0.02436283603310585,
      "rewards/accuracy_reward": 0.8373372554779053,
      "rewards/format_reward": 1.0,
      "step": 191
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 106.30409240722656,
      "epoch": 2.8235294117647056,
      "grad_norm": 1.92475811318894,
      "kl": 0.0361328125,
      "learning_rate": 8.588235294117646e-07,
      "loss": 0.0014,
      "reward": 1.8702090978622437,
      "reward_std": 0.026847295463085175,
      "rewards/accuracy_reward": 0.8714108467102051,
      "rewards/format_reward": 0.998798131942749,
      "step": 192
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 104.48197174072266,
      "epoch": 2.838235294117647,
      "grad_norm": 2.0449651951501866,
      "kl": 0.034423828125,
      "learning_rate": 8.580882352941176e-07,
      "loss": 0.0014,
      "reward": 1.843615174293518,
      "reward_std": 0.02285837009549141,
      "rewards/accuracy_reward": 0.8436151742935181,
      "rewards/format_reward": 1.0,
      "step": 193
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 104.01202392578125,
      "epoch": 2.8529411764705883,
      "grad_norm": 1.2736084553332587,
      "kl": 0.036865234375,
      "learning_rate": 8.573529411764705e-07,
      "loss": 0.0015,
      "reward": 1.8695170879364014,
      "reward_std": 0.022367505356669426,
      "rewards/accuracy_reward": 0.8695171475410461,
      "rewards/format_reward": 1.0,
      "step": 194
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 104.91827392578125,
      "epoch": 2.8676470588235294,
      "grad_norm": 1.9579889152211967,
      "kl": 0.047119140625,
      "learning_rate": 8.566176470588234e-07,
      "loss": 0.0019,
      "reward": 1.8573535680770874,
      "reward_std": 0.026318097487092018,
      "rewards/accuracy_reward": 0.8585554361343384,
      "rewards/format_reward": 0.998798131942749,
      "step": 195
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 105.89543914794922,
      "epoch": 2.8823529411764706,
      "grad_norm": 1.6397435637270048,
      "kl": 0.03173828125,
      "learning_rate": 8.558823529411764e-07,
      "loss": 0.0013,
      "reward": 1.8664311170578003,
      "reward_std": 0.02275273948907852,
      "rewards/accuracy_reward": 0.8664309978485107,
      "rewards/format_reward": 1.0,
      "step": 196
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 106.19351196289062,
      "epoch": 2.8970588235294117,
      "grad_norm": 1.3453321422064775,
      "kl": 0.03271484375,
      "learning_rate": 8.551470588235294e-07,
      "loss": 0.0013,
      "reward": 1.8688963651657104,
      "reward_std": 0.02163490653038025,
      "rewards/accuracy_reward": 0.8688963651657104,
      "rewards/format_reward": 1.0,
      "step": 197
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 107.34495544433594,
      "epoch": 2.911764705882353,
      "grad_norm": 1.6994019668289626,
      "kl": 0.034423828125,
      "learning_rate": 8.544117647058822e-07,
      "loss": 0.0014,
      "reward": 1.8594874143600464,
      "reward_std": 0.025675183162093163,
      "rewards/accuracy_reward": 0.8606892824172974,
      "rewards/format_reward": 0.998798131942749,
      "step": 198
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 106.97956848144531,
      "epoch": 2.9264705882352944,
      "grad_norm": 1.443393412750695,
      "kl": 0.033935546875,
      "learning_rate": 8.536764705882352e-07,
      "loss": 0.0014,
      "reward": 1.8734104633331299,
      "reward_std": 0.02485453523695469,
      "rewards/accuracy_reward": 0.8734104633331299,
      "rewards/format_reward": 1.0,
      "step": 199
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 111.54447174072266,
      "epoch": 2.9411764705882355,
      "grad_norm": 1.3185192214107238,
      "kl": 0.0322265625,
      "learning_rate": 8.529411764705882e-07,
      "loss": 0.0013,
      "reward": 1.8638544082641602,
      "reward_std": 0.024562494829297066,
      "rewards/accuracy_reward": 0.8650562167167664,
      "rewards/format_reward": 0.998798131942749,
      "step": 200
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 111.21995544433594,
      "epoch": 2.9558823529411766,
      "grad_norm": 1.9943222079885201,
      "kl": 0.032470703125,
      "learning_rate": 8.522058823529411e-07,
      "loss": 0.0013,
      "reward": 1.8713494539260864,
      "reward_std": 0.027820177376270294,
      "rewards/accuracy_reward": 0.8725512623786926,
      "rewards/format_reward": 0.998798131942749,
      "step": 201
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 110.25000762939453,
      "epoch": 2.9705882352941178,
      "grad_norm": 1.3647685703194712,
      "kl": 0.033203125,
      "learning_rate": 8.514705882352941e-07,
      "loss": 0.0013,
      "reward": 1.8598443269729614,
      "reward_std": 0.025986719876527786,
      "rewards/accuracy_reward": 0.8598442673683167,
      "rewards/format_reward": 1.0,
      "step": 202
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 111.90625762939453,
      "epoch": 2.985294117647059,
      "grad_norm": 1.6126487460912005,
      "kl": 0.037841796875,
      "learning_rate": 8.50735294117647e-07,
      "loss": 0.0015,
      "reward": 1.8783414363861084,
      "reward_std": 0.023664260283112526,
      "rewards/accuracy_reward": 0.8783413171768188,
      "rewards/format_reward": 1.0,
      "step": 203
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 111.15220642089844,
      "epoch": 3.0,
      "grad_norm": 1.780161633176048,
      "kl": 0.034423828125,
      "learning_rate": 8.499999999999999e-07,
      "loss": 0.0014,
      "reward": 1.8632652759552002,
      "reward_std": 0.02758597768843174,
      "rewards/accuracy_reward": 0.8647872805595398,
      "rewards/format_reward": 0.9984779357910156,
      "step": 204
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 113.47956848144531,
      "epoch": 3.014705882352941,
      "grad_norm": 1.9805220919990834,
      "kl": 0.03271484375,
      "learning_rate": 8.492647058823529e-07,
      "loss": 0.0013,
      "reward": 1.8707077503204346,
      "reward_std": 0.023023683577775955,
      "rewards/accuracy_reward": 0.8707077503204346,
      "rewards/format_reward": 1.0,
      "step": 205
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 114.53366088867188,
      "epoch": 3.0294117647058822,
      "grad_norm": 1.759835112928017,
      "kl": 0.03271484375,
      "learning_rate": 8.485294117647059e-07,
      "loss": 0.0013,
      "reward": 1.8758447170257568,
      "reward_std": 0.02334214560687542,
      "rewards/accuracy_reward": 0.8758447170257568,
      "rewards/format_reward": 1.0,
      "step": 206
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 115.1346206665039,
      "epoch": 3.0441176470588234,
      "grad_norm": 1.9299225316519943,
      "kl": 0.032958984375,
      "learning_rate": 8.477941176470587e-07,
      "loss": 0.0013,
      "reward": 1.8904192447662354,
      "reward_std": 0.021329201757907867,
      "rewards/accuracy_reward": 0.8904191255569458,
      "rewards/format_reward": 1.0,
      "step": 207
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 114.8221206665039,
      "epoch": 3.0588235294117645,
      "grad_norm": 1.5162706953943,
      "kl": 0.03759765625,
      "learning_rate": 8.470588235294117e-07,
      "loss": 0.0015,
      "reward": 1.8666112422943115,
      "reward_std": 0.02535596303641796,
      "rewards/accuracy_reward": 0.8666113615036011,
      "rewards/format_reward": 1.0,
      "step": 208
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 115.12019348144531,
      "epoch": 3.073529411764706,
      "grad_norm": 3.4426286843801193,
      "kl": 0.08642578125,
      "learning_rate": 8.463235294117647e-07,
      "loss": 0.0035,
      "reward": 1.8811273574829102,
      "reward_std": 0.022378716617822647,
      "rewards/accuracy_reward": 0.8811271786689758,
      "rewards/format_reward": 1.0,
      "step": 209
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 117.17909240722656,
      "epoch": 3.088235294117647,
      "grad_norm": 1.7571727846632814,
      "kl": 0.0380859375,
      "learning_rate": 8.455882352941176e-07,
      "loss": 0.0015,
      "reward": 1.8708646297454834,
      "reward_std": 0.027770236134529114,
      "rewards/accuracy_reward": 0.8732684254646301,
      "rewards/format_reward": 0.9975962042808533,
      "step": 210
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 117.16947174072266,
      "epoch": 3.1029411764705883,
      "grad_norm": 1.336840943007517,
      "kl": 0.041259765625,
      "learning_rate": 8.448529411764706e-07,
      "loss": 0.0016,
      "reward": 1.8811522722244263,
      "reward_std": 0.02304122783243656,
      "rewards/accuracy_reward": 0.8811522722244263,
      "rewards/format_reward": 1.0,
      "step": 211
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 116.03846740722656,
      "epoch": 3.1176470588235294,
      "grad_norm": 1.8533547410757114,
      "kl": 0.0400390625,
      "learning_rate": 8.441176470588235e-07,
      "loss": 0.0016,
      "reward": 1.8757656812667847,
      "reward_std": 0.027794932946562767,
      "rewards/accuracy_reward": 0.8769674897193909,
      "rewards/format_reward": 0.998798131942749,
      "step": 212
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 117.97356414794922,
      "epoch": 3.1323529411764706,
      "grad_norm": 1.2600737914736126,
      "kl": 0.0380859375,
      "learning_rate": 8.433823529411764e-07,
      "loss": 0.0015,
      "reward": 1.8561041355133057,
      "reward_std": 0.031202280893921852,
      "rewards/accuracy_reward": 0.8585077524185181,
      "rewards/format_reward": 0.9975962042808533,
      "step": 213
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 117.35938262939453,
      "epoch": 3.1470588235294117,
      "grad_norm": 1.7979464745878548,
      "kl": 0.043701171875,
      "learning_rate": 8.426470588235294e-07,
      "loss": 0.0017,
      "reward": 1.8608158826828003,
      "reward_std": 0.026735760271549225,
      "rewards/accuracy_reward": 0.8608157634735107,
      "rewards/format_reward": 1.0,
      "step": 214
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 115.00241088867188,
      "epoch": 3.161764705882353,
      "grad_norm": 5.398535880427292,
      "kl": 0.05322265625,
      "learning_rate": 8.419117647058824e-07,
      "loss": 0.0021,
      "reward": 1.8839178085327148,
      "reward_std": 0.02646607905626297,
      "rewards/accuracy_reward": 0.8851196765899658,
      "rewards/format_reward": 0.998798131942749,
      "step": 215
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 116.87620544433594,
      "epoch": 3.176470588235294,
      "grad_norm": 1.6689071059712295,
      "kl": 0.041748046875,
      "learning_rate": 8.411764705882352e-07,
      "loss": 0.0017,
      "reward": 1.8544951677322388,
      "reward_std": 0.02808697521686554,
      "rewards/accuracy_reward": 0.854495108127594,
      "rewards/format_reward": 1.0,
      "step": 216
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 113.43029022216797,
      "epoch": 3.1911764705882355,
      "grad_norm": 1.6258182876226766,
      "kl": 0.04296875,
      "learning_rate": 8.404411764705882e-07,
      "loss": 0.0017,
      "reward": 1.8663568496704102,
      "reward_std": 0.029328130185604095,
      "rewards/accuracy_reward": 0.8675586581230164,
      "rewards/format_reward": 0.998798131942749,
      "step": 217
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 114.61418914794922,
      "epoch": 3.2058823529411766,
      "grad_norm": 1.4035611828729841,
      "kl": 0.040771484375,
      "learning_rate": 8.397058823529412e-07,
      "loss": 0.0016,
      "reward": 1.8575928211212158,
      "reward_std": 0.02494001016020775,
      "rewards/accuracy_reward": 0.8575928211212158,
      "rewards/format_reward": 1.0,
      "step": 218
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 112.87620544433594,
      "epoch": 3.2205882352941178,
      "grad_norm": 1.5589742726291336,
      "kl": 0.0419921875,
      "learning_rate": 8.389705882352941e-07,
      "loss": 0.0017,
      "reward": 1.8885266780853271,
      "reward_std": 0.023614982143044472,
      "rewards/accuracy_reward": 0.8897286653518677,
      "rewards/format_reward": 0.998798131942749,
      "step": 219
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 109.81851196289062,
      "epoch": 3.235294117647059,
      "grad_norm": 1.385282340462598,
      "kl": 0.041259765625,
      "learning_rate": 8.38235294117647e-07,
      "loss": 0.0017,
      "reward": 1.855526328086853,
      "reward_std": 0.030946535989642143,
      "rewards/accuracy_reward": 0.8567281365394592,
      "rewards/format_reward": 0.998798131942749,
      "step": 220
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 106.7439956665039,
      "epoch": 3.25,
      "grad_norm": 2.962059838231098,
      "kl": 0.044189453125,
      "learning_rate": 8.375e-07,
      "loss": 0.0018,
      "reward": 1.855432152748108,
      "reward_std": 0.02485704980790615,
      "rewards/accuracy_reward": 0.8554319739341736,
      "rewards/format_reward": 1.0,
      "step": 221
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 103.8533706665039,
      "epoch": 3.264705882352941,
      "grad_norm": 1.3569137376665106,
      "kl": 0.04296875,
      "learning_rate": 8.367647058823529e-07,
      "loss": 0.0017,
      "reward": 1.8471739292144775,
      "reward_std": 0.028255922719836235,
      "rewards/accuracy_reward": 0.8483757972717285,
      "rewards/format_reward": 0.998798131942749,
      "step": 222
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 102.22956848144531,
      "epoch": 3.2794117647058822,
      "grad_norm": 1.5185134752025757,
      "kl": 0.04345703125,
      "learning_rate": 8.360294117647059e-07,
      "loss": 0.0017,
      "reward": 1.8899589776992798,
      "reward_std": 0.02284637838602066,
      "rewards/accuracy_reward": 0.889958918094635,
      "rewards/format_reward": 1.0,
      "step": 223
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 99.96394348144531,
      "epoch": 3.2941176470588234,
      "grad_norm": 2.4678802175191117,
      "kl": 0.0458984375,
      "learning_rate": 8.352941176470589e-07,
      "loss": 0.0018,
      "reward": 1.8883994817733765,
      "reward_std": 0.021819431334733963,
      "rewards/accuracy_reward": 0.8883993625640869,
      "rewards/format_reward": 1.0,
      "step": 224
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 97.47356414794922,
      "epoch": 3.3088235294117645,
      "grad_norm": 1.7983286178946878,
      "kl": 0.04736328125,
      "learning_rate": 8.345588235294117e-07,
      "loss": 0.0019,
      "reward": 1.8640559911727905,
      "reward_std": 0.023397305980324745,
      "rewards/accuracy_reward": 0.8640559315681458,
      "rewards/format_reward": 1.0,
      "step": 225
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 97.4675521850586,
      "epoch": 3.323529411764706,
      "grad_norm": 1.8613936498025156,
      "kl": 0.04736328125,
      "learning_rate": 8.338235294117647e-07,
      "loss": 0.0019,
      "reward": 1.8887497186660767,
      "reward_std": 0.022856805473566055,
      "rewards/accuracy_reward": 0.8887495994567871,
      "rewards/format_reward": 1.0,
      "step": 226
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 97.93029022216797,
      "epoch": 3.338235294117647,
      "grad_norm": 1.5411506435438795,
      "kl": 0.045654296875,
      "learning_rate": 8.330882352941177e-07,
      "loss": 0.0018,
      "reward": 1.8804612159729004,
      "reward_std": 0.02065611444413662,
      "rewards/accuracy_reward": 0.8804610371589661,
      "rewards/format_reward": 1.0,
      "step": 227
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 97.00601196289062,
      "epoch": 3.3529411764705883,
      "grad_norm": 2.851058373606194,
      "kl": 0.048583984375,
      "learning_rate": 8.323529411764706e-07,
      "loss": 0.0019,
      "reward": 1.8770747184753418,
      "reward_std": 0.0234384723007679,
      "rewards/accuracy_reward": 0.8770747184753418,
      "rewards/format_reward": 1.0,
      "step": 228
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 97.92909240722656,
      "epoch": 3.3676470588235294,
      "grad_norm": 1.437225684861633,
      "kl": 0.04345703125,
      "learning_rate": 8.316176470588235e-07,
      "loss": 0.0017,
      "reward": 1.8713281154632568,
      "reward_std": 0.02329498529434204,
      "rewards/accuracy_reward": 0.8713279962539673,
      "rewards/format_reward": 1.0,
      "step": 229
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 97.06610870361328,
      "epoch": 3.3823529411764706,
      "grad_norm": 1.9896299970328106,
      "kl": 0.04296875,
      "learning_rate": 8.308823529411765e-07,
      "loss": 0.0017,
      "reward": 1.878764271736145,
      "reward_std": 0.025371139869093895,
      "rewards/accuracy_reward": 0.8787642121315002,
      "rewards/format_reward": 1.0,
      "step": 230
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 98.38341522216797,
      "epoch": 3.3970588235294117,
      "grad_norm": 2.234055405089306,
      "kl": 0.04345703125,
      "learning_rate": 8.301470588235294e-07,
      "loss": 0.0017,
      "reward": 1.898535132408142,
      "reward_std": 0.02297838404774666,
      "rewards/accuracy_reward": 0.8985350131988525,
      "rewards/format_reward": 1.0,
      "step": 231
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 98.45072174072266,
      "epoch": 3.411764705882353,
      "grad_norm": 1.776208061156291,
      "kl": 0.04052734375,
      "learning_rate": 8.294117647058824e-07,
      "loss": 0.0016,
      "reward": 1.873674750328064,
      "reward_std": 0.022656356915831566,
      "rewards/accuracy_reward": 0.8736746907234192,
      "rewards/format_reward": 1.0,
      "step": 232
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 102.99038696289062,
      "epoch": 3.426470588235294,
      "grad_norm": 1.5187065914939428,
      "kl": 0.0390625,
      "learning_rate": 8.286764705882354e-07,
      "loss": 0.0016,
      "reward": 1.8770731687545776,
      "reward_std": 0.02269612066447735,
      "rewards/accuracy_reward": 0.8770731091499329,
      "rewards/format_reward": 1.0,
      "step": 233
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 105.18389892578125,
      "epoch": 3.4411764705882355,
      "grad_norm": 1.6681947750734276,
      "kl": 0.040771484375,
      "learning_rate": 8.279411764705881e-07,
      "loss": 0.0016,
      "reward": 1.8845463991165161,
      "reward_std": 0.02028636820614338,
      "rewards/accuracy_reward": 0.8845462203025818,
      "rewards/format_reward": 1.0,
      "step": 234
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 107.5613021850586,
      "epoch": 3.4558823529411766,
      "grad_norm": 1.4760305455453029,
      "kl": 0.03759765625,
      "learning_rate": 8.272058823529411e-07,
      "loss": 0.0015,
      "reward": 1.890356183052063,
      "reward_std": 0.02070949226617813,
      "rewards/accuracy_reward": 0.8903560042381287,
      "rewards/format_reward": 1.0,
      "step": 235
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 109.21394348144531,
      "epoch": 3.4705882352941178,
      "grad_norm": 16.169977810311337,
      "kl": 0.03759765625,
      "learning_rate": 8.264705882352941e-07,
      "loss": 0.0015,
      "reward": 1.8766230344772339,
      "reward_std": 0.02773961052298546,
      "rewards/accuracy_reward": 0.8778248429298401,
      "rewards/format_reward": 0.998798131942749,
      "step": 236
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 114.24759674072266,
      "epoch": 3.485294117647059,
      "grad_norm": 6.784504465196233,
      "kl": 0.046630859375,
      "learning_rate": 8.25735294117647e-07,
      "loss": 0.0019,
      "reward": 1.8772554397583008,
      "reward_std": 0.022827137261629105,
      "rewards/accuracy_reward": 0.8772553205490112,
      "rewards/format_reward": 1.0,
      "step": 237
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 115.28485870361328,
      "epoch": 3.5,
      "grad_norm": 1.5048421325605899,
      "kl": 0.040283203125,
      "learning_rate": 8.249999999999999e-07,
      "loss": 0.0016,
      "reward": 1.8868188858032227,
      "reward_std": 0.022883420810103416,
      "rewards/accuracy_reward": 0.8868188261985779,
      "rewards/format_reward": 1.0,
      "step": 238
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 115.67428588867188,
      "epoch": 3.514705882352941,
      "grad_norm": 1.9797082947474498,
      "kl": 0.041259765625,
      "learning_rate": 8.242647058823528e-07,
      "loss": 0.0017,
      "reward": 1.8824166059494019,
      "reward_std": 0.025011856108903885,
      "rewards/accuracy_reward": 0.8824164867401123,
      "rewards/format_reward": 1.0,
      "step": 239
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 115.68991088867188,
      "epoch": 3.5294117647058822,
      "grad_norm": 1.3808876653722937,
      "kl": 0.038818359375,
      "learning_rate": 8.235294117647058e-07,
      "loss": 0.0016,
      "reward": 1.8804829120635986,
      "reward_std": 0.019605407491326332,
      "rewards/accuracy_reward": 0.8804829716682434,
      "rewards/format_reward": 1.0,
      "step": 240
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 115.4675521850586,
      "epoch": 3.5441176470588234,
      "grad_norm": 2.2656927922991352,
      "kl": 0.03955078125,
      "learning_rate": 8.227941176470588e-07,
      "loss": 0.0016,
      "reward": 1.8745205402374268,
      "reward_std": 0.019595161080360413,
      "rewards/accuracy_reward": 0.8745204210281372,
      "rewards/format_reward": 1.0,
      "step": 241
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 117.08293914794922,
      "epoch": 3.5588235294117645,
      "grad_norm": 2.2969538457560796,
      "kl": 0.040771484375,
      "learning_rate": 8.220588235294116e-07,
      "loss": 0.0016,
      "reward": 1.8878697156906128,
      "reward_std": 0.02017536200582981,
      "rewards/accuracy_reward": 0.8878695964813232,
      "rewards/format_reward": 1.0,
      "step": 242
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 116.17788696289062,
      "epoch": 3.5735294117647056,
      "grad_norm": 1.3250386411482078,
      "kl": 0.036865234375,
      "learning_rate": 8.213235294117646e-07,
      "loss": 0.0015,
      "reward": 1.882023572921753,
      "reward_std": 0.019443001598119736,
      "rewards/accuracy_reward": 0.8820234537124634,
      "rewards/format_reward": 1.0,
      "step": 243
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 116.6863021850586,
      "epoch": 3.588235294117647,
      "grad_norm": 1.666209048217596,
      "kl": 0.041259765625,
      "learning_rate": 8.205882352941176e-07,
      "loss": 0.0016,
      "reward": 1.8930561542510986,
      "reward_std": 0.023166730999946594,
      "rewards/accuracy_reward": 0.8942580223083496,
      "rewards/format_reward": 0.998798131942749,
      "step": 244
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 113.91947174072266,
      "epoch": 3.6029411764705883,
      "grad_norm": 8.010516964998228,
      "kl": 0.0419921875,
      "learning_rate": 8.198529411764705e-07,
      "loss": 0.0017,
      "reward": 1.8905775547027588,
      "reward_std": 0.01961654983460903,
      "rewards/accuracy_reward": 0.890577495098114,
      "rewards/format_reward": 1.0,
      "step": 245
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 113.99759674072266,
      "epoch": 3.6176470588235294,
      "grad_norm": 2.0494021896720236,
      "kl": 0.07470703125,
      "learning_rate": 8.191176470588235e-07,
      "loss": 0.003,
      "reward": 1.8533637523651123,
      "reward_std": 0.022241346538066864,
      "rewards/accuracy_reward": 0.8533636331558228,
      "rewards/format_reward": 1.0,
      "step": 246
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 116.3377456665039,
      "epoch": 3.6323529411764706,
      "grad_norm": 2.0730725351705477,
      "kl": 0.041748046875,
      "learning_rate": 8.183823529411764e-07,
      "loss": 0.0017,
      "reward": 1.8775808811187744,
      "reward_std": 0.02502569369971752,
      "rewards/accuracy_reward": 0.8787827491760254,
      "rewards/format_reward": 0.998798131942749,
      "step": 247
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 114.89663696289062,
      "epoch": 3.6470588235294117,
      "grad_norm": 2.0409255064434504,
      "kl": 0.04931640625,
      "learning_rate": 8.176470588235293e-07,
      "loss": 0.002,
      "reward": 1.887112021446228,
      "reward_std": 0.02062002196907997,
      "rewards/accuracy_reward": 0.8871120810508728,
      "rewards/format_reward": 1.0,
      "step": 248
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 114.95913696289062,
      "epoch": 3.661764705882353,
      "grad_norm": 1.4309091507183769,
      "kl": 0.04638671875,
      "learning_rate": 8.169117647058823e-07,
      "loss": 0.0019,
      "reward": 1.8621455430984497,
      "reward_std": 0.02185475081205368,
      "rewards/accuracy_reward": 0.8621454834938049,
      "rewards/format_reward": 1.0,
      "step": 249
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 115.15144348144531,
      "epoch": 3.6764705882352944,
      "grad_norm": 2.5697762019528216,
      "kl": 0.0390625,
      "learning_rate": 8.161764705882353e-07,
      "loss": 0.0016,
      "reward": 1.885780692100525,
      "reward_std": 0.020246701315045357,
      "rewards/accuracy_reward": 0.8857808113098145,
      "rewards/format_reward": 1.0,
      "step": 250
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 113.49159240722656,
      "epoch": 3.6911764705882355,
      "grad_norm": 1.879132846743946,
      "kl": 0.040283203125,
      "learning_rate": 8.154411764705881e-07,
      "loss": 0.0016,
      "reward": 1.8713446855545044,
      "reward_std": 0.024630680680274963,
      "rewards/accuracy_reward": 0.8725464940071106,
      "rewards/format_reward": 0.998798131942749,
      "step": 251
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 114.16226196289062,
      "epoch": 3.7058823529411766,
      "grad_norm": 2.1971208610905966,
      "kl": 0.0400390625,
      "learning_rate": 8.147058823529411e-07,
      "loss": 0.0016,
      "reward": 1.8803523778915405,
      "reward_std": 0.02186383679509163,
      "rewards/accuracy_reward": 0.8803524971008301,
      "rewards/format_reward": 1.0,
      "step": 252
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 114.24279022216797,
      "epoch": 3.7205882352941178,
      "grad_norm": 1.8334229890682787,
      "kl": 0.0380859375,
      "learning_rate": 8.139705882352941e-07,
      "loss": 0.0015,
      "reward": 1.8805620670318604,
      "reward_std": 0.021345175802707672,
      "rewards/accuracy_reward": 0.880561888217926,
      "rewards/format_reward": 1.0,
      "step": 253
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 113.02764892578125,
      "epoch": 3.735294117647059,
      "grad_norm": 3.7258956882443703,
      "kl": 0.045654296875,
      "learning_rate": 8.13235294117647e-07,
      "loss": 0.0018,
      "reward": 1.9009848833084106,
      "reward_std": 0.02094642072916031,
      "rewards/accuracy_reward": 0.9009847044944763,
      "rewards/format_reward": 1.0,
      "step": 254
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 114.55769348144531,
      "epoch": 3.75,
      "grad_norm": 1.678601436345225,
      "kl": 0.045166015625,
      "learning_rate": 8.125e-07,
      "loss": 0.0018,
      "reward": 1.8814560174942017,
      "reward_std": 0.021061250939965248,
      "rewards/accuracy_reward": 0.8814561367034912,
      "rewards/format_reward": 1.0,
      "step": 255
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 111.65264892578125,
      "epoch": 3.764705882352941,
      "grad_norm": 1.745259980031886,
      "kl": 0.04052734375,
      "learning_rate": 8.117647058823529e-07,
      "loss": 0.0016,
      "reward": 1.8852721452713013,
      "reward_std": 0.020116155967116356,
      "rewards/accuracy_reward": 0.8852721452713013,
      "rewards/format_reward": 1.0,
      "step": 256
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 112.62019348144531,
      "epoch": 3.7794117647058822,
      "grad_norm": 2.461830018990095,
      "kl": 0.041015625,
      "learning_rate": 8.110294117647058e-07,
      "loss": 0.0016,
      "reward": 1.8814154863357544,
      "reward_std": 0.02218296006321907,
      "rewards/accuracy_reward": 0.8814153075218201,
      "rewards/format_reward": 1.0,
      "step": 257
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 110.7175521850586,
      "epoch": 3.7941176470588234,
      "grad_norm": 1.68704697360333,
      "kl": 0.03955078125,
      "learning_rate": 8.102941176470588e-07,
      "loss": 0.0016,
      "reward": 1.8427629470825195,
      "reward_std": 0.023804400116205215,
      "rewards/accuracy_reward": 0.8427630066871643,
      "rewards/format_reward": 1.0,
      "step": 258
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 109.58894348144531,
      "epoch": 3.8088235294117645,
      "grad_norm": 1.709855378144107,
      "kl": 0.04052734375,
      "learning_rate": 8.095588235294118e-07,
      "loss": 0.0016,
      "reward": 1.882066011428833,
      "reward_std": 0.024021144956350327,
      "rewards/accuracy_reward": 0.8832679986953735,
      "rewards/format_reward": 0.998798131942749,
      "step": 259
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 109.78245544433594,
      "epoch": 3.8235294117647056,
      "grad_norm": 1.8255695685831124,
      "kl": 0.04150390625,
      "learning_rate": 8.088235294117646e-07,
      "loss": 0.0016,
      "reward": 1.8689602613449097,
      "reward_std": 0.019115259870886803,
      "rewards/accuracy_reward": 0.8689602017402649,
      "rewards/format_reward": 1.0,
      "step": 260
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 108.55409240722656,
      "epoch": 3.838235294117647,
      "grad_norm": 3.3416809660726403,
      "kl": 0.04052734375,
      "learning_rate": 8.080882352941176e-07,
      "loss": 0.0017,
      "reward": 1.8896381855010986,
      "reward_std": 0.019461615011096,
      "rewards/accuracy_reward": 0.8896381855010986,
      "rewards/format_reward": 1.0,
      "step": 261
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 108.46634674072266,
      "epoch": 3.8529411764705883,
      "grad_norm": 2.630911489725864,
      "kl": 0.0380859375,
      "learning_rate": 8.073529411764706e-07,
      "loss": 0.0015,
      "reward": 1.87398362159729,
      "reward_std": 0.023873545229434967,
      "rewards/accuracy_reward": 0.8739834427833557,
      "rewards/format_reward": 1.0,
      "step": 262
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 108.01683044433594,
      "epoch": 3.8676470588235294,
      "grad_norm": 1.843021207465952,
      "kl": 0.040283203125,
      "learning_rate": 8.066176470588235e-07,
      "loss": 0.0016,
      "reward": 1.8928405046463013,
      "reward_std": 0.019875595346093178,
      "rewards/accuracy_reward": 0.892840564250946,
      "rewards/format_reward": 1.0,
      "step": 263
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 107.4627456665039,
      "epoch": 3.8823529411764706,
      "grad_norm": 2.191986984722032,
      "kl": 0.040283203125,
      "learning_rate": 8.058823529411764e-07,
      "loss": 0.0016,
      "reward": 1.868396282196045,
      "reward_std": 0.025705352425575256,
      "rewards/accuracy_reward": 0.8683962225914001,
      "rewards/format_reward": 1.0,
      "step": 264
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 105.62139892578125,
      "epoch": 3.8970588235294117,
      "grad_norm": 1.9987217362414245,
      "kl": 0.037353515625,
      "learning_rate": 8.051470588235294e-07,
      "loss": 0.0015,
      "reward": 1.864815592765808,
      "reward_std": 0.02225598692893982,
      "rewards/accuracy_reward": 0.8648155331611633,
      "rewards/format_reward": 1.0,
      "step": 265
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 107.4988021850586,
      "epoch": 3.911764705882353,
      "grad_norm": 2.050016597438298,
      "kl": 0.0400390625,
      "learning_rate": 8.044117647058823e-07,
      "loss": 0.0016,
      "reward": 1.8612146377563477,
      "reward_std": 0.02399851381778717,
      "rewards/accuracy_reward": 0.8612145781517029,
      "rewards/format_reward": 1.0,
      "step": 266
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 104.59375762939453,
      "epoch": 3.9264705882352944,
      "grad_norm": 1.6627115812158326,
      "kl": 0.043701171875,
      "learning_rate": 8.036764705882353e-07,
      "loss": 0.0017,
      "reward": 1.8737601041793823,
      "reward_std": 0.024240681901574135,
      "rewards/accuracy_reward": 0.8737599849700928,
      "rewards/format_reward": 1.0,
      "step": 267
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 104.94351196289062,
      "epoch": 3.9411764705882355,
      "grad_norm": 1.9515217712056225,
      "kl": 0.0439453125,
      "learning_rate": 8.029411764705883e-07,
      "loss": 0.0018,
      "reward": 1.866271734237671,
      "reward_std": 0.023485863581299782,
      "rewards/accuracy_reward": 0.8662715554237366,
      "rewards/format_reward": 1.0,
      "step": 268
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 104.50721740722656,
      "epoch": 3.9558823529411766,
      "grad_norm": 1.741853556080663,
      "kl": 0.05029296875,
      "learning_rate": 8.022058823529411e-07,
      "loss": 0.002,
      "reward": 1.88307523727417,
      "reward_std": 0.021244022995233536,
      "rewards/accuracy_reward": 0.8830749988555908,
      "rewards/format_reward": 1.0,
      "step": 269
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 107.47356414794922,
      "epoch": 3.9705882352941178,
      "grad_norm": 4.485332406331801,
      "kl": 0.05712890625,
      "learning_rate": 8.014705882352941e-07,
      "loss": 0.0023,
      "reward": 1.8939332962036133,
      "reward_std": 0.025166498497128487,
      "rewards/accuracy_reward": 0.8939332365989685,
      "rewards/format_reward": 1.0,
      "step": 270
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 105.5613021850586,
      "epoch": 3.985294117647059,
      "grad_norm": 1.6003108364548544,
      "kl": 0.060302734375,
      "learning_rate": 8.007352941176471e-07,
      "loss": 0.0024,
      "reward": 1.883571982383728,
      "reward_std": 0.025911405682563782,
      "rewards/accuracy_reward": 0.8847737312316895,
      "rewards/format_reward": 0.998798131942749,
      "step": 271
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 104.46575164794922,
      "epoch": 4.0,
      "grad_norm": 1.5983921955450255,
      "kl": 0.0537109375,
      "learning_rate": 8e-07,
      "loss": 0.0022,
      "reward": 1.8933547735214233,
      "reward_std": 0.02186509408056736,
      "rewards/accuracy_reward": 0.8933547735214233,
      "rewards/format_reward": 1.0,
      "step": 272
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 103.33413696289062,
      "epoch": 4.014705882352941,
      "grad_norm": 2.727093502391197,
      "kl": 0.053466796875,
      "learning_rate": 7.992647058823529e-07,
      "loss": 0.0021,
      "reward": 1.8750649690628052,
      "reward_std": 0.019625522196292877,
      "rewards/accuracy_reward": 0.8750649094581604,
      "rewards/format_reward": 1.0,
      "step": 273
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 103.62019348144531,
      "epoch": 4.029411764705882,
      "grad_norm": 1.8871571408393943,
      "kl": 0.06103515625,
      "learning_rate": 7.985294117647059e-07,
      "loss": 0.0024,
      "reward": 1.8743054866790771,
      "reward_std": 0.024585077539086342,
      "rewards/accuracy_reward": 0.8743056058883667,
      "rewards/format_reward": 1.0,
      "step": 274
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 101.87019348144531,
      "epoch": 4.044117647058823,
      "grad_norm": 2.102514327126164,
      "kl": 0.05810546875,
      "learning_rate": 7.977941176470588e-07,
      "loss": 0.0023,
      "reward": 1.89711594581604,
      "reward_std": 0.021427426487207413,
      "rewards/accuracy_reward": 0.8971157670021057,
      "rewards/format_reward": 1.0,
      "step": 275
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 101.05529022216797,
      "epoch": 4.0588235294117645,
      "grad_norm": 1.5123356737294997,
      "kl": 0.05810546875,
      "learning_rate": 7.970588235294118e-07,
      "loss": 0.0023,
      "reward": 1.8750731945037842,
      "reward_std": 0.022188987582921982,
      "rewards/accuracy_reward": 0.875073254108429,
      "rewards/format_reward": 1.0,
      "step": 276
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 100.95072174072266,
      "epoch": 4.073529411764706,
      "grad_norm": 1.4897094983434982,
      "kl": 0.060791015625,
      "learning_rate": 7.963235294117646e-07,
      "loss": 0.0024,
      "reward": 1.886474847793579,
      "reward_std": 0.021508391946554184,
      "rewards/accuracy_reward": 0.8864747881889343,
      "rewards/format_reward": 1.0,
      "step": 277
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 100.2800521850586,
      "epoch": 4.088235294117647,
      "grad_norm": 3.202169591157568,
      "kl": 0.05517578125,
      "learning_rate": 7.955882352941175e-07,
      "loss": 0.0022,
      "reward": 1.8837499618530273,
      "reward_std": 0.02077949419617653,
      "rewards/accuracy_reward": 0.8849518299102783,
      "rewards/format_reward": 0.998798131942749,
      "step": 278
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 102.98798370361328,
      "epoch": 4.102941176470588,
      "grad_norm": 2.3522518543200115,
      "kl": 0.0498046875,
      "learning_rate": 7.948529411764705e-07,
      "loss": 0.002,
      "reward": 1.8832675218582153,
      "reward_std": 0.02067939005792141,
      "rewards/accuracy_reward": 0.883267343044281,
      "rewards/format_reward": 1.0,
      "step": 279
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 103.70072174072266,
      "epoch": 4.117647058823529,
      "grad_norm": 1.4424360808940544,
      "kl": 0.0478515625,
      "learning_rate": 7.941176470588235e-07,
      "loss": 0.0019,
      "reward": 1.8915950059890747,
      "reward_std": 0.018967358395457268,
      "rewards/accuracy_reward": 0.8915950059890747,
      "rewards/format_reward": 1.0,
      "step": 280
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 105.0096206665039,
      "epoch": 4.132352941176471,
      "grad_norm": 1.6320380993577588,
      "kl": 0.057373046875,
      "learning_rate": 7.933823529411764e-07,
      "loss": 0.0023,
      "reward": 1.8841755390167236,
      "reward_std": 0.019956490024924278,
      "rewards/accuracy_reward": 0.8841755390167236,
      "rewards/format_reward": 1.0,
      "step": 281
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 106.82452392578125,
      "epoch": 4.147058823529412,
      "grad_norm": 1.6748728201505565,
      "kl": 0.0478515625,
      "learning_rate": 7.926470588235293e-07,
      "loss": 0.0019,
      "reward": 1.881099820137024,
      "reward_std": 0.01961349882185459,
      "rewards/accuracy_reward": 0.8810995817184448,
      "rewards/format_reward": 1.0,
      "step": 282
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 108.52404022216797,
      "epoch": 4.161764705882353,
      "grad_norm": 3.540072890075679,
      "kl": 0.04736328125,
      "learning_rate": 7.919117647058823e-07,
      "loss": 0.0019,
      "reward": 1.8972245454788208,
      "reward_std": 0.020341571420431137,
      "rewards/accuracy_reward": 0.897224485874176,
      "rewards/format_reward": 1.0,
      "step": 283
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 108.55288696289062,
      "epoch": 4.176470588235294,
      "grad_norm": 1.9653905551543096,
      "kl": 0.045654296875,
      "learning_rate": 7.911764705882352e-07,
      "loss": 0.0018,
      "reward": 1.8803445100784302,
      "reward_std": 0.022631661966443062,
      "rewards/accuracy_reward": 0.8803444504737854,
      "rewards/format_reward": 1.0,
      "step": 284
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 110.54688262939453,
      "epoch": 4.1911764705882355,
      "grad_norm": 3.95007121019553,
      "kl": 0.046875,
      "learning_rate": 7.904411764705882e-07,
      "loss": 0.0019,
      "reward": 1.8853232860565186,
      "reward_std": 0.02392488718032837,
      "rewards/accuracy_reward": 0.8865249752998352,
      "rewards/format_reward": 0.998798131942749,
      "step": 285
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 111.4627456665039,
      "epoch": 4.205882352941177,
      "grad_norm": 1.6574614891623431,
      "kl": 0.05078125,
      "learning_rate": 7.897058823529411e-07,
      "loss": 0.002,
      "reward": 1.8902332782745361,
      "reward_std": 0.024903936311602592,
      "rewards/accuracy_reward": 0.8914350867271423,
      "rewards/format_reward": 0.998798131942749,
      "step": 286
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 112.29688262939453,
      "epoch": 4.220588235294118,
      "grad_norm": 3.256913991460478,
      "kl": 0.046142578125,
      "learning_rate": 7.88970588235294e-07,
      "loss": 0.0018,
      "reward": 1.8858234882354736,
      "reward_std": 0.024734003469347954,
      "rewards/accuracy_reward": 0.8882272243499756,
      "rewards/format_reward": 0.9975962042808533,
      "step": 287
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 111.60697174072266,
      "epoch": 4.235294117647059,
      "grad_norm": 1.3114281230029299,
      "kl": 0.0458984375,
      "learning_rate": 7.88235294117647e-07,
      "loss": 0.0019,
      "reward": 1.890177607536316,
      "reward_std": 0.02220279537141323,
      "rewards/accuracy_reward": 0.8901773691177368,
      "rewards/format_reward": 1.0,
      "step": 288
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 116.62741088867188,
      "epoch": 4.25,
      "grad_norm": 1.4493002218382003,
      "kl": 0.049560546875,
      "learning_rate": 7.875e-07,
      "loss": 0.002,
      "reward": 1.8985555171966553,
      "reward_std": 0.020094698294997215,
      "rewards/accuracy_reward": 0.8985555768013,
      "rewards/format_reward": 1.0,
      "step": 289
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 116.68389892578125,
      "epoch": 4.264705882352941,
      "grad_norm": 4.352470640135198,
      "kl": 0.04931640625,
      "learning_rate": 7.867647058823529e-07,
      "loss": 0.002,
      "reward": 1.8766732215881348,
      "reward_std": 0.027829263359308243,
      "rewards/accuracy_reward": 0.8790771961212158,
      "rewards/format_reward": 0.9975962042808533,
      "step": 290
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 114.04688262939453,
      "epoch": 4.279411764705882,
      "grad_norm": 3.874276314718884,
      "kl": 0.0830078125,
      "learning_rate": 7.860294117647058e-07,
      "loss": 0.0033,
      "reward": 1.8896328210830688,
      "reward_std": 0.021389471367001534,
      "rewards/accuracy_reward": 0.8896327614784241,
      "rewards/format_reward": 1.0,
      "step": 291
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 111.61178588867188,
      "epoch": 4.294117647058823,
      "grad_norm": 4.875295082608794,
      "kl": 0.04541015625,
      "learning_rate": 7.852941176470588e-07,
      "loss": 0.0018,
      "reward": 1.8860790729522705,
      "reward_std": 0.021392209455370903,
      "rewards/accuracy_reward": 0.8872808814048767,
      "rewards/format_reward": 0.998798131942749,
      "step": 292
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 112.39543914794922,
      "epoch": 4.3088235294117645,
      "grad_norm": 1.5633011375678962,
      "kl": 0.04345703125,
      "learning_rate": 7.845588235294117e-07,
      "loss": 0.0017,
      "reward": 1.8786873817443848,
      "reward_std": 0.021603846922516823,
      "rewards/accuracy_reward": 0.8786874413490295,
      "rewards/format_reward": 1.0,
      "step": 293
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 114.50360870361328,
      "epoch": 4.323529411764706,
      "grad_norm": 1.4773757778833134,
      "kl": 0.047119140625,
      "learning_rate": 7.838235294117647e-07,
      "loss": 0.0019,
      "reward": 1.8730394840240479,
      "reward_std": 0.02080570161342621,
      "rewards/accuracy_reward": 0.8730394244194031,
      "rewards/format_reward": 1.0,
      "step": 294
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 112.4363021850586,
      "epoch": 4.338235294117647,
      "grad_norm": 1.8801422889509678,
      "kl": 0.04638671875,
      "learning_rate": 7.830882352941176e-07,
      "loss": 0.0019,
      "reward": 1.8871146440505981,
      "reward_std": 0.020325137302279472,
      "rewards/accuracy_reward": 0.8871145844459534,
      "rewards/format_reward": 1.0,
      "step": 295
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 112.09014892578125,
      "epoch": 4.352941176470588,
      "grad_norm": 2.5097065769254594,
      "kl": 0.0498046875,
      "learning_rate": 7.823529411764705e-07,
      "loss": 0.002,
      "reward": 1.868382453918457,
      "reward_std": 0.02320130541920662,
      "rewards/accuracy_reward": 0.8695842623710632,
      "rewards/format_reward": 0.998798131942749,
      "step": 296
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 114.3377456665039,
      "epoch": 4.367647058823529,
      "grad_norm": 2.553942778131699,
      "kl": 0.044189453125,
      "learning_rate": 7.816176470588235e-07,
      "loss": 0.0018,
      "reward": 1.9048007726669312,
      "reward_std": 0.019152075052261353,
      "rewards/accuracy_reward": 0.9048006534576416,
      "rewards/format_reward": 1.0,
      "step": 297
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 110.56610870361328,
      "epoch": 4.382352941176471,
      "grad_norm": 1.2815769671843946,
      "kl": 0.049560546875,
      "learning_rate": 7.808823529411765e-07,
      "loss": 0.002,
      "reward": 1.8882514238357544,
      "reward_std": 0.023224936798214912,
      "rewards/accuracy_reward": 0.8906552791595459,
      "rewards/format_reward": 0.9975962042808533,
      "step": 298
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 112.27043914794922,
      "epoch": 4.397058823529412,
      "grad_norm": 1.4236109919766304,
      "kl": 0.0439453125,
      "learning_rate": 7.801470588235293e-07,
      "loss": 0.0018,
      "reward": 1.8794277906417847,
      "reward_std": 0.022136587649583817,
      "rewards/accuracy_reward": 0.8818315863609314,
      "rewards/format_reward": 0.9975962042808533,
      "step": 299
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 112.25000762939453,
      "epoch": 4.411764705882353,
      "grad_norm": 1.4092988693440858,
      "kl": 0.044189453125,
      "learning_rate": 7.794117647058823e-07,
      "loss": 0.0018,
      "reward": 1.8663372993469238,
      "reward_std": 0.02092735283076763,
      "rewards/accuracy_reward": 0.8663372993469238,
      "rewards/format_reward": 1.0,
      "step": 300
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 115.10577392578125,
      "epoch": 4.426470588235294,
      "grad_norm": 2.054063342384727,
      "kl": 0.0517578125,
      "learning_rate": 7.786764705882353e-07,
      "loss": 0.0021,
      "reward": 1.8976588249206543,
      "reward_std": 0.01807982660830021,
      "rewards/accuracy_reward": 0.8976587653160095,
      "rewards/format_reward": 1.0,
      "step": 301
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 113.3377456665039,
      "epoch": 4.4411764705882355,
      "grad_norm": 1.9093857237441312,
      "kl": 0.047607421875,
      "learning_rate": 7.779411764705882e-07,
      "loss": 0.0019,
      "reward": 1.8833731412887573,
      "reward_std": 0.01974165067076683,
      "rewards/accuracy_reward": 0.8833730816841125,
      "rewards/format_reward": 1.0,
      "step": 302
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 112.30409240722656,
      "epoch": 4.455882352941177,
      "grad_norm": 2.6735018202140988,
      "kl": 0.04541015625,
      "learning_rate": 7.772058823529412e-07,
      "loss": 0.0018,
      "reward": 1.8876824378967285,
      "reward_std": 0.019306330010294914,
      "rewards/accuracy_reward": 0.8876824975013733,
      "rewards/format_reward": 1.0,
      "step": 303
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 111.91106414794922,
      "epoch": 4.470588235294118,
      "grad_norm": 1.4970895120586163,
      "kl": 0.0439453125,
      "learning_rate": 7.764705882352941e-07,
      "loss": 0.0018,
      "reward": 1.8828318119049072,
      "reward_std": 0.020149705931544304,
      "rewards/accuracy_reward": 0.8840336799621582,
      "rewards/format_reward": 0.998798131942749,
      "step": 304
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 111.24519348144531,
      "epoch": 4.485294117647059,
      "grad_norm": 5.117496046280143,
      "kl": 0.04296875,
      "learning_rate": 7.75735294117647e-07,
      "loss": 0.0017,
      "reward": 1.8947540521621704,
      "reward_std": 0.024016601964831352,
      "rewards/accuracy_reward": 0.8971578478813171,
      "rewards/format_reward": 0.9975962042808533,
      "step": 305
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 110.50360870361328,
      "epoch": 4.5,
      "grad_norm": 2.5813219086011028,
      "kl": 0.052001953125,
      "learning_rate": 7.75e-07,
      "loss": 0.0021,
      "reward": 1.8869359493255615,
      "reward_std": 0.02114337682723999,
      "rewards/accuracy_reward": 0.8869357705116272,
      "rewards/format_reward": 1.0,
      "step": 306
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 111.6550521850586,
      "epoch": 4.514705882352941,
      "grad_norm": 2.5546889887402235,
      "kl": 0.0458984375,
      "learning_rate": 7.74264705882353e-07,
      "loss": 0.0018,
      "reward": 1.8768377304077148,
      "reward_std": 0.020170308649539948,
      "rewards/accuracy_reward": 0.8768375515937805,
      "rewards/format_reward": 1.0,
      "step": 307
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 111.10216522216797,
      "epoch": 4.529411764705882,
      "grad_norm": 2.0444532213674775,
      "kl": 0.048828125,
      "learning_rate": 7.735294117647058e-07,
      "loss": 0.002,
      "reward": 1.8887766599655151,
      "reward_std": 0.019136058166623116,
      "rewards/accuracy_reward": 0.8887766599655151,
      "rewards/format_reward": 1.0,
      "step": 308
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 109.73918914794922,
      "epoch": 4.544117647058823,
      "grad_norm": 1.4865300788991052,
      "kl": 0.05419921875,
      "learning_rate": 7.727941176470588e-07,
      "loss": 0.0022,
      "reward": 1.8884928226470947,
      "reward_std": 0.018670309334993362,
      "rewards/accuracy_reward": 0.88849276304245,
      "rewards/format_reward": 1.0,
      "step": 309
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 109.81610870361328,
      "epoch": 4.5588235294117645,
      "grad_norm": 1.8317615570289676,
      "kl": 0.0498046875,
      "learning_rate": 7.720588235294118e-07,
      "loss": 0.002,
      "reward": 1.8673958778381348,
      "reward_std": 0.022479576990008354,
      "rewards/accuracy_reward": 0.8673959374427795,
      "rewards/format_reward": 1.0,
      "step": 310
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 110.82933044433594,
      "epoch": 4.573529411764706,
      "grad_norm": 1.3908315302560892,
      "kl": 0.047119140625,
      "learning_rate": 7.713235294117647e-07,
      "loss": 0.0019,
      "reward": 1.8690990209579468,
      "reward_std": 0.01971607096493244,
      "rewards/accuracy_reward": 0.8690990805625916,
      "rewards/format_reward": 1.0,
      "step": 311
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 108.58053588867188,
      "epoch": 4.588235294117647,
      "grad_norm": 2.2483220834096223,
      "kl": 0.0458984375,
      "learning_rate": 7.705882352941177e-07,
      "loss": 0.0018,
      "reward": 1.8699548244476318,
      "reward_std": 0.02238030731678009,
      "rewards/accuracy_reward": 0.8699548840522766,
      "rewards/format_reward": 1.0,
      "step": 312
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 108.37741088867188,
      "epoch": 4.602941176470588,
      "grad_norm": 1.9027749884419907,
      "kl": 0.049560546875,
      "learning_rate": 7.698529411764706e-07,
      "loss": 0.002,
      "reward": 1.8906946182250977,
      "reward_std": 0.02228682115674019,
      "rewards/accuracy_reward": 0.8918964266777039,
      "rewards/format_reward": 0.998798131942749,
      "step": 313
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 110.70673370361328,
      "epoch": 4.617647058823529,
      "grad_norm": 1.2672783381025527,
      "kl": 0.045654296875,
      "learning_rate": 7.691176470588235e-07,
      "loss": 0.0018,
      "reward": 1.8969817161560059,
      "reward_std": 0.01983834058046341,
      "rewards/accuracy_reward": 0.8969816565513611,
      "rewards/format_reward": 1.0,
      "step": 314
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 108.74279022216797,
      "epoch": 4.632352941176471,
      "grad_norm": 1.3268803478175482,
      "kl": 0.0458984375,
      "learning_rate": 7.683823529411765e-07,
      "loss": 0.0018,
      "reward": 1.884127140045166,
      "reward_std": 0.02158300392329693,
      "rewards/accuracy_reward": 0.884127140045166,
      "rewards/format_reward": 1.0,
      "step": 315
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 111.26683044433594,
      "epoch": 4.647058823529412,
      "grad_norm": 4.254278045186561,
      "kl": 0.045654296875,
      "learning_rate": 7.676470588235295e-07,
      "loss": 0.0018,
      "reward": 1.8847973346710205,
      "reward_std": 0.023614434525370598,
      "rewards/accuracy_reward": 0.887201189994812,
      "rewards/format_reward": 0.9975962042808533,
      "step": 316
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 113.02644348144531,
      "epoch": 4.661764705882353,
      "grad_norm": 1.4138173515058006,
      "kl": 0.045654296875,
      "learning_rate": 7.669117647058823e-07,
      "loss": 0.0018,
      "reward": 1.8916139602661133,
      "reward_std": 0.021671609953045845,
      "rewards/accuracy_reward": 0.8916139006614685,
      "rewards/format_reward": 1.0,
      "step": 317
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 111.03485870361328,
      "epoch": 4.676470588235294,
      "grad_norm": 1.4297077613699323,
      "kl": 0.041748046875,
      "learning_rate": 7.661764705882353e-07,
      "loss": 0.0017,
      "reward": 1.8865143060684204,
      "reward_std": 0.022390324622392654,
      "rewards/accuracy_reward": 0.8877161145210266,
      "rewards/format_reward": 0.998798131942749,
      "step": 318
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 110.99759674072266,
      "epoch": 4.6911764705882355,
      "grad_norm": 1.2033391337835728,
      "kl": 0.041259765625,
      "learning_rate": 7.654411764705882e-07,
      "loss": 0.0017,
      "reward": 1.8856085538864136,
      "reward_std": 0.024989847093820572,
      "rewards/accuracy_reward": 0.8868104815483093,
      "rewards/format_reward": 0.998798131942749,
      "step": 319
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 112.29447174072266,
      "epoch": 4.705882352941177,
      "grad_norm": 1.3733454266831897,
      "kl": 0.044189453125,
      "learning_rate": 7.647058823529411e-07,
      "loss": 0.0018,
      "reward": 1.878577709197998,
      "reward_std": 0.021778374910354614,
      "rewards/accuracy_reward": 0.8785776495933533,
      "rewards/format_reward": 1.0,
      "step": 320
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 113.90745544433594,
      "epoch": 4.720588235294118,
      "grad_norm": 1.5274339339441825,
      "kl": 0.039794921875,
      "learning_rate": 7.63970588235294e-07,
      "loss": 0.0016,
      "reward": 1.872501015663147,
      "reward_std": 0.01973523572087288,
      "rewards/accuracy_reward": 0.8725009560585022,
      "rewards/format_reward": 1.0,
      "step": 321
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 111.89063262939453,
      "epoch": 4.735294117647059,
      "grad_norm": 1.5962482033199081,
      "kl": 0.041015625,
      "learning_rate": 7.63235294117647e-07,
      "loss": 0.0016,
      "reward": 1.8869467973709106,
      "reward_std": 0.022963199764490128,
      "rewards/accuracy_reward": 0.8881486058235168,
      "rewards/format_reward": 0.998798131942749,
      "step": 322
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 111.35938262939453,
      "epoch": 4.75,
      "grad_norm": 2.017481820546224,
      "kl": 0.039794921875,
      "learning_rate": 7.624999999999999e-07,
      "loss": 0.0016,
      "reward": 1.8841745853424072,
      "reward_std": 0.025449533015489578,
      "rewards/accuracy_reward": 0.886578381061554,
      "rewards/format_reward": 0.9975962042808533,
      "step": 323
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 110.48918914794922,
      "epoch": 4.764705882352941,
      "grad_norm": 5.893277558364017,
      "kl": 0.04052734375,
      "learning_rate": 7.617647058823529e-07,
      "loss": 0.0016,
      "reward": 1.8966243267059326,
      "reward_std": 0.017657946795225143,
      "rewards/accuracy_reward": 0.8966242074966431,
      "rewards/format_reward": 1.0,
      "step": 324
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 109.59495544433594,
      "epoch": 4.779411764705882,
      "grad_norm": 1.9364427240337212,
      "kl": 0.03857421875,
      "learning_rate": 7.610294117647059e-07,
      "loss": 0.0015,
      "reward": 1.8959040641784668,
      "reward_std": 0.01980137825012207,
      "rewards/accuracy_reward": 0.8959041237831116,
      "rewards/format_reward": 1.0,
      "step": 325
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 110.75000762939453,
      "epoch": 4.794117647058823,
      "grad_norm": 1.8831495019330111,
      "kl": 0.042724609375,
      "learning_rate": 7.602941176470587e-07,
      "loss": 0.0017,
      "reward": 1.8923227787017822,
      "reward_std": 0.021822385489940643,
      "rewards/accuracy_reward": 0.8935245871543884,
      "rewards/format_reward": 0.998798131942749,
      "step": 326
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 110.97476196289062,
      "epoch": 4.8088235294117645,
      "grad_norm": 3.7686300331111924,
      "kl": 0.0478515625,
      "learning_rate": 7.595588235294117e-07,
      "loss": 0.0019,
      "reward": 1.9007014036178589,
      "reward_std": 0.020531214773654938,
      "rewards/accuracy_reward": 0.9031053781509399,
      "rewards/format_reward": 0.9975962042808533,
      "step": 327
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 111.31610870361328,
      "epoch": 4.823529411764706,
      "grad_norm": 2.3075627417580566,
      "kl": 0.043212890625,
      "learning_rate": 7.588235294117647e-07,
      "loss": 0.0017,
      "reward": 1.8857510089874268,
      "reward_std": 0.019290894269943237,
      "rewards/accuracy_reward": 0.8857510089874268,
      "rewards/format_reward": 1.0,
      "step": 328
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 110.52284240722656,
      "epoch": 4.838235294117647,
      "grad_norm": 2.4505241717663218,
      "kl": 0.044677734375,
      "learning_rate": 7.580882352941176e-07,
      "loss": 0.0018,
      "reward": 1.8795123100280762,
      "reward_std": 0.017949938774108887,
      "rewards/accuracy_reward": 0.8795123100280762,
      "rewards/format_reward": 1.0,
      "step": 329
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 111.17548370361328,
      "epoch": 4.852941176470588,
      "grad_norm": 1.715533751803751,
      "kl": 0.04443359375,
      "learning_rate": 7.573529411764705e-07,
      "loss": 0.0018,
      "reward": 1.8877590894699097,
      "reward_std": 0.01760478876531124,
      "rewards/accuracy_reward": 0.8877590298652649,
      "rewards/format_reward": 1.0,
      "step": 330
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 110.52644348144531,
      "epoch": 4.867647058823529,
      "grad_norm": 1.6176970942840496,
      "kl": 0.049560546875,
      "learning_rate": 7.566176470588235e-07,
      "loss": 0.002,
      "reward": 1.8978863954544067,
      "reward_std": 0.01671546697616577,
      "rewards/accuracy_reward": 0.8978862166404724,
      "rewards/format_reward": 1.0,
      "step": 331
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 108.68870544433594,
      "epoch": 4.882352941176471,
      "grad_norm": 1.7008680805466612,
      "kl": 0.04541015625,
      "learning_rate": 7.558823529411764e-07,
      "loss": 0.0018,
      "reward": 1.8799728155136108,
      "reward_std": 0.017163129523396492,
      "rewards/accuracy_reward": 0.8799727559089661,
      "rewards/format_reward": 1.0,
      "step": 332
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 109.3064956665039,
      "epoch": 4.897058823529412,
      "grad_norm": 5.121313120929418,
      "kl": 0.04443359375,
      "learning_rate": 7.551470588235294e-07,
      "loss": 0.0018,
      "reward": 1.8954551219940186,
      "reward_std": 0.01760135032236576,
      "rewards/accuracy_reward": 0.8954549431800842,
      "rewards/format_reward": 1.0,
      "step": 333
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 108.14784240722656,
      "epoch": 4.911764705882353,
      "grad_norm": 1.3552009766911794,
      "kl": 0.046142578125,
      "learning_rate": 7.544117647058824e-07,
      "loss": 0.0018,
      "reward": 1.8577083349227905,
      "reward_std": 0.018814481794834137,
      "rewards/accuracy_reward": 0.8577081561088562,
      "rewards/format_reward": 1.0,
      "step": 334
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 108.55889892578125,
      "epoch": 4.926470588235294,
      "grad_norm": 2.140682303302643,
      "kl": 0.05078125,
      "learning_rate": 7.536764705882352e-07,
      "loss": 0.002,
      "reward": 1.881057620048523,
      "reward_std": 0.019572759047150612,
      "rewards/accuracy_reward": 0.881057620048523,
      "rewards/format_reward": 1.0,
      "step": 335
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 107.62620544433594,
      "epoch": 4.9411764705882355,
      "grad_norm": 3.272070631910461,
      "kl": 0.046630859375,
      "learning_rate": 7.529411764705882e-07,
      "loss": 0.0019,
      "reward": 1.8901574611663818,
      "reward_std": 0.015406808815896511,
      "rewards/accuracy_reward": 0.8901574611663818,
      "rewards/format_reward": 1.0,
      "step": 336
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 109.64784240722656,
      "epoch": 4.955882352941177,
      "grad_norm": 3.6448062675563726,
      "kl": 0.053955078125,
      "learning_rate": 7.522058823529412e-07,
      "loss": 0.0022,
      "reward": 1.875773310661316,
      "reward_std": 0.02223069965839386,
      "rewards/accuracy_reward": 0.8769751191139221,
      "rewards/format_reward": 0.998798131942749,
      "step": 337
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 107.64904022216797,
      "epoch": 4.970588235294118,
      "grad_norm": 1.4354791062562573,
      "kl": 0.06103515625,
      "learning_rate": 7.514705882352941e-07,
      "loss": 0.0024,
      "reward": 1.8930749893188477,
      "reward_std": 0.018565351143479347,
      "rewards/accuracy_reward": 0.8930751085281372,
      "rewards/format_reward": 1.0,
      "step": 338
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 108.71995544433594,
      "epoch": 4.985294117647059,
      "grad_norm": 1.8660094880654952,
      "kl": 0.05810546875,
      "learning_rate": 7.50735294117647e-07,
      "loss": 0.0023,
      "reward": 1.8842915296554565,
      "reward_std": 0.02008167654275894,
      "rewards/accuracy_reward": 0.8842914700508118,
      "rewards/format_reward": 1.0,
      "step": 339
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 106.37290954589844,
      "epoch": 5.0,
      "grad_norm": 1.508014879105866,
      "kl": 0.05078125,
      "learning_rate": 7.5e-07,
      "loss": 0.002,
      "reward": 1.8771898746490479,
      "reward_std": 0.02064666524529457,
      "rewards/accuracy_reward": 0.8771898150444031,
      "rewards/format_reward": 1.0,
      "step": 340
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 107.55889892578125,
      "epoch": 5.014705882352941,
      "grad_norm": 1.6330775894781793,
      "kl": 0.0478515625,
      "learning_rate": 7.492647058823529e-07,
      "loss": 0.0019,
      "reward": 1.8954782485961914,
      "reward_std": 0.017391271889209747,
      "rewards/accuracy_reward": 0.8954780697822571,
      "rewards/format_reward": 1.0,
      "step": 341
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 107.3738021850586,
      "epoch": 5.029411764705882,
      "grad_norm": 1.5308902611838158,
      "kl": 0.04833984375,
      "learning_rate": 7.485294117647059e-07,
      "loss": 0.0019,
      "reward": 1.9155678749084473,
      "reward_std": 0.018119707703590393,
      "rewards/accuracy_reward": 0.9167699217796326,
      "rewards/format_reward": 0.998798131942749,
      "step": 342
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 104.8269271850586,
      "epoch": 5.044117647058823,
      "grad_norm": 3.6296477320844893,
      "kl": 0.06298828125,
      "learning_rate": 7.477941176470587e-07,
      "loss": 0.0025,
      "reward": 1.8854138851165771,
      "reward_std": 0.02051103487610817,
      "rewards/accuracy_reward": 0.8854138851165771,
      "rewards/format_reward": 1.0,
      "step": 343
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 104.56370544433594,
      "epoch": 5.0588235294117645,
      "grad_norm": 3.0646702385899562,
      "kl": 0.047119140625,
      "learning_rate": 7.470588235294117e-07,
      "loss": 0.0019,
      "reward": 1.9035041332244873,
      "reward_std": 0.0216043870896101,
      "rewards/accuracy_reward": 0.9047059416770935,
      "rewards/format_reward": 0.998798131942749,
      "step": 344
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 105.50241088867188,
      "epoch": 5.073529411764706,
      "grad_norm": 1.7545084393443575,
      "kl": 0.049560546875,
      "learning_rate": 7.463235294117647e-07,
      "loss": 0.002,
      "reward": 1.879274606704712,
      "reward_std": 0.020033851265907288,
      "rewards/accuracy_reward": 0.8792744278907776,
      "rewards/format_reward": 1.0,
      "step": 345
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 106.07572174072266,
      "epoch": 5.088235294117647,
      "grad_norm": 2.265810746797048,
      "kl": 0.0458984375,
      "learning_rate": 7.455882352941176e-07,
      "loss": 0.0018,
      "reward": 1.8713810443878174,
      "reward_std": 0.019391771405935287,
      "rewards/accuracy_reward": 0.8713809251785278,
      "rewards/format_reward": 1.0,
      "step": 346
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 105.33053588867188,
      "epoch": 5.102941176470588,
      "grad_norm": 4.678653380466847,
      "kl": 0.058349609375,
      "learning_rate": 7.448529411764706e-07,
      "loss": 0.0023,
      "reward": 1.8980027437210083,
      "reward_std": 0.019409772008657455,
      "rewards/accuracy_reward": 0.8992045521736145,
      "rewards/format_reward": 0.998798131942749,
      "step": 347
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 107.4519271850586,
      "epoch": 5.117647058823529,
      "grad_norm": 5.587657670792633,
      "kl": 0.044921875,
      "learning_rate": 7.441176470588235e-07,
      "loss": 0.0018,
      "reward": 1.8735015392303467,
      "reward_std": 0.019884636625647545,
      "rewards/accuracy_reward": 0.8735013604164124,
      "rewards/format_reward": 1.0,
      "step": 348
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 106.25000762939453,
      "epoch": 5.132352941176471,
      "grad_norm": 1.4732972699207496,
      "kl": 0.050048828125,
      "learning_rate": 7.433823529411764e-07,
      "loss": 0.002,
      "reward": 1.8633267879486084,
      "reward_std": 0.020572712644934654,
      "rewards/accuracy_reward": 0.8633267879486084,
      "rewards/format_reward": 1.0,
      "step": 349
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 109.14303588867188,
      "epoch": 5.147058823529412,
      "grad_norm": 1.4434229984530293,
      "kl": 0.051025390625,
      "learning_rate": 7.426470588235294e-07,
      "loss": 0.002,
      "reward": 1.8769383430480957,
      "reward_std": 0.02142389491200447,
      "rewards/accuracy_reward": 0.8781403303146362,
      "rewards/format_reward": 0.998798131942749,
      "step": 350
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 107.86418914794922,
      "epoch": 5.161764705882353,
      "grad_norm": 2.15469914353684,
      "kl": 0.04541015625,
      "learning_rate": 7.419117647058824e-07,
      "loss": 0.0018,
      "reward": 1.882927417755127,
      "reward_std": 0.019372547045350075,
      "rewards/accuracy_reward": 0.882927417755127,
      "rewards/format_reward": 1.0,
      "step": 351
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 110.43389892578125,
      "epoch": 5.176470588235294,
      "grad_norm": 1.7010413230089305,
      "kl": 0.04345703125,
      "learning_rate": 7.411764705882352e-07,
      "loss": 0.0017,
      "reward": 1.8917574882507324,
      "reward_std": 0.021130278706550598,
      "rewards/accuracy_reward": 0.8929594159126282,
      "rewards/format_reward": 0.998798131942749,
      "step": 352
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 109.03726196289062,
      "epoch": 5.1911764705882355,
      "grad_norm": 1.4178252033931626,
      "kl": 0.0458984375,
      "learning_rate": 7.404411764705882e-07,
      "loss": 0.0018,
      "reward": 1.902689814567566,
      "reward_std": 0.020760996267199516,
      "rewards/accuracy_reward": 0.9038916826248169,
      "rewards/format_reward": 0.998798131942749,
      "step": 353
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 110.5721206665039,
      "epoch": 5.205882352941177,
      "grad_norm": 3.030997635370593,
      "kl": 0.049072265625,
      "learning_rate": 7.397058823529412e-07,
      "loss": 0.002,
      "reward": 1.8927603960037231,
      "reward_std": 0.017168840393424034,
      "rewards/accuracy_reward": 0.8927604556083679,
      "rewards/format_reward": 1.0,
      "step": 354
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 111.91106414794922,
      "epoch": 5.220588235294118,
      "grad_norm": 1.680026683397978,
      "kl": 0.046630859375,
      "learning_rate": 7.389705882352941e-07,
      "loss": 0.0019,
      "reward": 1.9062864780426025,
      "reward_std": 0.021335626021027565,
      "rewards/accuracy_reward": 0.9074884653091431,
      "rewards/format_reward": 0.998798131942749,
      "step": 355
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 112.44231414794922,
      "epoch": 5.235294117647059,
      "grad_norm": 1.9153566616206694,
      "kl": 0.044189453125,
      "learning_rate": 7.382352941176471e-07,
      "loss": 0.0018,
      "reward": 1.8809791803359985,
      "reward_std": 0.02271600253880024,
      "rewards/accuracy_reward": 0.8821810483932495,
      "rewards/format_reward": 0.998798131942749,
      "step": 356
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 113.19231414794922,
      "epoch": 5.25,
      "grad_norm": 3.558694171062752,
      "kl": 0.048583984375,
      "learning_rate": 7.375e-07,
      "loss": 0.002,
      "reward": 1.9085917472839355,
      "reward_std": 0.01498875766992569,
      "rewards/accuracy_reward": 0.9085916876792908,
      "rewards/format_reward": 1.0,
      "step": 357
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 114.37019348144531,
      "epoch": 5.264705882352941,
      "grad_norm": 2.0945595152655208,
      "kl": 0.04443359375,
      "learning_rate": 7.367647058823529e-07,
      "loss": 0.0018,
      "reward": 1.8866630792617798,
      "reward_std": 0.021569877862930298,
      "rewards/accuracy_reward": 0.8890668749809265,
      "rewards/format_reward": 0.9975962042808533,
      "step": 358
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 113.05769348144531,
      "epoch": 5.279411764705882,
      "grad_norm": 1.8644120471415602,
      "kl": 0.048095703125,
      "learning_rate": 7.360294117647059e-07,
      "loss": 0.0019,
      "reward": 1.890414834022522,
      "reward_std": 0.021897265687584877,
      "rewards/accuracy_reward": 0.891616702079773,
      "rewards/format_reward": 0.998798131942749,
      "step": 359
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 112.48438262939453,
      "epoch": 5.294117647058823,
      "grad_norm": 2.9245490886556236,
      "kl": 0.044921875,
      "learning_rate": 7.352941176470589e-07,
      "loss": 0.0018,
      "reward": 1.8733696937561035,
      "reward_std": 0.019373353570699692,
      "rewards/accuracy_reward": 0.8745716214179993,
      "rewards/format_reward": 0.998798131942749,
      "step": 360
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 112.13101196289062,
      "epoch": 5.3088235294117645,
      "grad_norm": 2.4171192429296933,
      "kl": 0.04833984375,
      "learning_rate": 7.345588235294117e-07,
      "loss": 0.0019,
      "reward": 1.8899723291397095,
      "reward_std": 0.020467419177293777,
      "rewards/accuracy_reward": 0.8911741971969604,
      "rewards/format_reward": 0.998798131942749,
      "step": 361
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 112.25000762939453,
      "epoch": 5.323529411764706,
      "grad_norm": 2.1361138297509035,
      "kl": 0.0439453125,
      "learning_rate": 7.338235294117646e-07,
      "loss": 0.0018,
      "reward": 1.892791509628296,
      "reward_std": 0.015906570479273796,
      "rewards/accuracy_reward": 0.8927913308143616,
      "rewards/format_reward": 1.0,
      "step": 362
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 113.56370544433594,
      "epoch": 5.338235294117647,
      "grad_norm": 1.92362426869984,
      "kl": 0.045166015625,
      "learning_rate": 7.330882352941176e-07,
      "loss": 0.0018,
      "reward": 1.8666787147521973,
      "reward_std": 0.019119028002023697,
      "rewards/accuracy_reward": 0.8666786551475525,
      "rewards/format_reward": 1.0,
      "step": 363
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 112.6971206665039,
      "epoch": 5.352941176470588,
      "grad_norm": 1.6581703712574578,
      "kl": 0.049072265625,
      "learning_rate": 7.323529411764705e-07,
      "loss": 0.002,
      "reward": 1.874783992767334,
      "reward_std": 0.01916446164250374,
      "rewards/accuracy_reward": 0.8759857416152954,
      "rewards/format_reward": 0.998798131942749,
      "step": 364
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 112.17428588867188,
      "epoch": 5.367647058823529,
      "grad_norm": 1.316540945229491,
      "kl": 0.0478515625,
      "learning_rate": 7.316176470588234e-07,
      "loss": 0.0019,
      "reward": 1.8962125778198242,
      "reward_std": 0.01739468425512314,
      "rewards/accuracy_reward": 0.8962125182151794,
      "rewards/format_reward": 1.0,
      "step": 365
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 112.37620544433594,
      "epoch": 5.382352941176471,
      "grad_norm": 8.565938358324495,
      "kl": 0.046630859375,
      "learning_rate": 7.308823529411764e-07,
      "loss": 0.0019,
      "reward": 1.8905795812606812,
      "reward_std": 0.020930087193846703,
      "rewards/accuracy_reward": 0.890579342842102,
      "rewards/format_reward": 1.0,
      "step": 366
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 110.68991088867188,
      "epoch": 5.397058823529412,
      "grad_norm": 3.469819197381053,
      "kl": 0.05419921875,
      "learning_rate": 7.301470588235293e-07,
      "loss": 0.0022,
      "reward": 1.8995338678359985,
      "reward_std": 0.02121318317949772,
      "rewards/accuracy_reward": 0.9007357358932495,
      "rewards/format_reward": 0.998798131942749,
      "step": 367
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 110.24759674072266,
      "epoch": 5.411764705882353,
      "grad_norm": 3.6183994362346517,
      "kl": 0.04248046875,
      "learning_rate": 7.294117647058823e-07,
      "loss": 0.0017,
      "reward": 1.8944286108016968,
      "reward_std": 0.019341690465807915,
      "rewards/accuracy_reward": 0.894428551197052,
      "rewards/format_reward": 1.0,
      "step": 368
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 110.42188262939453,
      "epoch": 5.426470588235294,
      "grad_norm": 1.5985396075036147,
      "kl": 0.046630859375,
      "learning_rate": 7.286764705882353e-07,
      "loss": 0.0019,
      "reward": 1.9031792879104614,
      "reward_std": 0.01647799275815487,
      "rewards/accuracy_reward": 0.9031792283058167,
      "rewards/format_reward": 1.0,
      "step": 369
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 110.56851196289062,
      "epoch": 5.4411764705882355,
      "grad_norm": 1.2316701809022303,
      "kl": 0.0439453125,
      "learning_rate": 7.279411764705881e-07,
      "loss": 0.0018,
      "reward": 1.874247670173645,
      "reward_std": 0.021124592050909996,
      "rewards/accuracy_reward": 0.8754495978355408,
      "rewards/format_reward": 0.998798131942749,
      "step": 370
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 107.75360870361328,
      "epoch": 5.455882352941177,
      "grad_norm": 2.02457072563307,
      "kl": 0.046142578125,
      "learning_rate": 7.272058823529411e-07,
      "loss": 0.0018,
      "reward": 1.8876434564590454,
      "reward_std": 0.018463727086782455,
      "rewards/accuracy_reward": 0.887643575668335,
      "rewards/format_reward": 1.0,
      "step": 371
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 108.8894271850586,
      "epoch": 5.470588235294118,
      "grad_norm": 1.9670592101281013,
      "kl": 0.04296875,
      "learning_rate": 7.264705882352941e-07,
      "loss": 0.0017,
      "reward": 1.8814630508422852,
      "reward_std": 0.016402965411543846,
      "rewards/accuracy_reward": 0.8814628720283508,
      "rewards/format_reward": 1.0,
      "step": 372
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 110.15625762939453,
      "epoch": 5.485294117647059,
      "grad_norm": 1.4520387331980766,
      "kl": 0.044189453125,
      "learning_rate": 7.25735294117647e-07,
      "loss": 0.0018,
      "reward": 1.8984284400939941,
      "reward_std": 0.015249278396368027,
      "rewards/accuracy_reward": 0.8984284400939941,
      "rewards/format_reward": 1.0,
      "step": 373
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 107.3738021850586,
      "epoch": 5.5,
      "grad_norm": 1.5706724220082209,
      "kl": 0.04833984375,
      "learning_rate": 7.249999999999999e-07,
      "loss": 0.0019,
      "reward": 1.8929314613342285,
      "reward_std": 0.018274588510394096,
      "rewards/accuracy_reward": 0.8929312825202942,
      "rewards/format_reward": 1.0,
      "step": 374
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 109.56370544433594,
      "epoch": 5.514705882352941,
      "grad_norm": 3.532450016398957,
      "kl": 0.0439453125,
      "learning_rate": 7.242647058823529e-07,
      "loss": 0.0018,
      "reward": 1.892069935798645,
      "reward_std": 0.019071701914072037,
      "rewards/accuracy_reward": 0.8920699954032898,
      "rewards/format_reward": 1.0,
      "step": 375
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 108.27284240722656,
      "epoch": 5.529411764705882,
      "grad_norm": 2.9126439595526112,
      "kl": 0.052734375,
      "learning_rate": 7.235294117647058e-07,
      "loss": 0.0021,
      "reward": 1.8809665441513062,
      "reward_std": 0.01572202891111374,
      "rewards/accuracy_reward": 0.8809664845466614,
      "rewards/format_reward": 1.0,
      "step": 376
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 108.28846740722656,
      "epoch": 5.544117647058823,
      "grad_norm": 2.3143306804592423,
      "kl": 0.048095703125,
      "learning_rate": 7.227941176470588e-07,
      "loss": 0.0019,
      "reward": 1.905410885810852,
      "reward_std": 0.015785330906510353,
      "rewards/accuracy_reward": 0.9054108262062073,
      "rewards/format_reward": 1.0,
      "step": 377
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 108.16947174072266,
      "epoch": 5.5588235294117645,
      "grad_norm": 2.6475439861754095,
      "kl": 0.04638671875,
      "learning_rate": 7.220588235294118e-07,
      "loss": 0.0019,
      "reward": 1.8877785205841064,
      "reward_std": 0.01584857515990734,
      "rewards/accuracy_reward": 0.8877785205841064,
      "rewards/format_reward": 1.0,
      "step": 378
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 107.41346740722656,
      "epoch": 5.573529411764706,
      "grad_norm": 2.025239158313011,
      "kl": 0.04736328125,
      "learning_rate": 7.213235294117646e-07,
      "loss": 0.0019,
      "reward": 1.9032666683197021,
      "reward_std": 0.016153108328580856,
      "rewards/accuracy_reward": 0.9032665491104126,
      "rewards/format_reward": 1.0,
      "step": 379
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 109.09616088867188,
      "epoch": 5.588235294117647,
      "grad_norm": 4.81037382465536,
      "kl": 0.049072265625,
      "learning_rate": 7.205882352941176e-07,
      "loss": 0.002,
      "reward": 1.8973829746246338,
      "reward_std": 0.018959591165184975,
      "rewards/accuracy_reward": 0.8985849618911743,
      "rewards/format_reward": 0.998798131942749,
      "step": 380
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 109.60456848144531,
      "epoch": 5.602941176470588,
      "grad_norm": 2.0424310708762543,
      "kl": 0.0478515625,
      "learning_rate": 7.198529411764706e-07,
      "loss": 0.0019,
      "reward": 1.891837239265442,
      "reward_std": 0.017265740782022476,
      "rewards/accuracy_reward": 0.8918372392654419,
      "rewards/format_reward": 1.0,
      "step": 381
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 110.49639892578125,
      "epoch": 5.617647058823529,
      "grad_norm": 7.450414291777024,
      "kl": 0.040283203125,
      "learning_rate": 7.191176470588235e-07,
      "loss": 0.0016,
      "reward": 1.8869714736938477,
      "reward_std": 0.014449415728449821,
      "rewards/accuracy_reward": 0.8869713544845581,
      "rewards/format_reward": 1.0,
      "step": 382
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 111.50841522216797,
      "epoch": 5.632352941176471,
      "grad_norm": 2.035166070284095,
      "kl": 0.05029296875,
      "learning_rate": 7.183823529411764e-07,
      "loss": 0.002,
      "reward": 1.8704235553741455,
      "reward_std": 0.01924438774585724,
      "rewards/accuracy_reward": 0.8704236149787903,
      "rewards/format_reward": 1.0,
      "step": 383
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 109.93269348144531,
      "epoch": 5.647058823529412,
      "grad_norm": 1.5304768166279001,
      "kl": 0.04638671875,
      "learning_rate": 7.176470588235294e-07,
      "loss": 0.0019,
      "reward": 1.889725923538208,
      "reward_std": 0.020155081525444984,
      "rewards/accuracy_reward": 0.890927791595459,
      "rewards/format_reward": 0.998798131942749,
      "step": 384
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 110.50481414794922,
      "epoch": 5.661764705882353,
      "grad_norm": 3.3086513126052437,
      "kl": 0.046875,
      "learning_rate": 7.169117647058823e-07,
      "loss": 0.0019,
      "reward": 1.9017019271850586,
      "reward_std": 0.02102070115506649,
      "rewards/accuracy_reward": 0.9029037952423096,
      "rewards/format_reward": 0.998798131942749,
      "step": 385
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 111.59856414794922,
      "epoch": 5.676470588235294,
      "grad_norm": 1.9961992837605382,
      "kl": 0.04443359375,
      "learning_rate": 7.161764705882353e-07,
      "loss": 0.0018,
      "reward": 1.898403286933899,
      "reward_std": 0.01933371089398861,
      "rewards/accuracy_reward": 0.8996050953865051,
      "rewards/format_reward": 0.998798131942749,
      "step": 386
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 112.00360870361328,
      "epoch": 5.6911764705882355,
      "grad_norm": 5.861203308278023,
      "kl": 0.044189453125,
      "learning_rate": 7.154411764705882e-07,
      "loss": 0.0018,
      "reward": 1.9097044467926025,
      "reward_std": 0.017147168517112732,
      "rewards/accuracy_reward": 0.9097043871879578,
      "rewards/format_reward": 1.0,
      "step": 387
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 109.95433044433594,
      "epoch": 5.705882352941177,
      "grad_norm": 1.8272044694889817,
      "kl": 0.046142578125,
      "learning_rate": 7.147058823529411e-07,
      "loss": 0.0018,
      "reward": 1.897651195526123,
      "reward_std": 0.019564587622880936,
      "rewards/accuracy_reward": 0.897650957107544,
      "rewards/format_reward": 1.0,
      "step": 388
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 110.26202392578125,
      "epoch": 5.720588235294118,
      "grad_norm": 1.945047556411728,
      "kl": 0.053955078125,
      "learning_rate": 7.139705882352941e-07,
      "loss": 0.0022,
      "reward": 1.8845726251602173,
      "reward_std": 0.018275026232004166,
      "rewards/accuracy_reward": 0.8845725655555725,
      "rewards/format_reward": 1.0,
      "step": 389
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 108.27764892578125,
      "epoch": 5.735294117647059,
      "grad_norm": 1.872210801006099,
      "kl": 0.0419921875,
      "learning_rate": 7.132352941176471e-07,
      "loss": 0.0017,
      "reward": 1.8972781896591187,
      "reward_std": 0.019161643460392952,
      "rewards/accuracy_reward": 0.8972780704498291,
      "rewards/format_reward": 1.0,
      "step": 390
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 108.4206771850586,
      "epoch": 5.75,
      "grad_norm": 1.628706031121031,
      "kl": 0.044921875,
      "learning_rate": 7.125e-07,
      "loss": 0.0018,
      "reward": 1.8868787288665771,
      "reward_std": 0.017812801524996758,
      "rewards/accuracy_reward": 0.8868787884712219,
      "rewards/format_reward": 1.0,
      "step": 391
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 108.36538696289062,
      "epoch": 5.764705882352941,
      "grad_norm": 2.025154646985711,
      "kl": 0.043701171875,
      "learning_rate": 7.117647058823529e-07,
      "loss": 0.0017,
      "reward": 1.8947402238845825,
      "reward_std": 0.016441920772194862,
      "rewards/accuracy_reward": 0.8947402834892273,
      "rewards/format_reward": 1.0,
      "step": 392
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 107.59375762939453,
      "epoch": 5.779411764705882,
      "grad_norm": 3.1223862690686466,
      "kl": 0.04443359375,
      "learning_rate": 7.110294117647059e-07,
      "loss": 0.0018,
      "reward": 1.884828805923462,
      "reward_std": 0.02042386308312416,
      "rewards/accuracy_reward": 0.8848287463188171,
      "rewards/format_reward": 1.0,
      "step": 393
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 107.52163696289062,
      "epoch": 5.794117647058823,
      "grad_norm": 1.2474938760026444,
      "kl": 0.0439453125,
      "learning_rate": 7.102941176470588e-07,
      "loss": 0.0018,
      "reward": 1.894531488418579,
      "reward_std": 0.017643319442868233,
      "rewards/accuracy_reward": 0.8945314288139343,
      "rewards/format_reward": 1.0,
      "step": 394
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 106.53966522216797,
      "epoch": 5.8088235294117645,
      "grad_norm": 2.4031954689781556,
      "kl": 0.04443359375,
      "learning_rate": 7.095588235294118e-07,
      "loss": 0.0018,
      "reward": 1.906731367111206,
      "reward_std": 0.019448986276984215,
      "rewards/accuracy_reward": 0.9079332947731018,
      "rewards/format_reward": 0.998798131942749,
      "step": 395
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 106.2752456665039,
      "epoch": 5.823529411764706,
      "grad_norm": 2.3766132625090495,
      "kl": 0.06005859375,
      "learning_rate": 7.088235294117647e-07,
      "loss": 0.0024,
      "reward": 1.9001885652542114,
      "reward_std": 0.018285507336258888,
      "rewards/accuracy_reward": 0.9001885056495667,
      "rewards/format_reward": 1.0,
      "step": 396
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 108.21034240722656,
      "epoch": 5.838235294117647,
      "grad_norm": 1.3410708995130616,
      "kl": 0.049560546875,
      "learning_rate": 7.080882352941176e-07,
      "loss": 0.002,
      "reward": 1.8895319700241089,
      "reward_std": 0.019128086045384407,
      "rewards/accuracy_reward": 0.8907338976860046,
      "rewards/format_reward": 0.998798131942749,
      "step": 397
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 104.0300521850586,
      "epoch": 5.852941176470588,
      "grad_norm": 2.1806221304312188,
      "kl": 0.047607421875,
      "learning_rate": 7.073529411764706e-07,
      "loss": 0.0019,
      "reward": 1.8873209953308105,
      "reward_std": 0.015813617035746574,
      "rewards/accuracy_reward": 0.8873209357261658,
      "rewards/format_reward": 1.0,
      "step": 398
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 106.05889892578125,
      "epoch": 5.867647058823529,
      "grad_norm": 2.381149520093015,
      "kl": 0.05126953125,
      "learning_rate": 7.066176470588236e-07,
      "loss": 0.0021,
      "reward": 1.9061532020568848,
      "reward_std": 0.017387444153428078,
      "rewards/accuracy_reward": 0.9061532020568848,
      "rewards/format_reward": 1.0,
      "step": 399
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 106.60216522216797,
      "epoch": 5.882352941176471,
      "grad_norm": 2.0072217926486235,
      "kl": 0.048583984375,
      "learning_rate": 7.058823529411765e-07,
      "loss": 0.0019,
      "reward": 1.8860886096954346,
      "reward_std": 0.01758231595158577,
      "rewards/accuracy_reward": 0.8860886096954346,
      "rewards/format_reward": 1.0,
      "step": 400
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 106.2908706665039,
      "epoch": 5.897058823529412,
      "grad_norm": 3.7976903995467244,
      "kl": 0.043701171875,
      "learning_rate": 7.051470588235294e-07,
      "loss": 0.0017,
      "reward": 1.9081175327301025,
      "reward_std": 0.016382873058319092,
      "rewards/accuracy_reward": 0.9081174731254578,
      "rewards/format_reward": 1.0,
      "step": 401
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 106.97235870361328,
      "epoch": 5.911764705882353,
      "grad_norm": 1.6832136746582174,
      "kl": 0.05322265625,
      "learning_rate": 7.044117647058824e-07,
      "loss": 0.0021,
      "reward": 1.8958063125610352,
      "reward_std": 0.014682255685329437,
      "rewards/accuracy_reward": 0.8958064317703247,
      "rewards/format_reward": 1.0,
      "step": 402
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 106.90384674072266,
      "epoch": 5.926470588235294,
      "grad_norm": 2.1815402582773618,
      "kl": 0.0498046875,
      "learning_rate": 7.036764705882353e-07,
      "loss": 0.002,
      "reward": 1.8909062147140503,
      "reward_std": 0.018609097227454185,
      "rewards/accuracy_reward": 0.8909061551094055,
      "rewards/format_reward": 1.0,
      "step": 403
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 105.67788696289062,
      "epoch": 5.9411764705882355,
      "grad_norm": 1.2061130128497455,
      "kl": 0.047119140625,
      "learning_rate": 7.029411764705882e-07,
      "loss": 0.0019,
      "reward": 1.8931101560592651,
      "reward_std": 0.01808551885187626,
      "rewards/accuracy_reward": 0.8931100368499756,
      "rewards/format_reward": 1.0,
      "step": 404
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 106.68870544433594,
      "epoch": 5.955882352941177,
      "grad_norm": 1.9433195416220157,
      "kl": 0.046630859375,
      "learning_rate": 7.022058823529411e-07,
      "loss": 0.0019,
      "reward": 1.9073554277420044,
      "reward_std": 0.018337463960051537,
      "rewards/accuracy_reward": 0.9073554277420044,
      "rewards/format_reward": 1.0,
      "step": 405
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 107.47476196289062,
      "epoch": 5.970588235294118,
      "grad_norm": 1.8820064437816768,
      "kl": 0.060546875,
      "learning_rate": 7.01470588235294e-07,
      "loss": 0.0024,
      "reward": 1.894400715827942,
      "reward_std": 0.021061552688479424,
      "rewards/accuracy_reward": 0.8944007158279419,
      "rewards/format_reward": 1.0,
      "step": 406
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 107.52884674072266,
      "epoch": 5.985294117647059,
      "grad_norm": 1.4623484883692877,
      "kl": 0.05859375,
      "learning_rate": 7.00735294117647e-07,
      "loss": 0.0024,
      "reward": 1.9013231992721558,
      "reward_std": 0.02373763732612133,
      "rewards/accuracy_reward": 0.9037269949913025,
      "rewards/format_reward": 0.9975962042808533,
      "step": 407
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 106.86758422851562,
      "epoch": 6.0,
      "grad_norm": 2.1711922114519124,
      "kl": 0.048828125,
      "learning_rate": 7e-07,
      "loss": 0.002,
      "reward": 1.9016249179840088,
      "reward_std": 0.023924626410007477,
      "rewards/accuracy_reward": 0.9046690464019775,
      "rewards/format_reward": 0.9969558715820312,
      "step": 408
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 108.19110870361328,
      "epoch": 6.014705882352941,
      "grad_norm": 1.7310411415632332,
      "kl": 0.052490234375,
      "learning_rate": 6.992647058823528e-07,
      "loss": 0.0021,
      "reward": 1.8854323625564575,
      "reward_std": 0.023845864459872246,
      "rewards/accuracy_reward": 0.8866342902183533,
      "rewards/format_reward": 0.998798131942749,
      "step": 409
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 107.46875762939453,
      "epoch": 6.029411764705882,
      "grad_norm": 5.687861204362904,
      "kl": 0.0908203125,
      "learning_rate": 6.985294117647058e-07,
      "loss": 0.0036,
      "reward": 1.9103180170059204,
      "reward_std": 0.017014116048812866,
      "rewards/accuracy_reward": 0.9103179574012756,
      "rewards/format_reward": 1.0,
      "step": 410
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 106.84134674072266,
      "epoch": 6.044117647058823,
      "grad_norm": 1.8355942786177852,
      "kl": 0.048583984375,
      "learning_rate": 6.977941176470588e-07,
      "loss": 0.002,
      "reward": 1.8937817811965942,
      "reward_std": 0.018505675718188286,
      "rewards/accuracy_reward": 0.8937817215919495,
      "rewards/format_reward": 1.0,
      "step": 411
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 106.79688262939453,
      "epoch": 6.0588235294117645,
      "grad_norm": 1.9759679376426085,
      "kl": 0.052001953125,
      "learning_rate": 6.970588235294117e-07,
      "loss": 0.0021,
      "reward": 1.8942158222198486,
      "reward_std": 0.02279289998114109,
      "rewards/accuracy_reward": 0.8954177498817444,
      "rewards/format_reward": 0.998798131942749,
      "step": 412
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 107.37860870361328,
      "epoch": 6.073529411764706,
      "grad_norm": 2.300855046609347,
      "kl": 0.059814453125,
      "learning_rate": 6.963235294117647e-07,
      "loss": 0.0024,
      "reward": 1.8962349891662598,
      "reward_std": 0.020851602777838707,
      "rewards/accuracy_reward": 0.8962349891662598,
      "rewards/format_reward": 1.0,
      "step": 413
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 108.5252456665039,
      "epoch": 6.088235294117647,
      "grad_norm": 12.310047514562172,
      "kl": 0.047607421875,
      "learning_rate": 6.955882352941176e-07,
      "loss": 0.0019,
      "reward": 1.899317979812622,
      "reward_std": 0.020527588203549385,
      "rewards/accuracy_reward": 0.9005199074745178,
      "rewards/format_reward": 0.998798131942749,
      "step": 414
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 105.1238021850586,
      "epoch": 6.102941176470588,
      "grad_norm": 1.5229186035110283,
      "kl": 0.05224609375,
      "learning_rate": 6.948529411764705e-07,
      "loss": 0.0021,
      "reward": 1.8821978569030762,
      "reward_std": 0.022272054105997086,
      "rewards/accuracy_reward": 0.8833997249603271,
      "rewards/format_reward": 0.998798131942749,
      "step": 415
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 109.74279022216797,
      "epoch": 6.117647058823529,
      "grad_norm": 3.2878165148453236,
      "kl": 0.053955078125,
      "learning_rate": 6.941176470588235e-07,
      "loss": 0.0022,
      "reward": 1.86842679977417,
      "reward_std": 0.024334656074643135,
      "rewards/accuracy_reward": 0.8684268593788147,
      "rewards/format_reward": 1.0,
      "step": 416
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 108.48918914794922,
      "epoch": 6.132352941176471,
      "grad_norm": 1.2525615714604925,
      "kl": 0.05029296875,
      "learning_rate": 6.933823529411765e-07,
      "loss": 0.002,
      "reward": 1.914664387702942,
      "reward_std": 0.01958000287413597,
      "rewards/accuracy_reward": 0.9146643877029419,
      "rewards/format_reward": 1.0,
      "step": 417
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 107.93269348144531,
      "epoch": 6.147058823529412,
      "grad_norm": 2.561139147224278,
      "kl": 0.052734375,
      "learning_rate": 6.926470588235293e-07,
      "loss": 0.0021,
      "reward": 1.9154963493347168,
      "reward_std": 0.016698814928531647,
      "rewards/accuracy_reward": 0.9154961109161377,
      "rewards/format_reward": 1.0,
      "step": 418
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 106.98077392578125,
      "epoch": 6.161764705882353,
      "grad_norm": 2.1752639111351666,
      "kl": 0.058349609375,
      "learning_rate": 6.919117647058823e-07,
      "loss": 0.0023,
      "reward": 1.9047746658325195,
      "reward_std": 0.018907994031906128,
      "rewards/accuracy_reward": 0.9047747254371643,
      "rewards/format_reward": 1.0,
      "step": 419
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 109.78606414794922,
      "epoch": 6.176470588235294,
      "grad_norm": 2.0292217325556416,
      "kl": 0.0517578125,
      "learning_rate": 6.911764705882353e-07,
      "loss": 0.0021,
      "reward": 1.9034497737884521,
      "reward_std": 0.024114146828651428,
      "rewards/accuracy_reward": 0.9034495949745178,
      "rewards/format_reward": 1.0,
      "step": 420
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 108.4519271850586,
      "epoch": 6.1911764705882355,
      "grad_norm": 6.699279802691781,
      "kl": 0.055419921875,
      "learning_rate": 6.904411764705882e-07,
      "loss": 0.0022,
      "reward": 1.887842059135437,
      "reward_std": 0.020647389814257622,
      "rewards/accuracy_reward": 0.887842059135437,
      "rewards/format_reward": 1.0,
      "step": 421
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 106.42188262939453,
      "epoch": 6.205882352941177,
      "grad_norm": 1.534235488134726,
      "kl": 0.050537109375,
      "learning_rate": 6.897058823529411e-07,
      "loss": 0.002,
      "reward": 1.896329402923584,
      "reward_std": 0.018744876608252525,
      "rewards/accuracy_reward": 0.8963293433189392,
      "rewards/format_reward": 1.0,
      "step": 422
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 107.9831771850586,
      "epoch": 6.220588235294118,
      "grad_norm": 1.5766135493091737,
      "kl": 0.0478515625,
      "learning_rate": 6.889705882352941e-07,
      "loss": 0.0019,
      "reward": 1.8838109970092773,
      "reward_std": 0.02198025956749916,
      "rewards/accuracy_reward": 0.8850129246711731,
      "rewards/format_reward": 0.998798131942749,
      "step": 423
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 105.84616088867188,
      "epoch": 6.235294117647059,
      "grad_norm": 2.0708203706815853,
      "kl": 0.078125,
      "learning_rate": 6.88235294117647e-07,
      "loss": 0.0031,
      "reward": 1.8984771966934204,
      "reward_std": 0.021190447732806206,
      "rewards/accuracy_reward": 0.8984770178794861,
      "rewards/format_reward": 1.0,
      "step": 424
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 104.41226196289062,
      "epoch": 6.25,
      "grad_norm": 1.7348463789648598,
      "kl": 0.050048828125,
      "learning_rate": 6.875e-07,
      "loss": 0.002,
      "reward": 1.9097307920455933,
      "reward_std": 0.019229857251048088,
      "rewards/accuracy_reward": 0.9097307324409485,
      "rewards/format_reward": 1.0,
      "step": 425
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 104.37500762939453,
      "epoch": 6.264705882352941,
      "grad_norm": 15.785742386708213,
      "kl": 0.05224609375,
      "learning_rate": 6.86764705882353e-07,
      "loss": 0.0021,
      "reward": 1.8983619213104248,
      "reward_std": 0.018431352451443672,
      "rewards/accuracy_reward": 0.8983619213104248,
      "rewards/format_reward": 1.0,
      "step": 426
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 103.81009674072266,
      "epoch": 6.279411764705882,
      "grad_norm": 2.6324515777384017,
      "kl": 0.056396484375,
      "learning_rate": 6.860294117647058e-07,
      "loss": 0.0023,
      "reward": 1.8895386457443237,
      "reward_std": 0.02556595578789711,
      "rewards/accuracy_reward": 0.889538586139679,
      "rewards/format_reward": 1.0,
      "step": 427
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 102.29808044433594,
      "epoch": 6.294117647058823,
      "grad_norm": 1.6012975476940077,
      "kl": 0.04931640625,
      "learning_rate": 6.852941176470588e-07,
      "loss": 0.002,
      "reward": 1.9019221067428589,
      "reward_std": 0.023045556619763374,
      "rewards/accuracy_reward": 0.9031239748001099,
      "rewards/format_reward": 0.998798131942749,
      "step": 428
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 102.78125762939453,
      "epoch": 6.3088235294117645,
      "grad_norm": 2.761466006204092,
      "kl": 0.0732421875,
      "learning_rate": 6.845588235294118e-07,
      "loss": 0.0029,
      "reward": 1.8907809257507324,
      "reward_std": 0.021768517792224884,
      "rewards/accuracy_reward": 0.8907808661460876,
      "rewards/format_reward": 1.0,
      "step": 429
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 104.83413696289062,
      "epoch": 6.323529411764706,
      "grad_norm": 1.9867024053824853,
      "kl": 0.05517578125,
      "learning_rate": 6.838235294117647e-07,
      "loss": 0.0022,
      "reward": 1.8989307880401611,
      "reward_std": 0.02491842210292816,
      "rewards/accuracy_reward": 0.9001325964927673,
      "rewards/format_reward": 0.998798131942749,
      "step": 430
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 104.92548370361328,
      "epoch": 6.338235294117647,
      "grad_norm": 2.506151416718587,
      "kl": 0.060546875,
      "learning_rate": 6.830882352941176e-07,
      "loss": 0.0024,
      "reward": 1.8981561660766602,
      "reward_std": 0.022075144574046135,
      "rewards/accuracy_reward": 0.8981561064720154,
      "rewards/format_reward": 1.0,
      "step": 431
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 106.0300521850586,
      "epoch": 6.352941176470588,
      "grad_norm": 3.453658386620713,
      "kl": 0.052734375,
      "learning_rate": 6.823529411764706e-07,
      "loss": 0.0021,
      "reward": 1.8945103883743286,
      "reward_std": 0.02551359124481678,
      "rewards/accuracy_reward": 0.8969140648841858,
      "rewards/format_reward": 0.9975962042808533,
      "step": 432
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 106.82933044433594,
      "epoch": 6.367647058823529,
      "grad_norm": 1.8291376872404717,
      "kl": 0.049072265625,
      "learning_rate": 6.816176470588235e-07,
      "loss": 0.002,
      "reward": 1.8951138257980347,
      "reward_std": 0.02293815091252327,
      "rewards/accuracy_reward": 0.8951138257980347,
      "rewards/format_reward": 1.0,
      "step": 433
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 105.97356414794922,
      "epoch": 6.382352941176471,
      "grad_norm": 4.4657449515144965,
      "kl": 0.051025390625,
      "learning_rate": 6.808823529411765e-07,
      "loss": 0.002,
      "reward": 1.8871821165084839,
      "reward_std": 0.02225620485842228,
      "rewards/accuracy_reward": 0.8871821165084839,
      "rewards/format_reward": 1.0,
      "step": 434
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 108.1033706665039,
      "epoch": 6.397058823529412,
      "grad_norm": 1.0719524860130538,
      "kl": 0.05029296875,
      "learning_rate": 6.801470588235295e-07,
      "loss": 0.002,
      "reward": 1.8991506099700928,
      "reward_std": 0.024187587201595306,
      "rewards/accuracy_reward": 0.9003525376319885,
      "rewards/format_reward": 0.998798131942749,
      "step": 435
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 108.64063262939453,
      "epoch": 6.411764705882353,
      "grad_norm": 1.3740348079978744,
      "kl": 0.04736328125,
      "learning_rate": 6.794117647058823e-07,
      "loss": 0.0019,
      "reward": 1.8954823017120361,
      "reward_std": 0.024401018396019936,
      "rewards/accuracy_reward": 0.8954823017120361,
      "rewards/format_reward": 1.0,
      "step": 436
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 110.45433044433594,
      "epoch": 6.426470588235294,
      "grad_norm": 2.5851908946821287,
      "kl": 0.05126953125,
      "learning_rate": 6.786764705882353e-07,
      "loss": 0.002,
      "reward": 1.9073567390441895,
      "reward_std": 0.020375726744532585,
      "rewards/accuracy_reward": 0.9073567390441895,
      "rewards/format_reward": 1.0,
      "step": 437
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 111.93389892578125,
      "epoch": 6.4411764705882355,
      "grad_norm": 1.167566426650332,
      "kl": 0.048828125,
      "learning_rate": 6.779411764705883e-07,
      "loss": 0.0019,
      "reward": 1.8960213661193848,
      "reward_std": 0.024139244109392166,
      "rewards/accuracy_reward": 0.8960214257240295,
      "rewards/format_reward": 1.0,
      "step": 438
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 112.06610870361328,
      "epoch": 6.455882352941177,
      "grad_norm": 1.1659335106544957,
      "kl": 0.047607421875,
      "learning_rate": 6.772058823529412e-07,
      "loss": 0.0019,
      "reward": 1.8937411308288574,
      "reward_std": 0.02366737276315689,
      "rewards/accuracy_reward": 0.8949430584907532,
      "rewards/format_reward": 0.998798131942749,
      "step": 439
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 112.38341522216797,
      "epoch": 6.470588235294118,
      "grad_norm": 3.3358776773497176,
      "kl": 0.048095703125,
      "learning_rate": 6.764705882352941e-07,
      "loss": 0.0019,
      "reward": 1.8928595781326294,
      "reward_std": 0.021879835054278374,
      "rewards/accuracy_reward": 0.8928595781326294,
      "rewards/format_reward": 1.0,
      "step": 440
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 114.35216522216797,
      "epoch": 6.485294117647059,
      "grad_norm": 1.3906028161781399,
      "kl": 0.052490234375,
      "learning_rate": 6.757352941176471e-07,
      "loss": 0.0021,
      "reward": 1.8927258253097534,
      "reward_std": 0.020552638918161392,
      "rewards/accuracy_reward": 0.8927258849143982,
      "rewards/format_reward": 1.0,
      "step": 441
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 114.82091522216797,
      "epoch": 6.5,
      "grad_norm": 3.8548633882230567,
      "kl": 0.059814453125,
      "learning_rate": 6.75e-07,
      "loss": 0.0024,
      "reward": 1.9043219089508057,
      "reward_std": 0.02147606387734413,
      "rewards/accuracy_reward": 0.9055235981941223,
      "rewards/format_reward": 0.998798131942749,
      "step": 442
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 112.60456848144531,
      "epoch": 6.514705882352941,
      "grad_norm": 2.723744256999097,
      "kl": 0.05419921875,
      "learning_rate": 6.74264705882353e-07,
      "loss": 0.0022,
      "reward": 1.8985867500305176,
      "reward_std": 0.017505373805761337,
      "rewards/accuracy_reward": 0.8985867500305176,
      "rewards/format_reward": 1.0,
      "step": 443
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 113.91466522216797,
      "epoch": 6.529411764705882,
      "grad_norm": 2.467770138712217,
      "kl": 0.05126953125,
      "learning_rate": 6.735294117647058e-07,
      "loss": 0.0021,
      "reward": 1.8991594314575195,
      "reward_std": 0.0224330872297287,
      "rewards/accuracy_reward": 0.9003612995147705,
      "rewards/format_reward": 0.998798131942749,
      "step": 444
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 112.12981414794922,
      "epoch": 6.544117647058823,
      "grad_norm": 1.235445619018083,
      "kl": 0.056396484375,
      "learning_rate": 6.727941176470588e-07,
      "loss": 0.0023,
      "reward": 1.8991564512252808,
      "reward_std": 0.01990494504570961,
      "rewards/accuracy_reward": 0.900358259677887,
      "rewards/format_reward": 0.998798131942749,
      "step": 445
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 112.72716522216797,
      "epoch": 6.5588235294117645,
      "grad_norm": 4.57392237252639,
      "kl": 0.0576171875,
      "learning_rate": 6.720588235294118e-07,
      "loss": 0.0023,
      "reward": 1.8918448686599731,
      "reward_std": 0.018118849024176598,
      "rewards/accuracy_reward": 0.8918446898460388,
      "rewards/format_reward": 1.0,
      "step": 446
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 113.4519271850586,
      "epoch": 6.573529411764706,
      "grad_norm": 1.3515959803768587,
      "kl": 0.05908203125,
      "learning_rate": 6.713235294117646e-07,
      "loss": 0.0024,
      "reward": 1.8976571559906006,
      "reward_std": 0.019342336803674698,
      "rewards/accuracy_reward": 0.897657036781311,
      "rewards/format_reward": 1.0,
      "step": 447
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 113.30889892578125,
      "epoch": 6.588235294117647,
      "grad_norm": 2.000337423528591,
      "kl": 0.0654296875,
      "learning_rate": 6.705882352941176e-07,
      "loss": 0.0026,
      "reward": 1.9029864072799683,
      "reward_std": 0.01870027370750904,
      "rewards/accuracy_reward": 0.9029862880706787,
      "rewards/format_reward": 1.0,
      "step": 448
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 112.02764892578125,
      "epoch": 6.602941176470588,
      "grad_norm": 1.5831041809663697,
      "kl": 0.05908203125,
      "learning_rate": 6.698529411764705e-07,
      "loss": 0.0024,
      "reward": 1.8981215953826904,
      "reward_std": 0.023059705272316933,
      "rewards/accuracy_reward": 0.8993234038352966,
      "rewards/format_reward": 0.998798131942749,
      "step": 449
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 110.4471206665039,
      "epoch": 6.617647058823529,
      "grad_norm": 1.7615653292492308,
      "kl": 0.061279296875,
      "learning_rate": 6.691176470588234e-07,
      "loss": 0.0025,
      "reward": 1.8908604383468628,
      "reward_std": 0.02007007598876953,
      "rewards/accuracy_reward": 0.890860378742218,
      "rewards/format_reward": 1.0,
      "step": 450
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 112.1346206665039,
      "epoch": 6.632352941176471,
      "grad_norm": 1.6994254050705504,
      "kl": 0.06298828125,
      "learning_rate": 6.683823529411764e-07,
      "loss": 0.0025,
      "reward": 1.9058598279953003,
      "reward_std": 0.017709294334053993,
      "rewards/accuracy_reward": 0.9058597683906555,
      "rewards/format_reward": 1.0,
      "step": 451
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 110.54688262939453,
      "epoch": 6.647058823529412,
      "grad_norm": 2.143476953028863,
      "kl": 0.051513671875,
      "learning_rate": 6.676470588235294e-07,
      "loss": 0.0021,
      "reward": 1.9037673473358154,
      "reward_std": 0.01912897825241089,
      "rewards/accuracy_reward": 0.9037670493125916,
      "rewards/format_reward": 1.0,
      "step": 452
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 110.27644348144531,
      "epoch": 6.661764705882353,
      "grad_norm": 2.01478887608093,
      "kl": 0.05029296875,
      "learning_rate": 6.669117647058822e-07,
      "loss": 0.002,
      "reward": 1.8817764520645142,
      "reward_std": 0.0232061930000782,
      "rewards/accuracy_reward": 0.8829783201217651,
      "rewards/format_reward": 0.998798131942749,
      "step": 453
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 110.56731414794922,
      "epoch": 6.676470588235294,
      "grad_norm": 2.690844672468002,
      "kl": 0.052978515625,
      "learning_rate": 6.661764705882352e-07,
      "loss": 0.0021,
      "reward": 1.923115849494934,
      "reward_std": 0.014602301642298698,
      "rewards/accuracy_reward": 0.9231158494949341,
      "rewards/format_reward": 1.0,
      "step": 454
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 109.59495544433594,
      "epoch": 6.6911764705882355,
      "grad_norm": 1.7310069495348166,
      "kl": 0.048095703125,
      "learning_rate": 6.654411764705882e-07,
      "loss": 0.0019,
      "reward": 1.901757836341858,
      "reward_std": 0.019638314843177795,
      "rewards/accuracy_reward": 0.9017577767372131,
      "rewards/format_reward": 1.0,
      "step": 455
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 111.0877456665039,
      "epoch": 6.705882352941177,
      "grad_norm": 4.401458379456843,
      "kl": 0.052978515625,
      "learning_rate": 6.647058823529411e-07,
      "loss": 0.0021,
      "reward": 1.9064425230026245,
      "reward_std": 0.018589207902550697,
      "rewards/accuracy_reward": 0.9076444506645203,
      "rewards/format_reward": 0.998798131942749,
      "step": 456
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 110.69231414794922,
      "epoch": 6.720588235294118,
      "grad_norm": 2.0195398931401867,
      "kl": 0.04736328125,
      "learning_rate": 6.63970588235294e-07,
      "loss": 0.0019,
      "reward": 1.8995909690856934,
      "reward_std": 0.015642177313566208,
      "rewards/accuracy_reward": 0.8995909094810486,
      "rewards/format_reward": 1.0,
      "step": 457
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 110.81731414794922,
      "epoch": 6.735294117647059,
      "grad_norm": 1.5989334696664628,
      "kl": 0.05224609375,
      "learning_rate": 6.63235294117647e-07,
      "loss": 0.0021,
      "reward": 1.8981314897537231,
      "reward_std": 0.020380795001983643,
      "rewards/accuracy_reward": 0.8981314897537231,
      "rewards/format_reward": 1.0,
      "step": 458
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 111.00841522216797,
      "epoch": 6.75,
      "grad_norm": 3.993916385712523,
      "kl": 0.045166015625,
      "learning_rate": 6.624999999999999e-07,
      "loss": 0.0018,
      "reward": 1.9087324142456055,
      "reward_std": 0.018654441460967064,
      "rewards/accuracy_reward": 0.9087323546409607,
      "rewards/format_reward": 1.0,
      "step": 459
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 106.8689956665039,
      "epoch": 6.764705882352941,
      "grad_norm": 5.917555869011424,
      "kl": 0.0458984375,
      "learning_rate": 6.617647058823529e-07,
      "loss": 0.0018,
      "reward": 1.9015119075775146,
      "reward_std": 0.01625223271548748,
      "rewards/accuracy_reward": 0.9015120267868042,
      "rewards/format_reward": 1.0,
      "step": 460
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 109.99639892578125,
      "epoch": 6.779411764705882,
      "grad_norm": 2.4437828911603607,
      "kl": 0.047607421875,
      "learning_rate": 6.610294117647059e-07,
      "loss": 0.0019,
      "reward": 1.8872531652450562,
      "reward_std": 0.020600629970431328,
      "rewards/accuracy_reward": 0.8872530460357666,
      "rewards/format_reward": 1.0,
      "step": 461
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 110.85456848144531,
      "epoch": 6.794117647058823,
      "grad_norm": 1.220750465871548,
      "kl": 0.054931640625,
      "learning_rate": 6.602941176470587e-07,
      "loss": 0.0022,
      "reward": 1.902260661125183,
      "reward_std": 0.020439889281988144,
      "rewards/accuracy_reward": 0.9034625291824341,
      "rewards/format_reward": 0.998798131942749,
      "step": 462
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 110.22116088867188,
      "epoch": 6.8088235294117645,
      "grad_norm": 2.3522146924761804,
      "kl": 0.048583984375,
      "learning_rate": 6.595588235294117e-07,
      "loss": 0.0019,
      "reward": 1.8962863683700562,
      "reward_std": 0.0212849248200655,
      "rewards/accuracy_reward": 0.8962863087654114,
      "rewards/format_reward": 1.0,
      "step": 463
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 108.49038696289062,
      "epoch": 6.823529411764706,
      "grad_norm": 2.653649647480221,
      "kl": 0.054931640625,
      "learning_rate": 6.588235294117647e-07,
      "loss": 0.0022,
      "reward": 1.907329797744751,
      "reward_std": 0.01719941571354866,
      "rewards/accuracy_reward": 0.9073297381401062,
      "rewards/format_reward": 1.0,
      "step": 464
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 107.78606414794922,
      "epoch": 6.838235294117647,
      "grad_norm": 1.5113941947142104,
      "kl": 0.07421875,
      "learning_rate": 6.580882352941176e-07,
      "loss": 0.003,
      "reward": 1.905678153038025,
      "reward_std": 0.027055159211158752,
      "rewards/accuracy_reward": 0.9092837572097778,
      "rewards/format_reward": 0.9963942766189575,
      "step": 465
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 107.69831848144531,
      "epoch": 6.852941176470588,
      "grad_norm": 2.845621962915759,
      "kl": 0.054443359375,
      "learning_rate": 6.573529411764705e-07,
      "loss": 0.0022,
      "reward": 1.900158405303955,
      "reward_std": 0.0176254715770483,
      "rewards/accuracy_reward": 0.9001582264900208,
      "rewards/format_reward": 1.0,
      "step": 466
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 108.25721740722656,
      "epoch": 6.867647058823529,
      "grad_norm": 1.7461784259949287,
      "kl": 0.060546875,
      "learning_rate": 6.566176470588235e-07,
      "loss": 0.0024,
      "reward": 1.9130123853683472,
      "reward_std": 0.018750464543700218,
      "rewards/accuracy_reward": 0.9130123257637024,
      "rewards/format_reward": 1.0,
      "step": 467
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 108.20433044433594,
      "epoch": 6.882352941176471,
      "grad_norm": 2.1022701427819346,
      "kl": 0.051025390625,
      "learning_rate": 6.558823529411764e-07,
      "loss": 0.002,
      "reward": 1.9070162773132324,
      "reward_std": 0.01999017596244812,
      "rewards/accuracy_reward": 0.9070160388946533,
      "rewards/format_reward": 1.0,
      "step": 468
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 109.2596206665039,
      "epoch": 6.897058823529412,
      "grad_norm": 1.9884301824289226,
      "kl": 0.05322265625,
      "learning_rate": 6.551470588235294e-07,
      "loss": 0.0021,
      "reward": 1.8830878734588623,
      "reward_std": 0.025470944121479988,
      "rewards/accuracy_reward": 0.884289562702179,
      "rewards/format_reward": 0.998798131942749,
      "step": 469
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 109.24639892578125,
      "epoch": 6.911764705882353,
      "grad_norm": 2.363006531050007,
      "kl": 0.060791015625,
      "learning_rate": 6.544117647058824e-07,
      "loss": 0.0024,
      "reward": 1.899465799331665,
      "reward_std": 0.018820466473698616,
      "rewards/accuracy_reward": 0.8994657397270203,
      "rewards/format_reward": 1.0,
      "step": 470
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 109.1189956665039,
      "epoch": 6.926470588235294,
      "grad_norm": 1.955011720116878,
      "kl": 0.053466796875,
      "learning_rate": 6.536764705882352e-07,
      "loss": 0.0021,
      "reward": 1.9155514240264893,
      "reward_std": 0.017393523827195168,
      "rewards/accuracy_reward": 0.9155513048171997,
      "rewards/format_reward": 1.0,
      "step": 471
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 109.6971206665039,
      "epoch": 6.9411764705882355,
      "grad_norm": 1.9269477392518999,
      "kl": 0.050048828125,
      "learning_rate": 6.529411764705882e-07,
      "loss": 0.002,
      "reward": 1.8984498977661133,
      "reward_std": 0.019148575142025948,
      "rewards/accuracy_reward": 0.8996517658233643,
      "rewards/format_reward": 0.998798131942749,
      "step": 472
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 107.8221206665039,
      "epoch": 6.955882352941177,
      "grad_norm": 2.9581301695718514,
      "kl": 0.0537109375,
      "learning_rate": 6.522058823529412e-07,
      "loss": 0.0021,
      "reward": 1.8932065963745117,
      "reward_std": 0.0210660882294178,
      "rewards/accuracy_reward": 0.8932065367698669,
      "rewards/format_reward": 1.0,
      "step": 473
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 107.91947174072266,
      "epoch": 6.970588235294118,
      "grad_norm": 2.111290033794273,
      "kl": 0.059326171875,
      "learning_rate": 6.514705882352941e-07,
      "loss": 0.0024,
      "reward": 1.901350498199463,
      "reward_std": 0.021692074835300446,
      "rewards/accuracy_reward": 0.9025523662567139,
      "rewards/format_reward": 0.998798131942749,
      "step": 474
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 109.16947174072266,
      "epoch": 6.985294117647059,
      "grad_norm": 2.2057304416622108,
      "kl": 0.060791015625,
      "learning_rate": 6.50735294117647e-07,
      "loss": 0.0024,
      "reward": 1.9076955318450928,
      "reward_std": 0.021920649334788322,
      "rewards/accuracy_reward": 0.9088975191116333,
      "rewards/format_reward": 0.998798131942749,
      "step": 475
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 109.17960357666016,
      "epoch": 7.0,
      "grad_norm": 1.617597906920766,
      "kl": 0.056396484375,
      "learning_rate": 6.5e-07,
      "loss": 0.0023,
      "reward": 1.9078389406204224,
      "reward_std": 0.023063138127326965,
      "rewards/accuracy_reward": 0.9078390598297119,
      "rewards/format_reward": 1.0,
      "step": 476
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 108.60216522216797,
      "epoch": 7.014705882352941,
      "grad_norm": 1.6817279506404212,
      "kl": 0.0546875,
      "learning_rate": 6.492647058823529e-07,
      "loss": 0.0022,
      "reward": 1.9104756116867065,
      "reward_std": 0.020941542461514473,
      "rewards/accuracy_reward": 0.9116773009300232,
      "rewards/format_reward": 0.998798131942749,
      "step": 477
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 111.2752456665039,
      "epoch": 7.029411764705882,
      "grad_norm": 2.3781369604321836,
      "kl": 0.052490234375,
      "learning_rate": 6.485294117647059e-07,
      "loss": 0.0021,
      "reward": 1.8896749019622803,
      "reward_std": 0.020192930474877357,
      "rewards/accuracy_reward": 0.8896746635437012,
      "rewards/format_reward": 1.0,
      "step": 478
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 110.77163696289062,
      "epoch": 7.044117647058823,
      "grad_norm": 1.276439509093276,
      "kl": 0.059326171875,
      "learning_rate": 6.477941176470589e-07,
      "loss": 0.0024,
      "reward": 1.8859210014343262,
      "reward_std": 0.022135620936751366,
      "rewards/accuracy_reward": 0.8859207630157471,
      "rewards/format_reward": 1.0,
      "step": 479
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 110.27163696289062,
      "epoch": 7.0588235294117645,
      "grad_norm": 2.4358423376131815,
      "kl": 0.047607421875,
      "learning_rate": 6.470588235294117e-07,
      "loss": 0.0019,
      "reward": 1.8972910642623901,
      "reward_std": 0.020838702097535133,
      "rewards/accuracy_reward": 0.8972910642623901,
      "rewards/format_reward": 1.0,
      "step": 480
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 110.93991088867188,
      "epoch": 7.073529411764706,
      "grad_norm": 2.4539112139463777,
      "kl": 0.05078125,
      "learning_rate": 6.463235294117647e-07,
      "loss": 0.002,
      "reward": 1.8896875381469727,
      "reward_std": 0.0205745380371809,
      "rewards/accuracy_reward": 0.8896873593330383,
      "rewards/format_reward": 1.0,
      "step": 481
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 109.85697174072266,
      "epoch": 7.088235294117647,
      "grad_norm": 1.5146047739430706,
      "kl": 0.054931640625,
      "learning_rate": 6.455882352941177e-07,
      "loss": 0.0022,
      "reward": 1.9029262065887451,
      "reward_std": 0.020332474261522293,
      "rewards/accuracy_reward": 0.9029260873794556,
      "rewards/format_reward": 1.0,
      "step": 482
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 112.64543914794922,
      "epoch": 7.102941176470588,
      "grad_norm": 1.6263090300209622,
      "kl": 0.049560546875,
      "learning_rate": 6.448529411764706e-07,
      "loss": 0.002,
      "reward": 1.9050016403198242,
      "reward_std": 0.020645473152399063,
      "rewards/accuracy_reward": 0.9050015807151794,
      "rewards/format_reward": 1.0,
      "step": 483
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 112.32452392578125,
      "epoch": 7.117647058823529,
      "grad_norm": 2.386969995018498,
      "kl": 0.04833984375,
      "learning_rate": 6.441176470588235e-07,
      "loss": 0.0019,
      "reward": 1.8980265855789185,
      "reward_std": 0.02636340633034706,
      "rewards/accuracy_reward": 0.9016323685646057,
      "rewards/format_reward": 0.9963942766189575,
      "step": 484
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 113.62741088867188,
      "epoch": 7.132352941176471,
      "grad_norm": 2.488441055097933,
      "kl": 0.060302734375,
      "learning_rate": 6.433823529411765e-07,
      "loss": 0.0024,
      "reward": 1.8940205574035645,
      "reward_std": 0.021218711510300636,
      "rewards/accuracy_reward": 0.8940205574035645,
      "rewards/format_reward": 1.0,
      "step": 485
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 113.08173370361328,
      "epoch": 7.147058823529412,
      "grad_norm": 1.4885294609323232,
      "kl": 0.055419921875,
      "learning_rate": 6.426470588235294e-07,
      "loss": 0.0022,
      "reward": 1.9081767797470093,
      "reward_std": 0.021031972020864487,
      "rewards/accuracy_reward": 0.9093785881996155,
      "rewards/format_reward": 0.998798131942749,
      "step": 486
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 114.18029022216797,
      "epoch": 7.161764705882353,
      "grad_norm": 2.3558483279167257,
      "kl": 0.050048828125,
      "learning_rate": 6.419117647058824e-07,
      "loss": 0.002,
      "reward": 1.9089596271514893,
      "reward_std": 0.021060261875391006,
      "rewards/accuracy_reward": 0.910161554813385,
      "rewards/format_reward": 0.998798131942749,
      "step": 487
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 113.5300521850586,
      "epoch": 7.176470588235294,
      "grad_norm": 2.185868565772857,
      "kl": 0.051513671875,
      "learning_rate": 6.411764705882354e-07,
      "loss": 0.002,
      "reward": 1.895540475845337,
      "reward_std": 0.02588125877082348,
      "rewards/accuracy_reward": 0.8991460204124451,
      "rewards/format_reward": 0.9963942766189575,
      "step": 488
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 113.70313262939453,
      "epoch": 7.1911764705882355,
      "grad_norm": 13.781167768252143,
      "kl": 0.0556640625,
      "learning_rate": 6.404411764705881e-07,
      "loss": 0.0022,
      "reward": 1.900436520576477,
      "reward_std": 0.018973568454384804,
      "rewards/accuracy_reward": 0.9004363417625427,
      "rewards/format_reward": 1.0,
      "step": 489
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 113.69231414794922,
      "epoch": 7.205882352941177,
      "grad_norm": 1.787139009281532,
      "kl": 0.055908203125,
      "learning_rate": 6.397058823529411e-07,
      "loss": 0.0022,
      "reward": 1.914228916168213,
      "reward_std": 0.025157246738672256,
      "rewards/accuracy_reward": 0.9166327714920044,
      "rewards/format_reward": 0.9975962042808533,
      "step": 490
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 112.53606414794922,
      "epoch": 7.220588235294118,
      "grad_norm": 1.3383135392600143,
      "kl": 0.05322265625,
      "learning_rate": 6.389705882352941e-07,
      "loss": 0.0021,
      "reward": 1.9147520065307617,
      "reward_std": 0.01828881911933422,
      "rewards/accuracy_reward": 0.9147518277168274,
      "rewards/format_reward": 1.0,
      "step": 491
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 111.80769348144531,
      "epoch": 7.235294117647059,
      "grad_norm": 1.499587887382779,
      "kl": 0.049560546875,
      "learning_rate": 6.38235294117647e-07,
      "loss": 0.002,
      "reward": 1.9006352424621582,
      "reward_std": 0.021384483203291893,
      "rewards/accuracy_reward": 0.901837170124054,
      "rewards/format_reward": 0.998798131942749,
      "step": 492
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 111.85216522216797,
      "epoch": 7.25,
      "grad_norm": 9.51747929867123,
      "kl": 0.052978515625,
      "learning_rate": 6.374999999999999e-07,
      "loss": 0.0021,
      "reward": 1.9037039279937744,
      "reward_std": 0.017083317041397095,
      "rewards/accuracy_reward": 0.9037038683891296,
      "rewards/format_reward": 1.0,
      "step": 493
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 111.1863021850586,
      "epoch": 7.264705882352941,
      "grad_norm": 2.190318540866678,
      "kl": 0.052978515625,
      "learning_rate": 6.367647058823529e-07,
      "loss": 0.0021,
      "reward": 1.9090691804885864,
      "reward_std": 0.016836341470479965,
      "rewards/accuracy_reward": 0.9090691804885864,
      "rewards/format_reward": 1.0,
      "step": 494
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 110.6658706665039,
      "epoch": 7.279411764705882,
      "grad_norm": 16.171603319545223,
      "kl": 0.0576171875,
      "learning_rate": 6.360294117647058e-07,
      "loss": 0.0023,
      "reward": 1.8776507377624512,
      "reward_std": 0.018403498455882072,
      "rewards/accuracy_reward": 0.8776505589485168,
      "rewards/format_reward": 1.0,
      "step": 495
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 111.74519348144531,
      "epoch": 7.294117647058823,
      "grad_norm": 1.9202223901107958,
      "kl": 0.054931640625,
      "learning_rate": 6.352941176470588e-07,
      "loss": 0.0022,
      "reward": 1.883994698524475,
      "reward_std": 0.018577104434370995,
      "rewards/accuracy_reward": 0.8839945197105408,
      "rewards/format_reward": 1.0,
      "step": 496
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 111.7752456665039,
      "epoch": 7.3088235294117645,
      "grad_norm": 2.7194904114682545,
      "kl": 0.05908203125,
      "learning_rate": 6.345588235294117e-07,
      "loss": 0.0024,
      "reward": 1.9002341032028198,
      "reward_std": 0.02195299230515957,
      "rewards/accuracy_reward": 0.9026379585266113,
      "rewards/format_reward": 0.9975962042808533,
      "step": 497
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 111.41106414794922,
      "epoch": 7.323529411764706,
      "grad_norm": 2.818503081927554,
      "kl": 0.0546875,
      "learning_rate": 6.338235294117646e-07,
      "loss": 0.0022,
      "reward": 1.9031240940093994,
      "reward_std": 0.021893031895160675,
      "rewards/accuracy_reward": 0.9055278897285461,
      "rewards/format_reward": 0.9975962042808533,
      "step": 498
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 111.1658706665039,
      "epoch": 7.338235294117647,
      "grad_norm": 2.2904375345608683,
      "kl": 0.054443359375,
      "learning_rate": 6.330882352941176e-07,
      "loss": 0.0022,
      "reward": 1.8997858762741089,
      "reward_std": 0.01683872379362583,
      "rewards/accuracy_reward": 0.8997858762741089,
      "rewards/format_reward": 1.0,
      "step": 499
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 108.5877456665039,
      "epoch": 7.352941176470588,
      "grad_norm": 1.9874226919695666,
      "kl": 0.05712890625,
      "learning_rate": 6.323529411764706e-07,
      "loss": 0.0023,
      "reward": 1.9213850498199463,
      "reward_std": 0.016001444309949875,
      "rewards/accuracy_reward": 0.9213849306106567,
      "rewards/format_reward": 1.0,
      "step": 500
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 109.45433044433594,
      "epoch": 7.367647058823529,
      "grad_norm": 1.4810061564924804,
      "kl": 0.06005859375,
      "learning_rate": 6.316176470588234e-07,
      "loss": 0.0024,
      "reward": 1.9132663011550903,
      "reward_std": 0.016456911340355873,
      "rewards/accuracy_reward": 0.9132663011550903,
      "rewards/format_reward": 1.0,
      "step": 501
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 109.87259674072266,
      "epoch": 7.382352941176471,
      "grad_norm": 2.354230850080614,
      "kl": 0.0537109375,
      "learning_rate": 6.308823529411764e-07,
      "loss": 0.0021,
      "reward": 1.9061925411224365,
      "reward_std": 0.015117881819605827,
      "rewards/accuracy_reward": 0.906192421913147,
      "rewards/format_reward": 1.0,
      "step": 502
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 109.88221740722656,
      "epoch": 7.397058823529412,
      "grad_norm": 2.702198636610281,
      "kl": 0.055908203125,
      "learning_rate": 6.301470588235294e-07,
      "loss": 0.0023,
      "reward": 1.8956823348999023,
      "reward_std": 0.017957165837287903,
      "rewards/accuracy_reward": 0.8956822752952576,
      "rewards/format_reward": 1.0,
      "step": 503
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 107.91827392578125,
      "epoch": 7.411764705882353,
      "grad_norm": 2.707910090355551,
      "kl": 0.05517578125,
      "learning_rate": 6.294117647058823e-07,
      "loss": 0.0022,
      "reward": 1.8958656787872314,
      "reward_std": 0.01692720130085945,
      "rewards/accuracy_reward": 0.8958656787872314,
      "rewards/format_reward": 1.0,
      "step": 504
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 107.41106414794922,
      "epoch": 7.426470588235294,
      "grad_norm": 2.732018255754022,
      "kl": 0.06884765625,
      "learning_rate": 6.286764705882353e-07,
      "loss": 0.0028,
      "reward": 1.907556176185608,
      "reward_std": 0.016339732334017754,
      "rewards/accuracy_reward": 0.9075562953948975,
      "rewards/format_reward": 1.0,
      "step": 505
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 107.5252456665039,
      "epoch": 7.4411764705882355,
      "grad_norm": 1.9561974612826654,
      "kl": 0.061279296875,
      "learning_rate": 6.279411764705882e-07,
      "loss": 0.0025,
      "reward": 1.9172346591949463,
      "reward_std": 0.017858320847153664,
      "rewards/accuracy_reward": 0.9184365272521973,
      "rewards/format_reward": 0.998798131942749,
      "step": 506
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 108.40745544433594,
      "epoch": 7.455882352941177,
      "grad_norm": 4.587975830736582,
      "kl": 0.07080078125,
      "learning_rate": 6.272058823529411e-07,
      "loss": 0.0028,
      "reward": 1.9027947187423706,
      "reward_std": 0.014803332276642323,
      "rewards/accuracy_reward": 0.9027946591377258,
      "rewards/format_reward": 1.0,
      "step": 507
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 106.92909240722656,
      "epoch": 7.470588235294118,
      "grad_norm": 3.0606242226177085,
      "kl": 0.05859375,
      "learning_rate": 6.264705882352941e-07,
      "loss": 0.0023,
      "reward": 1.9170668125152588,
      "reward_std": 0.01804046705365181,
      "rewards/accuracy_reward": 0.9182686805725098,
      "rewards/format_reward": 0.998798131942749,
      "step": 508
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 105.75000762939453,
      "epoch": 7.485294117647059,
      "grad_norm": 1.565766574508637,
      "kl": 0.06005859375,
      "learning_rate": 6.257352941176471e-07,
      "loss": 0.0024,
      "reward": 1.9129410982131958,
      "reward_std": 0.016586054116487503,
      "rewards/accuracy_reward": 0.914142906665802,
      "rewards/format_reward": 0.998798131942749,
      "step": 509
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 108.11659240722656,
      "epoch": 7.5,
      "grad_norm": 2.188031838473879,
      "kl": 0.06494140625,
      "learning_rate": 6.249999999999999e-07,
      "loss": 0.0026,
      "reward": 1.9109857082366943,
      "reward_std": 0.01680620200932026,
      "rewards/accuracy_reward": 0.9109857082366943,
      "rewards/format_reward": 1.0,
      "step": 510
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 107.4206771850586,
      "epoch": 7.514705882352941,
      "grad_norm": 2.301709034695236,
      "kl": 0.05419921875,
      "learning_rate": 6.242647058823529e-07,
      "loss": 0.0022,
      "reward": 1.8962676525115967,
      "reward_std": 0.01672891154885292,
      "rewards/accuracy_reward": 0.8962674736976624,
      "rewards/format_reward": 1.0,
      "step": 511
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 109.53726196289062,
      "epoch": 7.529411764705882,
      "grad_norm": 2.3427553949456135,
      "kl": 0.057373046875,
      "learning_rate": 6.235294117647059e-07,
      "loss": 0.0023,
      "reward": 1.9040287733078003,
      "reward_std": 0.017709048464894295,
      "rewards/accuracy_reward": 0.9040287137031555,
      "rewards/format_reward": 1.0,
      "step": 512
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 108.12019348144531,
      "epoch": 7.544117647058823,
      "grad_norm": 1.665426870174403,
      "kl": 0.0537109375,
      "learning_rate": 6.227941176470588e-07,
      "loss": 0.0022,
      "reward": 1.88170325756073,
      "reward_std": 0.017501268535852432,
      "rewards/accuracy_reward": 0.8817031979560852,
      "rewards/format_reward": 1.0,
      "step": 513
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 108.15144348144531,
      "epoch": 7.5588235294117645,
      "grad_norm": 1.6488965895634309,
      "kl": 0.06591796875,
      "learning_rate": 6.220588235294118e-07,
      "loss": 0.0026,
      "reward": 1.8925837278366089,
      "reward_std": 0.015000036917626858,
      "rewards/accuracy_reward": 0.8925836682319641,
      "rewards/format_reward": 1.0,
      "step": 514
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 109.16106414794922,
      "epoch": 7.573529411764706,
      "grad_norm": 1.5981543745714246,
      "kl": 0.05517578125,
      "learning_rate": 6.213235294117647e-07,
      "loss": 0.0022,
      "reward": 1.920210599899292,
      "reward_std": 0.01567223109304905,
      "rewards/accuracy_reward": 0.9202104806900024,
      "rewards/format_reward": 1.0,
      "step": 515
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 110.27764892578125,
      "epoch": 7.588235294117647,
      "grad_norm": 1.914576473194542,
      "kl": 0.054931640625,
      "learning_rate": 6.205882352941176e-07,
      "loss": 0.0022,
      "reward": 1.9109389781951904,
      "reward_std": 0.01758970133960247,
      "rewards/accuracy_reward": 0.9121409058570862,
      "rewards/format_reward": 0.998798131942749,
      "step": 516
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 108.74639892578125,
      "epoch": 7.602941176470588,
      "grad_norm": 2.2056419921026444,
      "kl": 0.06591796875,
      "learning_rate": 6.198529411764706e-07,
      "loss": 0.0026,
      "reward": 1.909173846244812,
      "reward_std": 0.0163656584918499,
      "rewards/accuracy_reward": 0.909173846244812,
      "rewards/format_reward": 1.0,
      "step": 517
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 109.97956848144531,
      "epoch": 7.617647058823529,
      "grad_norm": 2.3280532068329505,
      "kl": 0.0517578125,
      "learning_rate": 6.191176470588236e-07,
      "loss": 0.0021,
      "reward": 1.8901021480560303,
      "reward_std": 0.01763245277106762,
      "rewards/accuracy_reward": 0.890101969242096,
      "rewards/format_reward": 1.0,
      "step": 518
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 111.45793914794922,
      "epoch": 7.632352941176471,
      "grad_norm": 1.402338629602267,
      "kl": 0.05712890625,
      "learning_rate": 6.183823529411764e-07,
      "loss": 0.0023,
      "reward": 1.9023847579956055,
      "reward_std": 0.016608012840151787,
      "rewards/accuracy_reward": 0.9023846387863159,
      "rewards/format_reward": 1.0,
      "step": 519
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 110.38341522216797,
      "epoch": 7.647058823529412,
      "grad_norm": 1.6835653034818958,
      "kl": 0.058837890625,
      "learning_rate": 6.176470588235294e-07,
      "loss": 0.0024,
      "reward": 1.8985044956207275,
      "reward_std": 0.01695878617465496,
      "rewards/accuracy_reward": 0.8985044360160828,
      "rewards/format_reward": 1.0,
      "step": 520
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 111.1346206665039,
      "epoch": 7.661764705882353,
      "grad_norm": 9.541516490873528,
      "kl": 0.05419921875,
      "learning_rate": 6.169117647058824e-07,
      "loss": 0.0022,
      "reward": 1.903184175491333,
      "reward_std": 0.01812390610575676,
      "rewards/accuracy_reward": 0.9031841158866882,
      "rewards/format_reward": 1.0,
      "step": 521
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 110.78125762939453,
      "epoch": 7.676470588235294,
      "grad_norm": 2.907795347878127,
      "kl": 0.056396484375,
      "learning_rate": 6.161764705882353e-07,
      "loss": 0.0022,
      "reward": 1.9058642387390137,
      "reward_std": 0.022106477990746498,
      "rewards/accuracy_reward": 0.9070661067962646,
      "rewards/format_reward": 0.998798131942749,
      "step": 522
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 110.3425521850586,
      "epoch": 7.6911764705882355,
      "grad_norm": 2.122588959449984,
      "kl": 0.047607421875,
      "learning_rate": 6.154411764705883e-07,
      "loss": 0.0019,
      "reward": 1.892153024673462,
      "reward_std": 0.017479056492447853,
      "rewards/accuracy_reward": 0.8921528458595276,
      "rewards/format_reward": 1.0,
      "step": 523
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 110.2019271850586,
      "epoch": 7.705882352941177,
      "grad_norm": 2.3564295353752986,
      "kl": 0.05615234375,
      "learning_rate": 6.147058823529412e-07,
      "loss": 0.0022,
      "reward": 1.9090198278427124,
      "reward_std": 0.01690274477005005,
      "rewards/accuracy_reward": 0.9090198874473572,
      "rewards/format_reward": 1.0,
      "step": 524
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 111.92188262939453,
      "epoch": 7.720588235294118,
      "grad_norm": 1.574872602292037,
      "kl": 0.055419921875,
      "learning_rate": 6.139705882352941e-07,
      "loss": 0.0022,
      "reward": 1.901806354522705,
      "reward_std": 0.020534634590148926,
      "rewards/accuracy_reward": 0.9030082821846008,
      "rewards/format_reward": 0.998798131942749,
      "step": 525
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 112.71634674072266,
      "epoch": 7.735294117647059,
      "grad_norm": 1.4642975468665929,
      "kl": 0.05615234375,
      "learning_rate": 6.132352941176471e-07,
      "loss": 0.0022,
      "reward": 1.8870961666107178,
      "reward_std": 0.0190273430198431,
      "rewards/accuracy_reward": 0.8870960474014282,
      "rewards/format_reward": 1.0,
      "step": 526
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 112.76683044433594,
      "epoch": 7.75,
      "grad_norm": 1.64957668506362,
      "kl": 0.055908203125,
      "learning_rate": 6.125000000000001e-07,
      "loss": 0.0022,
      "reward": 1.888306736946106,
      "reward_std": 0.018305247649550438,
      "rewards/accuracy_reward": 0.888306736946106,
      "rewards/format_reward": 1.0,
      "step": 527
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 109.36659240722656,
      "epoch": 7.764705882352941,
      "grad_norm": 1.8470414599040466,
      "kl": 0.054443359375,
      "learning_rate": 6.117647058823529e-07,
      "loss": 0.0022,
      "reward": 1.914186716079712,
      "reward_std": 0.018072694540023804,
      "rewards/accuracy_reward": 0.9141866564750671,
      "rewards/format_reward": 1.0,
      "step": 528
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 112.49279022216797,
      "epoch": 7.779411764705882,
      "grad_norm": 1.4757794429532862,
      "kl": 0.05419921875,
      "learning_rate": 6.110294117647059e-07,
      "loss": 0.0022,
      "reward": 1.8795055150985718,
      "reward_std": 0.022909091785550117,
      "rewards/accuracy_reward": 0.8807073831558228,
      "rewards/format_reward": 0.998798131942749,
      "step": 529
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 111.40384674072266,
      "epoch": 7.794117647058823,
      "grad_norm": 2.0338636064103333,
      "kl": 0.059326171875,
      "learning_rate": 6.102941176470589e-07,
      "loss": 0.0024,
      "reward": 1.9119820594787598,
      "reward_std": 0.01673622988164425,
      "rewards/accuracy_reward": 0.9119820594787598,
      "rewards/format_reward": 1.0,
      "step": 530
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 112.04808044433594,
      "epoch": 7.8088235294117645,
      "grad_norm": 1.5646667587468694,
      "kl": 0.0498046875,
      "learning_rate": 6.095588235294118e-07,
      "loss": 0.002,
      "reward": 1.9045634269714355,
      "reward_std": 0.019505947828292847,
      "rewards/accuracy_reward": 0.9045633673667908,
      "rewards/format_reward": 1.0,
      "step": 531
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 110.84495544433594,
      "epoch": 7.823529411764706,
      "grad_norm": 1.9910418315918015,
      "kl": 0.04736328125,
      "learning_rate": 6.088235294117646e-07,
      "loss": 0.0019,
      "reward": 1.9242442846298218,
      "reward_std": 0.015269236639142036,
      "rewards/accuracy_reward": 0.9242441058158875,
      "rewards/format_reward": 1.0,
      "step": 532
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 111.67909240722656,
      "epoch": 7.838235294117647,
      "grad_norm": 2.1921636670867213,
      "kl": 0.05126953125,
      "learning_rate": 6.080882352941175e-07,
      "loss": 0.0021,
      "reward": 1.9174097776412964,
      "reward_std": 0.016234705224633217,
      "rewards/accuracy_reward": 0.9174097776412964,
      "rewards/format_reward": 1.0,
      "step": 533
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 108.95433044433594,
      "epoch": 7.852941176470588,
      "grad_norm": 5.773358773291786,
      "kl": 0.052734375,
      "learning_rate": 6.073529411764705e-07,
      "loss": 0.0021,
      "reward": 1.904146671295166,
      "reward_std": 0.018309401348233223,
      "rewards/accuracy_reward": 0.904146671295166,
      "rewards/format_reward": 1.0,
      "step": 534
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 109.10216522216797,
      "epoch": 7.867647058823529,
      "grad_norm": 3.1463228619890646,
      "kl": 0.0498046875,
      "learning_rate": 6.066176470588235e-07,
      "loss": 0.002,
      "reward": 1.9100762605667114,
      "reward_std": 0.0184779055416584,
      "rewards/accuracy_reward": 0.9100762009620667,
      "rewards/format_reward": 1.0,
      "step": 535
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 108.77644348144531,
      "epoch": 7.882352941176471,
      "grad_norm": 4.843243565505217,
      "kl": 0.04736328125,
      "learning_rate": 6.058823529411763e-07,
      "loss": 0.0019,
      "reward": 1.9202395677566528,
      "reward_std": 0.0161683801561594,
      "rewards/accuracy_reward": 0.9202395677566528,
      "rewards/format_reward": 1.0,
      "step": 536
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 108.99159240722656,
      "epoch": 7.897058823529412,
      "grad_norm": 1.5367043322463625,
      "kl": 0.0546875,
      "learning_rate": 6.051470588235293e-07,
      "loss": 0.0022,
      "reward": 1.9017425775527954,
      "reward_std": 0.017361411824822426,
      "rewards/accuracy_reward": 0.901742696762085,
      "rewards/format_reward": 1.0,
      "step": 537
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 109.72716522216797,
      "epoch": 7.911764705882353,
      "grad_norm": 1.6647667695224042,
      "kl": 0.04736328125,
      "learning_rate": 6.044117647058823e-07,
      "loss": 0.0019,
      "reward": 1.9160679578781128,
      "reward_std": 0.01760725863277912,
      "rewards/accuracy_reward": 0.9160678386688232,
      "rewards/format_reward": 1.0,
      "step": 538
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 110.06971740722656,
      "epoch": 7.926470588235294,
      "grad_norm": 1.5196339741547849,
      "kl": 0.04931640625,
      "learning_rate": 6.036764705882352e-07,
      "loss": 0.002,
      "reward": 1.9151060581207275,
      "reward_std": 0.017051585018634796,
      "rewards/accuracy_reward": 0.9151060581207275,
      "rewards/format_reward": 1.0,
      "step": 539
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 109.31971740722656,
      "epoch": 7.9411764705882355,
      "grad_norm": 2.094866774315946,
      "kl": 0.04736328125,
      "learning_rate": 6.029411764705882e-07,
      "loss": 0.0019,
      "reward": 1.8982237577438354,
      "reward_std": 0.017682263627648354,
      "rewards/accuracy_reward": 0.8982236981391907,
      "rewards/format_reward": 1.0,
      "step": 540
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 108.39543914794922,
      "epoch": 7.955882352941177,
      "grad_norm": 3.260878136220598,
      "kl": 0.04638671875,
      "learning_rate": 6.022058823529411e-07,
      "loss": 0.0019,
      "reward": 1.8989781141281128,
      "reward_std": 0.016665568575263023,
      "rewards/accuracy_reward": 0.8989779949188232,
      "rewards/format_reward": 1.0,
      "step": 541
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 108.65625762939453,
      "epoch": 7.970588235294118,
      "grad_norm": 2.6615474581638856,
      "kl": 0.05126953125,
      "learning_rate": 6.01470588235294e-07,
      "loss": 0.002,
      "reward": 1.9191153049468994,
      "reward_std": 0.016191896051168442,
      "rewards/accuracy_reward": 0.9191152453422546,
      "rewards/format_reward": 1.0,
      "step": 542
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 108.72116088867188,
      "epoch": 7.985294117647059,
      "grad_norm": 2.0899244271104815,
      "kl": 0.05322265625,
      "learning_rate": 6.00735294117647e-07,
      "loss": 0.0021,
      "reward": 1.9136680364608765,
      "reward_std": 0.017099877819418907,
      "rewards/accuracy_reward": 0.9136679768562317,
      "rewards/format_reward": 1.0,
      "step": 543
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 108.3500747680664,
      "epoch": 8.0,
      "grad_norm": 3.3285479491467647,
      "kl": 0.050048828125,
      "learning_rate": 6e-07,
      "loss": 0.002,
      "reward": 1.921190857887268,
      "reward_std": 0.019811777397990227,
      "rewards/accuracy_reward": 0.9211909174919128,
      "rewards/format_reward": 1.0,
      "step": 544
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 107.65625762939453,
      "epoch": 8.014705882352942,
      "grad_norm": 2.8131573068708047,
      "kl": 0.0478515625,
      "learning_rate": 5.992647058823528e-07,
      "loss": 0.0019,
      "reward": 1.9018555879592896,
      "reward_std": 0.017298076301813126,
      "rewards/accuracy_reward": 0.9018555879592896,
      "rewards/format_reward": 1.0,
      "step": 545
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 109.14423370361328,
      "epoch": 8.029411764705882,
      "grad_norm": 1.7926326341266048,
      "kl": 0.048583984375,
      "learning_rate": 5.985294117647058e-07,
      "loss": 0.0019,
      "reward": 1.902543544769287,
      "reward_std": 0.018634334206581116,
      "rewards/accuracy_reward": 0.9025435447692871,
      "rewards/format_reward": 1.0,
      "step": 546
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 109.56370544433594,
      "epoch": 8.044117647058824,
      "grad_norm": 1.4449179097965148,
      "kl": 0.048583984375,
      "learning_rate": 5.977941176470588e-07,
      "loss": 0.0019,
      "reward": 1.916272759437561,
      "reward_std": 0.01657208427786827,
      "rewards/accuracy_reward": 0.916272759437561,
      "rewards/format_reward": 1.0,
      "step": 547
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 109.97356414794922,
      "epoch": 8.058823529411764,
      "grad_norm": 6.756394605908286,
      "kl": 0.05126953125,
      "learning_rate": 5.970588235294117e-07,
      "loss": 0.0021,
      "reward": 1.9202964305877686,
      "reward_std": 0.018207618966698647,
      "rewards/accuracy_reward": 0.920296311378479,
      "rewards/format_reward": 1.0,
      "step": 548
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 112.40745544433594,
      "epoch": 8.073529411764707,
      "grad_norm": 3.5136091347598986,
      "kl": 0.050537109375,
      "learning_rate": 5.963235294117647e-07,
      "loss": 0.002,
      "reward": 1.9019232988357544,
      "reward_std": 0.0198956411331892,
      "rewards/accuracy_reward": 0.9019232392311096,
      "rewards/format_reward": 1.0,
      "step": 549
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 110.43991088867188,
      "epoch": 8.088235294117647,
      "grad_norm": 2.2839304418519792,
      "kl": 0.052001953125,
      "learning_rate": 5.955882352941176e-07,
      "loss": 0.0021,
      "reward": 1.9206691980361938,
      "reward_std": 0.014682717621326447,
      "rewards/accuracy_reward": 0.9206691384315491,
      "rewards/format_reward": 1.0,
      "step": 550
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 113.02043914794922,
      "epoch": 8.102941176470589,
      "grad_norm": 2.484567856847274,
      "kl": 0.055419921875,
      "learning_rate": 5.948529411764705e-07,
      "loss": 0.0022,
      "reward": 1.9100967645645142,
      "reward_std": 0.018312010914087296,
      "rewards/accuracy_reward": 0.9100965857505798,
      "rewards/format_reward": 1.0,
      "step": 551
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 114.34375762939453,
      "epoch": 8.117647058823529,
      "grad_norm": 1.6432899488733537,
      "kl": 0.05419921875,
      "learning_rate": 5.941176470588235e-07,
      "loss": 0.0022,
      "reward": 1.8991518020629883,
      "reward_std": 0.021620534360408783,
      "rewards/accuracy_reward": 0.900353729724884,
      "rewards/format_reward": 0.998798131942749,
      "step": 552
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 113.63581848144531,
      "epoch": 8.132352941176471,
      "grad_norm": 2.0271204887130416,
      "kl": 0.05712890625,
      "learning_rate": 5.933823529411765e-07,
      "loss": 0.0023,
      "reward": 1.900107502937317,
      "reward_std": 0.019040938466787338,
      "rewards/accuracy_reward": 0.9001075029373169,
      "rewards/format_reward": 1.0,
      "step": 553
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 115.09616088867188,
      "epoch": 8.147058823529411,
      "grad_norm": 1.5813240400193374,
      "kl": 0.060791015625,
      "learning_rate": 5.926470588235293e-07,
      "loss": 0.0024,
      "reward": 1.915557622909546,
      "reward_std": 0.020517727360129356,
      "rewards/accuracy_reward": 0.9167595505714417,
      "rewards/format_reward": 0.998798131942749,
      "step": 554
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 115.71394348144531,
      "epoch": 8.161764705882353,
      "grad_norm": 4.637148825036712,
      "kl": 0.0576171875,
      "learning_rate": 5.919117647058823e-07,
      "loss": 0.0023,
      "reward": 1.9027869701385498,
      "reward_std": 0.019659575074911118,
      "rewards/accuracy_reward": 0.9027867317199707,
      "rewards/format_reward": 1.0,
      "step": 555
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 114.27404022216797,
      "epoch": 8.176470588235293,
      "grad_norm": 1.6509059104235309,
      "kl": 0.0556640625,
      "learning_rate": 5.911764705882353e-07,
      "loss": 0.0022,
      "reward": 1.898353934288025,
      "reward_std": 0.019414007663726807,
      "rewards/accuracy_reward": 0.8983538746833801,
      "rewards/format_reward": 1.0,
      "step": 556
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 117.88101196289062,
      "epoch": 8.191176470588236,
      "grad_norm": 1.2731867179847607,
      "kl": 0.0576171875,
      "learning_rate": 5.904411764705882e-07,
      "loss": 0.0023,
      "reward": 1.8892360925674438,
      "reward_std": 0.025776617228984833,
      "rewards/accuracy_reward": 0.8916398882865906,
      "rewards/format_reward": 0.9975962042808533,
      "step": 557
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 115.42188262939453,
      "epoch": 8.205882352941176,
      "grad_norm": 2.6655315489926967,
      "kl": 0.059326171875,
      "learning_rate": 5.897058823529412e-07,
      "loss": 0.0024,
      "reward": 1.9022037982940674,
      "reward_std": 0.01893787272274494,
      "rewards/accuracy_reward": 0.9034057855606079,
      "rewards/format_reward": 0.998798131942749,
      "step": 558
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 115.95793914794922,
      "epoch": 8.220588235294118,
      "grad_norm": 1.7443518844175923,
      "kl": 0.053955078125,
      "learning_rate": 5.889705882352941e-07,
      "loss": 0.0022,
      "reward": 1.9148838520050049,
      "reward_std": 0.016492504626512527,
      "rewards/accuracy_reward": 0.9148839116096497,
      "rewards/format_reward": 1.0,
      "step": 559
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 114.52043914794922,
      "epoch": 8.235294117647058,
      "grad_norm": 2.0265041434888853,
      "kl": 0.050048828125,
      "learning_rate": 5.88235294117647e-07,
      "loss": 0.002,
      "reward": 1.90397047996521,
      "reward_std": 0.020762760192155838,
      "rewards/accuracy_reward": 0.90397047996521,
      "rewards/format_reward": 1.0,
      "step": 560
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 116.28966522216797,
      "epoch": 8.25,
      "grad_norm": 1.5715106635469254,
      "kl": 0.055419921875,
      "learning_rate": 5.875e-07,
      "loss": 0.0022,
      "reward": 1.911298394203186,
      "reward_std": 0.019145453348755836,
      "rewards/accuracy_reward": 0.9112983345985413,
      "rewards/format_reward": 1.0,
      "step": 561
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 112.29808044433594,
      "epoch": 8.264705882352942,
      "grad_norm": 1.4932683274744363,
      "kl": 0.057373046875,
      "learning_rate": 5.86764705882353e-07,
      "loss": 0.0023,
      "reward": 1.9099814891815186,
      "reward_std": 0.025356175377964973,
      "rewards/accuracy_reward": 0.9123852252960205,
      "rewards/format_reward": 0.9975962042808533,
      "step": 562
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 113.23197174072266,
      "epoch": 8.279411764705882,
      "grad_norm": 7.280740226777266,
      "kl": 0.055419921875,
      "learning_rate": 5.860294117647058e-07,
      "loss": 0.0022,
      "reward": 1.9085097312927246,
      "reward_std": 0.018148187547922134,
      "rewards/accuracy_reward": 0.9085096716880798,
      "rewards/format_reward": 1.0,
      "step": 563
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 112.80048370361328,
      "epoch": 8.294117647058824,
      "grad_norm": 2.9076060047905163,
      "kl": 0.05419921875,
      "learning_rate": 5.852941176470588e-07,
      "loss": 0.0022,
      "reward": 1.909620761871338,
      "reward_std": 0.018346594646573067,
      "rewards/accuracy_reward": 0.9096207618713379,
      "rewards/format_reward": 1.0,
      "step": 564
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 112.26081848144531,
      "epoch": 8.308823529411764,
      "grad_norm": 3.2289837306279168,
      "kl": 0.060791015625,
      "learning_rate": 5.845588235294118e-07,
      "loss": 0.0024,
      "reward": 1.9069983959197998,
      "reward_std": 0.0201736968010664,
      "rewards/accuracy_reward": 0.9082001447677612,
      "rewards/format_reward": 0.998798131942749,
      "step": 565
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 112.25120544433594,
      "epoch": 8.323529411764707,
      "grad_norm": 1.7199349309033736,
      "kl": 0.05517578125,
      "learning_rate": 5.838235294117647e-07,
      "loss": 0.0022,
      "reward": 1.887500286102295,
      "reward_std": 0.020135780796408653,
      "rewards/accuracy_reward": 0.8887021541595459,
      "rewards/format_reward": 0.998798131942749,
      "step": 566
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 111.41226196289062,
      "epoch": 8.338235294117647,
      "grad_norm": 6.802575033519044,
      "kl": 0.0576171875,
      "learning_rate": 5.830882352941177e-07,
      "loss": 0.0023,
      "reward": 1.8974536657333374,
      "reward_std": 0.020134514197707176,
      "rewards/accuracy_reward": 0.8974535465240479,
      "rewards/format_reward": 1.0,
      "step": 567
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 111.51683044433594,
      "epoch": 8.352941176470589,
      "grad_norm": 36.18405306346145,
      "kl": 0.054931640625,
      "learning_rate": 5.823529411764706e-07,
      "loss": 0.0022,
      "reward": 1.8917019367218018,
      "reward_std": 0.01857663132250309,
      "rewards/accuracy_reward": 0.8929038643836975,
      "rewards/format_reward": 0.998798131942749,
      "step": 568
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 111.5769271850586,
      "epoch": 8.367647058823529,
      "grad_norm": 2.3459848810177975,
      "kl": 0.052734375,
      "learning_rate": 5.816176470588235e-07,
      "loss": 0.0021,
      "reward": 1.9242991209030151,
      "reward_std": 0.018164673820137978,
      "rewards/accuracy_reward": 0.9255009293556213,
      "rewards/format_reward": 0.998798131942749,
      "step": 569
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 109.03485870361328,
      "epoch": 8.382352941176471,
      "grad_norm": 1.6032225663505273,
      "kl": 0.06103515625,
      "learning_rate": 5.808823529411765e-07,
      "loss": 0.0025,
      "reward": 1.9081476926803589,
      "reward_std": 0.019294267520308495,
      "rewards/accuracy_reward": 0.9081474542617798,
      "rewards/format_reward": 1.0,
      "step": 570
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 110.40866088867188,
      "epoch": 8.397058823529411,
      "grad_norm": 2.653883841862032,
      "kl": 0.05908203125,
      "learning_rate": 5.801470588235295e-07,
      "loss": 0.0024,
      "reward": 1.904270052909851,
      "reward_std": 0.017766019329428673,
      "rewards/accuracy_reward": 0.9042699933052063,
      "rewards/format_reward": 1.0,
      "step": 571
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 108.58173370361328,
      "epoch": 8.411764705882353,
      "grad_norm": 1.8098644267187556,
      "kl": 0.05908203125,
      "learning_rate": 5.794117647058823e-07,
      "loss": 0.0024,
      "reward": 1.9107073545455933,
      "reward_std": 0.017658641561865807,
      "rewards/accuracy_reward": 0.9107073545455933,
      "rewards/format_reward": 1.0,
      "step": 572
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 108.56250762939453,
      "epoch": 8.426470588235293,
      "grad_norm": 3.2259185749098784,
      "kl": 0.06005859375,
      "learning_rate": 5.786764705882353e-07,
      "loss": 0.0024,
      "reward": 1.899427056312561,
      "reward_std": 0.016428910195827484,
      "rewards/accuracy_reward": 0.8994268774986267,
      "rewards/format_reward": 1.0,
      "step": 573
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 108.02764892578125,
      "epoch": 8.441176470588236,
      "grad_norm": 3.282308261215477,
      "kl": 0.05859375,
      "learning_rate": 5.779411764705882e-07,
      "loss": 0.0023,
      "reward": 1.9036839008331299,
      "reward_std": 0.020171254873275757,
      "rewards/accuracy_reward": 0.9036839008331299,
      "rewards/format_reward": 1.0,
      "step": 574
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 108.05048370361328,
      "epoch": 8.455882352941176,
      "grad_norm": 1.7713296358743136,
      "kl": 0.061767578125,
      "learning_rate": 5.772058823529411e-07,
      "loss": 0.0025,
      "reward": 1.9190058708190918,
      "reward_std": 0.018266772851347923,
      "rewards/accuracy_reward": 0.919005811214447,
      "rewards/format_reward": 1.0,
      "step": 575
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 107.00241088867188,
      "epoch": 8.470588235294118,
      "grad_norm": 1.9874750258288356,
      "kl": 0.055908203125,
      "learning_rate": 5.76470588235294e-07,
      "loss": 0.0023,
      "reward": 1.9230592250823975,
      "reward_std": 0.013289669528603554,
      "rewards/accuracy_reward": 0.9230592250823975,
      "rewards/format_reward": 1.0,
      "step": 576
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 107.89063262939453,
      "epoch": 8.485294117647058,
      "grad_norm": 2.286169294844814,
      "kl": 0.0576171875,
      "learning_rate": 5.75735294117647e-07,
      "loss": 0.0023,
      "reward": 1.8973294496536255,
      "reward_std": 0.01872727833688259,
      "rewards/accuracy_reward": 0.8973293900489807,
      "rewards/format_reward": 1.0,
      "step": 577
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 106.49279022216797,
      "epoch": 8.5,
      "grad_norm": 2.4071803200242416,
      "kl": 0.0654296875,
      "learning_rate": 5.749999999999999e-07,
      "loss": 0.0026,
      "reward": 1.9316534996032715,
      "reward_std": 0.017379358410835266,
      "rewards/accuracy_reward": 0.9316534399986267,
      "rewards/format_reward": 1.0,
      "step": 578
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 106.60216522216797,
      "epoch": 8.514705882352942,
      "grad_norm": 1.79762667011906,
      "kl": 0.064453125,
      "learning_rate": 5.742647058823529e-07,
      "loss": 0.0026,
      "reward": 1.9131649732589722,
      "reward_std": 0.020887164399027824,
      "rewards/accuracy_reward": 0.9143666625022888,
      "rewards/format_reward": 0.998798131942749,
      "step": 579
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 108.56971740722656,
      "epoch": 8.529411764705882,
      "grad_norm": 2.3602888807544526,
      "kl": 0.053955078125,
      "learning_rate": 5.735294117647059e-07,
      "loss": 0.0022,
      "reward": 1.9001418352127075,
      "reward_std": 0.01754673384130001,
      "rewards/accuracy_reward": 0.9001417756080627,
      "rewards/format_reward": 1.0,
      "step": 580
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 109.32091522216797,
      "epoch": 8.544117647058824,
      "grad_norm": 1.8304863518291064,
      "kl": 0.05615234375,
      "learning_rate": 5.727941176470587e-07,
      "loss": 0.0022,
      "reward": 1.9000111818313599,
      "reward_std": 0.017091548070311546,
      "rewards/accuracy_reward": 0.9000111222267151,
      "rewards/format_reward": 1.0,
      "step": 581
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 106.98558044433594,
      "epoch": 8.558823529411764,
      "grad_norm": 2.667131501264164,
      "kl": 0.0517578125,
      "learning_rate": 5.720588235294117e-07,
      "loss": 0.0021,
      "reward": 1.9127933979034424,
      "reward_std": 0.01638048328459263,
      "rewards/accuracy_reward": 0.9127933979034424,
      "rewards/format_reward": 1.0,
      "step": 582
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 108.97116088867188,
      "epoch": 8.573529411764707,
      "grad_norm": 3.838255993972307,
      "kl": 0.05322265625,
      "learning_rate": 5.713235294117647e-07,
      "loss": 0.0021,
      "reward": 1.9052388668060303,
      "reward_std": 0.015317871235311031,
      "rewards/accuracy_reward": 0.9052388668060303,
      "rewards/format_reward": 1.0,
      "step": 583
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 107.7800521850586,
      "epoch": 8.588235294117647,
      "grad_norm": 1.8912826875619027,
      "kl": 0.0517578125,
      "learning_rate": 5.705882352941176e-07,
      "loss": 0.0021,
      "reward": 1.9118890762329102,
      "reward_std": 0.016504064202308655,
      "rewards/accuracy_reward": 0.9118890166282654,
      "rewards/format_reward": 1.0,
      "step": 584
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 107.36538696289062,
      "epoch": 8.602941176470589,
      "grad_norm": 1.9564532336203913,
      "kl": 0.05517578125,
      "learning_rate": 5.698529411764705e-07,
      "loss": 0.0022,
      "reward": 1.9083279371261597,
      "reward_std": 0.01711435243487358,
      "rewards/accuracy_reward": 0.9083278775215149,
      "rewards/format_reward": 1.0,
      "step": 585
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 108.89784240722656,
      "epoch": 8.617647058823529,
      "grad_norm": 1.411173293415255,
      "kl": 0.049072265625,
      "learning_rate": 5.691176470588235e-07,
      "loss": 0.002,
      "reward": 1.9171125888824463,
      "reward_std": 0.016976075246930122,
      "rewards/accuracy_reward": 0.9183142781257629,
      "rewards/format_reward": 0.998798131942749,
      "step": 586
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 109.58053588867188,
      "epoch": 8.632352941176471,
      "grad_norm": 1.8153636392557528,
      "kl": 0.050537109375,
      "learning_rate": 5.683823529411764e-07,
      "loss": 0.002,
      "reward": 1.9182407855987549,
      "reward_std": 0.013324742205440998,
      "rewards/accuracy_reward": 0.9182407855987549,
      "rewards/format_reward": 1.0,
      "step": 587
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 109.02644348144531,
      "epoch": 8.647058823529411,
      "grad_norm": 2.991122791360467,
      "kl": 0.049072265625,
      "learning_rate": 5.676470588235294e-07,
      "loss": 0.002,
      "reward": 1.9190689325332642,
      "reward_std": 0.01496278028935194,
      "rewards/accuracy_reward": 0.9190687537193298,
      "rewards/format_reward": 1.0,
      "step": 588
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 110.1971206665039,
      "epoch": 8.661764705882353,
      "grad_norm": 1.5660894873733406,
      "kl": 0.05078125,
      "learning_rate": 5.669117647058823e-07,
      "loss": 0.002,
      "reward": 1.8968241214752197,
      "reward_std": 0.015287560410797596,
      "rewards/accuracy_reward": 0.896824061870575,
      "rewards/format_reward": 1.0,
      "step": 589
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 109.95072174072266,
      "epoch": 8.676470588235293,
      "grad_norm": 1.7772209705137463,
      "kl": 0.054443359375,
      "learning_rate": 5.661764705882352e-07,
      "loss": 0.0022,
      "reward": 1.9128403663635254,
      "reward_std": 0.020346660166978836,
      "rewards/accuracy_reward": 0.9140422940254211,
      "rewards/format_reward": 0.998798131942749,
      "step": 590
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 111.78726196289062,
      "epoch": 8.691176470588236,
      "grad_norm": 1.6721267407160185,
      "kl": 0.046630859375,
      "learning_rate": 5.654411764705882e-07,
      "loss": 0.0019,
      "reward": 1.9050085544586182,
      "reward_std": 0.018483325839042664,
      "rewards/accuracy_reward": 0.9050085544586182,
      "rewards/format_reward": 1.0,
      "step": 591
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 110.97956848144531,
      "epoch": 8.705882352941176,
      "grad_norm": 1.6709044867050522,
      "kl": 0.052490234375,
      "learning_rate": 5.647058823529412e-07,
      "loss": 0.0021,
      "reward": 1.924027442932129,
      "reward_std": 0.019660327583551407,
      "rewards/accuracy_reward": 0.9252294301986694,
      "rewards/format_reward": 0.998798131942749,
      "step": 592
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 110.9988021850586,
      "epoch": 8.720588235294118,
      "grad_norm": 4.487069114218076,
      "kl": 0.05029296875,
      "learning_rate": 5.639705882352941e-07,
      "loss": 0.002,
      "reward": 1.8979933261871338,
      "reward_std": 0.01683800294995308,
      "rewards/accuracy_reward": 0.897993266582489,
      "rewards/format_reward": 1.0,
      "step": 593
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 111.0769271850586,
      "epoch": 8.735294117647058,
      "grad_norm": 2.4615353009495817,
      "kl": 0.07763671875,
      "learning_rate": 5.63235294117647e-07,
      "loss": 0.0031,
      "reward": 1.8970861434936523,
      "reward_std": 0.019765671342611313,
      "rewards/accuracy_reward": 0.8982879519462585,
      "rewards/format_reward": 0.998798131942749,
      "step": 594
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 110.85216522216797,
      "epoch": 8.75,
      "grad_norm": 1.735078503716415,
      "kl": 0.051025390625,
      "learning_rate": 5.625e-07,
      "loss": 0.002,
      "reward": 1.9034225940704346,
      "reward_std": 0.018201464787125587,
      "rewards/accuracy_reward": 0.9034226536750793,
      "rewards/format_reward": 1.0,
      "step": 595
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 112.47116088867188,
      "epoch": 8.764705882352942,
      "grad_norm": 2.193668601424281,
      "kl": 0.048095703125,
      "learning_rate": 5.617647058823529e-07,
      "loss": 0.0019,
      "reward": 1.8874914646148682,
      "reward_std": 0.024996280670166016,
      "rewards/accuracy_reward": 0.8898953795433044,
      "rewards/format_reward": 0.9975962042808533,
      "step": 596
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 111.53366088867188,
      "epoch": 8.779411764705882,
      "grad_norm": 1.5231562376577301,
      "kl": 0.0498046875,
      "learning_rate": 5.610294117647059e-07,
      "loss": 0.002,
      "reward": 1.910040259361267,
      "reward_std": 0.016859347000718117,
      "rewards/accuracy_reward": 0.9100401997566223,
      "rewards/format_reward": 1.0,
      "step": 597
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 110.86058044433594,
      "epoch": 8.794117647058824,
      "grad_norm": 6.4583154297757455,
      "kl": 0.05029296875,
      "learning_rate": 5.602941176470588e-07,
      "loss": 0.002,
      "reward": 1.9111155271530151,
      "reward_std": 0.016216028481721878,
      "rewards/accuracy_reward": 0.9111157059669495,
      "rewards/format_reward": 1.0,
      "step": 598
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 112.24279022216797,
      "epoch": 8.808823529411764,
      "grad_norm": 1.775836022535222,
      "kl": 0.05224609375,
      "learning_rate": 5.595588235294117e-07,
      "loss": 0.0021,
      "reward": 1.917399287223816,
      "reward_std": 0.016161981970071793,
      "rewards/accuracy_reward": 0.9173993468284607,
      "rewards/format_reward": 1.0,
      "step": 599
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 112.46514892578125,
      "epoch": 8.823529411764707,
      "grad_norm": 1.374775331428731,
      "kl": 0.052001953125,
      "learning_rate": 5.588235294117647e-07,
      "loss": 0.0021,
      "reward": 1.8994345664978027,
      "reward_std": 0.017103036865592003,
      "rewards/accuracy_reward": 0.8994343876838684,
      "rewards/format_reward": 1.0,
      "step": 600
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 111.3894271850586,
      "epoch": 8.838235294117647,
      "grad_norm": 2.9229408253654365,
      "kl": 0.05029296875,
      "learning_rate": 5.580882352941177e-07,
      "loss": 0.002,
      "reward": 1.9245597124099731,
      "reward_std": 0.016064738854765892,
      "rewards/accuracy_reward": 0.9245598316192627,
      "rewards/format_reward": 1.0,
      "step": 601
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 111.09134674072266,
      "epoch": 8.852941176470589,
      "grad_norm": 1.761051200827917,
      "kl": 0.053955078125,
      "learning_rate": 5.573529411764706e-07,
      "loss": 0.0022,
      "reward": 1.9057958126068115,
      "reward_std": 0.018393708392977715,
      "rewards/accuracy_reward": 0.9057958126068115,
      "rewards/format_reward": 1.0,
      "step": 602
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 109.97956848144531,
      "epoch": 8.867647058823529,
      "grad_norm": 1.9085242419393635,
      "kl": 0.05419921875,
      "learning_rate": 5.566176470588235e-07,
      "loss": 0.0022,
      "reward": 1.9057040214538574,
      "reward_std": 0.02528255060315132,
      "rewards/accuracy_reward": 0.9069057106971741,
      "rewards/format_reward": 0.998798131942749,
      "step": 603
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 109.55409240722656,
      "epoch": 8.882352941176471,
      "grad_norm": 2.2056191207524742,
      "kl": 0.050537109375,
      "learning_rate": 5.558823529411765e-07,
      "loss": 0.002,
      "reward": 1.9001765251159668,
      "reward_std": 0.01882914826273918,
      "rewards/accuracy_reward": 0.9001763463020325,
      "rewards/format_reward": 1.0,
      "step": 604
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 109.1189956665039,
      "epoch": 8.897058823529411,
      "grad_norm": 1.6684830952892549,
      "kl": 0.049072265625,
      "learning_rate": 5.551470588235294e-07,
      "loss": 0.002,
      "reward": 1.9099537134170532,
      "reward_std": 0.015918085351586342,
      "rewards/accuracy_reward": 0.9099536538124084,
      "rewards/format_reward": 1.0,
      "step": 605
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 108.1081771850586,
      "epoch": 8.911764705882353,
      "grad_norm": 2.437963685533989,
      "kl": 0.052001953125,
      "learning_rate": 5.544117647058824e-07,
      "loss": 0.0021,
      "reward": 1.9115108251571655,
      "reward_std": 0.01843283697962761,
      "rewards/accuracy_reward": 0.911510705947876,
      "rewards/format_reward": 1.0,
      "step": 606
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 108.92188262939453,
      "epoch": 8.926470588235293,
      "grad_norm": 5.175318731979124,
      "kl": 0.053466796875,
      "learning_rate": 5.536764705882353e-07,
      "loss": 0.0021,
      "reward": 1.9014861583709717,
      "reward_std": 0.016052136197686195,
      "rewards/accuracy_reward": 0.9014859199523926,
      "rewards/format_reward": 1.0,
      "step": 607
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 107.3377456665039,
      "epoch": 8.941176470588236,
      "grad_norm": 1.466199052775766,
      "kl": 0.055419921875,
      "learning_rate": 5.529411764705882e-07,
      "loss": 0.0022,
      "reward": 1.8980929851531982,
      "reward_std": 0.020213749259710312,
      "rewards/accuracy_reward": 0.8992946743965149,
      "rewards/format_reward": 0.998798131942749,
      "step": 608
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 108.5252456665039,
      "epoch": 8.955882352941176,
      "grad_norm": 2.4803683445732196,
      "kl": 0.05224609375,
      "learning_rate": 5.522058823529412e-07,
      "loss": 0.0021,
      "reward": 1.8946930170059204,
      "reward_std": 0.020943893119692802,
      "rewards/accuracy_reward": 0.8946928381919861,
      "rewards/format_reward": 1.0,
      "step": 609
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 107.81851196289062,
      "epoch": 8.970588235294118,
      "grad_norm": 2.199347709692285,
      "kl": 0.056396484375,
      "learning_rate": 5.514705882352942e-07,
      "loss": 0.0023,
      "reward": 1.9163153171539307,
      "reward_std": 0.019964834675192833,
      "rewards/accuracy_reward": 0.9175171852111816,
      "rewards/format_reward": 0.998798131942749,
      "step": 610
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 106.43389892578125,
      "epoch": 8.985294117647058,
      "grad_norm": 1.3598803521986345,
      "kl": 0.058349609375,
      "learning_rate": 5.50735294117647e-07,
      "loss": 0.0023,
      "reward": 1.9314730167388916,
      "reward_std": 0.01666920818388462,
      "rewards/accuracy_reward": 0.9314729571342468,
      "rewards/format_reward": 1.0,
      "step": 611
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 105.38204193115234,
      "epoch": 9.0,
      "grad_norm": 2.5292498675314574,
      "kl": 0.0546875,
      "learning_rate": 5.5e-07,
      "loss": 0.0022,
      "reward": 1.9147766828536987,
      "reward_std": 0.012550020590424538,
      "rewards/accuracy_reward": 0.9147767424583435,
      "rewards/format_reward": 1.0,
      "step": 612
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 106.68269348144531,
      "epoch": 9.014705882352942,
      "grad_norm": 1.359576415697596,
      "kl": 0.05419921875,
      "learning_rate": 5.49264705882353e-07,
      "loss": 0.0022,
      "reward": 1.9160057306289673,
      "reward_std": 0.018689963966608047,
      "rewards/accuracy_reward": 0.9172075986862183,
      "rewards/format_reward": 0.998798131942749,
      "step": 613
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 107.44831848144531,
      "epoch": 9.029411764705882,
      "grad_norm": 1.5142397828555845,
      "kl": 0.0595703125,
      "learning_rate": 5.485294117647059e-07,
      "loss": 0.0024,
      "reward": 1.8921860456466675,
      "reward_std": 0.021658875048160553,
      "rewards/accuracy_reward": 0.8933879137039185,
      "rewards/format_reward": 0.998798131942749,
      "step": 614
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 106.91226196289062,
      "epoch": 9.044117647058824,
      "grad_norm": 5.469469688964059,
      "kl": 0.05615234375,
      "learning_rate": 5.477941176470589e-07,
      "loss": 0.0023,
      "reward": 1.9030665159225464,
      "reward_std": 0.02024255506694317,
      "rewards/accuracy_reward": 0.9042685627937317,
      "rewards/format_reward": 0.998798131942749,
      "step": 615
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 106.23197174072266,
      "epoch": 9.058823529411764,
      "grad_norm": 2.869773876567515,
      "kl": 0.07080078125,
      "learning_rate": 5.470588235294118e-07,
      "loss": 0.0028,
      "reward": 1.912828803062439,
      "reward_std": 0.018641209229826927,
      "rewards/accuracy_reward": 0.9128286838531494,
      "rewards/format_reward": 1.0,
      "step": 616
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 107.40625762939453,
      "epoch": 9.073529411764707,
      "grad_norm": 1.880572098436719,
      "kl": 0.05712890625,
      "learning_rate": 5.463235294117646e-07,
      "loss": 0.0023,
      "reward": 1.9045672416687012,
      "reward_std": 0.01760552078485489,
      "rewards/accuracy_reward": 0.9045671820640564,
      "rewards/format_reward": 1.0,
      "step": 617
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 106.60216522216797,
      "epoch": 9.088235294117647,
      "grad_norm": 1.6132861487685206,
      "kl": 0.056884765625,
      "learning_rate": 5.455882352941176e-07,
      "loss": 0.0023,
      "reward": 1.9064825773239136,
      "reward_std": 0.02047489397227764,
      "rewards/accuracy_reward": 0.9064825773239136,
      "rewards/format_reward": 1.0,
      "step": 618
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 108.12139892578125,
      "epoch": 9.102941176470589,
      "grad_norm": 1.8352294433327585,
      "kl": 0.061279296875,
      "learning_rate": 5.448529411764706e-07,
      "loss": 0.0024,
      "reward": 1.9166762828826904,
      "reward_std": 0.017683234065771103,
      "rewards/accuracy_reward": 0.9166761636734009,
      "rewards/format_reward": 1.0,
      "step": 619
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 107.82813262939453,
      "epoch": 9.117647058823529,
      "grad_norm": 1.9584175345112422,
      "kl": 0.05810546875,
      "learning_rate": 5.441176470588234e-07,
      "loss": 0.0023,
      "reward": 1.9098379611968994,
      "reward_std": 0.019358208402991295,
      "rewards/accuracy_reward": 0.9098379015922546,
      "rewards/format_reward": 1.0,
      "step": 620
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 109.54928588867188,
      "epoch": 9.132352941176471,
      "grad_norm": 1.681570036282453,
      "kl": 0.05908203125,
      "learning_rate": 5.433823529411764e-07,
      "loss": 0.0024,
      "reward": 1.903365135192871,
      "reward_std": 0.018463198095560074,
      "rewards/accuracy_reward": 0.9045670032501221,
      "rewards/format_reward": 0.998798131942749,
      "step": 621
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 110.81370544433594,
      "epoch": 9.147058823529411,
      "grad_norm": 2.175735273009701,
      "kl": 0.06787109375,
      "learning_rate": 5.426470588235294e-07,
      "loss": 0.0027,
      "reward": 1.9245383739471436,
      "reward_std": 0.015885451808571815,
      "rewards/accuracy_reward": 0.9245381355285645,
      "rewards/format_reward": 1.0,
      "step": 622
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 111.20313262939453,
      "epoch": 9.161764705882353,
      "grad_norm": 2.20406419419079,
      "kl": 0.05810546875,
      "learning_rate": 5.419117647058823e-07,
      "loss": 0.0023,
      "reward": 1.9137063026428223,
      "reward_std": 0.018180424347519875,
      "rewards/accuracy_reward": 0.9137064218521118,
      "rewards/format_reward": 1.0,
      "step": 623
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 112.13221740722656,
      "epoch": 9.176470588235293,
      "grad_norm": 1.7531001551644594,
      "kl": 0.057373046875,
      "learning_rate": 5.411764705882353e-07,
      "loss": 0.0023,
      "reward": 1.9103367328643799,
      "reward_std": 0.01696370542049408,
      "rewards/accuracy_reward": 0.9103366136550903,
      "rewards/format_reward": 1.0,
      "step": 624
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 112.02163696289062,
      "epoch": 9.191176470588236,
      "grad_norm": 1.9594882556521003,
      "kl": 0.06201171875,
      "learning_rate": 5.404411764705882e-07,
      "loss": 0.0025,
      "reward": 1.9130682945251465,
      "reward_std": 0.014644093811511993,
      "rewards/accuracy_reward": 0.9130681157112122,
      "rewards/format_reward": 1.0,
      "step": 625
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 113.65866088867188,
      "epoch": 9.205882352941176,
      "grad_norm": 2.6647032908164303,
      "kl": 0.0595703125,
      "learning_rate": 5.397058823529411e-07,
      "loss": 0.0024,
      "reward": 1.9141494035720825,
      "reward_std": 0.013022142462432384,
      "rewards/accuracy_reward": 0.9141494035720825,
      "rewards/format_reward": 1.0,
      "step": 626
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 113.36298370361328,
      "epoch": 9.220588235294118,
      "grad_norm": 2.5371748740188753,
      "kl": 0.060791015625,
      "learning_rate": 5.389705882352941e-07,
      "loss": 0.0025,
      "reward": 1.91664719581604,
      "reward_std": 0.017335418611764908,
      "rewards/accuracy_reward": 0.9166471362113953,
      "rewards/format_reward": 1.0,
      "step": 627
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 114.4627456665039,
      "epoch": 9.235294117647058,
      "grad_norm": 1.7615752856014142,
      "kl": 0.05712890625,
      "learning_rate": 5.382352941176471e-07,
      "loss": 0.0023,
      "reward": 1.8938990831375122,
      "reward_std": 0.016765251755714417,
      "rewards/accuracy_reward": 0.893899142742157,
      "rewards/format_reward": 1.0,
      "step": 628
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 113.77404022216797,
      "epoch": 9.25,
      "grad_norm": 2.0562724198467914,
      "kl": 0.0634765625,
      "learning_rate": 5.374999999999999e-07,
      "loss": 0.0026,
      "reward": 1.9214463233947754,
      "reward_std": 0.014558774419128895,
      "rewards/accuracy_reward": 0.9214461445808411,
      "rewards/format_reward": 1.0,
      "step": 629
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 113.44351196289062,
      "epoch": 9.264705882352942,
      "grad_norm": 2.1195056229903555,
      "kl": 0.0576171875,
      "learning_rate": 5.367647058823529e-07,
      "loss": 0.0023,
      "reward": 1.9065834283828735,
      "reward_std": 0.016932621598243713,
      "rewards/accuracy_reward": 0.906583309173584,
      "rewards/format_reward": 1.0,
      "step": 630
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 113.56610870361328,
      "epoch": 9.279411764705882,
      "grad_norm": 1.5239442402080228,
      "kl": 0.0546875,
      "learning_rate": 5.360294117647058e-07,
      "loss": 0.0022,
      "reward": 1.9177383184432983,
      "reward_std": 0.014962482266128063,
      "rewards/accuracy_reward": 0.9177382588386536,
      "rewards/format_reward": 1.0,
      "step": 631
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 113.86538696289062,
      "epoch": 9.294117647058824,
      "grad_norm": 2.027623877286765,
      "kl": 0.05517578125,
      "learning_rate": 5.352941176470588e-07,
      "loss": 0.0022,
      "reward": 1.9050103425979614,
      "reward_std": 0.014742089435458183,
      "rewards/accuracy_reward": 0.9050101637840271,
      "rewards/format_reward": 1.0,
      "step": 632
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 113.48558044433594,
      "epoch": 9.308823529411764,
      "grad_norm": 1.9258858099537597,
      "kl": 0.060546875,
      "learning_rate": 5.345588235294117e-07,
      "loss": 0.0024,
      "reward": 1.9041670560836792,
      "reward_std": 0.017647940665483475,
      "rewards/accuracy_reward": 0.9053689241409302,
      "rewards/format_reward": 0.998798131942749,
      "step": 633
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 113.88221740722656,
      "epoch": 9.323529411764707,
      "grad_norm": 5.481626944028736,
      "kl": 0.05419921875,
      "learning_rate": 5.338235294117646e-07,
      "loss": 0.0022,
      "reward": 1.8989014625549316,
      "reward_std": 0.017911560833454132,
      "rewards/accuracy_reward": 0.8989012837409973,
      "rewards/format_reward": 1.0,
      "step": 634
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 114.16947174072266,
      "epoch": 9.338235294117647,
      "grad_norm": 1.5838252460154945,
      "kl": 0.052734375,
      "learning_rate": 5.330882352941176e-07,
      "loss": 0.0021,
      "reward": 1.9077017307281494,
      "reward_std": 0.01683455891907215,
      "rewards/accuracy_reward": 0.9089035987854004,
      "rewards/format_reward": 0.998798131942749,
      "step": 635
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 112.02764892578125,
      "epoch": 9.352941176470589,
      "grad_norm": 20.072264259712355,
      "kl": 0.05908203125,
      "learning_rate": 5.323529411764706e-07,
      "loss": 0.0024,
      "reward": 1.9148465394973755,
      "reward_std": 0.01663627289235592,
      "rewards/accuracy_reward": 0.9160483479499817,
      "rewards/format_reward": 0.998798131942749,
      "step": 636
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 114.4158706665039,
      "epoch": 9.367647058823529,
      "grad_norm": 1.5736397571374754,
      "kl": 0.05712890625,
      "learning_rate": 5.316176470588235e-07,
      "loss": 0.0023,
      "reward": 1.9175165891647339,
      "reward_std": 0.017536688596010208,
      "rewards/accuracy_reward": 0.9175165891647339,
      "rewards/format_reward": 1.0,
      "step": 637
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 111.81971740722656,
      "epoch": 9.382352941176471,
      "grad_norm": 6.15992463353084,
      "kl": 0.054931640625,
      "learning_rate": 5.308823529411764e-07,
      "loss": 0.0022,
      "reward": 1.9007567167282104,
      "reward_std": 0.015532239340245724,
      "rewards/accuracy_reward": 0.9007567167282104,
      "rewards/format_reward": 1.0,
      "step": 638
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 111.92668914794922,
      "epoch": 9.397058823529411,
      "grad_norm": 1.9267289588312566,
      "kl": 0.05712890625,
      "learning_rate": 5.301470588235294e-07,
      "loss": 0.0023,
      "reward": 1.910290241241455,
      "reward_std": 0.017649637535214424,
      "rewards/accuracy_reward": 0.9114922285079956,
      "rewards/format_reward": 0.998798131942749,
      "step": 639
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 113.70553588867188,
      "epoch": 9.411764705882353,
      "grad_norm": 2.331055894294996,
      "kl": 0.057861328125,
      "learning_rate": 5.294117647058823e-07,
      "loss": 0.0023,
      "reward": 1.9117798805236816,
      "reward_std": 0.014000487513840199,
      "rewards/accuracy_reward": 0.9117798209190369,
      "rewards/format_reward": 1.0,
      "step": 640
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 114.7644271850586,
      "epoch": 9.426470588235293,
      "grad_norm": 2.9782765384820302,
      "kl": 0.05908203125,
      "learning_rate": 5.286764705882353e-07,
      "loss": 0.0024,
      "reward": 1.8985432386398315,
      "reward_std": 0.016746515408158302,
      "rewards/accuracy_reward": 0.8985430002212524,
      "rewards/format_reward": 1.0,
      "step": 641
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 113.16106414794922,
      "epoch": 9.441176470588236,
      "grad_norm": 3.0263333302618585,
      "kl": 0.0576171875,
      "learning_rate": 5.279411764705882e-07,
      "loss": 0.0023,
      "reward": 1.9077398777008057,
      "reward_std": 0.016522472724318504,
      "rewards/accuracy_reward": 0.9089417457580566,
      "rewards/format_reward": 0.998798131942749,
      "step": 642
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 112.0252456665039,
      "epoch": 9.455882352941176,
      "grad_norm": 3.6088229354945858,
      "kl": 0.05615234375,
      "learning_rate": 5.272058823529411e-07,
      "loss": 0.0023,
      "reward": 1.912189245223999,
      "reward_std": 0.01710669696331024,
      "rewards/accuracy_reward": 0.9133911728858948,
      "rewards/format_reward": 0.998798131942749,
      "step": 643
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 113.2908706665039,
      "epoch": 9.470588235294118,
      "grad_norm": 8.188981831182572,
      "kl": 0.053466796875,
      "learning_rate": 5.264705882352941e-07,
      "loss": 0.0021,
      "reward": 1.9111279249191284,
      "reward_std": 0.0145622119307518,
      "rewards/accuracy_reward": 0.9111279249191284,
      "rewards/format_reward": 1.0,
      "step": 644
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 111.99639892578125,
      "epoch": 9.485294117647058,
      "grad_norm": 3.3433584553752866,
      "kl": 0.054931640625,
      "learning_rate": 5.257352941176471e-07,
      "loss": 0.0022,
      "reward": 1.9080722332000732,
      "reward_std": 0.01804247312247753,
      "rewards/accuracy_reward": 0.9092740416526794,
      "rewards/format_reward": 0.998798131942749,
      "step": 645
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 113.66346740722656,
      "epoch": 9.5,
      "grad_norm": 1.3581128929115824,
      "kl": 0.05908203125,
      "learning_rate": 5.25e-07,
      "loss": 0.0024,
      "reward": 1.9173364639282227,
      "reward_std": 0.0154867609962821,
      "rewards/accuracy_reward": 0.9173362851142883,
      "rewards/format_reward": 1.0,
      "step": 646
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 111.97956848144531,
      "epoch": 9.514705882352942,
      "grad_norm": 3.583185511646481,
      "kl": 0.0517578125,
      "learning_rate": 5.242647058823529e-07,
      "loss": 0.0021,
      "reward": 1.913078784942627,
      "reward_std": 0.01793815754354,
      "rewards/accuracy_reward": 0.9142807126045227,
      "rewards/format_reward": 0.998798131942749,
      "step": 647
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 112.81971740722656,
      "epoch": 9.529411764705882,
      "grad_norm": 2.2219404022193494,
      "kl": 0.058349609375,
      "learning_rate": 5.235294117647059e-07,
      "loss": 0.0023,
      "reward": 1.9022469520568848,
      "reward_std": 0.019146382808685303,
      "rewards/accuracy_reward": 0.9046508073806763,
      "rewards/format_reward": 0.9975962042808533,
      "step": 648
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 112.07813262939453,
      "epoch": 9.544117647058824,
      "grad_norm": 2.825487501905413,
      "kl": 0.0625,
      "learning_rate": 5.227941176470588e-07,
      "loss": 0.0025,
      "reward": 1.916589617729187,
      "reward_std": 0.016064953058958054,
      "rewards/accuracy_reward": 0.9177915453910828,
      "rewards/format_reward": 0.998798131942749,
      "step": 649
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 110.50601196289062,
      "epoch": 9.558823529411764,
      "grad_norm": 5.158018889820331,
      "kl": 0.051513671875,
      "learning_rate": 5.220588235294118e-07,
      "loss": 0.0021,
      "reward": 1.9119791984558105,
      "reward_std": 0.015298433601856232,
      "rewards/accuracy_reward": 0.9119791388511658,
      "rewards/format_reward": 1.0,
      "step": 650
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 110.58413696289062,
      "epoch": 9.573529411764707,
      "grad_norm": 2.329896463825327,
      "kl": 0.047119140625,
      "learning_rate": 5.213235294117647e-07,
      "loss": 0.0019,
      "reward": 1.9101742506027222,
      "reward_std": 0.015595993027091026,
      "rewards/accuracy_reward": 0.9101741909980774,
      "rewards/format_reward": 1.0,
      "step": 651
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 110.80769348144531,
      "epoch": 9.588235294117647,
      "grad_norm": 1.93503535510301,
      "kl": 0.05322265625,
      "learning_rate": 5.205882352941176e-07,
      "loss": 0.0021,
      "reward": 1.9048891067504883,
      "reward_std": 0.014783341437578201,
      "rewards/accuracy_reward": 0.904888927936554,
      "rewards/format_reward": 1.0,
      "step": 652
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 109.08534240722656,
      "epoch": 9.602941176470589,
      "grad_norm": 2.84495334769509,
      "kl": 0.060791015625,
      "learning_rate": 5.198529411764706e-07,
      "loss": 0.0024,
      "reward": 1.9240138530731201,
      "reward_std": 0.014201764948666096,
      "rewards/accuracy_reward": 0.9240137338638306,
      "rewards/format_reward": 1.0,
      "step": 653
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 109.84856414794922,
      "epoch": 9.617647058823529,
      "grad_norm": 3.012251558930328,
      "kl": 0.05419921875,
      "learning_rate": 5.191176470588236e-07,
      "loss": 0.0022,
      "reward": 1.905051350593567,
      "reward_std": 0.013432993553578854,
      "rewards/accuracy_reward": 0.9050513505935669,
      "rewards/format_reward": 1.0,
      "step": 654
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 108.21394348144531,
      "epoch": 9.632352941176471,
      "grad_norm": 2.1316420678211956,
      "kl": 0.056640625,
      "learning_rate": 5.183823529411764e-07,
      "loss": 0.0023,
      "reward": 1.9092304706573486,
      "reward_std": 0.01678118109703064,
      "rewards/accuracy_reward": 0.9092304706573486,
      "rewards/format_reward": 1.0,
      "step": 655
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 108.33654022216797,
      "epoch": 9.647058823529411,
      "grad_norm": 2.2956788687873826,
      "kl": 0.06298828125,
      "learning_rate": 5.176470588235294e-07,
      "loss": 0.0025,
      "reward": 1.92042076587677,
      "reward_std": 0.016655659303069115,
      "rewards/accuracy_reward": 0.92042076587677,
      "rewards/format_reward": 1.0,
      "step": 656
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 109.62259674072266,
      "epoch": 9.661764705882353,
      "grad_norm": 2.467082612046906,
      "kl": 0.054931640625,
      "learning_rate": 5.169117647058824e-07,
      "loss": 0.0022,
      "reward": 1.9160475730895996,
      "reward_std": 0.01348184235394001,
      "rewards/accuracy_reward": 0.9160475134849548,
      "rewards/format_reward": 1.0,
      "step": 657
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 107.80529022216797,
      "epoch": 9.676470588235293,
      "grad_norm": 7.66545161952161,
      "kl": 0.06396484375,
      "learning_rate": 5.161764705882353e-07,
      "loss": 0.0026,
      "reward": 1.9231846332550049,
      "reward_std": 0.016615498811006546,
      "rewards/accuracy_reward": 0.9231847524642944,
      "rewards/format_reward": 1.0,
      "step": 658
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 109.1189956665039,
      "epoch": 9.691176470588236,
      "grad_norm": 2.341042431156855,
      "kl": 0.057373046875,
      "learning_rate": 5.154411764705882e-07,
      "loss": 0.0023,
      "reward": 1.912707805633545,
      "reward_std": 0.018813416361808777,
      "rewards/accuracy_reward": 0.9127077460289001,
      "rewards/format_reward": 1.0,
      "step": 659
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 109.10938262939453,
      "epoch": 9.705882352941176,
      "grad_norm": 2.1614288925748966,
      "kl": 0.0654296875,
      "learning_rate": 5.147058823529411e-07,
      "loss": 0.0026,
      "reward": 1.9185718297958374,
      "reward_std": 0.01388857513666153,
      "rewards/accuracy_reward": 0.9185718894004822,
      "rewards/format_reward": 1.0,
      "step": 660
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 108.92788696289062,
      "epoch": 9.720588235294118,
      "grad_norm": 2.80121444579991,
      "kl": 0.06396484375,
      "learning_rate": 5.13970588235294e-07,
      "loss": 0.0026,
      "reward": 1.9083857536315918,
      "reward_std": 0.01726207509636879,
      "rewards/accuracy_reward": 0.9083855152130127,
      "rewards/format_reward": 1.0,
      "step": 661
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 108.50721740722656,
      "epoch": 9.735294117647058,
      "grad_norm": 2.7135290225206936,
      "kl": 0.060546875,
      "learning_rate": 5.13235294117647e-07,
      "loss": 0.0024,
      "reward": 1.8957988023757935,
      "reward_std": 0.017370108515024185,
      "rewards/accuracy_reward": 0.8957986235618591,
      "rewards/format_reward": 1.0,
      "step": 662
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 109.86298370361328,
      "epoch": 9.75,
      "grad_norm": 1.6206042221923098,
      "kl": 0.054931640625,
      "learning_rate": 5.125e-07,
      "loss": 0.0022,
      "reward": 1.9014039039611816,
      "reward_std": 0.0170334093272686,
      "rewards/accuracy_reward": 0.9014038443565369,
      "rewards/format_reward": 1.0,
      "step": 663
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 109.10577392578125,
      "epoch": 9.764705882352942,
      "grad_norm": 2.1238141084640287,
      "kl": 0.056396484375,
      "learning_rate": 5.117647058823528e-07,
      "loss": 0.0022,
      "reward": 1.9057307243347168,
      "reward_std": 0.017774611711502075,
      "rewards/accuracy_reward": 0.9057307243347168,
      "rewards/format_reward": 1.0,
      "step": 664
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 109.60096740722656,
      "epoch": 9.779411764705882,
      "grad_norm": 1.6218363054606357,
      "kl": 0.052490234375,
      "learning_rate": 5.110294117647058e-07,
      "loss": 0.0021,
      "reward": 1.892624855041504,
      "reward_std": 0.018729185685515404,
      "rewards/accuracy_reward": 0.8926247954368591,
      "rewards/format_reward": 1.0,
      "step": 665
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 108.73798370361328,
      "epoch": 9.794117647058824,
      "grad_norm": 4.41894162926749,
      "kl": 0.061279296875,
      "learning_rate": 5.102941176470588e-07,
      "loss": 0.0024,
      "reward": 1.8940479755401611,
      "reward_std": 0.019051702693104744,
      "rewards/accuracy_reward": 0.8940479159355164,
      "rewards/format_reward": 1.0,
      "step": 666
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 109.1706771850586,
      "epoch": 9.808823529411764,
      "grad_norm": 2.5204373524492754,
      "kl": 0.0625,
      "learning_rate": 5.095588235294117e-07,
      "loss": 0.0025,
      "reward": 1.9303815364837646,
      "reward_std": 0.016049683094024658,
      "rewards/accuracy_reward": 0.9303815364837646,
      "rewards/format_reward": 1.0,
      "step": 667
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 110.2596206665039,
      "epoch": 9.823529411764707,
      "grad_norm": 2.247808117324336,
      "kl": 0.062255859375,
      "learning_rate": 5.088235294117646e-07,
      "loss": 0.0025,
      "reward": 1.905713677406311,
      "reward_std": 0.0260589811950922,
      "rewards/accuracy_reward": 0.9069154858589172,
      "rewards/format_reward": 0.998798131942749,
      "step": 668
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 109.00360870361328,
      "epoch": 9.838235294117647,
      "grad_norm": 2.9378173829289542,
      "kl": 0.05517578125,
      "learning_rate": 5.080882352941176e-07,
      "loss": 0.0022,
      "reward": 1.9220634698867798,
      "reward_std": 0.018444770947098732,
      "rewards/accuracy_reward": 0.923265278339386,
      "rewards/format_reward": 0.998798131942749,
      "step": 669
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 109.38341522216797,
      "epoch": 9.852941176470589,
      "grad_norm": 3.1381468730562396,
      "kl": 0.055908203125,
      "learning_rate": 5.073529411764705e-07,
      "loss": 0.0022,
      "reward": 1.9113494157791138,
      "reward_std": 0.017814362421631813,
      "rewards/accuracy_reward": 0.91255122423172,
      "rewards/format_reward": 0.998798131942749,
      "step": 670
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 110.08654022216797,
      "epoch": 9.867647058823529,
      "grad_norm": 1.8503806680078807,
      "kl": 0.053466796875,
      "learning_rate": 5.066176470588235e-07,
      "loss": 0.0021,
      "reward": 1.9079095125198364,
      "reward_std": 0.01949225552380085,
      "rewards/accuracy_reward": 0.9091113209724426,
      "rewards/format_reward": 0.998798131942749,
      "step": 671
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 110.08654022216797,
      "epoch": 9.882352941176471,
      "grad_norm": 1.6795640500896751,
      "kl": 0.051025390625,
      "learning_rate": 5.058823529411765e-07,
      "loss": 0.002,
      "reward": 1.9217894077301025,
      "reward_std": 0.01603555493056774,
      "rewards/accuracy_reward": 0.9217892289161682,
      "rewards/format_reward": 1.0,
      "step": 672
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 109.29327392578125,
      "epoch": 9.897058823529411,
      "grad_norm": 1.9285682708702305,
      "kl": 0.05126953125,
      "learning_rate": 5.051470588235293e-07,
      "loss": 0.002,
      "reward": 1.9053161144256592,
      "reward_std": 0.01650179550051689,
      "rewards/accuracy_reward": 0.9053158760070801,
      "rewards/format_reward": 1.0,
      "step": 673
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 109.87981414794922,
      "epoch": 9.911764705882353,
      "grad_norm": 2.038439014531085,
      "kl": 0.05126953125,
      "learning_rate": 5.044117647058823e-07,
      "loss": 0.002,
      "reward": 1.9242396354675293,
      "reward_std": 0.016753748059272766,
      "rewards/accuracy_reward": 0.9242396354675293,
      "rewards/format_reward": 1.0,
      "step": 674
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 110.26202392578125,
      "epoch": 9.926470588235293,
      "grad_norm": 1.6234604098074579,
      "kl": 0.053955078125,
      "learning_rate": 5.036764705882353e-07,
      "loss": 0.0022,
      "reward": 1.914339542388916,
      "reward_std": 0.01580924354493618,
      "rewards/accuracy_reward": 0.9143396019935608,
      "rewards/format_reward": 1.0,
      "step": 675
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 111.72116088867188,
      "epoch": 9.941176470588236,
      "grad_norm": 5.62812065845287,
      "kl": 0.0537109375,
      "learning_rate": 5.029411764705882e-07,
      "loss": 0.0021,
      "reward": 1.905260443687439,
      "reward_std": 0.018505161628127098,
      "rewards/accuracy_reward": 0.9052603244781494,
      "rewards/format_reward": 1.0,
      "step": 676
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 111.61538696289062,
      "epoch": 9.955882352941176,
      "grad_norm": 1.7952493866544998,
      "kl": 0.048095703125,
      "learning_rate": 5.022058823529411e-07,
      "loss": 0.0019,
      "reward": 1.9118279218673706,
      "reward_std": 0.01776190474629402,
      "rewards/accuracy_reward": 0.9118278622627258,
      "rewards/format_reward": 1.0,
      "step": 677
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 112.45433044433594,
      "epoch": 9.970588235294118,
      "grad_norm": 1.7192070770567416,
      "kl": 0.059814453125,
      "learning_rate": 5.014705882352941e-07,
      "loss": 0.0024,
      "reward": 1.9066780805587769,
      "reward_std": 0.01788855344057083,
      "rewards/accuracy_reward": 0.9066779613494873,
      "rewards/format_reward": 1.0,
      "step": 678
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 113.34856414794922,
      "epoch": 9.985294117647058,
      "grad_norm": 2.906612944219061,
      "kl": 0.06005859375,
      "learning_rate": 5.00735294117647e-07,
      "loss": 0.0024,
      "reward": 1.9021302461624146,
      "reward_std": 0.022055475041270256,
      "rewards/accuracy_reward": 0.9021300673484802,
      "rewards/format_reward": 1.0,
      "step": 679
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 109.98021697998047,
      "epoch": 10.0,
      "grad_norm": 1.6562395375859271,
      "kl": 0.047607421875,
      "learning_rate": 5e-07,
      "loss": 0.0019,
      "reward": 1.9186270236968994,
      "reward_std": 0.019354015588760376,
      "rewards/accuracy_reward": 0.9201489090919495,
      "rewards/format_reward": 0.9984779357910156,
      "step": 680
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 111.89423370361328,
      "epoch": 10.014705882352942,
      "grad_norm": 2.9068911534041764,
      "kl": 0.05712890625,
      "learning_rate": 4.99264705882353e-07,
      "loss": 0.0023,
      "reward": 1.907537817955017,
      "reward_std": 0.01922653429210186,
      "rewards/accuracy_reward": 0.9075377583503723,
      "rewards/format_reward": 1.0,
      "step": 681
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 111.22356414794922,
      "epoch": 10.029411764705882,
      "grad_norm": 3.5541747826461805,
      "kl": 0.055419921875,
      "learning_rate": 4.985294117647058e-07,
      "loss": 0.0022,
      "reward": 1.917733073234558,
      "reward_std": 0.01717350259423256,
      "rewards/accuracy_reward": 0.9177329540252686,
      "rewards/format_reward": 1.0,
      "step": 682
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 112.38341522216797,
      "epoch": 10.044117647058824,
      "grad_norm": 1.903386359782145,
      "kl": 0.057373046875,
      "learning_rate": 4.977941176470588e-07,
      "loss": 0.0023,
      "reward": 1.9117200374603271,
      "reward_std": 0.01899932697415352,
      "rewards/accuracy_reward": 0.9129219651222229,
      "rewards/format_reward": 0.998798131942749,
      "step": 683
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 111.29688262939453,
      "epoch": 10.058823529411764,
      "grad_norm": 1.8861273273154469,
      "kl": 0.05810546875,
      "learning_rate": 4.970588235294118e-07,
      "loss": 0.0023,
      "reward": 1.9124337434768677,
      "reward_std": 0.02022276073694229,
      "rewards/accuracy_reward": 0.9136357307434082,
      "rewards/format_reward": 0.998798131942749,
      "step": 684
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 112.86058044433594,
      "epoch": 10.073529411764707,
      "grad_norm": 2.8430111341562823,
      "kl": 0.05078125,
      "learning_rate": 4.963235294117647e-07,
      "loss": 0.002,
      "reward": 1.9203444719314575,
      "reward_std": 0.016688454896211624,
      "rewards/accuracy_reward": 0.9203445315361023,
      "rewards/format_reward": 1.0,
      "step": 685
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 112.33053588867188,
      "epoch": 10.088235294117647,
      "grad_norm": 2.803964635402021,
      "kl": 0.05859375,
      "learning_rate": 4.955882352941176e-07,
      "loss": 0.0023,
      "reward": 1.9130260944366455,
      "reward_std": 0.019971825182437897,
      "rewards/accuracy_reward": 0.9142280220985413,
      "rewards/format_reward": 0.998798131942749,
      "step": 686
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 113.25481414794922,
      "epoch": 10.102941176470589,
      "grad_norm": 2.1290481897859066,
      "kl": 0.05517578125,
      "learning_rate": 4.948529411764706e-07,
      "loss": 0.0022,
      "reward": 1.9154521226882935,
      "reward_std": 0.019178370013833046,
      "rewards/accuracy_reward": 0.9166539311408997,
      "rewards/format_reward": 0.998798131942749,
      "step": 687
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 115.15264892578125,
      "epoch": 10.117647058823529,
      "grad_norm": 2.2315042906400944,
      "kl": 0.0556640625,
      "learning_rate": 4.941176470588235e-07,
      "loss": 0.0022,
      "reward": 1.9171024560928345,
      "reward_std": 0.015508743934333324,
      "rewards/accuracy_reward": 0.9171023368835449,
      "rewards/format_reward": 1.0,
      "step": 688
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 114.18991088867188,
      "epoch": 10.132352941176471,
      "grad_norm": 8.899062168348006,
      "kl": 0.05859375,
      "learning_rate": 4.933823529411765e-07,
      "loss": 0.0023,
      "reward": 1.918565034866333,
      "reward_std": 0.016680248081684113,
      "rewards/accuracy_reward": 0.9185649156570435,
      "rewards/format_reward": 1.0,
      "step": 689
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 114.71154022216797,
      "epoch": 10.147058823529411,
      "grad_norm": 1.684126014415895,
      "kl": 0.05712890625,
      "learning_rate": 4.926470588235295e-07,
      "loss": 0.0023,
      "reward": 1.8880369663238525,
      "reward_std": 0.01872415281832218,
      "rewards/accuracy_reward": 0.8880367875099182,
      "rewards/format_reward": 1.0,
      "step": 690
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 114.60697174072266,
      "epoch": 10.161764705882353,
      "grad_norm": 2.3449846328839596,
      "kl": 0.05712890625,
      "learning_rate": 4.919117647058823e-07,
      "loss": 0.0023,
      "reward": 1.910075306892395,
      "reward_std": 0.024183494970202446,
      "rewards/accuracy_reward": 0.9124791622161865,
      "rewards/format_reward": 0.9975962042808533,
      "step": 691
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 113.82572174072266,
      "epoch": 10.176470588235293,
      "grad_norm": 3.23331569595653,
      "kl": 0.06298828125,
      "learning_rate": 4.911764705882352e-07,
      "loss": 0.0025,
      "reward": 1.9134478569030762,
      "reward_std": 0.01637120544910431,
      "rewards/accuracy_reward": 0.9134476184844971,
      "rewards/format_reward": 1.0,
      "step": 692
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 112.63702392578125,
      "epoch": 10.191176470588236,
      "grad_norm": 2.3313697919069027,
      "kl": 0.054931640625,
      "learning_rate": 4.904411764705882e-07,
      "loss": 0.0022,
      "reward": 1.905639410018921,
      "reward_std": 0.016094274818897247,
      "rewards/accuracy_reward": 0.9056394100189209,
      "rewards/format_reward": 1.0,
      "step": 693
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 112.92188262939453,
      "epoch": 10.205882352941176,
      "grad_norm": 2.6050118805934894,
      "kl": 0.061767578125,
      "learning_rate": 4.897058823529412e-07,
      "loss": 0.0025,
      "reward": 1.9092425107955933,
      "reward_std": 0.017230795696377754,
      "rewards/accuracy_reward": 0.9092425107955933,
      "rewards/format_reward": 1.0,
      "step": 694
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 112.80409240722656,
      "epoch": 10.220588235294118,
      "grad_norm": 2.25480395310822,
      "kl": 0.05615234375,
      "learning_rate": 4.88970588235294e-07,
      "loss": 0.0023,
      "reward": 1.9153361320495605,
      "reward_std": 0.01601433753967285,
      "rewards/accuracy_reward": 0.9153360724449158,
      "rewards/format_reward": 1.0,
      "step": 695
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 113.41226196289062,
      "epoch": 10.235294117647058,
      "grad_norm": 1.8257275879557004,
      "kl": 0.05712890625,
      "learning_rate": 4.88235294117647e-07,
      "loss": 0.0023,
      "reward": 1.9105335474014282,
      "reward_std": 0.014649370685219765,
      "rewards/accuracy_reward": 0.9105334877967834,
      "rewards/format_reward": 1.0,
      "step": 696
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 113.64063262939453,
      "epoch": 10.25,
      "grad_norm": 1.904503451542197,
      "kl": 0.05712890625,
      "learning_rate": 4.875e-07,
      "loss": 0.0023,
      "reward": 1.9262598752975464,
      "reward_std": 0.01650756038725376,
      "rewards/accuracy_reward": 0.9262598752975464,
      "rewards/format_reward": 1.0,
      "step": 697
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 110.5144271850586,
      "epoch": 10.264705882352942,
      "grad_norm": 2.0347260430337584,
      "kl": 0.05712890625,
      "learning_rate": 4.867647058823529e-07,
      "loss": 0.0023,
      "reward": 1.9039766788482666,
      "reward_std": 0.01813569664955139,
      "rewards/accuracy_reward": 0.9051785469055176,
      "rewards/format_reward": 0.998798131942749,
      "step": 698
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 112.98798370361328,
      "epoch": 10.279411764705882,
      "grad_norm": 3.903638499409672,
      "kl": 0.055908203125,
      "learning_rate": 4.860294117647058e-07,
      "loss": 0.0022,
      "reward": 1.9150954484939575,
      "reward_std": 0.014970256946980953,
      "rewards/accuracy_reward": 0.9150952696800232,
      "rewards/format_reward": 1.0,
      "step": 699
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 111.92308044433594,
      "epoch": 10.294117647058824,
      "grad_norm": 2.18266269445753,
      "kl": 0.0595703125,
      "learning_rate": 4.852941176470588e-07,
      "loss": 0.0024,
      "reward": 1.9265812635421753,
      "reward_std": 0.013946098275482655,
      "rewards/accuracy_reward": 0.9265811443328857,
      "rewards/format_reward": 1.0,
      "step": 700
    }
  ],
  "logging_steps": 1.0,
  "max_steps": 1360,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 20,
  "save_steps": 100,
  "stateful_callbacks": {
    "TrainerControl": {
      "args": {
        "should_epoch_stop": false,
        "should_evaluate": false,
        "should_log": false,
        "should_save": true,
        "should_training_stop": false
      },
      "attributes": {}
    }
  },
  "total_flos": 0.0,
  "train_batch_size": 52,
  "trial_name": null,
  "trial_params": null
}