{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 0.045714285714285714,
  "eval_steps": 500,
  "global_step": 200,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "completion_length": 60.5390625,
      "epoch": 0.00022857142857142857,
      "grad_norm": 10.703285191093784,
      "kl": 0.0,
      "learning_rate": 9.998857142857142e-07,
      "loss": -0.0,
      "reward": 0.8828125,
      "reward_std": 0.6644280552864075,
      "rewards/accuracy_reward": 0.5390625,
      "rewards/format_reward": 0.34375,
      "step": 1
    },
    {
      "completion_length": 57.21875,
      "epoch": 0.00045714285714285713,
      "grad_norm": 22.66626440357455,
      "kl": 0.00278472900390625,
      "learning_rate": 9.997714285714284e-07,
      "loss": 0.0001,
      "reward": 0.859375,
      "reward_std": 0.5827889442443848,
      "rewards/accuracy_reward": 0.609375,
      "rewards/format_reward": 0.25,
      "step": 2
    },
    {
      "completion_length": 67.8046875,
      "epoch": 0.0006857142857142857,
      "grad_norm": 10.690573424380442,
      "kl": 0.002536773681640625,
      "learning_rate": 9.996571428571428e-07,
      "loss": 0.0001,
      "reward": 0.96875,
      "reward_std": 0.6606916189193726,
      "rewards/accuracy_reward": 0.59375,
      "rewards/format_reward": 0.375,
      "step": 3
    },
    {
      "completion_length": 67.453125,
      "epoch": 0.0009142857142857143,
      "grad_norm": 12.689291178005337,
      "kl": 0.0084381103515625,
      "learning_rate": 9.99542857142857e-07,
      "loss": 0.0003,
      "reward": 1.140625,
      "reward_std": 0.653181254863739,
      "rewards/accuracy_reward": 0.6796875,
      "rewards/format_reward": 0.4609375,
      "step": 4
    },
    {
      "completion_length": 73.7578125,
      "epoch": 0.001142857142857143,
      "grad_norm": 21.665928823719504,
      "kl": 0.01983642578125,
      "learning_rate": 9.994285714285715e-07,
      "loss": 0.0008,
      "reward": 1.078125,
      "reward_std": 0.6801478266716003,
      "rewards/accuracy_reward": 0.6171875,
      "rewards/format_reward": 0.4609375,
      "step": 5
    },
    {
      "completion_length": 71.5,
      "epoch": 0.0013714285714285714,
      "grad_norm": 13.644383775583323,
      "kl": 0.15087890625,
      "learning_rate": 9.993142857142857e-07,
      "loss": 0.006,
      "reward": 1.359375,
      "reward_std": 0.7011792361736298,
      "rewards/accuracy_reward": 0.71875,
      "rewards/format_reward": 0.640625,
      "step": 6
    },
    {
      "completion_length": 70.3125,
      "epoch": 0.0016,
      "grad_norm": 7.649875433185531,
      "kl": 0.076171875,
      "learning_rate": 9.992e-07,
      "loss": 0.003,
      "reward": 1.21875,
      "reward_std": 0.6829855144023895,
      "rewards/accuracy_reward": 0.578125,
      "rewards/format_reward": 0.640625,
      "step": 7
    },
    {
      "completion_length": 68.59375,
      "epoch": 0.0018285714285714285,
      "grad_norm": 7.022608082329708,
      "kl": 0.01971435546875,
      "learning_rate": 9.990857142857141e-07,
      "loss": 0.0008,
      "reward": 1.265625,
      "reward_std": 0.7393486499786377,
      "rewards/accuracy_reward": 0.5859375,
      "rewards/format_reward": 0.6796875,
      "step": 8
    },
    {
      "completion_length": 61.3203125,
      "epoch": 0.002057142857142857,
      "grad_norm": 9.199767598881943,
      "kl": 0.03564453125,
      "learning_rate": 9.989714285714285e-07,
      "loss": 0.0014,
      "reward": 1.4140625,
      "reward_std": 0.6415750086307526,
      "rewards/accuracy_reward": 0.625,
      "rewards/format_reward": 0.7890625,
      "step": 9
    },
    {
      "completion_length": 39.0859375,
      "epoch": 0.002285714285714286,
      "grad_norm": 8.277586397113673,
      "kl": 0.079833984375,
      "learning_rate": 9.988571428571428e-07,
      "loss": 0.0032,
      "reward": 1.65625,
      "reward_std": 0.4268445074558258,
      "rewards/accuracy_reward": 0.7109375,
      "rewards/format_reward": 0.9453125,
      "step": 10
    },
    {
      "completion_length": 34.0625,
      "epoch": 0.002514285714285714,
      "grad_norm": 32.71996883582455,
      "kl": 0.35009765625,
      "learning_rate": 9.987428571428572e-07,
      "loss": 0.014,
      "reward": 1.84375,
      "reward_std": 0.34242063760757446,
      "rewards/accuracy_reward": 0.8828125,
      "rewards/format_reward": 0.9609375,
      "step": 11
    },
    {
      "completion_length": 30.515625,
      "epoch": 0.002742857142857143,
      "grad_norm": 6.939538623922572,
      "kl": 0.14404296875,
      "learning_rate": 9.986285714285714e-07,
      "loss": 0.0058,
      "reward": 1.8125,
      "reward_std": 0.2973729074001312,
      "rewards/accuracy_reward": 0.828125,
      "rewards/format_reward": 0.984375,
      "step": 12
    },
    {
      "completion_length": 28.71875,
      "epoch": 0.0029714285714285715,
      "grad_norm": 13.847555179940393,
      "kl": 0.125244140625,
      "learning_rate": 9.985142857142856e-07,
      "loss": 0.005,
      "reward": 1.78125,
      "reward_std": 0.34511563181877136,
      "rewards/accuracy_reward": 0.8203125,
      "rewards/format_reward": 0.9609375,
      "step": 13
    },
    {
      "completion_length": 26.546875,
      "epoch": 0.0032,
      "grad_norm": 9.845787547823988,
      "kl": 0.1943359375,
      "learning_rate": 9.983999999999998e-07,
      "loss": 0.0078,
      "reward": 1.875,
      "reward_std": 0.25513993203639984,
      "rewards/accuracy_reward": 0.8828125,
      "rewards/format_reward": 0.9921875,
      "step": 14
    },
    {
      "completion_length": 24.375,
      "epoch": 0.0034285714285714284,
      "grad_norm": 6.605027050176717,
      "kl": 0.18017578125,
      "learning_rate": 9.982857142857143e-07,
      "loss": 0.0072,
      "reward": 1.9453125,
      "reward_std": 0.13941730558872223,
      "rewards/accuracy_reward": 0.96875,
      "rewards/format_reward": 0.9765625,
      "step": 15
    },
    {
      "completion_length": 25.03125,
      "epoch": 0.003657142857142857,
      "grad_norm": 6.444383169792875,
      "kl": 0.1826171875,
      "learning_rate": 9.981714285714285e-07,
      "loss": 0.0073,
      "reward": 1.9296875,
      "reward_std": 0.14855942130088806,
      "rewards/accuracy_reward": 0.9375,
      "rewards/format_reward": 0.9921875,
      "step": 16
    },
    {
      "completion_length": 22.890625,
      "epoch": 0.0038857142857142857,
      "grad_norm": 5.855107670376587,
      "kl": 0.2314453125,
      "learning_rate": 9.98057142857143e-07,
      "loss": 0.0092,
      "reward": 1.9609375,
      "reward_std": 0.11048543266952038,
      "rewards/accuracy_reward": 0.96875,
      "rewards/format_reward": 0.9921875,
      "step": 17
    },
    {
      "completion_length": 25.6015625,
      "epoch": 0.004114285714285714,
      "grad_norm": 4.239263893707161,
      "kl": 0.146240234375,
      "learning_rate": 9.979428571428571e-07,
      "loss": 0.0058,
      "reward": 1.890625,
      "reward_std": 0.16097760759294033,
      "rewards/accuracy_reward": 0.8984375,
      "rewards/format_reward": 0.9921875,
      "step": 18
    },
    {
      "completion_length": 24.6796875,
      "epoch": 0.004342857142857143,
      "grad_norm": 85.87272813883176,
      "kl": 1.6513671875,
      "learning_rate": 9.978285714285713e-07,
      "loss": 0.0663,
      "reward": 1.9453125,
      "reward_std": 0.15467960387468338,
      "rewards/accuracy_reward": 0.953125,
      "rewards/format_reward": 0.9921875,
      "step": 19
    },
    {
      "completion_length": 25.21875,
      "epoch": 0.004571428571428572,
      "grad_norm": 7.352112071445704,
      "kl": 0.121826171875,
      "learning_rate": 9.977142857142856e-07,
      "loss": 0.0049,
      "reward": 1.921875,
      "reward_std": 0.14283225685358047,
      "rewards/accuracy_reward": 0.9375,
      "rewards/format_reward": 0.984375,
      "step": 20
    },
    {
      "completion_length": 32.171875,
      "epoch": 0.0048,
      "grad_norm": 3.616286943831982,
      "kl": 0.116943359375,
      "learning_rate": 9.976e-07,
      "loss": 0.0047,
      "reward": 1.9453125,
      "reward_std": 0.13941730558872223,
      "rewards/accuracy_reward": 0.953125,
      "rewards/format_reward": 0.9921875,
      "step": 21
    },
    {
      "completion_length": 37.6015625,
      "epoch": 0.005028571428571428,
      "grad_norm": 7.274515372329607,
      "kl": 0.091552734375,
      "learning_rate": 9.974857142857142e-07,
      "loss": 0.0037,
      "reward": 1.8515625,
      "reward_std": 0.24452663213014603,
      "rewards/accuracy_reward": 0.8828125,
      "rewards/format_reward": 0.96875,
      "step": 22
    },
    {
      "completion_length": 44.0078125,
      "epoch": 0.005257142857142857,
      "grad_norm": 5.947417900695541,
      "kl": 0.0751953125,
      "learning_rate": 9.973714285714286e-07,
      "loss": 0.003,
      "reward": 1.8203125,
      "reward_std": 0.2909065783023834,
      "rewards/accuracy_reward": 0.828125,
      "rewards/format_reward": 0.9921875,
      "step": 23
    },
    {
      "completion_length": 43.109375,
      "epoch": 0.005485714285714286,
      "grad_norm": 3.8347490932566988,
      "kl": 0.0712890625,
      "learning_rate": 9.972571428571428e-07,
      "loss": 0.0028,
      "reward": 1.8515625,
      "reward_std": 0.2421332448720932,
      "rewards/accuracy_reward": 0.875,
      "rewards/format_reward": 0.9765625,
      "step": 24
    },
    {
      "completion_length": 41.0390625,
      "epoch": 0.005714285714285714,
      "grad_norm": 8.800620022314726,
      "kl": 0.0908203125,
      "learning_rate": 9.97142857142857e-07,
      "loss": 0.0036,
      "reward": 1.859375,
      "reward_std": 0.30617378652095795,
      "rewards/accuracy_reward": 0.8984375,
      "rewards/format_reward": 0.9609375,
      "step": 25
    },
    {
      "completion_length": 46.34375,
      "epoch": 0.005942857142857143,
      "grad_norm": 7.875592660314657,
      "kl": 0.095703125,
      "learning_rate": 9.970285714285713e-07,
      "loss": 0.0038,
      "reward": 1.8359375,
      "reward_std": 0.3247993439435959,
      "rewards/accuracy_reward": 0.8671875,
      "rewards/format_reward": 0.96875,
      "step": 26
    },
    {
      "completion_length": 36.5703125,
      "epoch": 0.006171428571428572,
      "grad_norm": 4.596628477713428,
      "kl": 0.080810546875,
      "learning_rate": 9.969142857142857e-07,
      "loss": 0.0032,
      "reward": 1.9296875,
      "reward_std": 0.15308690071105957,
      "rewards/accuracy_reward": 0.953125,
      "rewards/format_reward": 0.9765625,
      "step": 27
    },
    {
      "completion_length": 36.671875,
      "epoch": 0.0064,
      "grad_norm": 4.277075434635652,
      "kl": 0.143310546875,
      "learning_rate": 9.968e-07,
      "loss": 0.0057,
      "reward": 1.9140625,
      "reward_std": 0.19728107750415802,
      "rewards/accuracy_reward": 0.9453125,
      "rewards/format_reward": 0.96875,
      "step": 28
    },
    {
      "completion_length": 33.03125,
      "epoch": 0.006628571428571429,
      "grad_norm": 5.060462099806738,
      "kl": 0.1181640625,
      "learning_rate": 9.966857142857141e-07,
      "loss": 0.0047,
      "reward": 1.9609375,
      "reward_std": 0.05550474114716053,
      "rewards/accuracy_reward": 0.96875,
      "rewards/format_reward": 0.9921875,
      "step": 29
    },
    {
      "completion_length": 30.328125,
      "epoch": 0.006857142857142857,
      "grad_norm": 7.160561542512279,
      "kl": 0.130859375,
      "learning_rate": 9.965714285714286e-07,
      "loss": 0.0052,
      "reward": 1.890625,
      "reward_std": 0.1841355413198471,
      "rewards/accuracy_reward": 0.921875,
      "rewards/format_reward": 0.96875,
      "step": 30
    },
    {
      "completion_length": 30.4609375,
      "epoch": 0.0070857142857142855,
      "grad_norm": 7.652933666919937,
      "kl": 0.15185546875,
      "learning_rate": 9.964571428571428e-07,
      "loss": 0.0061,
      "reward": 1.9609375,
      "reward_std": 0.09522313997149467,
      "rewards/accuracy_reward": 0.984375,
      "rewards/format_reward": 0.9765625,
      "step": 31
    },
    {
      "completion_length": 27.9765625,
      "epoch": 0.007314285714285714,
      "grad_norm": 4.358139729170779,
      "kl": 0.22509765625,
      "learning_rate": 9.963428571428572e-07,
      "loss": 0.009,
      "reward": 1.921875,
      "reward_std": 0.1751839816570282,
      "rewards/accuracy_reward": 0.984375,
      "rewards/format_reward": 0.9375,
      "step": 32
    },
    {
      "completion_length": 24.65625,
      "epoch": 0.007542857142857143,
      "grad_norm": 5.8837498672235276,
      "kl": 0.23095703125,
      "learning_rate": 9.962285714285714e-07,
      "loss": 0.0092,
      "reward": 1.9296875,
      "reward_std": 0.1649293415248394,
      "rewards/accuracy_reward": 0.9609375,
      "rewards/format_reward": 0.96875,
      "step": 33
    },
    {
      "completion_length": 23.703125,
      "epoch": 0.0077714285714285715,
      "grad_norm": 9.238459245997138,
      "kl": 0.17822265625,
      "learning_rate": 9.961142857142856e-07,
      "loss": 0.0071,
      "reward": 1.9375,
      "reward_std": 0.12756996601819992,
      "rewards/accuracy_reward": 0.9453125,
      "rewards/format_reward": 0.9921875,
      "step": 34
    },
    {
      "completion_length": 23.4609375,
      "epoch": 0.008,
      "grad_norm": 4.868805817754177,
      "kl": 0.13671875,
      "learning_rate": 9.959999999999999e-07,
      "loss": 0.0055,
      "reward": 1.8984375,
      "reward_std": 0.0765409953892231,
      "rewards/accuracy_reward": 0.90625,
      "rewards/format_reward": 0.9921875,
      "step": 35
    },
    {
      "completion_length": 23.3125,
      "epoch": 0.008228571428571429,
      "grad_norm": 3.226487511067401,
      "kl": 0.118896484375,
      "learning_rate": 9.958857142857143e-07,
      "loss": 0.0048,
      "reward": 1.96875,
      "reward_std": 0.07312605530023575,
      "rewards/accuracy_reward": 0.96875,
      "rewards/format_reward": 1.0,
      "step": 36
    },
    {
      "completion_length": 24.09375,
      "epoch": 0.008457142857142858,
      "grad_norm": 7.80114791539136,
      "kl": 0.19775390625,
      "learning_rate": 9.957714285714285e-07,
      "loss": 0.0079,
      "reward": 1.9609375,
      "reward_std": 0.09522313997149467,
      "rewards/accuracy_reward": 0.96875,
      "rewards/format_reward": 0.9921875,
      "step": 37
    },
    {
      "completion_length": 25.1484375,
      "epoch": 0.008685714285714286,
      "grad_norm": 4.18932865913775,
      "kl": 0.14453125,
      "learning_rate": 9.95657142857143e-07,
      "loss": 0.0058,
      "reward": 1.9765625,
      "reward_std": 0.06629125773906708,
      "rewards/accuracy_reward": 0.9921875,
      "rewards/format_reward": 0.984375,
      "step": 38
    },
    {
      "completion_length": 26.9453125,
      "epoch": 0.008914285714285715,
      "grad_norm": 8.487604109803861,
      "kl": 0.1162109375,
      "learning_rate": 9.955428571428572e-07,
      "loss": 0.0046,
      "reward": 1.9609375,
      "reward_std": 0.09522314183413982,
      "rewards/accuracy_reward": 0.9765625,
      "rewards/format_reward": 0.984375,
      "step": 39
    },
    {
      "completion_length": 29.3828125,
      "epoch": 0.009142857142857144,
      "grad_norm": 2.0968680354159215,
      "kl": 0.10986328125,
      "learning_rate": 9.954285714285714e-07,
      "loss": 0.0044,
      "reward": 1.96875,
      "reward_std": 0.0731260534375906,
      "rewards/accuracy_reward": 0.9765625,
      "rewards/format_reward": 0.9921875,
      "step": 40
    },
    {
      "completion_length": 36.53125,
      "epoch": 0.009371428571428572,
      "grad_norm": 4.5085311018708305,
      "kl": 0.183837890625,
      "learning_rate": 9.953142857142856e-07,
      "loss": 0.0074,
      "reward": 1.9375,
      "reward_std": 0.1767766922712326,
      "rewards/accuracy_reward": 0.9609375,
      "rewards/format_reward": 0.9765625,
      "step": 41
    },
    {
      "completion_length": 39.3671875,
      "epoch": 0.0096,
      "grad_norm": 2.6069274036950754,
      "kl": 0.1064453125,
      "learning_rate": 9.952e-07,
      "loss": 0.0043,
      "reward": 1.9296875,
      "reward_std": 0.12169159017503262,
      "rewards/accuracy_reward": 0.9453125,
      "rewards/format_reward": 0.984375,
      "step": 42
    },
    {
      "completion_length": 41.4296875,
      "epoch": 0.009828571428571428,
      "grad_norm": 5.290738825829395,
      "kl": 0.10888671875,
      "learning_rate": 9.950857142857142e-07,
      "loss": 0.0043,
      "reward": 1.9140625,
      "reward_std": 0.21254335343837738,
      "rewards/accuracy_reward": 0.9296875,
      "rewards/format_reward": 0.984375,
      "step": 43
    },
    {
      "completion_length": 43.9375,
      "epoch": 0.010057142857142857,
      "grad_norm": 2.814643157199769,
      "kl": 0.1064453125,
      "learning_rate": 9.949714285714287e-07,
      "loss": 0.0043,
      "reward": 1.8984375,
      "reward_std": 0.14211230725049973,
      "rewards/accuracy_reward": 0.921875,
      "rewards/format_reward": 0.9765625,
      "step": 44
    },
    {
      "completion_length": 47.71875,
      "epoch": 0.010285714285714285,
      "grad_norm": 4.0575918220030704,
      "kl": 0.0743408203125,
      "learning_rate": 9.948571428571429e-07,
      "loss": 0.003,
      "reward": 1.9609375,
      "reward_std": 0.11048543080687523,
      "rewards/accuracy_reward": 0.9765625,
      "rewards/format_reward": 0.984375,
      "step": 45
    },
    {
      "completion_length": 52.921875,
      "epoch": 0.010514285714285714,
      "grad_norm": 2.9896163996026632,
      "kl": 0.0635986328125,
      "learning_rate": 9.94742857142857e-07,
      "loss": 0.0026,
      "reward": 1.921875,
      "reward_std": 0.22097086906433105,
      "rewards/accuracy_reward": 0.9609375,
      "rewards/format_reward": 0.9609375,
      "step": 46
    },
    {
      "completion_length": 49.09375,
      "epoch": 0.010742857142857143,
      "grad_norm": 4.580069062633702,
      "kl": 0.078857421875,
      "learning_rate": 9.946285714285713e-07,
      "loss": 0.0032,
      "reward": 1.96875,
      "reward_std": 0.07312605530023575,
      "rewards/accuracy_reward": 0.9765625,
      "rewards/format_reward": 0.9921875,
      "step": 47
    },
    {
      "completion_length": 45.3671875,
      "epoch": 0.010971428571428571,
      "grad_norm": 3.913659787689115,
      "kl": 0.083984375,
      "learning_rate": 9.945142857142857e-07,
      "loss": 0.0034,
      "reward": 1.9375,
      "reward_std": 0.12179600074887276,
      "rewards/accuracy_reward": 0.9375,
      "rewards/format_reward": 1.0,
      "step": 48
    },
    {
      "completion_length": 45.1171875,
      "epoch": 0.0112,
      "grad_norm": 9.965961895484126,
      "kl": 0.061767578125,
      "learning_rate": 9.944e-07,
      "loss": 0.0025,
      "reward": 1.859375,
      "reward_std": 0.19568345695734024,
      "rewards/accuracy_reward": 0.859375,
      "rewards/format_reward": 1.0,
      "step": 49
    },
    {
      "completion_length": 49.3671875,
      "epoch": 0.011428571428571429,
      "grad_norm": 1.4992015134527492,
      "kl": 0.06396484375,
      "learning_rate": 9.942857142857144e-07,
      "loss": 0.0026,
      "reward": 1.9609375,
      "reward_std": 0.0765409953892231,
      "rewards/accuracy_reward": 0.96875,
      "rewards/format_reward": 0.9921875,
      "step": 50
    },
    {
      "completion_length": 46.3359375,
      "epoch": 0.011657142857142857,
      "grad_norm": 1.981987333134919,
      "kl": 0.06787109375,
      "learning_rate": 9.941714285714286e-07,
      "loss": 0.0027,
      "reward": 1.9375,
      "reward_std": 0.07312605157494545,
      "rewards/accuracy_reward": 0.9453125,
      "rewards/format_reward": 0.9921875,
      "step": 51
    },
    {
      "completion_length": 49.15625,
      "epoch": 0.011885714285714286,
      "grad_norm": 2.4535899096268925,
      "kl": 0.0771484375,
      "learning_rate": 9.940571428571428e-07,
      "loss": 0.0031,
      "reward": 1.953125,
      "reward_std": 0.11732023023068905,
      "rewards/accuracy_reward": 0.9765625,
      "rewards/format_reward": 0.9765625,
      "step": 52
    },
    {
      "completion_length": 48.890625,
      "epoch": 0.012114285714285715,
      "grad_norm": 4.043109782609877,
      "kl": 0.071044921875,
      "learning_rate": 9.93942857142857e-07,
      "loss": 0.0028,
      "reward": 1.9375,
      "reward_std": 0.1767766885459423,
      "rewards/accuracy_reward": 0.953125,
      "rewards/format_reward": 0.984375,
      "step": 53
    },
    {
      "completion_length": 48.71875,
      "epoch": 0.012342857142857143,
      "grad_norm": 7.098653343228014,
      "kl": 0.0904541015625,
      "learning_rate": 9.938285714285713e-07,
      "loss": 0.0036,
      "reward": 1.953125,
      "reward_std": 0.1173202246427536,
      "rewards/accuracy_reward": 0.9765625,
      "rewards/format_reward": 0.9765625,
      "step": 54
    },
    {
      "completion_length": 47.2890625,
      "epoch": 0.012571428571428572,
      "grad_norm": 2.8731592604291927,
      "kl": 0.07080078125,
      "learning_rate": 9.937142857142857e-07,
      "loss": 0.0028,
      "reward": 1.9609375,
      "reward_std": 0.07654099725186825,
      "rewards/accuracy_reward": 0.9765625,
      "rewards/format_reward": 0.984375,
      "step": 55
    },
    {
      "completion_length": 48.6640625,
      "epoch": 0.0128,
      "grad_norm": 3.860584082839684,
      "kl": 0.1083984375,
      "learning_rate": 9.936e-07,
      "loss": 0.0043,
      "reward": 1.9453125,
      "reward_std": 0.15467960387468338,
      "rewards/accuracy_reward": 0.984375,
      "rewards/format_reward": 0.9609375,
      "step": 56
    },
    {
      "completion_length": 46.15625,
      "epoch": 0.01302857142857143,
      "grad_norm": 2.7032112024339936,
      "kl": 0.072021484375,
      "learning_rate": 9.934857142857143e-07,
      "loss": 0.0029,
      "reward": 1.953125,
      "reward_std": 0.09863808378577232,
      "rewards/accuracy_reward": 0.953125,
      "rewards/format_reward": 1.0,
      "step": 57
    },
    {
      "completion_length": 45.109375,
      "epoch": 0.013257142857142858,
      "grad_norm": 4.222613991207365,
      "kl": 0.080078125,
      "learning_rate": 9.933714285714285e-07,
      "loss": 0.0032,
      "reward": 1.8984375,
      "reward_std": 0.18361148983240128,
      "rewards/accuracy_reward": 0.921875,
      "rewards/format_reward": 0.9765625,
      "step": 58
    },
    {
      "completion_length": 45.8359375,
      "epoch": 0.013485714285714285,
      "grad_norm": 6.641195678009211,
      "kl": 0.093017578125,
      "learning_rate": 9.932571428571428e-07,
      "loss": 0.0037,
      "reward": 1.9609375,
      "reward_std": 0.09522313997149467,
      "rewards/accuracy_reward": 0.9765625,
      "rewards/format_reward": 0.984375,
      "step": 59
    },
    {
      "completion_length": 37.1171875,
      "epoch": 0.013714285714285714,
      "grad_norm": 4.401146336869325,
      "kl": 0.074462890625,
      "learning_rate": 9.93142857142857e-07,
      "loss": 0.003,
      "reward": 1.9140625,
      "reward_std": 0.06629125773906708,
      "rewards/accuracy_reward": 0.921875,
      "rewards/format_reward": 0.9921875,
      "step": 60
    },
    {
      "completion_length": 37.1953125,
      "epoch": 0.013942857142857142,
      "grad_norm": 4.618099832772619,
      "kl": 0.0694580078125,
      "learning_rate": 9.930285714285714e-07,
      "loss": 0.0028,
      "reward": 1.96875,
      "reward_std": 0.0883883461356163,
      "rewards/accuracy_reward": 0.96875,
      "rewards/format_reward": 1.0,
      "step": 61
    },
    {
      "completion_length": 39.765625,
      "epoch": 0.014171428571428571,
      "grad_norm": 3.918295900405684,
      "kl": 0.089599609375,
      "learning_rate": 9.929142857142856e-07,
      "loss": 0.0036,
      "reward": 1.9375,
      "reward_std": 0.0731260534375906,
      "rewards/accuracy_reward": 0.9375,
      "rewards/format_reward": 1.0,
      "step": 62
    },
    {
      "completion_length": 35.4765625,
      "epoch": 0.0144,
      "grad_norm": 3.5814825628476172,
      "kl": 0.10546875,
      "learning_rate": 9.928e-07,
      "loss": 0.0042,
      "reward": 1.96875,
      "reward_std": 0.05444390885531902,
      "rewards/accuracy_reward": 0.9765625,
      "rewards/format_reward": 0.9921875,
      "step": 63
    },
    {
      "completion_length": 49.8828125,
      "epoch": 0.014628571428571428,
      "grad_norm": 6.056662632723462,
      "kl": 0.25390625,
      "learning_rate": 9.926857142857143e-07,
      "loss": 0.0102,
      "reward": 1.96875,
      "reward_std": 0.0883883461356163,
      "rewards/accuracy_reward": 0.984375,
      "rewards/format_reward": 0.984375,
      "step": 64
    },
    {
      "completion_length": 42.5625,
      "epoch": 0.014857142857142857,
      "grad_norm": 1.5534296827884366,
      "kl": 0.07666015625,
      "learning_rate": 9.925714285714285e-07,
      "loss": 0.0031,
      "reward": 1.9765625,
      "reward_std": 0.05102896690368652,
      "rewards/accuracy_reward": 0.984375,
      "rewards/format_reward": 0.9921875,
      "step": 65
    },
    {
      "completion_length": 40.90625,
      "epoch": 0.015085714285714286,
      "grad_norm": 2.574114014533703,
      "kl": 0.0791015625,
      "learning_rate": 9.924571428571427e-07,
      "loss": 0.0032,
      "reward": 1.9140625,
      "reward_std": 0.06629125960171223,
      "rewards/accuracy_reward": 0.921875,
      "rewards/format_reward": 0.9921875,
      "step": 66
    },
    {
      "completion_length": 44.1640625,
      "epoch": 0.015314285714285714,
      "grad_norm": 1.9115486983004006,
      "kl": 0.088623046875,
      "learning_rate": 9.923428571428571e-07,
      "loss": 0.0036,
      "reward": 1.9765625,
      "reward_std": 0.06629125960171223,
      "rewards/accuracy_reward": 0.9921875,
      "rewards/format_reward": 0.984375,
      "step": 67
    },
    {
      "completion_length": 44.2578125,
      "epoch": 0.015542857142857143,
      "grad_norm": 2.030214286624162,
      "kl": 0.088623046875,
      "learning_rate": 9.922285714285713e-07,
      "loss": 0.0035,
      "reward": 1.9375,
      "reward_std": 0.11230766773223877,
      "rewards/accuracy_reward": 0.953125,
      "rewards/format_reward": 0.984375,
      "step": 68
    },
    {
      "completion_length": 47.9921875,
      "epoch": 0.01577142857142857,
      "grad_norm": 0.8186959964971635,
      "kl": 0.072509765625,
      "learning_rate": 9.921142857142858e-07,
      "loss": 0.0029,
      "reward": 1.9921875,
      "reward_std": 0.022097086533904076,
      "rewards/accuracy_reward": 1.0,
      "rewards/format_reward": 0.9921875,
      "step": 69
    },
    {
      "completion_length": 52.1328125,
      "epoch": 0.016,
      "grad_norm": 37.88639051831449,
      "kl": 0.0625,
      "learning_rate": 9.92e-07,
      "loss": 0.0025,
      "reward": 1.9140625,
      "reward_std": 0.1541428230702877,
      "rewards/accuracy_reward": 0.9296875,
      "rewards/format_reward": 0.984375,
      "step": 70
    },
    {
      "completion_length": 48.828125,
      "epoch": 0.01622857142857143,
      "grad_norm": 7.284082315391348,
      "kl": 0.10693359375,
      "learning_rate": 9.918857142857142e-07,
      "loss": 0.0043,
      "reward": 1.9296875,
      "reward_std": 0.1649293377995491,
      "rewards/accuracy_reward": 0.9375,
      "rewards/format_reward": 0.9921875,
      "step": 71
    },
    {
      "completion_length": 42.6640625,
      "epoch": 0.016457142857142858,
      "grad_norm": 8.385406536876486,
      "kl": 0.08544921875,
      "learning_rate": 9.917714285714284e-07,
      "loss": 0.0034,
      "reward": 1.96875,
      "reward_std": 0.0883883461356163,
      "rewards/accuracy_reward": 0.984375,
      "rewards/format_reward": 0.984375,
      "step": 72
    },
    {
      "completion_length": 49.1875,
      "epoch": 0.016685714285714286,
      "grad_norm": 1.5426801294464958,
      "kl": 0.057373046875,
      "learning_rate": 9.916571428571429e-07,
      "loss": 0.0023,
      "reward": 1.9765625,
      "reward_std": 0.06629125960171223,
      "rewards/accuracy_reward": 0.984375,
      "rewards/format_reward": 0.9921875,
      "step": 73
    },
    {
      "completion_length": 44.5078125,
      "epoch": 0.016914285714285715,
      "grad_norm": 9.678117437857079,
      "kl": 0.110107421875,
      "learning_rate": 9.91542857142857e-07,
      "loss": 0.0044,
      "reward": 1.9453125,
      "reward_std": 0.15467960201203823,
      "rewards/accuracy_reward": 0.96875,
      "rewards/format_reward": 0.9765625,
      "step": 74
    },
    {
      "completion_length": 54.71875,
      "epoch": 0.017142857142857144,
      "grad_norm": 3.959183231169419,
      "kl": 0.061279296875,
      "learning_rate": 9.914285714285715e-07,
      "loss": 0.0025,
      "reward": 1.9375,
      "reward_std": 0.1462521031498909,
      "rewards/accuracy_reward": 0.96875,
      "rewards/format_reward": 0.96875,
      "step": 75
    },
    {
      "completion_length": 51.0625,
      "epoch": 0.017371428571428572,
      "grad_norm": 1.7301352947580328,
      "kl": 0.0645751953125,
      "learning_rate": 9.913142857142857e-07,
      "loss": 0.0026,
      "reward": 1.9609375,
      "reward_std": 0.11048543266952038,
      "rewards/accuracy_reward": 0.984375,
      "rewards/format_reward": 0.9765625,
      "step": 76
    },
    {
      "completion_length": 49.7578125,
      "epoch": 0.0176,
      "grad_norm": 5.095778506090335,
      "kl": 0.1181640625,
      "learning_rate": 9.912e-07,
      "loss": 0.0047,
      "reward": 1.890625,
      "reward_std": 0.20069601386785507,
      "rewards/accuracy_reward": 0.921875,
      "rewards/format_reward": 0.96875,
      "step": 77
    },
    {
      "completion_length": 50.6171875,
      "epoch": 0.01782857142857143,
      "grad_norm": 7.086357422436246,
      "kl": 0.196044921875,
      "learning_rate": 9.910857142857141e-07,
      "loss": 0.0079,
      "reward": 1.953125,
      "reward_std": 0.1173202246427536,
      "rewards/accuracy_reward": 0.96875,
      "rewards/format_reward": 0.984375,
      "step": 78
    },
    {
      "completion_length": 50.0078125,
      "epoch": 0.01805714285714286,
      "grad_norm": 2.0574955692290704,
      "kl": 0.06787109375,
      "learning_rate": 9.909714285714286e-07,
      "loss": 0.0027,
      "reward": 1.9140625,
      "reward_std": 0.06629125960171223,
      "rewards/accuracy_reward": 0.9296875,
      "rewards/format_reward": 0.984375,
      "step": 79
    },
    {
      "completion_length": 45.3046875,
      "epoch": 0.018285714285714287,
      "grad_norm": 4.20999498130068,
      "kl": 0.086669921875,
      "learning_rate": 9.908571428571428e-07,
      "loss": 0.0035,
      "reward": 1.90625,
      "reward_std": 0.1309898104518652,
      "rewards/accuracy_reward": 0.921875,
      "rewards/format_reward": 0.984375,
      "step": 80
    },
    {
      "completion_length": 45.6328125,
      "epoch": 0.018514285714285716,
      "grad_norm": 6.228175833600017,
      "kl": 0.053466796875,
      "learning_rate": 9.90742857142857e-07,
      "loss": 0.0021,
      "reward": 1.9765625,
      "reward_std": 0.06629125960171223,
      "rewards/accuracy_reward": 0.9921875,
      "rewards/format_reward": 0.984375,
      "step": 81
    },
    {
      "completion_length": 37.2421875,
      "epoch": 0.018742857142857144,
      "grad_norm": 4.907370551522916,
      "kl": 0.14599609375,
      "learning_rate": 9.906285714285714e-07,
      "loss": 0.0058,
      "reward": 1.953125,
      "reward_std": 0.13258251547813416,
      "rewards/accuracy_reward": 0.984375,
      "rewards/format_reward": 0.96875,
      "step": 82
    },
    {
      "completion_length": 40.4296875,
      "epoch": 0.018971428571428573,
      "grad_norm": 2.287520018590471,
      "kl": 0.08935546875,
      "learning_rate": 9.905142857142857e-07,
      "loss": 0.0036,
      "reward": 1.9609375,
      "reward_std": 0.11048543080687523,
      "rewards/accuracy_reward": 0.9921875,
      "rewards/format_reward": 0.96875,
      "step": 83
    },
    {
      "completion_length": 41.421875,
      "epoch": 0.0192,
      "grad_norm": 1.2993082824617244,
      "kl": 0.0966796875,
      "learning_rate": 9.903999999999999e-07,
      "loss": 0.0039,
      "reward": 1.9765625,
      "reward_std": 0.06629125773906708,
      "rewards/accuracy_reward": 0.984375,
      "rewards/format_reward": 0.9921875,
      "step": 84
    },
    {
      "completion_length": 38.0,
      "epoch": 0.019428571428571427,
      "grad_norm": 2.185657680975154,
      "kl": 0.0751953125,
      "learning_rate": 9.902857142857143e-07,
      "loss": 0.003,
      "reward": 1.9765625,
      "reward_std": 0.051028965041041374,
      "rewards/accuracy_reward": 0.9765625,
      "rewards/format_reward": 1.0,
      "step": 85
    },
    {
      "completion_length": 39.7890625,
      "epoch": 0.019657142857142856,
      "grad_norm": 4.630024984937059,
      "kl": 0.16455078125,
      "learning_rate": 9.901714285714285e-07,
      "loss": 0.0066,
      "reward": 1.9765625,
      "reward_std": 0.06629125960171223,
      "rewards/accuracy_reward": 0.9765625,
      "rewards/format_reward": 1.0,
      "step": 86
    },
    {
      "completion_length": 42.9765625,
      "epoch": 0.019885714285714284,
      "grad_norm": 2.7685341517574553,
      "kl": 0.070556640625,
      "learning_rate": 9.900571428571427e-07,
      "loss": 0.0028,
      "reward": 1.984375,
      "reward_std": 0.04419417306780815,
      "rewards/accuracy_reward": 0.9921875,
      "rewards/format_reward": 0.9921875,
      "step": 87
    },
    {
      "completion_length": 42.984375,
      "epoch": 0.020114285714285713,
      "grad_norm": 1.7321605977483483,
      "kl": 0.07080078125,
      "learning_rate": 9.899428571428572e-07,
      "loss": 0.0028,
      "reward": 1.9921875,
      "reward_std": 0.022097086533904076,
      "rewards/accuracy_reward": 0.9921875,
      "rewards/format_reward": 1.0,
      "step": 88
    },
    {
      "completion_length": 46.9609375,
      "epoch": 0.02034285714285714,
      "grad_norm": 3.652542929106875,
      "kl": 0.05859375,
      "learning_rate": 9.898285714285714e-07,
      "loss": 0.0023,
      "reward": 1.8203125,
      "reward_std": 0.061278700828552246,
      "rewards/accuracy_reward": 0.8203125,
      "rewards/format_reward": 1.0,
      "step": 89
    },
    {
      "completion_length": 44.40625,
      "epoch": 0.02057142857142857,
      "grad_norm": 4.080955701693836,
      "kl": 0.075439453125,
      "learning_rate": 9.897142857142858e-07,
      "loss": 0.003,
      "reward": 1.9140625,
      "reward_std": 0.1054728776216507,
      "rewards/accuracy_reward": 0.9140625,
      "rewards/format_reward": 1.0,
      "step": 90
    },
    {
      "completion_length": 50.3359375,
      "epoch": 0.0208,
      "grad_norm": 4.076514993388891,
      "kl": 0.058349609375,
      "learning_rate": 9.896e-07,
      "loss": 0.0023,
      "reward": 1.9609375,
      "reward_std": 0.09522313624620438,
      "rewards/accuracy_reward": 0.96875,
      "rewards/format_reward": 0.9921875,
      "step": 91
    },
    {
      "completion_length": 53.1875,
      "epoch": 0.021028571428571428,
      "grad_norm": 2.2084704364472465,
      "kl": 0.0506591796875,
      "learning_rate": 9.894857142857142e-07,
      "loss": 0.002,
      "reward": 1.96875,
      "reward_std": 0.08838834427297115,
      "rewards/accuracy_reward": 0.9765625,
      "rewards/format_reward": 0.9921875,
      "step": 92
    },
    {
      "completion_length": 55.6640625,
      "epoch": 0.021257142857142856,
      "grad_norm": 1.0308928297613575,
      "kl": 0.0540771484375,
      "learning_rate": 9.893714285714285e-07,
      "loss": 0.0022,
      "reward": 1.9765625,
      "reward_std": 0.06629125960171223,
      "rewards/accuracy_reward": 0.984375,
      "rewards/format_reward": 0.9921875,
      "step": 93
    },
    {
      "completion_length": 56.9375,
      "epoch": 0.021485714285714285,
      "grad_norm": 2.727636329496369,
      "kl": 0.0565185546875,
      "learning_rate": 9.892571428571429e-07,
      "loss": 0.0023,
      "reward": 1.984375,
      "reward_std": 0.04419417306780815,
      "rewards/accuracy_reward": 0.9921875,
      "rewards/format_reward": 0.9921875,
      "step": 94
    },
    {
      "completion_length": 52.703125,
      "epoch": 0.021714285714285714,
      "grad_norm": 1.98405985935024,
      "kl": 0.0543212890625,
      "learning_rate": 9.89142857142857e-07,
      "loss": 0.0022,
      "reward": 1.96875,
      "reward_std": 0.0883883461356163,
      "rewards/accuracy_reward": 0.9765625,
      "rewards/format_reward": 0.9921875,
      "step": 95
    },
    {
      "completion_length": 50.7421875,
      "epoch": 0.021942857142857142,
      "grad_norm": 3.5118456105802536,
      "kl": 0.064453125,
      "learning_rate": 9.890285714285715e-07,
      "loss": 0.0026,
      "reward": 1.9375,
      "reward_std": 0.12179600074887276,
      "rewards/accuracy_reward": 0.953125,
      "rewards/format_reward": 0.984375,
      "step": 96
    },
    {
      "completion_length": 55.796875,
      "epoch": 0.02217142857142857,
      "grad_norm": 0.9492411481914552,
      "kl": 0.0418701171875,
      "learning_rate": 9.889142857142857e-07,
      "loss": 0.0017,
      "reward": 1.9921875,
      "reward_std": 0.022097086533904076,
      "rewards/accuracy_reward": 0.9921875,
      "rewards/format_reward": 1.0,
      "step": 97
    },
    {
      "completion_length": 52.2421875,
      "epoch": 0.0224,
      "grad_norm": 4.4767687974188295,
      "kl": 0.056884765625,
      "learning_rate": 9.888e-07,
      "loss": 0.0023,
      "reward": 1.9453125,
      "reward_std": 0.0657544769346714,
      "rewards/accuracy_reward": 0.953125,
      "rewards/format_reward": 0.9921875,
      "step": 98
    },
    {
      "completion_length": 57.59375,
      "epoch": 0.02262857142857143,
      "grad_norm": 8.44635910203149,
      "kl": 0.044677734375,
      "learning_rate": 9.886857142857142e-07,
      "loss": 0.0018,
      "reward": 1.9453125,
      "reward_std": 0.1054728701710701,
      "rewards/accuracy_reward": 0.953125,
      "rewards/format_reward": 0.9921875,
      "step": 99
    },
    {
      "completion_length": 54.3828125,
      "epoch": 0.022857142857142857,
      "grad_norm": 2.738295397895998,
      "kl": 0.04345703125,
      "learning_rate": 9.885714285714286e-07,
      "loss": 0.0017,
      "reward": 1.96875,
      "reward_std": 0.0883883461356163,
      "rewards/accuracy_reward": 0.984375,
      "rewards/format_reward": 0.984375,
      "step": 100
    },
    {
      "completion_length": 56.0703125,
      "epoch": 0.023085714285714286,
      "grad_norm": 19.889247062567375,
      "kl": 0.4180908203125,
      "learning_rate": 9.884571428571428e-07,
      "loss": 0.0167,
      "reward": 1.9453125,
      "reward_std": 0.13488983735442162,
      "rewards/accuracy_reward": 0.96875,
      "rewards/format_reward": 0.9765625,
      "step": 101
    },
    {
      "completion_length": 58.109375,
      "epoch": 0.023314285714285714,
      "grad_norm": 0.6015018526440337,
      "kl": 0.0450439453125,
      "learning_rate": 9.883428571428573e-07,
      "loss": 0.0018,
      "reward": 1.9921875,
      "reward_std": 0.022097086533904076,
      "rewards/accuracy_reward": 1.0,
      "rewards/format_reward": 0.9921875,
      "step": 102
    },
    {
      "completion_length": 53.5078125,
      "epoch": 0.023542857142857143,
      "grad_norm": 2.899178837752513,
      "kl": 0.098388671875,
      "learning_rate": 9.882285714285715e-07,
      "loss": 0.0039,
      "reward": 1.953125,
      "reward_std": 0.11279274895787239,
      "rewards/accuracy_reward": 0.9921875,
      "rewards/format_reward": 0.9609375,
      "step": 103
    },
    {
      "completion_length": 55.3125,
      "epoch": 0.023771428571428572,
      "grad_norm": 3.210337010301059,
      "kl": 0.0418701171875,
      "learning_rate": 9.881142857142857e-07,
      "loss": 0.0017,
      "reward": 1.9765625,
      "reward_std": 0.06629125960171223,
      "rewards/accuracy_reward": 0.9765625,
      "rewards/format_reward": 1.0,
      "step": 104
    },
    {
      "completion_length": 61.796875,
      "epoch": 0.024,
      "grad_norm": 1.9886765402697992,
      "kl": 0.087158203125,
      "learning_rate": 9.88e-07,
      "loss": 0.0035,
      "reward": 1.9609375,
      "reward_std": 0.09522313624620438,
      "rewards/accuracy_reward": 0.984375,
      "rewards/format_reward": 0.9765625,
      "step": 105
    },
    {
      "completion_length": 59.796875,
      "epoch": 0.02422857142857143,
      "grad_norm": 2.438071186691161,
      "kl": 0.048095703125,
      "learning_rate": 9.878857142857141e-07,
      "loss": 0.0019,
      "reward": 1.96875,
      "reward_std": 0.0731260534375906,
      "rewards/accuracy_reward": 0.9921875,
      "rewards/format_reward": 0.9765625,
      "step": 106
    },
    {
      "completion_length": 62.125,
      "epoch": 0.024457142857142858,
      "grad_norm": 1.4774862723677509,
      "kl": 0.0423583984375,
      "learning_rate": 9.877714285714285e-07,
      "loss": 0.0017,
      "reward": 1.9140625,
      "reward_std": 0.06629125773906708,
      "rewards/accuracy_reward": 0.921875,
      "rewards/format_reward": 0.9921875,
      "step": 107
    },
    {
      "completion_length": 67.71875,
      "epoch": 0.024685714285714287,
      "grad_norm": 2.8540539552056017,
      "kl": 0.052734375,
      "learning_rate": 9.876571428571428e-07,
      "loss": 0.0021,
      "reward": 1.9140625,
      "reward_std": 0.20912351459264755,
      "rewards/accuracy_reward": 0.9609375,
      "rewards/format_reward": 0.953125,
      "step": 108
    },
    {
      "completion_length": 64.6875,
      "epoch": 0.024914285714285715,
      "grad_norm": 41.35522637112842,
      "kl": 0.7977294921875,
      "learning_rate": 9.875428571428572e-07,
      "loss": 0.032,
      "reward": 1.8828125,
      "reward_std": 0.2766144573688507,
      "rewards/accuracy_reward": 0.9453125,
      "rewards/format_reward": 0.9375,
      "step": 109
    },
    {
      "completion_length": 59.8046875,
      "epoch": 0.025142857142857144,
      "grad_norm": 5.061271286444709,
      "kl": 0.1610107421875,
      "learning_rate": 9.874285714285714e-07,
      "loss": 0.0065,
      "reward": 1.9765625,
      "reward_std": 0.06629125773906708,
      "rewards/accuracy_reward": 0.984375,
      "rewards/format_reward": 0.9921875,
      "step": 110
    },
    {
      "completion_length": 63.03125,
      "epoch": 0.025371428571428573,
      "grad_norm": 3.10509300208902,
      "kl": 0.0562744140625,
      "learning_rate": 9.873142857142856e-07,
      "loss": 0.0022,
      "reward": 1.9453125,
      "reward_std": 0.15467960387468338,
      "rewards/accuracy_reward": 0.96875,
      "rewards/format_reward": 0.9765625,
      "step": 111
    },
    {
      "completion_length": 69.9296875,
      "epoch": 0.0256,
      "grad_norm": 2.204259487144025,
      "kl": 0.0458984375,
      "learning_rate": 9.871999999999998e-07,
      "loss": 0.0018,
      "reward": 1.953125,
      "reward_std": 0.13258251547813416,
      "rewards/accuracy_reward": 0.9765625,
      "rewards/format_reward": 0.9765625,
      "step": 112
    },
    {
      "completion_length": 72.1484375,
      "epoch": 0.02582857142857143,
      "grad_norm": 1.7406307467407076,
      "kl": 0.0572509765625,
      "learning_rate": 9.870857142857143e-07,
      "loss": 0.0023,
      "reward": 1.9453125,
      "reward_std": 0.11353681981563568,
      "rewards/accuracy_reward": 0.984375,
      "rewards/format_reward": 0.9609375,
      "step": 113
    },
    {
      "completion_length": 72.46875,
      "epoch": 0.02605714285714286,
      "grad_norm": 2.8201327045980826,
      "kl": 0.0682373046875,
      "learning_rate": 9.869714285714285e-07,
      "loss": 0.0027,
      "reward": 1.875,
      "reward_std": 0.21485067903995514,
      "rewards/accuracy_reward": 0.921875,
      "rewards/format_reward": 0.953125,
      "step": 114
    },
    {
      "completion_length": 67.9921875,
      "epoch": 0.026285714285714287,
      "grad_norm": 10.499453046694777,
      "kl": 0.052490234375,
      "learning_rate": 9.86857142857143e-07,
      "loss": 0.0021,
      "reward": 1.921875,
      "reward_std": 0.18702643364667892,
      "rewards/accuracy_reward": 0.9765625,
      "rewards/format_reward": 0.9453125,
      "step": 115
    },
    {
      "completion_length": 72.7265625,
      "epoch": 0.026514285714285716,
      "grad_norm": 5.822450748289075,
      "kl": 0.07470703125,
      "learning_rate": 9.867428571428571e-07,
      "loss": 0.003,
      "reward": 1.8515625,
      "reward_std": 0.17457806318998337,
      "rewards/accuracy_reward": 0.90625,
      "rewards/format_reward": 0.9453125,
      "step": 116
    },
    {
      "completion_length": 76.2109375,
      "epoch": 0.02674285714285714,
      "grad_norm": 1.6423673269385712,
      "kl": 0.0455322265625,
      "learning_rate": 9.866285714285713e-07,
      "loss": 0.0018,
      "reward": 1.8984375,
      "reward_std": 0.11962753906846046,
      "rewards/accuracy_reward": 0.921875,
      "rewards/format_reward": 0.9765625,
      "step": 117
    },
    {
      "completion_length": 70.34375,
      "epoch": 0.02697142857142857,
      "grad_norm": 2.583556041531364,
      "kl": 0.0413818359375,
      "learning_rate": 9.865142857142856e-07,
      "loss": 0.0017,
      "reward": 1.9453125,
      "reward_std": 0.08443661965429783,
      "rewards/accuracy_reward": 0.9609375,
      "rewards/format_reward": 0.984375,
      "step": 118
    },
    {
      "completion_length": 70.5,
      "epoch": 0.0272,
      "grad_norm": 3.218976427901654,
      "kl": 0.048828125,
      "learning_rate": 9.864e-07,
      "loss": 0.002,
      "reward": 1.96875,
      "reward_std": 0.0883883461356163,
      "rewards/accuracy_reward": 0.96875,
      "rewards/format_reward": 1.0,
      "step": 119
    },
    {
      "completion_length": 76.4609375,
      "epoch": 0.027428571428571427,
      "grad_norm": 18.538087066810288,
      "kl": 0.0823974609375,
      "learning_rate": 9.862857142857142e-07,
      "loss": 0.0033,
      "reward": 1.9609375,
      "reward_std": 0.11048543266952038,
      "rewards/accuracy_reward": 0.984375,
      "rewards/format_reward": 0.9765625,
      "step": 120
    },
    {
      "completion_length": 75.5078125,
      "epoch": 0.027657142857142856,
      "grad_norm": 1.4503224642217518,
      "kl": 0.03277587890625,
      "learning_rate": 9.861714285714286e-07,
      "loss": 0.0013,
      "reward": 1.953125,
      "reward_std": 0.1325825210660696,
      "rewards/accuracy_reward": 0.9765625,
      "rewards/format_reward": 0.9765625,
      "step": 121
    },
    {
      "completion_length": 66.5,
      "epoch": 0.027885714285714285,
      "grad_norm": 1.7736676916817882,
      "kl": 0.0667724609375,
      "learning_rate": 9.860571428571429e-07,
      "loss": 0.0027,
      "reward": 1.9609375,
      "reward_std": 0.05550473928451538,
      "rewards/accuracy_reward": 0.96875,
      "rewards/format_reward": 0.9921875,
      "step": 122
    },
    {
      "completion_length": 67.2265625,
      "epoch": 0.028114285714285713,
      "grad_norm": 3.7493103117551034,
      "kl": 0.1533203125,
      "learning_rate": 9.85942857142857e-07,
      "loss": 0.0061,
      "reward": 1.953125,
      "reward_std": 0.13258251547813416,
      "rewards/accuracy_reward": 0.984375,
      "rewards/format_reward": 0.96875,
      "step": 123
    },
    {
      "completion_length": 68.546875,
      "epoch": 0.028342857142857142,
      "grad_norm": 3.7867393511529692,
      "kl": 0.0511474609375,
      "learning_rate": 9.858285714285713e-07,
      "loss": 0.002,
      "reward": 1.953125,
      "reward_std": 0.10205793380737305,
      "rewards/accuracy_reward": 0.96875,
      "rewards/format_reward": 0.984375,
      "step": 124
    },
    {
      "completion_length": 62.546875,
      "epoch": 0.02857142857142857,
      "grad_norm": 4.434035147257405,
      "kl": 0.049560546875,
      "learning_rate": 9.857142857142857e-07,
      "loss": 0.002,
      "reward": 1.96875,
      "reward_std": 0.05444390885531902,
      "rewards/accuracy_reward": 0.96875,
      "rewards/format_reward": 1.0,
      "step": 125
    },
    {
      "completion_length": 64.6015625,
      "epoch": 0.0288,
      "grad_norm": 3.432165144734931,
      "kl": 0.08203125,
      "learning_rate": 9.856e-07,
      "loss": 0.0033,
      "reward": 1.9609375,
      "reward_std": 0.11048543266952038,
      "rewards/accuracy_reward": 0.9921875,
      "rewards/format_reward": 0.96875,
      "step": 126
    },
    {
      "completion_length": 66.9921875,
      "epoch": 0.029028571428571428,
      "grad_norm": 8.759490699043806,
      "kl": 0.305419921875,
      "learning_rate": 9.854857142857141e-07,
      "loss": 0.0122,
      "reward": 1.96875,
      "reward_std": 0.0883883461356163,
      "rewards/accuracy_reward": 0.984375,
      "rewards/format_reward": 0.984375,
      "step": 127
    },
    {
      "completion_length": 64.4609375,
      "epoch": 0.029257142857142857,
      "grad_norm": 1.4750338857861727,
      "kl": 0.044921875,
      "learning_rate": 9.853714285714286e-07,
      "loss": 0.0018,
      "reward": 1.90625,
      "reward_std": 0.06859857589006424,
      "rewards/accuracy_reward": 0.921875,
      "rewards/format_reward": 0.984375,
      "step": 128
    },
    {
      "completion_length": 67.3046875,
      "epoch": 0.029485714285714285,
      "grad_norm": 1.6563256869075114,
      "kl": 0.0606689453125,
      "learning_rate": 9.852571428571428e-07,
      "loss": 0.0024,
      "reward": 1.8828125,
      "reward_std": 0.15467960387468338,
      "rewards/accuracy_reward": 0.921875,
      "rewards/format_reward": 0.9609375,
      "step": 129
    },
    {
      "completion_length": 61.8203125,
      "epoch": 0.029714285714285714,
      "grad_norm": 1.9126809051119988,
      "kl": 0.07568359375,
      "learning_rate": 9.85142857142857e-07,
      "loss": 0.003,
      "reward": 1.96875,
      "reward_std": 0.08838834427297115,
      "rewards/accuracy_reward": 1.0,
      "rewards/format_reward": 0.96875,
      "step": 130
    },
    {
      "completion_length": 64.4921875,
      "epoch": 0.029942857142857143,
      "grad_norm": 5.229409294998544,
      "kl": 0.05419921875,
      "learning_rate": 9.850285714285714e-07,
      "loss": 0.0022,
      "reward": 1.9609375,
      "reward_std": 0.09522313997149467,
      "rewards/accuracy_reward": 0.9765625,
      "rewards/format_reward": 0.984375,
      "step": 131
    },
    {
      "completion_length": 69.71875,
      "epoch": 0.03017142857142857,
      "grad_norm": 2.251583112781417,
      "kl": 0.049072265625,
      "learning_rate": 9.849142857142857e-07,
      "loss": 0.002,
      "reward": 1.890625,
      "reward_std": 0.1173202209174633,
      "rewards/accuracy_reward": 0.921875,
      "rewards/format_reward": 0.96875,
      "step": 132
    },
    {
      "completion_length": 59.859375,
      "epoch": 0.0304,
      "grad_norm": 4.611873710615758,
      "kl": 0.08935546875,
      "learning_rate": 9.847999999999999e-07,
      "loss": 0.0036,
      "reward": 1.953125,
      "reward_std": 0.13258251547813416,
      "rewards/accuracy_reward": 0.984375,
      "rewards/format_reward": 0.96875,
      "step": 133
    },
    {
      "completion_length": 60.53125,
      "epoch": 0.03062857142857143,
      "grad_norm": 1.3999158725235017,
      "kl": 0.063720703125,
      "learning_rate": 9.846857142857143e-07,
      "loss": 0.0026,
      "reward": 1.9765625,
      "reward_std": 0.06629125960171223,
      "rewards/accuracy_reward": 0.9921875,
      "rewards/format_reward": 0.984375,
      "step": 134
    },
    {
      "completion_length": 67.9765625,
      "epoch": 0.030857142857142857,
      "grad_norm": 2.9816800608489764,
      "kl": 0.0570068359375,
      "learning_rate": 9.845714285714285e-07,
      "loss": 0.0023,
      "reward": 1.9453125,
      "reward_std": 0.15467960387468338,
      "rewards/accuracy_reward": 0.984375,
      "rewards/format_reward": 0.9609375,
      "step": 135
    },
    {
      "completion_length": 57.5703125,
      "epoch": 0.031085714285714286,
      "grad_norm": 1.8745969838575054,
      "kl": 0.0557861328125,
      "learning_rate": 9.844571428571427e-07,
      "loss": 0.0022,
      "reward": 1.984375,
      "reward_std": 0.04419417306780815,
      "rewards/accuracy_reward": 0.9921875,
      "rewards/format_reward": 0.9921875,
      "step": 136
    },
    {
      "completion_length": 64.4921875,
      "epoch": 0.03131428571428571,
      "grad_norm": 2.806402992422446,
      "kl": 0.068115234375,
      "learning_rate": 9.843428571428572e-07,
      "loss": 0.0027,
      "reward": 1.953125,
      "reward_std": 0.10205793380737305,
      "rewards/accuracy_reward": 0.9921875,
      "rewards/format_reward": 0.9609375,
      "step": 137
    },
    {
      "completion_length": 68.796875,
      "epoch": 0.03154285714285714,
      "grad_norm": 1.9958622497789609,
      "kl": 0.0560302734375,
      "learning_rate": 9.842285714285714e-07,
      "loss": 0.0022,
      "reward": 1.9609375,
      "reward_std": 0.11048543080687523,
      "rewards/accuracy_reward": 0.9921875,
      "rewards/format_reward": 0.96875,
      "step": 138
    },
    {
      "completion_length": 66.2109375,
      "epoch": 0.03177142857142857,
      "grad_norm": 3.0601097234576216,
      "kl": 0.0869140625,
      "learning_rate": 9.841142857142856e-07,
      "loss": 0.0035,
      "reward": 1.9296875,
      "reward_std": 0.19887378066778183,
      "rewards/accuracy_reward": 0.9765625,
      "rewards/format_reward": 0.953125,
      "step": 139
    },
    {
      "completion_length": 66.3984375,
      "epoch": 0.032,
      "grad_norm": 1.465517819831982,
      "kl": 0.0537109375,
      "learning_rate": 9.84e-07,
      "loss": 0.0021,
      "reward": 1.96875,
      "reward_std": 0.07312604784965515,
      "rewards/accuracy_reward": 0.9921875,
      "rewards/format_reward": 0.9765625,
      "step": 140
    },
    {
      "completion_length": 67.2109375,
      "epoch": 0.032228571428571426,
      "grad_norm": 2.3453447260795888,
      "kl": 0.06396484375,
      "learning_rate": 9.838857142857142e-07,
      "loss": 0.0026,
      "reward": 1.9296875,
      "reward_std": 0.12073516845703125,
      "rewards/accuracy_reward": 0.953125,
      "rewards/format_reward": 0.9765625,
      "step": 141
    },
    {
      "completion_length": 65.2265625,
      "epoch": 0.03245714285714286,
      "grad_norm": 4.532132473033689,
      "kl": 0.0845947265625,
      "learning_rate": 9.837714285714285e-07,
      "loss": 0.0034,
      "reward": 1.9375,
      "reward_std": 0.12756996229290962,
      "rewards/accuracy_reward": 0.96875,
      "rewards/format_reward": 0.96875,
      "step": 142
    },
    {
      "completion_length": 64.5078125,
      "epoch": 0.03268571428571428,
      "grad_norm": 3.4518249250566075,
      "kl": 0.05224609375,
      "learning_rate": 9.836571428571429e-07,
      "loss": 0.0021,
      "reward": 1.9375,
      "reward_std": 0.08337578736245632,
      "rewards/accuracy_reward": 0.953125,
      "rewards/format_reward": 0.984375,
      "step": 143
    },
    {
      "completion_length": 62.2265625,
      "epoch": 0.032914285714285715,
      "grad_norm": 0.7846089598394096,
      "kl": 0.0675048828125,
      "learning_rate": 9.83542857142857e-07,
      "loss": 0.0027,
      "reward": 1.9765625,
      "reward_std": 0.06629125773906708,
      "rewards/accuracy_reward": 0.9921875,
      "rewards/format_reward": 0.984375,
      "step": 144
    },
    {
      "completion_length": 61.7421875,
      "epoch": 0.03314285714285714,
      "grad_norm": 2.87124391463782,
      "kl": 0.067626953125,
      "learning_rate": 9.834285714285713e-07,
      "loss": 0.0027,
      "reward": 1.9921875,
      "reward_std": 0.022097086533904076,
      "rewards/accuracy_reward": 0.9921875,
      "rewards/format_reward": 1.0,
      "step": 145
    },
    {
      "completion_length": 62.25,
      "epoch": 0.03337142857142857,
      "grad_norm": 5.596334282187468,
      "kl": 0.0621337890625,
      "learning_rate": 9.833142857142857e-07,
      "loss": 0.0025,
      "reward": 1.8984375,
      "reward_std": 0.13488983735442162,
      "rewards/accuracy_reward": 0.921875,
      "rewards/format_reward": 0.9765625,
      "step": 146
    },
    {
      "completion_length": 62.3359375,
      "epoch": 0.0336,
      "grad_norm": 1.6541721983374376,
      "kl": 0.05712890625,
      "learning_rate": 9.832e-07,
      "loss": 0.0023,
      "reward": 1.96875,
      "reward_std": 0.0883883461356163,
      "rewards/accuracy_reward": 0.9765625,
      "rewards/format_reward": 0.9921875,
      "step": 147
    },
    {
      "completion_length": 61.546875,
      "epoch": 0.03382857142857143,
      "grad_norm": 2.334708796051361,
      "kl": 0.048095703125,
      "learning_rate": 9.830857142857144e-07,
      "loss": 0.0019,
      "reward": 1.9765625,
      "reward_std": 0.06629125960171223,
      "rewards/accuracy_reward": 0.9921875,
      "rewards/format_reward": 0.984375,
      "step": 148
    },
    {
      "completion_length": 58.1484375,
      "epoch": 0.034057142857142855,
      "grad_norm": 2.7209211148765946,
      "kl": 0.0694580078125,
      "learning_rate": 9.829714285714286e-07,
      "loss": 0.0028,
      "reward": 1.9609375,
      "reward_std": 0.07654099725186825,
      "rewards/accuracy_reward": 0.9609375,
      "rewards/format_reward": 1.0,
      "step": 149
    },
    {
      "completion_length": 59.96875,
      "epoch": 0.03428571428571429,
      "grad_norm": 0.17950070397896042,
      "kl": 0.0494384765625,
      "learning_rate": 9.828571428571428e-07,
      "loss": 0.002,
      "reward": 2.0,
      "reward_std": 0.0,
      "rewards/accuracy_reward": 1.0,
      "rewards/format_reward": 1.0,
      "step": 150
    },
    {
      "completion_length": 50.71875,
      "epoch": 0.03451428571428571,
      "grad_norm": 8.591213011576272,
      "kl": 0.093994140625,
      "learning_rate": 9.82742857142857e-07,
      "loss": 0.0038,
      "reward": 1.9375,
      "reward_std": 0.12179600074887276,
      "rewards/accuracy_reward": 0.953125,
      "rewards/format_reward": 0.984375,
      "step": 151
    },
    {
      "completion_length": 54.125,
      "epoch": 0.034742857142857145,
      "grad_norm": 1.6912179368628009,
      "kl": 0.096435546875,
      "learning_rate": 9.826285714285713e-07,
      "loss": 0.0039,
      "reward": 1.984375,
      "reward_std": 0.04419417306780815,
      "rewards/accuracy_reward": 0.9921875,
      "rewards/format_reward": 0.9921875,
      "step": 152
    },
    {
      "completion_length": 49.9375,
      "epoch": 0.03497142857142857,
      "grad_norm": 1.4857684912304752,
      "kl": 0.064208984375,
      "learning_rate": 9.825142857142857e-07,
      "loss": 0.0026,
      "reward": 1.984375,
      "reward_std": 0.04419417306780815,
      "rewards/accuracy_reward": 0.9921875,
      "rewards/format_reward": 0.9921875,
      "step": 153
    },
    {
      "completion_length": 54.359375,
      "epoch": 0.0352,
      "grad_norm": 1.4778745148631223,
      "kl": 0.07861328125,
      "learning_rate": 9.824e-07,
      "loss": 0.0032,
      "reward": 1.96875,
      "reward_std": 0.0883883461356163,
      "rewards/accuracy_reward": 0.9921875,
      "rewards/format_reward": 0.9765625,
      "step": 154
    },
    {
      "completion_length": 49.40625,
      "epoch": 0.03542857142857143,
      "grad_norm": 1.0280017752226174,
      "kl": 0.06298828125,
      "learning_rate": 9.822857142857143e-07,
      "loss": 0.0025,
      "reward": 1.9921875,
      "reward_std": 0.022097086533904076,
      "rewards/accuracy_reward": 1.0,
      "rewards/format_reward": 0.9921875,
      "step": 155
    },
    {
      "completion_length": 52.5703125,
      "epoch": 0.03565714285714286,
      "grad_norm": 0.5700446884979251,
      "kl": 0.0560302734375,
      "learning_rate": 9.821714285714285e-07,
      "loss": 0.0022,
      "reward": 1.984375,
      "reward_std": 0.04419417306780815,
      "rewards/accuracy_reward": 0.9921875,
      "rewards/format_reward": 0.9921875,
      "step": 156
    },
    {
      "completion_length": 54.421875,
      "epoch": 0.035885714285714285,
      "grad_norm": 1.2039727433634695,
      "kl": 0.0555419921875,
      "learning_rate": 9.820571428571428e-07,
      "loss": 0.0022,
      "reward": 1.9140625,
      "reward_std": 0.06629125960171223,
      "rewards/accuracy_reward": 0.921875,
      "rewards/format_reward": 0.9921875,
      "step": 157
    },
    {
      "completion_length": 56.953125,
      "epoch": 0.03611428571428572,
      "grad_norm": 2.486542472756636,
      "kl": 0.1021728515625,
      "learning_rate": 9.81942857142857e-07,
      "loss": 0.0041,
      "reward": 1.890625,
      "reward_std": 0.13258251547813416,
      "rewards/accuracy_reward": 0.9296875,
      "rewards/format_reward": 0.9609375,
      "step": 158
    },
    {
      "completion_length": 53.1875,
      "epoch": 0.03634285714285714,
      "grad_norm": 0.36756014846848356,
      "kl": 0.059326171875,
      "learning_rate": 9.818285714285714e-07,
      "loss": 0.0024,
      "reward": 2.0,
      "reward_std": 0.0,
      "rewards/accuracy_reward": 1.0,
      "rewards/format_reward": 1.0,
      "step": 159
    },
    {
      "completion_length": 49.625,
      "epoch": 0.036571428571428574,
      "grad_norm": 1.0982616965504457,
      "kl": 0.057861328125,
      "learning_rate": 9.817142857142856e-07,
      "loss": 0.0023,
      "reward": 1.8203125,
      "reward_std": 0.051028965041041374,
      "rewards/accuracy_reward": 0.828125,
      "rewards/format_reward": 0.9921875,
      "step": 160
    },
    {
      "completion_length": 48.265625,
      "epoch": 0.0368,
      "grad_norm": 3.8440532897426567,
      "kl": 0.067626953125,
      "learning_rate": 9.816e-07,
      "loss": 0.0027,
      "reward": 1.984375,
      "reward_std": 0.04419417306780815,
      "rewards/accuracy_reward": 0.9921875,
      "rewards/format_reward": 0.9921875,
      "step": 161
    },
    {
      "completion_length": 57.515625,
      "epoch": 0.03702857142857143,
      "grad_norm": 2.3727442099698814,
      "kl": 0.066650390625,
      "learning_rate": 9.814857142857143e-07,
      "loss": 0.0027,
      "reward": 1.84375,
      "reward_std": 0.1933293491601944,
      "rewards/accuracy_reward": 0.8828125,
      "rewards/format_reward": 0.9609375,
      "step": 162
    },
    {
      "completion_length": 53.7578125,
      "epoch": 0.03725714285714286,
      "grad_norm": 3.919835189657342,
      "kl": 0.067626953125,
      "learning_rate": 9.813714285714285e-07,
      "loss": 0.0027,
      "reward": 1.96875,
      "reward_std": 0.0731260534375906,
      "rewards/accuracy_reward": 0.9765625,
      "rewards/format_reward": 0.9921875,
      "step": 163
    },
    {
      "completion_length": 51.796875,
      "epoch": 0.03748571428571429,
      "grad_norm": 1.8092711437883862,
      "kl": 0.095947265625,
      "learning_rate": 9.812571428571427e-07,
      "loss": 0.0038,
      "reward": 1.9453125,
      "reward_std": 0.15467960759997368,
      "rewards/accuracy_reward": 0.984375,
      "rewards/format_reward": 0.9609375,
      "step": 164
    },
    {
      "completion_length": 50.453125,
      "epoch": 0.037714285714285714,
      "grad_norm": 1.4266781340843808,
      "kl": 0.120849609375,
      "learning_rate": 9.811428571428571e-07,
      "loss": 0.0048,
      "reward": 1.921875,
      "reward_std": 0.04419417306780815,
      "rewards/accuracy_reward": 0.9375,
      "rewards/format_reward": 0.984375,
      "step": 165
    },
    {
      "completion_length": 56.1015625,
      "epoch": 0.037942857142857146,
      "grad_norm": 1.7232141062911286,
      "kl": 0.090087890625,
      "learning_rate": 9.810285714285713e-07,
      "loss": 0.0036,
      "reward": 1.953125,
      "reward_std": 0.13258251920342445,
      "rewards/accuracy_reward": 0.984375,
      "rewards/format_reward": 0.96875,
      "step": 166
    },
    {
      "completion_length": 57.0234375,
      "epoch": 0.03817142857142857,
      "grad_norm": 2.67041493625045,
      "kl": 0.068603515625,
      "learning_rate": 9.809142857142858e-07,
      "loss": 0.0027,
      "reward": 1.9609375,
      "reward_std": 0.09522313438355923,
      "rewards/accuracy_reward": 0.9765625,
      "rewards/format_reward": 0.984375,
      "step": 167
    },
    {
      "completion_length": 49.03125,
      "epoch": 0.0384,
      "grad_norm": 2.955127668589717,
      "kl": 0.089111328125,
      "learning_rate": 9.808e-07,
      "loss": 0.0036,
      "reward": 1.9375,
      "reward_std": 0.14283225685358047,
      "rewards/accuracy_reward": 0.953125,
      "rewards/format_reward": 0.984375,
      "step": 168
    },
    {
      "completion_length": 57.09375,
      "epoch": 0.03862857142857143,
      "grad_norm": 2.3263914055363983,
      "kl": 0.091796875,
      "learning_rate": 9.806857142857142e-07,
      "loss": 0.0037,
      "reward": 1.9453125,
      "reward_std": 0.11962753534317017,
      "rewards/accuracy_reward": 0.9609375,
      "rewards/format_reward": 0.984375,
      "step": 169
    },
    {
      "completion_length": 48.7578125,
      "epoch": 0.038857142857142854,
      "grad_norm": 1.699697368726255,
      "kl": 0.10693359375,
      "learning_rate": 9.805714285714284e-07,
      "loss": 0.0043,
      "reward": 1.9609375,
      "reward_std": 0.11048543266952038,
      "rewards/accuracy_reward": 0.9921875,
      "rewards/format_reward": 0.96875,
      "step": 170
    },
    {
      "completion_length": 53.4765625,
      "epoch": 0.039085714285714286,
      "grad_norm": 7.06754718664756,
      "kl": 0.093017578125,
      "learning_rate": 9.804571428571429e-07,
      "loss": 0.0037,
      "reward": 1.9609375,
      "reward_std": 0.09522313438355923,
      "rewards/accuracy_reward": 0.984375,
      "rewards/format_reward": 0.9765625,
      "step": 171
    },
    {
      "completion_length": 45.015625,
      "epoch": 0.03931428571428571,
      "grad_norm": 5.026629712169161,
      "kl": 0.184814453125,
      "learning_rate": 9.80342857142857e-07,
      "loss": 0.0074,
      "reward": 1.9140625,
      "reward_std": 0.11048543453216553,
      "rewards/accuracy_reward": 0.9375,
      "rewards/format_reward": 0.9765625,
      "step": 172
    },
    {
      "completion_length": 41.90625,
      "epoch": 0.039542857142857143,
      "grad_norm": 1.4718412493336714,
      "kl": 0.108642578125,
      "learning_rate": 9.802285714285715e-07,
      "loss": 0.0043,
      "reward": 1.9296875,
      "reward_std": 0.16834918782114983,
      "rewards/accuracy_reward": 0.9765625,
      "rewards/format_reward": 0.953125,
      "step": 173
    },
    {
      "completion_length": 40.9453125,
      "epoch": 0.03977142857142857,
      "grad_norm": 1.6092594821902402,
      "kl": 0.140380859375,
      "learning_rate": 9.801142857142857e-07,
      "loss": 0.0056,
      "reward": 1.9609375,
      "reward_std": 0.09522314369678497,
      "rewards/accuracy_reward": 0.984375,
      "rewards/format_reward": 0.9765625,
      "step": 174
    },
    {
      "completion_length": 37.0546875,
      "epoch": 0.04,
      "grad_norm": 0.8453774068177028,
      "kl": 0.09375,
      "learning_rate": 9.8e-07,
      "loss": 0.0038,
      "reward": 1.9921875,
      "reward_std": 0.022097086533904076,
      "rewards/accuracy_reward": 1.0,
      "rewards/format_reward": 0.9921875,
      "step": 175
    },
    {
      "completion_length": 38.0546875,
      "epoch": 0.040228571428571426,
      "grad_norm": 2.718352991552224,
      "kl": 0.103759765625,
      "learning_rate": 9.798857142857142e-07,
      "loss": 0.0042,
      "reward": 1.9375,
      "reward_std": 0.0731260534375906,
      "rewards/accuracy_reward": 0.9453125,
      "rewards/format_reward": 0.9921875,
      "step": 176
    },
    {
      "completion_length": 38.046875,
      "epoch": 0.04045714285714286,
      "grad_norm": 3.7699079150042327,
      "kl": 0.10009765625,
      "learning_rate": 9.797714285714286e-07,
      "loss": 0.004,
      "reward": 1.96875,
      "reward_std": 0.05444391071796417,
      "rewards/accuracy_reward": 0.9765625,
      "rewards/format_reward": 0.9921875,
      "step": 177
    },
    {
      "completion_length": 41.140625,
      "epoch": 0.04068571428571428,
      "grad_norm": 0.9310201790289476,
      "kl": 0.092529296875,
      "learning_rate": 9.796571428571428e-07,
      "loss": 0.0037,
      "reward": 1.9921875,
      "reward_std": 0.022097086533904076,
      "rewards/accuracy_reward": 1.0,
      "rewards/format_reward": 0.9921875,
      "step": 178
    },
    {
      "completion_length": 36.625,
      "epoch": 0.040914285714285716,
      "grad_norm": 1.124282484763535,
      "kl": 0.077392578125,
      "learning_rate": 9.79542857142857e-07,
      "loss": 0.0031,
      "reward": 1.984375,
      "reward_std": 0.04419417306780815,
      "rewards/accuracy_reward": 0.9921875,
      "rewards/format_reward": 0.9921875,
      "step": 179
    },
    {
      "completion_length": 45.90625,
      "epoch": 0.04114285714285714,
      "grad_norm": 4.86588469905471,
      "kl": 0.078369140625,
      "learning_rate": 9.794285714285714e-07,
      "loss": 0.0031,
      "reward": 1.9609375,
      "reward_std": 0.061278700828552246,
      "rewards/accuracy_reward": 0.9609375,
      "rewards/format_reward": 1.0,
      "step": 180
    },
    {
      "completion_length": 52.1640625,
      "epoch": 0.04137142857142857,
      "grad_norm": 10.348692278913578,
      "kl": 0.509033203125,
      "learning_rate": 9.793142857142857e-07,
      "loss": 0.0204,
      "reward": 1.96875,
      "reward_std": 0.08838834427297115,
      "rewards/accuracy_reward": 0.9765625,
      "rewards/format_reward": 0.9921875,
      "step": 181
    },
    {
      "completion_length": 55.8515625,
      "epoch": 0.0416,
      "grad_norm": 0.867301778951744,
      "kl": 0.0771484375,
      "learning_rate": 9.791999999999999e-07,
      "loss": 0.0031,
      "reward": 1.984375,
      "reward_std": 0.0289318785071373,
      "rewards/accuracy_reward": 1.0,
      "rewards/format_reward": 0.984375,
      "step": 182
    },
    {
      "completion_length": 55.2265625,
      "epoch": 0.04182857142857143,
      "grad_norm": 1.4759500808698434,
      "kl": 0.100830078125,
      "learning_rate": 9.790857142857143e-07,
      "loss": 0.004,
      "reward": 1.984375,
      "reward_std": 0.04419417306780815,
      "rewards/accuracy_reward": 0.9921875,
      "rewards/format_reward": 0.9921875,
      "step": 183
    },
    {
      "completion_length": 62.109375,
      "epoch": 0.042057142857142855,
      "grad_norm": 2.7422696754579405,
      "kl": 0.08203125,
      "learning_rate": 9.789714285714285e-07,
      "loss": 0.0033,
      "reward": 1.890625,
      "reward_std": 0.16599016450345516,
      "rewards/accuracy_reward": 0.9296875,
      "rewards/format_reward": 0.9609375,
      "step": 184
    },
    {
      "completion_length": 66.0859375,
      "epoch": 0.04228571428571429,
      "grad_norm": 3.26807146819625,
      "kl": 0.075927734375,
      "learning_rate": 9.788571428571427e-07,
      "loss": 0.003,
      "reward": 1.96875,
      "reward_std": 0.0731260534375906,
      "rewards/accuracy_reward": 0.96875,
      "rewards/format_reward": 1.0,
      "step": 185
    },
    {
      "completion_length": 70.5625,
      "epoch": 0.04251428571428571,
      "grad_norm": 1.3451531440825353,
      "kl": 0.0623779296875,
      "learning_rate": 9.787428571428572e-07,
      "loss": 0.0025,
      "reward": 1.9765625,
      "reward_std": 0.06629125960171223,
      "rewards/accuracy_reward": 0.9921875,
      "rewards/format_reward": 0.984375,
      "step": 186
    },
    {
      "completion_length": 68.6640625,
      "epoch": 0.042742857142857145,
      "grad_norm": 3.080268039189956,
      "kl": 0.1949462890625,
      "learning_rate": 9.786285714285714e-07,
      "loss": 0.0078,
      "reward": 1.9609375,
      "reward_std": 0.09522313997149467,
      "rewards/accuracy_reward": 0.984375,
      "rewards/format_reward": 0.9765625,
      "step": 187
    },
    {
      "completion_length": 71.4296875,
      "epoch": 0.04297142857142857,
      "grad_norm": 1.214840371275306,
      "kl": 0.062255859375,
      "learning_rate": 9.785142857142856e-07,
      "loss": 0.0025,
      "reward": 1.9765625,
      "reward_std": 0.06629125960171223,
      "rewards/accuracy_reward": 0.9921875,
      "rewards/format_reward": 0.984375,
      "step": 188
    },
    {
      "completion_length": 73.140625,
      "epoch": 0.0432,
      "grad_norm": 1.6582893525712101,
      "kl": 0.06103515625,
      "learning_rate": 9.784e-07,
      "loss": 0.0024,
      "reward": 1.9609375,
      "reward_std": 0.11048543266952038,
      "rewards/accuracy_reward": 0.984375,
      "rewards/format_reward": 0.9765625,
      "step": 189
    },
    {
      "completion_length": 71.421875,
      "epoch": 0.04342857142857143,
      "grad_norm": 18.296674974106892,
      "kl": 0.761474609375,
      "learning_rate": 9.782857142857142e-07,
      "loss": 0.0306,
      "reward": 1.96875,
      "reward_std": 0.0883883461356163,
      "rewards/accuracy_reward": 0.9921875,
      "rewards/format_reward": 0.9765625,
      "step": 190
    },
    {
      "completion_length": 75.125,
      "epoch": 0.04365714285714286,
      "grad_norm": 1.9725252271658114,
      "kl": 0.079345703125,
      "learning_rate": 9.781714285714285e-07,
      "loss": 0.0032,
      "reward": 1.859375,
      "reward_std": 0.27430714666843414,
      "rewards/accuracy_reward": 0.90625,
      "rewards/format_reward": 0.953125,
      "step": 191
    },
    {
      "completion_length": 69.4921875,
      "epoch": 0.043885714285714285,
      "grad_norm": 4.443079937809978,
      "kl": 0.0584716796875,
      "learning_rate": 9.780571428571429e-07,
      "loss": 0.0023,
      "reward": 1.984375,
      "reward_std": 0.04419417306780815,
      "rewards/accuracy_reward": 0.9921875,
      "rewards/format_reward": 0.9921875,
      "step": 192
    },
    {
      "completion_length": 74.9609375,
      "epoch": 0.04411428571428572,
      "grad_norm": 1.9773324366901752,
      "kl": 0.072265625,
      "learning_rate": 9.77942857142857e-07,
      "loss": 0.0029,
      "reward": 1.96875,
      "reward_std": 0.0883883461356163,
      "rewards/accuracy_reward": 0.9921875,
      "rewards/format_reward": 0.9765625,
      "step": 193
    },
    {
      "completion_length": 67.2265625,
      "epoch": 0.04434285714285714,
      "grad_norm": 2.3979979311517843,
      "kl": 0.090087890625,
      "learning_rate": 9.778285714285713e-07,
      "loss": 0.0036,
      "reward": 1.90625,
      "reward_std": 0.13258251547813416,
      "rewards/accuracy_reward": 0.9296875,
      "rewards/format_reward": 0.9765625,
      "step": 194
    },
    {
      "completion_length": 78.5625,
      "epoch": 0.044571428571428574,
      "grad_norm": 5.012233254750768,
      "kl": 0.07177734375,
      "learning_rate": 9.777142857142857e-07,
      "loss": 0.0029,
      "reward": 1.9296875,
      "reward_std": 0.18361148238182068,
      "rewards/accuracy_reward": 0.984375,
      "rewards/format_reward": 0.9453125,
      "step": 195
    },
    {
      "completion_length": 67.265625,
      "epoch": 0.0448,
      "grad_norm": 2.290392211026634,
      "kl": 0.0670166015625,
      "learning_rate": 9.776e-07,
      "loss": 0.0027,
      "reward": 1.9609375,
      "reward_std": 0.11048543080687523,
      "rewards/accuracy_reward": 0.9921875,
      "rewards/format_reward": 0.96875,
      "step": 196
    },
    {
      "completion_length": 70.7734375,
      "epoch": 0.04502857142857143,
      "grad_norm": 1.2457447109442346,
      "kl": 0.0606689453125,
      "learning_rate": 9.774857142857142e-07,
      "loss": 0.0024,
      "reward": 1.9453125,
      "reward_std": 0.13488983735442162,
      "rewards/accuracy_reward": 0.9921875,
      "rewards/format_reward": 0.953125,
      "step": 197
    },
    {
      "completion_length": 73.5546875,
      "epoch": 0.04525714285714286,
      "grad_norm": 1.33067152867832,
      "kl": 0.0545654296875,
      "learning_rate": 9.773714285714286e-07,
      "loss": 0.0022,
      "reward": 1.9765625,
      "reward_std": 0.06629125960171223,
      "rewards/accuracy_reward": 1.0,
      "rewards/format_reward": 0.9765625,
      "step": 198
    },
    {
      "completion_length": 68.296875,
      "epoch": 0.04548571428571429,
      "grad_norm": 3.60596935251695,
      "kl": 0.0687255859375,
      "learning_rate": 9.772571428571428e-07,
      "loss": 0.0028,
      "reward": 1.953125,
      "reward_std": 0.09863808378577232,
      "rewards/accuracy_reward": 0.96875,
      "rewards/format_reward": 0.984375,
      "step": 199
    },
    {
      "completion_length": 69.2578125,
      "epoch": 0.045714285714285714,
      "grad_norm": 1.5741330243580274,
      "kl": 0.06787109375,
      "learning_rate": 9.77142857142857e-07,
      "loss": 0.0027,
      "reward": 1.9609375,
      "reward_std": 0.11048543266952038,
      "rewards/accuracy_reward": 0.984375,
      "rewards/format_reward": 0.9765625,
      "step": 200
    }
  ],
  "logging_steps": 1.0,
  "max_steps": 8750,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 2,
  "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": 1,
  "trial_name": null,
  "trial_params": null
}