{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 0.13733905579399142,
  "eval_steps": 500,
  "global_step": 160,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "advantages": 1.862645149230957e-08,
      "completion_length": 129.0625,
      "epoch": 0.0008583690987124463,
      "grad_norm": 174.6410675048828,
      "kl": 4.34375,
      "learning_rate": 9.991416309012877e-07,
      "loss": 0.434,
      "reward": 0.15625,
      "reward_mean": 0.15625,
      "reward_std": 0.3198433816432953,
      "rewards/accuracy_reward": 0.15625,
      "step": 1
    },
    {
      "advantages": 1.30385160446167e-08,
      "completion_length": 138.0625,
      "epoch": 0.0017167381974248926,
      "grad_norm": 332.6672668457031,
      "kl": 5.21875,
      "learning_rate": 9.982832618025751e-07,
      "loss": 0.5223,
      "reward": 0.34375,
      "reward_mean": 0.34375,
      "reward_std": 0.4628904461860657,
      "rewards/accuracy_reward": 0.34375,
      "step": 2
    },
    {
      "advantages": 2.421438694000244e-08,
      "completion_length": 117.4375,
      "epoch": 0.002575107296137339,
      "grad_norm": 166.6440887451172,
      "kl": 3.5,
      "learning_rate": 9.974248927038626e-07,
      "loss": 0.3496,
      "reward": 0.21875,
      "reward_mean": 0.21875,
      "reward_std": 0.3608423173427582,
      "rewards/accuracy_reward": 0.21875,
      "step": 3
    },
    {
      "advantages": 7.450580596923828e-09,
      "completion_length": 126.59375,
      "epoch": 0.0034334763948497852,
      "grad_norm": 111.89189910888672,
      "kl": 1.5078125,
      "learning_rate": 9.965665236051501e-07,
      "loss": 0.1505,
      "reward": 0.0625,
      "reward_mean": 0.0625,
      "reward_std": 0.1767766922712326,
      "rewards/accuracy_reward": 0.0625,
      "step": 4
    },
    {
      "advantages": 1.4901161193847656e-08,
      "completion_length": 164.875,
      "epoch": 0.004291845493562232,
      "grad_norm": 38.75446701049805,
      "kl": 1.359375,
      "learning_rate": 9.957081545064378e-07,
      "loss": 0.1359,
      "reward": 0.25,
      "reward_mean": 0.25,
      "reward_std": 0.4261348247528076,
      "rewards/accuracy_reward": 0.25,
      "step": 5
    },
    {
      "advantages": 2.0489096641540527e-08,
      "completion_length": 136.9375,
      "epoch": 0.005150214592274678,
      "grad_norm": 278.40093994140625,
      "kl": 4.5,
      "learning_rate": 9.948497854077253e-07,
      "loss": 0.4497,
      "reward": 0.1875,
      "reward_mean": 0.1875,
      "reward_std": 0.3945523500442505,
      "rewards/accuracy_reward": 0.1875,
      "step": 6
    },
    {
      "advantages": 2.60770320892334e-08,
      "completion_length": 150.8125,
      "epoch": 0.006008583690987125,
      "grad_norm": 189.5176544189453,
      "kl": 2.75,
      "learning_rate": 9.939914163090128e-07,
      "loss": 0.2753,
      "reward": 0.375,
      "reward_mean": 0.375,
      "reward_std": 0.5081326961517334,
      "rewards/accuracy_reward": 0.375,
      "step": 7
    },
    {
      "advantages": 1.30385160446167e-08,
      "completion_length": 135.5,
      "epoch": 0.0068669527896995704,
      "grad_norm": 1704.9647216796875,
      "kl": 4.34375,
      "learning_rate": 9.931330472103003e-07,
      "loss": 0.4358,
      "reward": 0.125,
      "reward_mean": 0.125,
      "reward_std": 0.2177756428718567,
      "rewards/accuracy_reward": 0.125,
      "step": 8
    },
    {
      "advantages": 1.1175870895385742e-08,
      "completion_length": 116.96875,
      "epoch": 0.007725321888412017,
      "grad_norm": 26.27825927734375,
      "kl": 0.8671875,
      "learning_rate": 9.92274678111588e-07,
      "loss": 0.0868,
      "reward": 0.09375,
      "reward_mean": 0.09375,
      "reward_std": 0.2651650309562683,
      "rewards/accuracy_reward": 0.09375,
      "step": 9
    },
    {
      "advantages": -9.313225746154785e-09,
      "completion_length": 105.625,
      "epoch": 0.008583690987124463,
      "grad_norm": 115.19660949707031,
      "kl": 1.28125,
      "learning_rate": 9.914163090128755e-07,
      "loss": 0.1278,
      "reward": 0.53125,
      "reward_mean": 0.53125,
      "reward_std": 0.521792471408844,
      "rewards/accuracy_reward": 0.53125,
      "step": 10
    },
    {
      "advantages": 1.30385160446167e-08,
      "completion_length": 125.78125,
      "epoch": 0.00944206008583691,
      "grad_norm": 242.439453125,
      "kl": 3.75,
      "learning_rate": 9.90557939914163e-07,
      "loss": 0.3733,
      "reward": 0.125,
      "reward_mean": 0.125,
      "reward_std": 0.2177756428718567,
      "rewards/accuracy_reward": 0.125,
      "step": 11
    },
    {
      "advantages": 2.421438694000244e-08,
      "completion_length": 140.65625,
      "epoch": 0.010300429184549357,
      "grad_norm": 57.61454772949219,
      "kl": 1.5625,
      "learning_rate": 9.896995708154506e-07,
      "loss": 0.1566,
      "reward": 0.21875,
      "reward_mean": 0.21875,
      "reward_std": 0.4218915104866028,
      "rewards/accuracy_reward": 0.21875,
      "step": 12
    },
    {
      "advantages": 1.4901161193847656e-08,
      "completion_length": 122.3125,
      "epoch": 0.011158798283261802,
      "grad_norm": 9.250263214111328,
      "kl": 0.416015625,
      "learning_rate": 9.888412017167381e-07,
      "loss": 0.0417,
      "reward": 0.125,
      "reward_mean": 0.125,
      "reward_std": 0.292504221200943,
      "rewards/accuracy_reward": 0.125,
      "step": 13
    },
    {
      "advantages": 7.450580596923828e-09,
      "completion_length": 134.34375,
      "epoch": 0.01201716738197425,
      "grad_norm": 29.975547790527344,
      "kl": 0.76171875,
      "learning_rate": 9.879828326180258e-07,
      "loss": 0.0763,
      "reward": 0.1875,
      "reward_mean": 0.1875,
      "reward_std": 0.3104073107242584,
      "rewards/accuracy_reward": 0.1875,
      "step": 14
    },
    {
      "advantages": -5.587935447692871e-09,
      "completion_length": 136.5,
      "epoch": 0.012875536480686695,
      "grad_norm": 35.626949310302734,
      "kl": 0.9140625,
      "learning_rate": 9.871244635193133e-07,
      "loss": 0.0915,
      "reward": 0.3125,
      "reward_mean": 0.3125,
      "reward_std": 0.3514062464237213,
      "rewards/accuracy_reward": 0.3125,
      "step": 15
    },
    {
      "advantages": 2.421438694000244e-08,
      "completion_length": 133.84375,
      "epoch": 0.013733905579399141,
      "grad_norm": 127.25942993164062,
      "kl": 1.765625,
      "learning_rate": 9.862660944206008e-07,
      "loss": 0.176,
      "reward": 0.21875,
      "reward_mean": 0.21875,
      "reward_std": 0.3608423173427582,
      "rewards/accuracy_reward": 0.21875,
      "step": 16
    },
    {
      "advantages": 2.0489096641540527e-08,
      "completion_length": 110.25,
      "epoch": 0.014592274678111588,
      "grad_norm": 51.83127975463867,
      "kl": 0.52734375,
      "learning_rate": 9.854077253218883e-07,
      "loss": 0.0526,
      "reward": 0.1875,
      "reward_mean": 0.1875,
      "reward_std": 0.3335031569004059,
      "rewards/accuracy_reward": 0.1875,
      "step": 17
    },
    {
      "advantages": 1.6763806343078613e-08,
      "completion_length": 132.71875,
      "epoch": 0.015450643776824034,
      "grad_norm": 71.73090362548828,
      "kl": 1.375,
      "learning_rate": 9.84549356223176e-07,
      "loss": 0.1369,
      "reward": 0.4375,
      "reward_mean": 0.4375,
      "reward_std": 0.5260357856750488,
      "rewards/accuracy_reward": 0.4375,
      "step": 18
    },
    {
      "advantages": 1.4901161193847656e-08,
      "completion_length": 139.96875,
      "epoch": 0.01630901287553648,
      "grad_norm": 47.92875289916992,
      "kl": 0.92578125,
      "learning_rate": 9.836909871244635e-07,
      "loss": 0.0925,
      "reward": 0.15625,
      "reward_mean": 0.15625,
      "reward_std": 0.24511480331420898,
      "rewards/accuracy_reward": 0.15625,
      "step": 19
    },
    {
      "advantages": 2.60770320892334e-08,
      "completion_length": 139.8125,
      "epoch": 0.017167381974248927,
      "grad_norm": 32.31055450439453,
      "kl": 0.703125,
      "learning_rate": 9.82832618025751e-07,
      "loss": 0.0701,
      "reward": 0.25,
      "reward_mean": 0.25,
      "reward_std": 0.4355512857437134,
      "rewards/accuracy_reward": 0.25,
      "step": 20
    },
    {
      "advantages": 1.1175870895385742e-08,
      "completion_length": 122.3125,
      "epoch": 0.018025751072961373,
      "grad_norm": 12.536384582519531,
      "kl": 0.423828125,
      "learning_rate": 9.819742489270387e-07,
      "loss": 0.0424,
      "reward": 0.09375,
      "reward_mean": 0.09375,
      "reward_std": 0.2651650309562683,
      "rewards/accuracy_reward": 0.09375,
      "step": 21
    },
    {
      "advantages": 1.4901161193847656e-08,
      "completion_length": 155.6875,
      "epoch": 0.01888412017167382,
      "grad_norm": 9.963027954101562,
      "kl": 0.421875,
      "learning_rate": 9.811158798283261e-07,
      "loss": 0.0421,
      "reward": 0.125,
      "reward_mean": 0.125,
      "reward_std": 0.3535533845424652,
      "rewards/accuracy_reward": 0.125,
      "step": 22
    },
    {
      "advantages": 1.1175870895385742e-08,
      "completion_length": 137.15625,
      "epoch": 0.019742489270386267,
      "grad_norm": 6.500565052032471,
      "kl": 0.25,
      "learning_rate": 9.802575107296136e-07,
      "loss": 0.025,
      "reward": 0.09375,
      "reward_mean": 0.09375,
      "reward_std": 0.2651650309562683,
      "rewards/accuracy_reward": 0.09375,
      "step": 23
    },
    {
      "advantages": 5.587935447692871e-09,
      "completion_length": 143.4375,
      "epoch": 0.020600858369098713,
      "grad_norm": 4.852046012878418,
      "kl": 0.208984375,
      "learning_rate": 9.793991416309011e-07,
      "loss": 0.021,
      "reward": 0.28125,
      "reward_mean": 0.28125,
      "reward_std": 0.3608423173427582,
      "rewards/accuracy_reward": 0.28125,
      "step": 24
    },
    {
      "advantages": 1.4901161193847656e-08,
      "completion_length": 130.03125,
      "epoch": 0.02145922746781116,
      "grad_norm": 11.683233261108398,
      "kl": 0.396484375,
      "learning_rate": 9.785407725321888e-07,
      "loss": 0.0395,
      "reward": 0.125,
      "reward_mean": 0.125,
      "reward_std": 0.2925041913986206,
      "rewards/accuracy_reward": 0.125,
      "step": 25
    },
    {
      "advantages": 3.166496753692627e-08,
      "completion_length": 150.59375,
      "epoch": 0.022317596566523604,
      "grad_norm": 7.088483810424805,
      "kl": 0.29296875,
      "learning_rate": 9.776824034334763e-07,
      "loss": 0.0294,
      "reward": 0.28125,
      "reward_mean": 0.28125,
      "reward_std": 0.4765698313713074,
      "rewards/accuracy_reward": 0.28125,
      "step": 26
    },
    {
      "advantages": 2.9802322387695312e-08,
      "completion_length": 125.1875,
      "epoch": 0.02317596566523605,
      "grad_norm": 9.142675399780273,
      "kl": 0.41015625,
      "learning_rate": 9.76824034334764e-07,
      "loss": 0.0412,
      "reward": 0.28125,
      "reward_mean": 0.28125,
      "reward_std": 0.4628904461860657,
      "rewards/accuracy_reward": 0.28125,
      "step": 27
    },
    {
      "advantages": 1.30385160446167e-08,
      "completion_length": 146.21875,
      "epoch": 0.0240343347639485,
      "grad_norm": 7.925714492797852,
      "kl": 0.22265625,
      "learning_rate": 9.759656652360515e-07,
      "loss": 0.0222,
      "reward": 0.375,
      "reward_mean": 0.375,
      "reward_std": 0.47655022144317627,
      "rewards/accuracy_reward": 0.375,
      "step": 28
    },
    {
      "advantages": 1.1175870895385742e-08,
      "completion_length": 156.46875,
      "epoch": 0.024892703862660945,
      "grad_norm": 9.395242691040039,
      "kl": 0.3984375,
      "learning_rate": 9.75107296137339e-07,
      "loss": 0.0397,
      "reward": 0.09375,
      "reward_mean": 0.09375,
      "reward_std": 0.2041158676147461,
      "rewards/accuracy_reward": 0.09375,
      "step": 29
    },
    {
      "advantages": 2.9802322387695312e-08,
      "completion_length": 148.65625,
      "epoch": 0.02575107296137339,
      "grad_norm": 8.077410697937012,
      "kl": 0.2890625,
      "learning_rate": 9.742489270386267e-07,
      "loss": 0.0289,
      "reward": 0.28125,
      "reward_mean": 0.28125,
      "reward_std": 0.4628904461860657,
      "rewards/accuracy_reward": 0.28125,
      "step": 30
    },
    {
      "advantages": 1.6763806343078613e-08,
      "completion_length": 137.3125,
      "epoch": 0.026609442060085836,
      "grad_norm": 8.573564529418945,
      "kl": 0.263671875,
      "learning_rate": 9.733905579399142e-07,
      "loss": 0.0264,
      "reward": 0.28125,
      "reward_mean": 0.28125,
      "reward_std": 0.378745436668396,
      "rewards/accuracy_reward": 0.28125,
      "step": 31
    },
    {
      "advantages": 2.9802322387695312e-08,
      "completion_length": 148.25,
      "epoch": 0.027467811158798282,
      "grad_norm": 6.53264045715332,
      "kl": 0.265625,
      "learning_rate": 9.725321888412016e-07,
      "loss": 0.0265,
      "reward": 0.28125,
      "reward_mean": 0.28125,
      "reward_std": 0.4628904461860657,
      "rewards/accuracy_reward": 0.28125,
      "step": 32
    },
    {
      "advantages": 3.166496753692627e-08,
      "completion_length": 133.0,
      "epoch": 0.02832618025751073,
      "grad_norm": 8.72734260559082,
      "kl": 0.36328125,
      "learning_rate": 9.716738197424891e-07,
      "loss": 0.0363,
      "reward": 0.3125,
      "reward_mean": 0.3125,
      "reward_std": 0.47655022144317627,
      "rewards/accuracy_reward": 0.3125,
      "step": 33
    },
    {
      "advantages": 9.313225746154785e-09,
      "completion_length": 133.3125,
      "epoch": 0.029184549356223177,
      "grad_norm": 11.700161933898926,
      "kl": 0.30859375,
      "learning_rate": 9.708154506437768e-07,
      "loss": 0.031,
      "reward": 0.09375,
      "reward_mean": 0.09375,
      "reward_std": 0.1293872892856598,
      "rewards/accuracy_reward": 0.09375,
      "step": 34
    },
    {
      "advantages": 2.9802322387695312e-08,
      "completion_length": 145.5625,
      "epoch": 0.030042918454935622,
      "grad_norm": 25.358783721923828,
      "kl": 0.486328125,
      "learning_rate": 9.699570815450643e-07,
      "loss": 0.0486,
      "reward": 0.28125,
      "reward_mean": 0.28125,
      "reward_std": 0.4628904461860657,
      "rewards/accuracy_reward": 0.28125,
      "step": 35
    },
    {
      "advantages": 1.862645149230957e-08,
      "completion_length": 132.625,
      "epoch": 0.030901287553648068,
      "grad_norm": 13.985993385314941,
      "kl": 0.298828125,
      "learning_rate": 9.690987124463518e-07,
      "loss": 0.0298,
      "reward": 0.15625,
      "reward_mean": 0.15625,
      "reward_std": 0.3808925747871399,
      "rewards/accuracy_reward": 0.15625,
      "step": 36
    },
    {
      "advantages": 7.450580596923828e-09,
      "completion_length": 160.90625,
      "epoch": 0.03175965665236052,
      "grad_norm": 18.302053451538086,
      "kl": 0.447265625,
      "learning_rate": 9.682403433476395e-07,
      "loss": 0.0447,
      "reward": 0.0625,
      "reward_mean": 0.0625,
      "reward_std": 0.1767766922712326,
      "rewards/accuracy_reward": 0.0625,
      "step": 37
    },
    {
      "advantages": 1.1175870895385742e-08,
      "completion_length": 144.875,
      "epoch": 0.03261802575107296,
      "grad_norm": 4.443456172943115,
      "kl": 0.248046875,
      "learning_rate": 9.67381974248927e-07,
      "loss": 0.0247,
      "reward": 0.09375,
      "reward_mean": 0.09375,
      "reward_std": 0.2041158676147461,
      "rewards/accuracy_reward": 0.09375,
      "step": 38
    },
    {
      "advantages": 1.6763806343078613e-08,
      "completion_length": 157.96875,
      "epoch": 0.03347639484978541,
      "grad_norm": 6.555863380432129,
      "kl": 0.279296875,
      "learning_rate": 9.665236051502147e-07,
      "loss": 0.0279,
      "reward": 0.15625,
      "reward_mean": 0.15625,
      "reward_std": 0.3061639666557312,
      "rewards/accuracy_reward": 0.15625,
      "step": 39
    },
    {
      "advantages": 3.725290298461914e-09,
      "completion_length": 131.03125,
      "epoch": 0.034334763948497854,
      "grad_norm": 7.260156631469727,
      "kl": 0.1865234375,
      "learning_rate": 9.656652360515022e-07,
      "loss": 0.0186,
      "reward": 0.28125,
      "reward_mean": 0.28125,
      "reward_std": 0.35564959049224854,
      "rewards/accuracy_reward": 0.28125,
      "step": 40
    },
    {
      "advantages": 1.6763806343078613e-08,
      "completion_length": 155.75,
      "epoch": 0.0351931330472103,
      "grad_norm": 9.163714408874512,
      "kl": 0.19140625,
      "learning_rate": 9.648068669527897e-07,
      "loss": 0.0191,
      "reward": 0.15625,
      "reward_mean": 0.15625,
      "reward_std": 0.3061639666557312,
      "rewards/accuracy_reward": 0.15625,
      "step": 41
    },
    {
      "advantages": 1.862645149230957e-08,
      "completion_length": 153.09375,
      "epoch": 0.036051502145922745,
      "grad_norm": 9.94912052154541,
      "kl": 0.345703125,
      "learning_rate": 9.639484978540771e-07,
      "loss": 0.0345,
      "reward": 0.15625,
      "reward_mean": 0.15625,
      "reward_std": 0.3198433816432953,
      "rewards/accuracy_reward": 0.15625,
      "step": 42
    },
    {
      "advantages": 2.0489096641540527e-08,
      "completion_length": 163.28125,
      "epoch": 0.03690987124463519,
      "grad_norm": 4.065970420837402,
      "kl": 0.2119140625,
      "learning_rate": 9.630901287553648e-07,
      "loss": 0.0212,
      "reward": 0.1875,
      "reward_mean": 0.1875,
      "reward_std": 0.3335031569004059,
      "rewards/accuracy_reward": 0.1875,
      "step": 43
    },
    {
      "advantages": 2.60770320892334e-08,
      "completion_length": 149.03125,
      "epoch": 0.03776824034334764,
      "grad_norm": 6.335684776306152,
      "kl": 0.330078125,
      "learning_rate": 9.622317596566523e-07,
      "loss": 0.033,
      "reward": 0.25,
      "reward_mean": 0.25,
      "reward_std": 0.3745020925998688,
      "rewards/accuracy_reward": 0.25,
      "step": 44
    },
    {
      "advantages": 1.862645149230957e-08,
      "completion_length": 168.0625,
      "epoch": 0.03862660944206009,
      "grad_norm": 5.854466438293457,
      "kl": 0.224609375,
      "learning_rate": 9.613733905579398e-07,
      "loss": 0.0225,
      "reward": 0.15625,
      "reward_mean": 0.15625,
      "reward_std": 0.3808925747871399,
      "rewards/accuracy_reward": 0.15625,
      "step": 45
    },
    {
      "advantages": 2.0489096641540527e-08,
      "completion_length": 159.3125,
      "epoch": 0.039484978540772535,
      "grad_norm": 59.691341400146484,
      "kl": 0.67578125,
      "learning_rate": 9.605150214592275e-07,
      "loss": 0.0677,
      "reward": 0.1875,
      "reward_mean": 0.1875,
      "reward_std": 0.3945523500442505,
      "rewards/accuracy_reward": 0.1875,
      "step": 46
    },
    {
      "advantages": 2.60770320892334e-08,
      "completion_length": 149.21875,
      "epoch": 0.04034334763948498,
      "grad_norm": 5.02371883392334,
      "kl": 0.1845703125,
      "learning_rate": 9.59656652360515e-07,
      "loss": 0.0185,
      "reward": 0.21875,
      "reward_mean": 0.21875,
      "reward_std": 0.4355708956718445,
      "rewards/accuracy_reward": 0.21875,
      "step": 47
    },
    {
      "advantages": 2.421438694000244e-08,
      "completion_length": 167.875,
      "epoch": 0.041201716738197426,
      "grad_norm": 4.878015041351318,
      "kl": 0.185546875,
      "learning_rate": 9.587982832618025e-07,
      "loss": 0.0185,
      "reward": 0.21875,
      "reward_mean": 0.21875,
      "reward_std": 0.4218915104866028,
      "rewards/accuracy_reward": 0.21875,
      "step": 48
    },
    {
      "advantages": 2.0489096641540527e-08,
      "completion_length": 152.96875,
      "epoch": 0.04206008583690987,
      "grad_norm": 5.698736667633057,
      "kl": 0.234375,
      "learning_rate": 9.5793991416309e-07,
      "loss": 0.0234,
      "reward": 0.21875,
      "reward_mean": 0.21875,
      "reward_std": 0.3471629321575165,
      "rewards/accuracy_reward": 0.21875,
      "step": 49
    },
    {
      "advantages": 7.450580596923828e-09,
      "completion_length": 157.8125,
      "epoch": 0.04291845493562232,
      "grad_norm": 5.776604652404785,
      "kl": 0.1474609375,
      "learning_rate": 9.570815450643777e-07,
      "loss": 0.0148,
      "reward": 0.3125,
      "reward_mean": 0.3125,
      "reward_std": 0.3745020925998688,
      "rewards/accuracy_reward": 0.3125,
      "step": 50
    },
    {
      "advantages": 2.2351741790771484e-08,
      "completion_length": 148.78125,
      "epoch": 0.04377682403433476,
      "grad_norm": 4.41421365737915,
      "kl": 0.169921875,
      "learning_rate": 9.562231759656652e-07,
      "loss": 0.0169,
      "reward": 0.21875,
      "reward_mean": 0.21875,
      "reward_std": 0.3608423173427582,
      "rewards/accuracy_reward": 0.21875,
      "step": 51
    },
    {
      "advantages": 1.1175870895385742e-08,
      "completion_length": 146.90625,
      "epoch": 0.04463519313304721,
      "grad_norm": 4.667245864868164,
      "kl": 0.189453125,
      "learning_rate": 9.553648068669528e-07,
      "loss": 0.0189,
      "reward": 0.21875,
      "reward_mean": 0.21875,
      "reward_std": 0.3377464711666107,
      "rewards/accuracy_reward": 0.21875,
      "step": 52
    },
    {
      "advantages": -3.725290298461914e-09,
      "completion_length": 158.625,
      "epoch": 0.045493562231759654,
      "grad_norm": 4.415360927581787,
      "kl": 0.1259765625,
      "learning_rate": 9.545064377682403e-07,
      "loss": 0.0126,
      "reward": 0.46875,
      "reward_mean": 0.46875,
      "reward_std": 0.3808925747871399,
      "rewards/accuracy_reward": 0.46875,
      "step": 53
    },
    {
      "advantages": 1.4901161193847656e-08,
      "completion_length": 155.65625,
      "epoch": 0.0463519313304721,
      "grad_norm": 12.489053726196289,
      "kl": 0.37109375,
      "learning_rate": 9.536480686695278e-07,
      "loss": 0.0371,
      "reward": 0.125,
      "reward_mean": 0.125,
      "reward_std": 0.2925041913986206,
      "rewards/accuracy_reward": 0.125,
      "step": 54
    },
    {
      "advantages": 2.9802322387695312e-08,
      "completion_length": 172.9375,
      "epoch": 0.04721030042918455,
      "grad_norm": 5.111673831939697,
      "kl": 0.173828125,
      "learning_rate": 9.527896995708154e-07,
      "loss": 0.0174,
      "reward": 0.25,
      "reward_mean": 0.25,
      "reward_std": 0.4629100561141968,
      "rewards/accuracy_reward": 0.25,
      "step": 55
    },
    {
      "advantages": 2.9802322387695312e-08,
      "completion_length": 153.4375,
      "epoch": 0.048068669527897,
      "grad_norm": 149.73927307128906,
      "kl": 2.0,
      "learning_rate": 9.519313304721029e-07,
      "loss": 0.2008,
      "reward": 0.28125,
      "reward_mean": 0.28125,
      "reward_std": 0.4628904461860657,
      "rewards/accuracy_reward": 0.28125,
      "step": 56
    },
    {
      "advantages": 1.4901161193847656e-08,
      "completion_length": 133.0625,
      "epoch": 0.048927038626609444,
      "grad_norm": 13.399458885192871,
      "kl": 0.37109375,
      "learning_rate": 9.510729613733906e-07,
      "loss": 0.0371,
      "reward": 0.15625,
      "reward_mean": 0.15625,
      "reward_std": 0.3061639964580536,
      "rewards/accuracy_reward": 0.15625,
      "step": 57
    },
    {
      "advantages": 1.6763806343078613e-08,
      "completion_length": 163.125,
      "epoch": 0.04978540772532189,
      "grad_norm": 6.9010210037231445,
      "kl": 0.294921875,
      "learning_rate": 9.502145922746781e-07,
      "loss": 0.0295,
      "reward": 0.15625,
      "reward_mean": 0.15625,
      "reward_std": 0.3061639964580536,
      "rewards/accuracy_reward": 0.15625,
      "step": 58
    },
    {
      "advantages": 1.6763806343078613e-08,
      "completion_length": 146.90625,
      "epoch": 0.050643776824034335,
      "grad_norm": 6.625538349151611,
      "kl": 0.166015625,
      "learning_rate": 9.493562231759657e-07,
      "loss": 0.0166,
      "reward": 0.28125,
      "reward_mean": 0.28125,
      "reward_std": 0.4397946000099182,
      "rewards/accuracy_reward": 0.28125,
      "step": 59
    },
    {
      "advantages": 3.725290298461914e-09,
      "completion_length": 151.03125,
      "epoch": 0.05150214592274678,
      "grad_norm": 9.018912315368652,
      "kl": 0.169921875,
      "learning_rate": 9.484978540772532e-07,
      "loss": 0.017,
      "reward": 0.28125,
      "reward_mean": 0.28125,
      "reward_std": 0.3471629321575165,
      "rewards/accuracy_reward": 0.28125,
      "step": 60
    },
    {
      "advantages": 1.30385160446167e-08,
      "completion_length": 121.25,
      "epoch": 0.05236051502145923,
      "grad_norm": 24.219348907470703,
      "kl": 0.47265625,
      "learning_rate": 9.476394849785408e-07,
      "loss": 0.0473,
      "reward": 0.125,
      "reward_mean": 0.125,
      "reward_std": 0.2177756428718567,
      "rewards/accuracy_reward": 0.125,
      "step": 61
    },
    {
      "advantages": 1.4901161193847656e-08,
      "completion_length": 142.5625,
      "epoch": 0.05321888412017167,
      "grad_norm": 5.040563106536865,
      "kl": 0.201171875,
      "learning_rate": 9.467811158798282e-07,
      "loss": 0.0201,
      "reward": 0.125,
      "reward_mean": 0.125,
      "reward_std": 0.2925041913986206,
      "rewards/accuracy_reward": 0.125,
      "step": 62
    },
    {
      "advantages": -1.862645149230957e-09,
      "completion_length": 145.53125,
      "epoch": 0.05407725321888412,
      "grad_norm": 9.56372356414795,
      "kl": 0.279296875,
      "learning_rate": 9.459227467811158e-07,
      "loss": 0.0279,
      "reward": 0.4375,
      "reward_mean": 0.4375,
      "reward_std": 0.49022960662841797,
      "rewards/accuracy_reward": 0.4375,
      "step": 63
    },
    {
      "advantages": 7.450580596923828e-09,
      "completion_length": 110.75,
      "epoch": 0.054935622317596564,
      "grad_norm": 5.647745132446289,
      "kl": 0.1884765625,
      "learning_rate": 9.450643776824034e-07,
      "loss": 0.0189,
      "reward": 0.3125,
      "reward_mean": 0.3125,
      "reward_std": 0.4082317352294922,
      "rewards/accuracy_reward": 0.3125,
      "step": 64
    },
    {
      "advantages": -1.30385160446167e-08,
      "completion_length": 152.0,
      "epoch": 0.055793991416309016,
      "grad_norm": 5.315371513366699,
      "kl": 0.17578125,
      "learning_rate": 9.442060085836909e-07,
      "loss": 0.0176,
      "reward": 0.46875,
      "reward_mean": 0.46875,
      "reward_std": 0.4807935357093811,
      "rewards/accuracy_reward": 0.46875,
      "step": 65
    },
    {
      "advantages": 1.6763806343078613e-08,
      "completion_length": 117.6875,
      "epoch": 0.05665236051502146,
      "grad_norm": 4.9794840812683105,
      "kl": 0.2275390625,
      "learning_rate": 9.433476394849785e-07,
      "loss": 0.0228,
      "reward": 0.28125,
      "reward_mean": 0.28125,
      "reward_std": 0.378745436668396,
      "rewards/accuracy_reward": 0.28125,
      "step": 66
    },
    {
      "advantages": 2.9802322387695312e-08,
      "completion_length": 140.15625,
      "epoch": 0.05751072961373391,
      "grad_norm": 4.917365074157715,
      "kl": 0.171875,
      "learning_rate": 9.42489270386266e-07,
      "loss": 0.0172,
      "reward": 0.28125,
      "reward_mean": 0.28125,
      "reward_std": 0.4628904461860657,
      "rewards/accuracy_reward": 0.28125,
      "step": 67
    },
    {
      "advantages": 1.1175870895385742e-08,
      "completion_length": 134.09375,
      "epoch": 0.05836909871244635,
      "grad_norm": 4.92598819732666,
      "kl": 0.1640625,
      "learning_rate": 9.416309012875536e-07,
      "loss": 0.0164,
      "reward": 0.34375,
      "reward_mean": 0.34375,
      "reward_std": 0.4628904461860657,
      "rewards/accuracy_reward": 0.34375,
      "step": 68
    },
    {
      "advantages": 1.4901161193847656e-08,
      "completion_length": 134.0625,
      "epoch": 0.0592274678111588,
      "grad_norm": 11.736459732055664,
      "kl": 0.44921875,
      "learning_rate": 9.407725321888411e-07,
      "loss": 0.0449,
      "reward": 0.25,
      "reward_mean": 0.25,
      "reward_std": 0.3650856614112854,
      "rewards/accuracy_reward": 0.25,
      "step": 69
    },
    {
      "advantages": 2.2351741790771484e-08,
      "completion_length": 131.5,
      "epoch": 0.060085836909871244,
      "grad_norm": 21.64668083190918,
      "kl": 0.47265625,
      "learning_rate": 9.399141630901288e-07,
      "loss": 0.0474,
      "reward": 0.21875,
      "reward_mean": 0.21875,
      "reward_std": 0.4218915104866028,
      "rewards/accuracy_reward": 0.21875,
      "step": 70
    },
    {
      "advantages": 2.0489096641540527e-08,
      "completion_length": 137.1875,
      "epoch": 0.06094420600858369,
      "grad_norm": 4.1953444480896,
      "kl": 0.1826171875,
      "learning_rate": 9.390557939914163e-07,
      "loss": 0.0182,
      "reward": 0.1875,
      "reward_mean": 0.1875,
      "reward_std": 0.3335031569004059,
      "rewards/accuracy_reward": 0.1875,
      "step": 71
    },
    {
      "advantages": 3.725290298461914e-09,
      "completion_length": 138.46875,
      "epoch": 0.061802575107296136,
      "grad_norm": 54.577999114990234,
      "kl": 0.73828125,
      "learning_rate": 9.381974248927038e-07,
      "loss": 0.0742,
      "reward": 0.40625,
      "reward_mean": 0.40625,
      "reward_std": 0.4807935357093811,
      "rewards/accuracy_reward": 0.40625,
      "step": 72
    },
    {
      "advantages": 0.0,
      "completion_length": 148.0,
      "epoch": 0.06266094420600858,
      "grad_norm": 24.432819366455078,
      "kl": 0.1826171875,
      "learning_rate": 9.373390557939914e-07,
      "loss": 0.0183,
      "reward": 0.25,
      "reward_mean": 0.25,
      "reward_std": 0.2587745785713196,
      "rewards/accuracy_reward": 0.25,
      "step": 73
    },
    {
      "advantages": 2.421438694000244e-08,
      "completion_length": 145.84375,
      "epoch": 0.06351931330472103,
      "grad_norm": 5.361104965209961,
      "kl": 0.150390625,
      "learning_rate": 9.364806866952789e-07,
      "loss": 0.015,
      "reward": 0.21875,
      "reward_mean": 0.21875,
      "reward_std": 0.4218915104866028,
      "rewards/accuracy_reward": 0.21875,
      "step": 74
    },
    {
      "advantages": 1.862645149230957e-08,
      "completion_length": 148.625,
      "epoch": 0.06437768240343347,
      "grad_norm": 4.662086009979248,
      "kl": 0.1630859375,
      "learning_rate": 9.356223175965665e-07,
      "loss": 0.0163,
      "reward": 0.28125,
      "reward_mean": 0.28125,
      "reward_std": 0.4534739851951599,
      "rewards/accuracy_reward": 0.28125,
      "step": 75
    },
    {
      "advantages": 1.30385160446167e-08,
      "completion_length": 138.375,
      "epoch": 0.06523605150214593,
      "grad_norm": 17.23464012145996,
      "kl": 0.404296875,
      "learning_rate": 9.34763948497854e-07,
      "loss": 0.0406,
      "reward": 0.125,
      "reward_mean": 0.125,
      "reward_std": 0.2177756428718567,
      "rewards/accuracy_reward": 0.125,
      "step": 76
    },
    {
      "advantages": 3.166496753692627e-08,
      "completion_length": 161.4375,
      "epoch": 0.06609442060085836,
      "grad_norm": 4.420433521270752,
      "kl": 0.162109375,
      "learning_rate": 9.339055793991416e-07,
      "loss": 0.0162,
      "reward": 0.3125,
      "reward_mean": 0.3125,
      "reward_std": 0.49022960662841797,
      "rewards/accuracy_reward": 0.3125,
      "step": 77
    },
    {
      "advantages": 2.2351741790771484e-08,
      "completion_length": 145.125,
      "epoch": 0.06695278969957082,
      "grad_norm": 11.806068420410156,
      "kl": 0.294921875,
      "learning_rate": 9.330472103004291e-07,
      "loss": 0.0294,
      "reward": 0.1875,
      "reward_mean": 0.1875,
      "reward_std": 0.4082317352294922,
      "rewards/accuracy_reward": 0.1875,
      "step": 78
    },
    {
      "advantages": 9.313225746154785e-09,
      "completion_length": 146.21875,
      "epoch": 0.06781115879828326,
      "grad_norm": 5.149102210998535,
      "kl": 0.201171875,
      "learning_rate": 9.321888412017167e-07,
      "loss": 0.0201,
      "reward": 0.34375,
      "reward_mean": 0.34375,
      "reward_std": 0.3966485261917114,
      "rewards/accuracy_reward": 0.34375,
      "step": 79
    },
    {
      "advantages": 1.6763806343078613e-08,
      "completion_length": 133.0,
      "epoch": 0.06866952789699571,
      "grad_norm": 4.339179992675781,
      "kl": 0.232421875,
      "learning_rate": 9.313304721030042e-07,
      "loss": 0.0233,
      "reward": 0.15625,
      "reward_mean": 0.15625,
      "reward_std": 0.24511480331420898,
      "rewards/accuracy_reward": 0.15625,
      "step": 80
    },
    {
      "advantages": 7.450580596923828e-09,
      "completion_length": 150.8125,
      "epoch": 0.06952789699570816,
      "grad_norm": 7.417867183685303,
      "kl": 0.166015625,
      "learning_rate": 9.304721030042918e-07,
      "loss": 0.0166,
      "reward": 0.28125,
      "reward_mean": 0.28125,
      "reward_std": 0.3608423173427582,
      "rewards/accuracy_reward": 0.28125,
      "step": 81
    },
    {
      "advantages": 1.862645149230957e-08,
      "completion_length": 150.1875,
      "epoch": 0.0703862660944206,
      "grad_norm": 4.655648708343506,
      "kl": 0.1748046875,
      "learning_rate": 9.296137339055793e-07,
      "loss": 0.0175,
      "reward": 0.15625,
      "reward_mean": 0.15625,
      "reward_std": 0.3198433816432953,
      "rewards/accuracy_reward": 0.15625,
      "step": 82
    },
    {
      "advantages": 1.30385160446167e-08,
      "completion_length": 144.1875,
      "epoch": 0.07124463519313305,
      "grad_norm": 6.301512241363525,
      "kl": 0.140625,
      "learning_rate": 9.287553648068669e-07,
      "loss": 0.014,
      "reward": 0.375,
      "reward_mean": 0.375,
      "reward_std": 0.48503684997558594,
      "rewards/accuracy_reward": 0.375,
      "step": 83
    },
    {
      "advantages": 7.450580596923828e-09,
      "completion_length": 155.75,
      "epoch": 0.07210300429184549,
      "grad_norm": 4.552245616912842,
      "kl": 0.1494140625,
      "learning_rate": 9.278969957081545e-07,
      "loss": 0.015,
      "reward": 0.46875,
      "reward_mean": 0.46875,
      "reward_std": 0.5302791595458984,
      "rewards/accuracy_reward": 0.46875,
      "step": 84
    },
    {
      "advantages": 1.862645149230957e-08,
      "completion_length": 163.59375,
      "epoch": 0.07296137339055794,
      "grad_norm": 5.0817060470581055,
      "kl": 0.1953125,
      "learning_rate": 9.27038626609442e-07,
      "loss": 0.0195,
      "reward": 0.28125,
      "reward_mean": 0.28125,
      "reward_std": 0.4534739851951599,
      "rewards/accuracy_reward": 0.28125,
      "step": 85
    },
    {
      "advantages": 7.450580596923828e-09,
      "completion_length": 167.5,
      "epoch": 0.07381974248927038,
      "grad_norm": 3.2791318893432617,
      "kl": 0.16796875,
      "learning_rate": 9.261802575107296e-07,
      "loss": 0.0168,
      "reward": 0.0625,
      "reward_mean": 0.0625,
      "reward_std": 0.1767766922712326,
      "rewards/accuracy_reward": 0.0625,
      "step": 86
    },
    {
      "advantages": 5.587935447692871e-09,
      "completion_length": 152.03125,
      "epoch": 0.07467811158798283,
      "grad_norm": 5.571934700012207,
      "kl": 0.2421875,
      "learning_rate": 9.253218884120171e-07,
      "loss": 0.0242,
      "reward": 0.28125,
      "reward_mean": 0.28125,
      "reward_std": 0.3608423173427582,
      "rewards/accuracy_reward": 0.28125,
      "step": 87
    },
    {
      "advantages": 1.862645149230957e-08,
      "completion_length": 171.15625,
      "epoch": 0.07553648068669527,
      "grad_norm": 4.619121074676514,
      "kl": 0.2109375,
      "learning_rate": 9.244635193133047e-07,
      "loss": 0.0211,
      "reward": 0.15625,
      "reward_mean": 0.15625,
      "reward_std": 0.3808925747871399,
      "rewards/accuracy_reward": 0.15625,
      "step": 88
    },
    {
      "advantages": 3.725290298461914e-09,
      "completion_length": 153.03125,
      "epoch": 0.07639484978540773,
      "grad_norm": 4.043124198913574,
      "kl": 0.13671875,
      "learning_rate": 9.236051502145923e-07,
      "loss": 0.0137,
      "reward": 0.28125,
      "reward_mean": 0.28125,
      "reward_std": 0.3471629321575165,
      "rewards/accuracy_reward": 0.28125,
      "step": 89
    },
    {
      "advantages": 3.725290298461914e-09,
      "completion_length": 147.53125,
      "epoch": 0.07725321888412018,
      "grad_norm": 2.962092638015747,
      "kl": 0.1513671875,
      "learning_rate": 9.227467811158798e-07,
      "loss": 0.0152,
      "reward": 0.15625,
      "reward_mean": 0.15625,
      "reward_std": 0.22201895713806152,
      "rewards/accuracy_reward": 0.15625,
      "step": 90
    },
    {
      "advantages": 3.3527612686157227e-08,
      "completion_length": 176.9375,
      "epoch": 0.07811158798283262,
      "grad_norm": 5.911281585693359,
      "kl": 0.1484375,
      "learning_rate": 9.218884120171674e-07,
      "loss": 0.0148,
      "reward": 0.3125,
      "reward_mean": 0.3125,
      "reward_std": 0.49022960662841797,
      "rewards/accuracy_reward": 0.3125,
      "step": 91
    },
    {
      "advantages": 0.0,
      "completion_length": 160.875,
      "epoch": 0.07896995708154507,
      "grad_norm": 3.9937198162078857,
      "kl": 0.173828125,
      "learning_rate": 9.210300429184548e-07,
      "loss": 0.0174,
      "reward": 0.25,
      "reward_mean": 0.25,
      "reward_std": 0.2925041913986206,
      "rewards/accuracy_reward": 0.25,
      "step": 92
    },
    {
      "advantages": 1.1175870895385742e-08,
      "completion_length": 167.0,
      "epoch": 0.07982832618025751,
      "grad_norm": 3.8715927600860596,
      "kl": 0.134765625,
      "learning_rate": 9.201716738197424e-07,
      "loss": 0.0135,
      "reward": 0.25,
      "reward_mean": 0.25,
      "reward_std": 0.3514062464237213,
      "rewards/accuracy_reward": 0.25,
      "step": 93
    },
    {
      "advantages": 3.166496753692627e-08,
      "completion_length": 155.53125,
      "epoch": 0.08068669527896996,
      "grad_norm": 4.316589832305908,
      "kl": 0.1640625,
      "learning_rate": 9.193133047210299e-07,
      "loss": 0.0164,
      "reward": 0.28125,
      "reward_mean": 0.28125,
      "reward_std": 0.4765698313713074,
      "rewards/accuracy_reward": 0.28125,
      "step": 94
    },
    {
      "advantages": 2.60770320892334e-08,
      "completion_length": 158.15625,
      "epoch": 0.0815450643776824,
      "grad_norm": 4.352808952331543,
      "kl": 0.1416015625,
      "learning_rate": 9.184549356223176e-07,
      "loss": 0.0142,
      "reward": 0.25,
      "reward_mean": 0.25,
      "reward_std": 0.4355512857437134,
      "rewards/accuracy_reward": 0.25,
      "step": 95
    },
    {
      "advantages": 1.6763806343078613e-08,
      "completion_length": 142.0625,
      "epoch": 0.08240343347639485,
      "grad_norm": 6.171362400054932,
      "kl": 0.1787109375,
      "learning_rate": 9.175965665236051e-07,
      "loss": 0.0178,
      "reward": 0.15625,
      "reward_mean": 0.15625,
      "reward_std": 0.3061639666557312,
      "rewards/accuracy_reward": 0.15625,
      "step": 96
    },
    {
      "advantages": 1.30385160446167e-08,
      "completion_length": 163.78125,
      "epoch": 0.08326180257510729,
      "grad_norm": 3.2692599296569824,
      "kl": 0.1796875,
      "learning_rate": 9.167381974248927e-07,
      "loss": 0.0179,
      "reward": 0.125,
      "reward_mean": 0.125,
      "reward_std": 0.2177756428718567,
      "rewards/accuracy_reward": 0.125,
      "step": 97
    },
    {
      "advantages": 1.30385160446167e-08,
      "completion_length": 158.0625,
      "epoch": 0.08412017167381974,
      "grad_norm": 3.954564094543457,
      "kl": 0.123046875,
      "learning_rate": 9.158798283261803e-07,
      "loss": 0.0123,
      "reward": 0.25,
      "reward_mean": 0.25,
      "reward_std": 0.3514062464237213,
      "rewards/accuracy_reward": 0.25,
      "step": 98
    },
    {
      "advantages": 0.0,
      "completion_length": 140.09375,
      "epoch": 0.08497854077253218,
      "grad_norm": 4.733702659606934,
      "kl": 0.166015625,
      "learning_rate": 9.150214592274678e-07,
      "loss": 0.0166,
      "reward": 0.25,
      "reward_mean": 0.25,
      "reward_std": 0.292504221200943,
      "rewards/accuracy_reward": 0.25,
      "step": 99
    },
    {
      "advantages": 1.6763806343078613e-08,
      "completion_length": 139.0625,
      "epoch": 0.08583690987124463,
      "grad_norm": 4.053472995758057,
      "kl": 0.15625,
      "learning_rate": 9.141630901287554e-07,
      "loss": 0.0157,
      "reward": 0.15625,
      "reward_mean": 0.15625,
      "reward_std": 0.3061639666557312,
      "rewards/accuracy_reward": 0.15625,
      "step": 100
    },
    {
      "advantages": 2.421438694000244e-08,
      "completion_length": 135.96875,
      "epoch": 0.08669527896995709,
      "grad_norm": 5.235788822174072,
      "kl": 0.18359375,
      "learning_rate": 9.133047210300429e-07,
      "loss": 0.0184,
      "reward": 0.34375,
      "reward_mean": 0.34375,
      "reward_std": 0.4944729208946228,
      "rewards/accuracy_reward": 0.34375,
      "step": 101
    },
    {
      "advantages": 1.862645149230957e-09,
      "completion_length": 164.1875,
      "epoch": 0.08755364806866953,
      "grad_norm": 4.011264801025391,
      "kl": 0.158203125,
      "learning_rate": 9.124463519313305e-07,
      "loss": 0.0158,
      "reward": 0.25,
      "reward_mean": 0.25,
      "reward_std": 0.3335031569004059,
      "rewards/accuracy_reward": 0.25,
      "step": 102
    },
    {
      "advantages": 1.30385160446167e-08,
      "completion_length": 141.59375,
      "epoch": 0.08841201716738198,
      "grad_norm": 4.756768226623535,
      "kl": 0.12109375,
      "learning_rate": 9.115879828326179e-07,
      "loss": 0.0121,
      "reward": 0.375,
      "reward_mean": 0.375,
      "reward_std": 0.4492306709289551,
      "rewards/accuracy_reward": 0.375,
      "step": 103
    },
    {
      "advantages": 2.9802322387695312e-08,
      "completion_length": 157.4375,
      "epoch": 0.08927038626609442,
      "grad_norm": 4.457541465759277,
      "kl": 0.1767578125,
      "learning_rate": 9.107296137339055e-07,
      "loss": 0.0177,
      "reward": 0.28125,
      "reward_mean": 0.28125,
      "reward_std": 0.4628904461860657,
      "rewards/accuracy_reward": 0.28125,
      "step": 104
    },
    {
      "advantages": 1.4901161193847656e-08,
      "completion_length": 174.65625,
      "epoch": 0.09012875536480687,
      "grad_norm": 3.6339664459228516,
      "kl": 0.1484375,
      "learning_rate": 9.09871244635193e-07,
      "loss": 0.0148,
      "reward": 0.125,
      "reward_mean": 0.125,
      "reward_std": 0.292504221200943,
      "rewards/accuracy_reward": 0.125,
      "step": 105
    },
    {
      "advantages": 2.0489096641540527e-08,
      "completion_length": 160.625,
      "epoch": 0.09098712446351931,
      "grad_norm": 5.5795159339904785,
      "kl": 0.1474609375,
      "learning_rate": 9.090128755364806e-07,
      "loss": 0.0147,
      "reward": 0.1875,
      "reward_mean": 0.1875,
      "reward_std": 0.3945523500442505,
      "rewards/accuracy_reward": 0.1875,
      "step": 106
    },
    {
      "advantages": -1.862645149230957e-09,
      "completion_length": 175.09375,
      "epoch": 0.09184549356223176,
      "grad_norm": 4.58608341217041,
      "kl": 0.20703125,
      "learning_rate": 9.081545064377682e-07,
      "loss": 0.0207,
      "reward": 0.21875,
      "reward_mean": 0.21875,
      "reward_std": 0.24511480331420898,
      "rewards/accuracy_reward": 0.21875,
      "step": 107
    },
    {
      "advantages": 1.4901161193847656e-08,
      "completion_length": 164.40625,
      "epoch": 0.0927038626609442,
      "grad_norm": 4.733726978302002,
      "kl": 0.13671875,
      "learning_rate": 9.072961373390558e-07,
      "loss": 0.0137,
      "reward": 0.25,
      "reward_mean": 0.25,
      "reward_std": 0.4261348247528076,
      "rewards/accuracy_reward": 0.25,
      "step": 108
    },
    {
      "advantages": 1.30385160446167e-08,
      "completion_length": 137.53125,
      "epoch": 0.09356223175965665,
      "grad_norm": 4.860609531402588,
      "kl": 0.1376953125,
      "learning_rate": 9.064377682403434e-07,
      "loss": 0.0138,
      "reward": 0.375,
      "reward_mean": 0.375,
      "reward_std": 0.4492306709289551,
      "rewards/accuracy_reward": 0.375,
      "step": 109
    },
    {
      "advantages": 2.60770320892334e-08,
      "completion_length": 143.1875,
      "epoch": 0.0944206008583691,
      "grad_norm": 5.030947685241699,
      "kl": 0.216796875,
      "learning_rate": 9.055793991416309e-07,
      "loss": 0.0217,
      "reward": 0.25,
      "reward_mean": 0.25,
      "reward_std": 0.4492306709289551,
      "rewards/accuracy_reward": 0.25,
      "step": 110
    },
    {
      "advantages": 2.0489096641540527e-08,
      "completion_length": 154.6875,
      "epoch": 0.09527896995708154,
      "grad_norm": 5.2177534103393555,
      "kl": 0.138671875,
      "learning_rate": 9.047210300429185e-07,
      "loss": 0.0139,
      "reward": 0.4375,
      "reward_mean": 0.4375,
      "reward_std": 0.3335031569004059,
      "rewards/accuracy_reward": 0.4375,
      "step": 111
    },
    {
      "advantages": 9.313225746154785e-09,
      "completion_length": 127.625,
      "epoch": 0.096137339055794,
      "grad_norm": 5.624181747436523,
      "kl": 0.1787109375,
      "learning_rate": 9.03862660944206e-07,
      "loss": 0.0179,
      "reward": 0.46875,
      "reward_mean": 0.46875,
      "reward_std": 0.5302791595458984,
      "rewards/accuracy_reward": 0.46875,
      "step": 112
    },
    {
      "advantages": 7.450580596923828e-09,
      "completion_length": 153.84375,
      "epoch": 0.09699570815450644,
      "grad_norm": 3.743622303009033,
      "kl": 0.1396484375,
      "learning_rate": 9.030042918454935e-07,
      "loss": 0.0139,
      "reward": 0.1875,
      "reward_mean": 0.1875,
      "reward_std": 0.3104073107242584,
      "rewards/accuracy_reward": 0.1875,
      "step": 113
    },
    {
      "advantages": -1.1175870895385742e-08,
      "completion_length": 162.5625,
      "epoch": 0.09785407725321889,
      "grad_norm": 4.363542556762695,
      "kl": 0.181640625,
      "learning_rate": 9.02145922746781e-07,
      "loss": 0.0182,
      "reward": 0.375,
      "reward_mean": 0.375,
      "reward_std": 0.3745020925998688,
      "rewards/accuracy_reward": 0.375,
      "step": 114
    },
    {
      "advantages": 7.450580596923828e-09,
      "completion_length": 154.3125,
      "epoch": 0.09871244635193133,
      "grad_norm": 5.912485599517822,
      "kl": 0.169921875,
      "learning_rate": 9.012875536480686e-07,
      "loss": 0.017,
      "reward": 0.34375,
      "reward_mean": 0.34375,
      "reward_std": 0.3608423173427582,
      "rewards/accuracy_reward": 0.34375,
      "step": 115
    },
    {
      "advantages": 1.4901161193847656e-08,
      "completion_length": 153.03125,
      "epoch": 0.09957081545064378,
      "grad_norm": 4.207987308502197,
      "kl": 0.11962890625,
      "learning_rate": 9.004291845493562e-07,
      "loss": 0.0119,
      "reward": 0.25,
      "reward_mean": 0.25,
      "reward_std": 0.4261348247528076,
      "rewards/accuracy_reward": 0.25,
      "step": 116
    },
    {
      "advantages": -5.587935447692871e-09,
      "completion_length": 149.09375,
      "epoch": 0.10042918454935622,
      "grad_norm": 4.171358108520508,
      "kl": 0.181640625,
      "learning_rate": 8.995708154506437e-07,
      "loss": 0.0181,
      "reward": 0.1875,
      "reward_mean": 0.1875,
      "reward_std": 0.2177756428718567,
      "rewards/accuracy_reward": 0.1875,
      "step": 117
    },
    {
      "advantages": 1.6763806343078613e-08,
      "completion_length": 135.25,
      "epoch": 0.10128755364806867,
      "grad_norm": 3.7221720218658447,
      "kl": 0.1357421875,
      "learning_rate": 8.987124463519313e-07,
      "loss": 0.0136,
      "reward": 0.15625,
      "reward_mean": 0.15625,
      "reward_std": 0.3061639964580536,
      "rewards/accuracy_reward": 0.15625,
      "step": 118
    },
    {
      "advantages": 1.4901161193847656e-08,
      "completion_length": 154.40625,
      "epoch": 0.10214592274678111,
      "grad_norm": 7.550022602081299,
      "kl": 0.318359375,
      "learning_rate": 8.978540772532188e-07,
      "loss": 0.0318,
      "reward": 0.25,
      "reward_mean": 0.25,
      "reward_std": 0.3650856614112854,
      "rewards/accuracy_reward": 0.25,
      "step": 119
    },
    {
      "advantages": 2.9802322387695312e-08,
      "completion_length": 139.15625,
      "epoch": 0.10300429184549356,
      "grad_norm": 5.675111293792725,
      "kl": 0.1552734375,
      "learning_rate": 8.969957081545064e-07,
      "loss": 0.0155,
      "reward": 0.28125,
      "reward_mean": 0.28125,
      "reward_std": 0.4628904461860657,
      "rewards/accuracy_reward": 0.28125,
      "step": 120
    },
    {
      "advantages": 1.1175870895385742e-08,
      "completion_length": 138.125,
      "epoch": 0.10386266094420601,
      "grad_norm": 4.683244228363037,
      "kl": 0.142578125,
      "learning_rate": 8.96137339055794e-07,
      "loss": 0.0142,
      "reward": 0.34375,
      "reward_mean": 0.34375,
      "reward_std": 0.4628904461860657,
      "rewards/accuracy_reward": 0.34375,
      "step": 121
    },
    {
      "advantages": -1.862645149230957e-09,
      "completion_length": 129.4375,
      "epoch": 0.10472103004291845,
      "grad_norm": 4.976833820343018,
      "kl": 0.1767578125,
      "learning_rate": 8.952789699570816e-07,
      "loss": 0.0177,
      "reward": 0.46875,
      "reward_mean": 0.46875,
      "reward_std": 0.5038893818855286,
      "rewards/accuracy_reward": 0.46875,
      "step": 122
    },
    {
      "advantages": 1.862645149230957e-09,
      "completion_length": 153.03125,
      "epoch": 0.1055793991416309,
      "grad_norm": 4.796186447143555,
      "kl": 0.1201171875,
      "learning_rate": 8.94420600858369e-07,
      "loss": 0.012,
      "reward": 0.375,
      "reward_mean": 0.375,
      "reward_std": 0.4671337604522705,
      "rewards/accuracy_reward": 0.375,
      "step": 123
    },
    {
      "advantages": -1.862645149230957e-09,
      "completion_length": 109.25,
      "epoch": 0.10643776824034334,
      "grad_norm": 4.939033031463623,
      "kl": 0.2236328125,
      "learning_rate": 8.935622317596566e-07,
      "loss": 0.0223,
      "reward": 0.46875,
      "reward_mean": 0.46875,
      "reward_std": 0.3061639964580536,
      "rewards/accuracy_reward": 0.46875,
      "step": 124
    },
    {
      "advantages": 1.30385160446167e-08,
      "completion_length": 131.90625,
      "epoch": 0.1072961373390558,
      "grad_norm": 6.3975114822387695,
      "kl": 0.1953125,
      "learning_rate": 8.927038626609442e-07,
      "loss": 0.0195,
      "reward": 0.375,
      "reward_mean": 0.375,
      "reward_std": 0.48503684997558594,
      "rewards/accuracy_reward": 0.375,
      "step": 125
    },
    {
      "advantages": 1.30385160446167e-08,
      "completion_length": 108.375,
      "epoch": 0.10815450643776824,
      "grad_norm": 5.976772308349609,
      "kl": 0.1650390625,
      "learning_rate": 8.918454935622317e-07,
      "loss": 0.0165,
      "reward": 0.375,
      "reward_mean": 0.375,
      "reward_std": 0.49022960662841797,
      "rewards/accuracy_reward": 0.375,
      "step": 126
    },
    {
      "advantages": 1.4901161193847656e-08,
      "completion_length": 129.25,
      "epoch": 0.10901287553648069,
      "grad_norm": 5.26469087600708,
      "kl": 0.14453125,
      "learning_rate": 8.909871244635193e-07,
      "loss": 0.0144,
      "reward": 0.375,
      "reward_mean": 0.375,
      "reward_std": 0.4629100561141968,
      "rewards/accuracy_reward": 0.375,
      "step": 127
    },
    {
      "advantages": -1.1175870895385742e-08,
      "completion_length": 128.3125,
      "epoch": 0.10987124463519313,
      "grad_norm": 5.042163848876953,
      "kl": 0.1328125,
      "learning_rate": 8.901287553648068e-07,
      "loss": 0.0133,
      "reward": 0.5,
      "reward_mean": 0.5,
      "reward_std": 0.5081326961517334,
      "rewards/accuracy_reward": 0.5,
      "step": 128
    },
    {
      "advantages": 1.862645149230957e-08,
      "completion_length": 150.875,
      "epoch": 0.11072961373390558,
      "grad_norm": 4.708669662475586,
      "kl": 0.12353515625,
      "learning_rate": 8.892703862660944e-07,
      "loss": 0.0124,
      "reward": 0.15625,
      "reward_mean": 0.15625,
      "reward_std": 0.3808925747871399,
      "rewards/accuracy_reward": 0.15625,
      "step": 129
    },
    {
      "advantages": 0.0,
      "completion_length": 120.40625,
      "epoch": 0.11158798283261803,
      "grad_norm": 5.916502952575684,
      "kl": 0.1708984375,
      "learning_rate": 8.884120171673819e-07,
      "loss": 0.0171,
      "reward": 0.34375,
      "reward_mean": 0.34375,
      "reward_std": 0.4397946000099182,
      "rewards/accuracy_reward": 0.34375,
      "step": 130
    },
    {
      "advantages": 1.4901161193847656e-08,
      "completion_length": 136.90625,
      "epoch": 0.11244635193133047,
      "grad_norm": 4.78549861907959,
      "kl": 0.1708984375,
      "learning_rate": 8.875536480686695e-07,
      "loss": 0.0171,
      "reward": 0.40625,
      "reward_mean": 0.40625,
      "reward_std": 0.5038893818855286,
      "rewards/accuracy_reward": 0.40625,
      "step": 131
    },
    {
      "advantages": -9.313225746154785e-09,
      "completion_length": 109.71875,
      "epoch": 0.11330472103004292,
      "grad_norm": 6.360599040985107,
      "kl": 0.25390625,
      "learning_rate": 8.86695278969957e-07,
      "loss": 0.0254,
      "reward": 0.40625,
      "reward_mean": 0.40625,
      "reward_std": 0.3608423173427582,
      "rewards/accuracy_reward": 0.40625,
      "step": 132
    },
    {
      "advantages": 1.4901161193847656e-08,
      "completion_length": 135.375,
      "epoch": 0.11416309012875536,
      "grad_norm": 4.24266242980957,
      "kl": 0.1650390625,
      "learning_rate": 8.858369098712447e-07,
      "loss": 0.0165,
      "reward": 0.28125,
      "reward_mean": 0.28125,
      "reward_std": 0.378745436668396,
      "rewards/accuracy_reward": 0.28125,
      "step": 133
    },
    {
      "advantages": 0.0,
      "completion_length": 97.59375,
      "epoch": 0.11502145922746781,
      "grad_norm": 6.252864360809326,
      "kl": 0.2177734375,
      "learning_rate": 8.849785407725322e-07,
      "loss": 0.0217,
      "reward": 0.5,
      "reward_mean": 0.5,
      "reward_std": 0.4082317352294922,
      "rewards/accuracy_reward": 0.5,
      "step": 134
    },
    {
      "advantages": 2.0489096641540527e-08,
      "completion_length": 118.8125,
      "epoch": 0.11587982832618025,
      "grad_norm": 5.049542427062988,
      "kl": 0.2060546875,
      "learning_rate": 8.841201716738197e-07,
      "loss": 0.0206,
      "reward": 0.1875,
      "reward_mean": 0.1875,
      "reward_std": 0.3335031569004059,
      "rewards/accuracy_reward": 0.1875,
      "step": 135
    },
    {
      "advantages": -3.725290298461914e-09,
      "completion_length": 120.5625,
      "epoch": 0.1167381974248927,
      "grad_norm": 5.062263011932373,
      "kl": 0.189453125,
      "learning_rate": 8.832618025751073e-07,
      "loss": 0.0189,
      "reward": 0.59375,
      "reward_mean": 0.59375,
      "reward_std": 0.4807935357093811,
      "rewards/accuracy_reward": 0.59375,
      "step": 136
    },
    {
      "advantages": 2.2351741790771484e-08,
      "completion_length": 122.96875,
      "epoch": 0.11759656652360514,
      "grad_norm": 6.447961330413818,
      "kl": 0.19921875,
      "learning_rate": 8.824034334763948e-07,
      "loss": 0.0199,
      "reward": 0.34375,
      "reward_mean": 0.34375,
      "reward_std": 0.4944729208946228,
      "rewards/accuracy_reward": 0.34375,
      "step": 137
    },
    {
      "advantages": 1.862645149230957e-08,
      "completion_length": 125.3125,
      "epoch": 0.1184549356223176,
      "grad_norm": 6.920899868011475,
      "kl": 0.208984375,
      "learning_rate": 8.815450643776824e-07,
      "loss": 0.0209,
      "reward": 0.28125,
      "reward_mean": 0.28125,
      "reward_std": 0.4534739851951599,
      "rewards/accuracy_reward": 0.28125,
      "step": 138
    },
    {
      "advantages": -3.725290298461914e-09,
      "completion_length": 128.4375,
      "epoch": 0.11931330472103004,
      "grad_norm": 4.800882339477539,
      "kl": 0.173828125,
      "learning_rate": 8.806866952789699e-07,
      "loss": 0.0173,
      "reward": 0.5,
      "reward_mean": 0.5,
      "reward_std": 0.4492306709289551,
      "rewards/accuracy_reward": 0.5,
      "step": 139
    },
    {
      "advantages": 1.862645149230957e-08,
      "completion_length": 108.1875,
      "epoch": 0.12017167381974249,
      "grad_norm": 22.795358657836914,
      "kl": 0.419921875,
      "learning_rate": 8.798283261802575e-07,
      "loss": 0.042,
      "reward": 0.15625,
      "reward_mean": 0.15625,
      "reward_std": 0.3808925747871399,
      "rewards/accuracy_reward": 0.15625,
      "step": 140
    },
    {
      "advantages": 1.1175870895385742e-08,
      "completion_length": 141.84375,
      "epoch": 0.12103004291845494,
      "grad_norm": 6.3895745277404785,
      "kl": 0.1962890625,
      "learning_rate": 8.78969957081545e-07,
      "loss": 0.0196,
      "reward": 0.375,
      "reward_mean": 0.375,
      "reward_std": 0.47655022144317627,
      "rewards/accuracy_reward": 0.375,
      "step": 141
    },
    {
      "advantages": 7.450580596923828e-09,
      "completion_length": 115.3125,
      "epoch": 0.12188841201716738,
      "grad_norm": 4.0062103271484375,
      "kl": 0.1669921875,
      "learning_rate": 8.781115879828326e-07,
      "loss": 0.0167,
      "reward": 0.1875,
      "reward_mean": 0.1875,
      "reward_std": 0.249358132481575,
      "rewards/accuracy_reward": 0.1875,
      "step": 142
    },
    {
      "advantages": -3.725290298461914e-09,
      "completion_length": 131.84375,
      "epoch": 0.12274678111587983,
      "grad_norm": 6.299867630004883,
      "kl": 0.2451171875,
      "learning_rate": 8.772532188841201e-07,
      "loss": 0.0246,
      "reward": 0.625,
      "reward_mean": 0.625,
      "reward_std": 0.4671337604522705,
      "rewards/accuracy_reward": 0.625,
      "step": 143
    },
    {
      "advantages": -1.30385160446167e-08,
      "completion_length": 134.3125,
      "epoch": 0.12360515021459227,
      "grad_norm": 5.634101390838623,
      "kl": 0.18359375,
      "learning_rate": 8.763948497854076e-07,
      "loss": 0.0184,
      "reward": 0.625,
      "reward_mean": 0.625,
      "reward_std": 0.4850368797779083,
      "rewards/accuracy_reward": 0.625,
      "step": 144
    },
    {
      "advantages": -2.0489096641540527e-08,
      "completion_length": 124.90625,
      "epoch": 0.12446351931330472,
      "grad_norm": 5.229036331176758,
      "kl": 0.1435546875,
      "learning_rate": 8.755364806866952e-07,
      "loss": 0.0143,
      "reward": 0.53125,
      "reward_mean": 0.53125,
      "reward_std": 0.4628904461860657,
      "rewards/accuracy_reward": 0.53125,
      "step": 145
    },
    {
      "advantages": 7.450580596923828e-09,
      "completion_length": 121.59375,
      "epoch": 0.12532188841201716,
      "grad_norm": 4.288768768310547,
      "kl": 0.1875,
      "learning_rate": 8.746781115879828e-07,
      "loss": 0.0188,
      "reward": 0.3125,
      "reward_mean": 0.3125,
      "reward_std": 0.3745020925998688,
      "rewards/accuracy_reward": 0.3125,
      "step": 146
    },
    {
      "advantages": 2.60770320892334e-08,
      "completion_length": 145.53125,
      "epoch": 0.12618025751072962,
      "grad_norm": 4.884040355682373,
      "kl": 0.189453125,
      "learning_rate": 8.738197424892704e-07,
      "loss": 0.0189,
      "reward": 0.375,
      "reward_mean": 0.375,
      "reward_std": 0.5081326961517334,
      "rewards/accuracy_reward": 0.375,
      "step": 147
    },
    {
      "advantages": 2.60770320892334e-08,
      "completion_length": 129.34375,
      "epoch": 0.12703862660944207,
      "grad_norm": 5.560000419616699,
      "kl": 0.1904296875,
      "learning_rate": 8.729613733905579e-07,
      "loss": 0.019,
      "reward": 0.375,
      "reward_mean": 0.375,
      "reward_std": 0.5081326961517334,
      "rewards/accuracy_reward": 0.375,
      "step": 148
    },
    {
      "advantages": 9.313225746154785e-09,
      "completion_length": 125.46875,
      "epoch": 0.1278969957081545,
      "grad_norm": 5.247032642364502,
      "kl": 0.14453125,
      "learning_rate": 8.721030042918455e-07,
      "loss": 0.0145,
      "reward": 0.46875,
      "reward_mean": 0.46875,
      "reward_std": 0.5302791595458984,
      "rewards/accuracy_reward": 0.46875,
      "step": 149
    },
    {
      "advantages": 1.1175870895385742e-08,
      "completion_length": 123.5625,
      "epoch": 0.12875536480686695,
      "grad_norm": 6.392980098724365,
      "kl": 0.193359375,
      "learning_rate": 8.71244635193133e-07,
      "loss": 0.0193,
      "reward": 0.5,
      "reward_mean": 0.5,
      "reward_std": 0.5081326961517334,
      "rewards/accuracy_reward": 0.5,
      "step": 150
    },
    {
      "advantages": -1.4901161193847656e-08,
      "completion_length": 125.84375,
      "epoch": 0.1296137339055794,
      "grad_norm": 5.506472587585449,
      "kl": 0.2099609375,
      "learning_rate": 8.703862660944206e-07,
      "loss": 0.021,
      "reward": 0.5625,
      "reward_mean": 0.5625,
      "reward_std": 0.5260357856750488,
      "rewards/accuracy_reward": 0.5625,
      "step": 151
    },
    {
      "advantages": 5.587935447692871e-09,
      "completion_length": 111.53125,
      "epoch": 0.13047210300429185,
      "grad_norm": 8.247237205505371,
      "kl": 0.2578125,
      "learning_rate": 8.695278969957082e-07,
      "loss": 0.0258,
      "reward": 0.40625,
      "reward_mean": 0.40625,
      "reward_std": 0.4944729208946228,
      "rewards/accuracy_reward": 0.40625,
      "step": 152
    },
    {
      "advantages": -1.862645149230957e-08,
      "completion_length": 135.6875,
      "epoch": 0.1313304721030043,
      "grad_norm": 5.90345573425293,
      "kl": 0.197265625,
      "learning_rate": 8.686695278969956e-07,
      "loss": 0.0197,
      "reward": 0.5625,
      "reward_mean": 0.5625,
      "reward_std": 0.49022960662841797,
      "rewards/accuracy_reward": 0.5625,
      "step": 153
    },
    {
      "advantages": 1.4901161193847656e-08,
      "completion_length": 134.53125,
      "epoch": 0.13218884120171673,
      "grad_norm": 5.609891891479492,
      "kl": 0.171875,
      "learning_rate": 8.678111587982832e-07,
      "loss": 0.0172,
      "reward": 0.40625,
      "reward_mean": 0.40625,
      "reward_std": 0.5123760104179382,
      "rewards/accuracy_reward": 0.40625,
      "step": 154
    },
    {
      "advantages": -1.1175870895385742e-08,
      "completion_length": 122.59375,
      "epoch": 0.13304721030042918,
      "grad_norm": 4.156961917877197,
      "kl": 0.1435546875,
      "learning_rate": 8.669527896995707e-07,
      "loss": 0.0143,
      "reward": 0.53125,
      "reward_mean": 0.53125,
      "reward_std": 0.3377464711666107,
      "rewards/accuracy_reward": 0.53125,
      "step": 155
    },
    {
      "advantages": -1.862645149230957e-09,
      "completion_length": 104.71875,
      "epoch": 0.13390557939914163,
      "grad_norm": 7.677206993103027,
      "kl": 0.291015625,
      "learning_rate": 8.660944206008583e-07,
      "loss": 0.0291,
      "reward": 0.21875,
      "reward_mean": 0.21875,
      "reward_std": 0.24511480331420898,
      "rewards/accuracy_reward": 0.21875,
      "step": 156
    },
    {
      "advantages": -1.30385160446167e-08,
      "completion_length": 130.375,
      "epoch": 0.13476394849785409,
      "grad_norm": 4.416824817657471,
      "kl": 0.166015625,
      "learning_rate": 8.652360515021458e-07,
      "loss": 0.0166,
      "reward": 0.625,
      "reward_mean": 0.625,
      "reward_std": 0.4492306709289551,
      "rewards/accuracy_reward": 0.625,
      "step": 157
    },
    {
      "advantages": 1.6763806343078613e-08,
      "completion_length": 119.46875,
      "epoch": 0.1356223175965665,
      "grad_norm": 3.8171494007110596,
      "kl": 0.1728515625,
      "learning_rate": 8.643776824034334e-07,
      "loss": 0.0173,
      "reward": 0.15625,
      "reward_mean": 0.15625,
      "reward_std": 0.24511480331420898,
      "rewards/accuracy_reward": 0.15625,
      "step": 158
    },
    {
      "advantages": 9.313225746154785e-09,
      "completion_length": 107.59375,
      "epoch": 0.13648068669527896,
      "grad_norm": 5.097965717315674,
      "kl": 0.1787109375,
      "learning_rate": 8.63519313304721e-07,
      "loss": 0.0179,
      "reward": 0.59375,
      "reward_mean": 0.59375,
      "reward_std": 0.38816186785697937,
      "rewards/accuracy_reward": 0.59375,
      "step": 159
    },
    {
      "advantages": 1.30385160446167e-08,
      "completion_length": 110.875,
      "epoch": 0.13733905579399142,
      "grad_norm": 5.293290138244629,
      "kl": 0.169921875,
      "learning_rate": 8.626609442060086e-07,
      "loss": 0.017,
      "reward": 0.375,
      "reward_mean": 0.375,
      "reward_std": 0.4492306709289551,
      "rewards/accuracy_reward": 0.375,
      "step": 160
    }
  ],
  "logging_steps": 1.0,
  "max_steps": 1165,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 1,
  "save_steps": 10,
  "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
}