{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 3.16,
  "eval_steps": 500,
  "global_step": 100,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "completion_length": 130.62890625,
      "epoch": 0.001587459073320766,
      "grad_norm": 0.95703125,
      "kl": 0.0,
      "learning_rate": 1e-05,
      "loss": 0.0,
      "reward": 1.0234375,
      "reward_std": 0.620339147746563,
      "rewards/correctness_reward_func": 1.0234375,
      "step": 1
    },
    {
      "completion_length": 150.61328125,
      "epoch": 0.003174918146641532,
      "grad_norm": 0.62109375,
      "kl": 0.061338113620877266,
      "learning_rate": 1e-05,
      "loss": 0.0003,
      "reward": 1.2734375,
      "reward_std": 0.3329593911767006,
      "rewards/correctness_reward_func": 1.2734375,
      "step": 2
    },
    {
      "completion_length": 171.515625,
      "epoch": 0.0047623772199622974,
      "grad_norm": 0.609375,
      "kl": 0.08757861610502005,
      "learning_rate": 1e-05,
      "loss": 0.0004,
      "reward": 1.25,
      "reward_std": 0.20200317353010178,
      "rewards/correctness_reward_func": 1.25,
      "step": 3
    },
    {
      "completion_length": 179.25390625,
      "epoch": 0.006349836293283064,
      "grad_norm": 14.625,
      "kl": 0.6993596660904586,
      "learning_rate": 1e-05,
      "loss": 0.0035,
      "reward": 1.265625,
      "reward_std": 0.2416265867650509,
      "rewards/correctness_reward_func": 1.265625,
      "step": 4
    },
    {
      "completion_length": 189.1015625,
      "epoch": 0.00793729536660383,
      "grad_norm": 0.484375,
      "kl": 0.053785258904099464,
      "learning_rate": 1e-05,
      "loss": 0.0003,
      "reward": 1.0703125,
      "reward_std": 0.2572515867650509,
      "rewards/correctness_reward_func": 1.0703125,
      "step": 5
    },
    {
      "completion_length": 234.6875,
      "epoch": 0.009524754439924595,
      "grad_norm": 0.59765625,
      "kl": 0.041797632817178965,
      "learning_rate": 1e-05,
      "loss": 0.0002,
      "reward": 0.984375,
      "reward_std": 0.3895031735301018,
      "rewards/correctness_reward_func": 0.984375,
      "step": 6
    },
    {
      "completion_length": 190.66015625,
      "epoch": 0.011112213513245362,
      "grad_norm": 0.8359375,
      "kl": 0.05031784961465746,
      "learning_rate": 1e-05,
      "loss": 0.0003,
      "reward": 1.0,
      "reward_std": 0.4339609779417515,
      "rewards/correctness_reward_func": 1.0,
      "step": 7
    },
    {
      "completion_length": 203.046875,
      "epoch": 0.012699672586566128,
      "grad_norm": 0.52734375,
      "kl": 0.05922417342662811,
      "learning_rate": 1e-05,
      "loss": 0.0003,
      "reward": 1.0625,
      "reward_std": 0.34987976029515266,
      "rewards/correctness_reward_func": 1.0625,
      "step": 8
    },
    {
      "completion_length": 231.3203125,
      "epoch": 0.014287131659886893,
      "grad_norm": 0.412109375,
      "kl": 0.038438764633610845,
      "learning_rate": 1e-05,
      "loss": 0.0002,
      "reward": 1.1015625,
      "reward_std": 0.3245859779417515,
      "rewards/correctness_reward_func": 1.1015625,
      "step": 9
    },
    {
      "completion_length": 228.765625,
      "epoch": 0.01587459073320766,
      "grad_norm": 0.466796875,
      "kl": 0.06533525174017996,
      "learning_rate": 1e-05,
      "loss": 0.0003,
      "reward": 1.0,
      "reward_std": 0.2777109779417515,
      "rewards/correctness_reward_func": 1.0,
      "step": 10
    },
    {
      "completion_length": 217.0546875,
      "epoch": 0.017462049806528424,
      "grad_norm": 0.388671875,
      "kl": 0.04172722063958645,
      "learning_rate": 1e-05,
      "loss": 0.0002,
      "reward": 1.1796875,
      "reward_std": 0.2885015867650509,
      "rewards/correctness_reward_func": 1.1796875,
      "step": 11
    },
    {
      "completion_length": 279.97265625,
      "epoch": 0.01904950887984919,
      "grad_norm": 0.578125,
      "kl": 0.05215206788852811,
      "learning_rate": 1e-05,
      "loss": 0.0003,
      "reward": 0.984375,
      "reward_std": 0.36308756470680237,
      "rewards/correctness_reward_func": 0.984375,
      "step": 12
    },
    {
      "completion_length": 288.18359375,
      "epoch": 0.02063696795316996,
      "grad_norm": 0.419921875,
      "kl": 0.04899335908703506,
      "learning_rate": 1e-05,
      "loss": 0.0002,
      "reward": 0.9140625,
      "reward_std": 0.32942036911845207,
      "rewards/correctness_reward_func": 0.9140625,
      "step": 13
    },
    {
      "completion_length": 280.36328125,
      "epoch": 0.022224427026490724,
      "grad_norm": 0.400390625,
      "kl": 0.0435270715970546,
      "learning_rate": 1e-05,
      "loss": 0.0002,
      "reward": 0.8984375,
      "reward_std": 0.2572515830397606,
      "rewards/correctness_reward_func": 0.8984375,
      "step": 14
    },
    {
      "completion_length": 252.9453125,
      "epoch": 0.02381188609981149,
      "grad_norm": 0.99609375,
      "kl": 0.07505812356248498,
      "learning_rate": 1e-05,
      "loss": 0.0004,
      "reward": 1.0234375,
      "reward_std": 0.30300476029515266,
      "rewards/correctness_reward_func": 1.0234375,
      "step": 15
    },
    {
      "completion_length": 311.01171875,
      "epoch": 0.025399345173132255,
      "grad_norm": 0.4375,
      "kl": 0.09632595372386277,
      "learning_rate": 1e-05,
      "loss": 0.0005,
      "reward": 0.9921875,
      "reward_std": 0.32104695588350296,
      "rewards/correctness_reward_func": 0.9921875,
      "step": 16
    },
    {
      "completion_length": 278.70703125,
      "epoch": 0.02698680424645302,
      "grad_norm": 0.478515625,
      "kl": 0.04982293304055929,
      "learning_rate": 1e-05,
      "loss": 0.0002,
      "reward": 0.8671875,
      "reward_std": 0.34746256470680237,
      "rewards/correctness_reward_func": 0.8671875,
      "step": 17
    },
    {
      "completion_length": 276.03515625,
      "epoch": 0.028574263319773786,
      "grad_norm": 0.48046875,
      "kl": 0.058841129299253225,
      "learning_rate": 1e-05,
      "loss": 0.0003,
      "reward": 0.9140625,
      "reward_std": 0.34746256470680237,
      "rewards/correctness_reward_func": 0.9140625,
      "step": 18
    },
    {
      "completion_length": 299.88671875,
      "epoch": 0.030161722393094552,
      "grad_norm": 0.43359375,
      "kl": 0.05037166504189372,
      "learning_rate": 1e-05,
      "loss": 0.0003,
      "reward": 0.96875,
      "reward_std": 0.31862976029515266,
      "rewards/correctness_reward_func": 0.96875,
      "step": 19
    },
    {
      "completion_length": 302.046875,
      "epoch": 0.03174918146641532,
      "grad_norm": 0.4140625,
      "kl": 0.051571789546869695,
      "learning_rate": 1e-05,
      "loss": 0.0003,
      "reward": 0.8515625,
      "reward_std": 0.37033915519714355,
      "rewards/correctness_reward_func": 0.8515625,
      "step": 20
    },
    {
      "completion_length": 384.6328125,
      "epoch": 0.03333664053973608,
      "grad_norm": 0.31640625,
      "kl": 0.04484600038267672,
      "learning_rate": 1e-05,
      "loss": 0.0002,
      "reward": 0.6875,
      "reward_std": 0.2416265867650509,
      "rewards/correctness_reward_func": 0.6875,
      "step": 21
    },
    {
      "completion_length": 326.0078125,
      "epoch": 0.03492409961305685,
      "grad_norm": 0.4296875,
      "kl": 0.04701656638644636,
      "learning_rate": 1e-05,
      "loss": 0.0002,
      "reward": 0.8984375,
      "reward_std": 0.4183359779417515,
      "rewards/correctness_reward_func": 0.8984375,
      "step": 22
    },
    {
      "completion_length": 298.75390625,
      "epoch": 0.036511558686377614,
      "grad_norm": 0.384765625,
      "kl": 0.05655176565051079,
      "learning_rate": 1e-05,
      "loss": 0.0003,
      "reward": 0.90625,
      "reward_std": 0.3270031735301018,
      "rewards/correctness_reward_func": 0.90625,
      "step": 23
    },
    {
      "completion_length": 371.8984375,
      "epoch": 0.03809901775969838,
      "grad_norm": 0.380859375,
      "kl": 0.06988459394779056,
      "learning_rate": 1e-05,
      "loss": 0.0003,
      "reward": 0.8203125,
      "reward_std": 0.30783915147185326,
      "rewards/correctness_reward_func": 0.8203125,
      "step": 24
    },
    {
      "completion_length": 367.21484375,
      "epoch": 0.03968647683301915,
      "grad_norm": 0.400390625,
      "kl": 0.061523064388893545,
      "learning_rate": 1e-05,
      "loss": 0.0003,
      "reward": 0.8203125,
      "reward_std": 0.2572515867650509,
      "rewards/correctness_reward_func": 0.8203125,
      "step": 25
    },
    {
      "completion_length": 317.98828125,
      "epoch": 0.04127393590633992,
      "grad_norm": 0.419921875,
      "kl": 0.05494794622063637,
      "learning_rate": 1e-05,
      "loss": 0.0003,
      "reward": 1.078125,
      "reward_std": 0.2909187823534012,
      "rewards/correctness_reward_func": 1.078125,
      "step": 26
    },
    {
      "completion_length": 359.47265625,
      "epoch": 0.04286139497966068,
      "grad_norm": 0.337890625,
      "kl": 0.052700593834742904,
      "learning_rate": 1e-05,
      "loss": 0.0003,
      "reward": 0.6953125,
      "reward_std": 0.3113781735301018,
      "rewards/correctness_reward_func": 0.6953125,
      "step": 27
    },
    {
      "completion_length": 359.359375,
      "epoch": 0.04444885405298145,
      "grad_norm": 0.392578125,
      "kl": 0.06072323571424931,
      "learning_rate": 1e-05,
      "loss": 0.0003,
      "reward": 0.6796875,
      "reward_std": 0.2620859779417515,
      "rewards/correctness_reward_func": 0.6796875,
      "step": 28
    },
    {
      "completion_length": 309.65625,
      "epoch": 0.046036313126302214,
      "grad_norm": 0.392578125,
      "kl": 0.06208631256595254,
      "learning_rate": 1e-05,
      "loss": 0.0003,
      "reward": 0.7578125,
      "reward_std": 0.34746256470680237,
      "rewards/correctness_reward_func": 0.7578125,
      "step": 29
    },
    {
      "completion_length": 324.359375,
      "epoch": 0.04762377219962298,
      "grad_norm": 0.419921875,
      "kl": 0.054743261309340596,
      "learning_rate": 1e-05,
      "loss": 0.0003,
      "reward": 0.7421875,
      "reward_std": 0.2752937823534012,
      "rewards/correctness_reward_func": 0.7421875,
      "step": 30
    },
    {
      "completion_length": 336.78125,
      "epoch": 0.049211231272943745,
      "grad_norm": 0.62890625,
      "kl": 0.09580778249073774,
      "learning_rate": 1e-05,
      "loss": 0.0005,
      "reward": 0.9140625,
      "reward_std": 0.3113781735301018,
      "rewards/correctness_reward_func": 0.9140625,
      "step": 31
    },
    {
      "completion_length": 382.821875,
      "epoch": 1.032,
      "grad_norm": 0.498046875,
      "kl": 0.05092890365049243,
      "learning_rate": 1e-05,
      "loss": 0.0003,
      "reward": 0.5125,
      "reward_std": 0.3471687823534012,
      "rewards/correctness_reward_func": 0.5125,
      "step": 32
    },
    {
      "completion_length": 293.07421875,
      "epoch": 1.064,
      "grad_norm": 0.5546875,
      "kl": 0.10762864979915321,
      "learning_rate": 1e-05,
      "loss": 0.0005,
      "reward": 0.875,
      "reward_std": 0.28254536911845207,
      "rewards/correctness_reward_func": 0.875,
      "step": 33
    },
    {
      "completion_length": 202.1875,
      "epoch": 1.096,
      "grad_norm": 2.03125,
      "kl": 0.1804919212590903,
      "learning_rate": 1e-05,
      "loss": 0.0009,
      "reward": 0.953125,
      "reward_std": 0.43042195588350296,
      "rewards/correctness_reward_func": 0.953125,
      "step": 34
    },
    {
      "completion_length": 298.62890625,
      "epoch": 1.1280000000000001,
      "grad_norm": 0.85546875,
      "kl": 0.14085538033396006,
      "learning_rate": 1e-05,
      "loss": 0.0007,
      "reward": 0.640625,
      "reward_std": 0.36792195588350296,
      "rewards/correctness_reward_func": 0.640625,
      "step": 35
    },
    {
      "completion_length": 274.4140625,
      "epoch": 1.16,
      "grad_norm": 0.59765625,
      "kl": 0.07066078553907573,
      "learning_rate": 1e-05,
      "loss": 0.0004,
      "reward": 0.859375,
      "reward_std": 0.4339609779417515,
      "rewards/correctness_reward_func": 0.859375,
      "step": 36
    },
    {
      "completion_length": 285.171875,
      "epoch": 1.192,
      "grad_norm": 0.58203125,
      "kl": 0.06347297015599906,
      "learning_rate": 1e-05,
      "loss": 0.0003,
      "reward": 0.796875,
      "reward_std": 0.4423343911767006,
      "rewards/correctness_reward_func": 0.796875,
      "step": 37
    },
    {
      "completion_length": 263.57421875,
      "epoch": 1.224,
      "grad_norm": 0.51953125,
      "kl": 0.0734753671567887,
      "learning_rate": 1e-05,
      "loss": 0.0004,
      "reward": 0.75,
      "reward_std": 0.4207531735301018,
      "rewards/correctness_reward_func": 0.75,
      "step": 38
    },
    {
      "completion_length": 348.9296875,
      "epoch": 1.256,
      "grad_norm": 0.625,
      "kl": 0.08903565420769155,
      "learning_rate": 1e-05,
      "loss": 0.0004,
      "reward": 0.65625,
      "reward_std": 0.3798343911767006,
      "rewards/correctness_reward_func": 0.65625,
      "step": 39
    },
    {
      "completion_length": 352.82421875,
      "epoch": 1.288,
      "grad_norm": 0.5390625,
      "kl": 0.06843947665765882,
      "learning_rate": 1e-05,
      "loss": 0.0003,
      "reward": 0.6484375,
      "reward_std": 0.4051281735301018,
      "rewards/correctness_reward_func": 0.6484375,
      "step": 40
    },
    {
      "completion_length": 329.1328125,
      "epoch": 1.32,
      "grad_norm": 0.46875,
      "kl": 0.057105657644569874,
      "learning_rate": 1e-05,
      "loss": 0.0003,
      "reward": 0.671875,
      "reward_std": 0.4520031735301018,
      "rewards/correctness_reward_func": 0.671875,
      "step": 41
    },
    {
      "completion_length": 322.95703125,
      "epoch": 1.3519999999999999,
      "grad_norm": 0.53515625,
      "kl": 0.05887855147011578,
      "learning_rate": 1e-05,
      "loss": 0.0003,
      "reward": 0.578125,
      "reward_std": 0.43042195588350296,
      "rewards/correctness_reward_func": 0.578125,
      "step": 42
    },
    {
      "completion_length": 373.6796875,
      "epoch": 1.384,
      "grad_norm": 0.375,
      "kl": 0.05490247346460819,
      "learning_rate": 1e-05,
      "loss": 0.0003,
      "reward": 0.7265625,
      "reward_std": 0.3426281735301018,
      "rewards/correctness_reward_func": 0.7265625,
      "step": 43
    },
    {
      "completion_length": 338.2109375,
      "epoch": 1.416,
      "grad_norm": 0.474609375,
      "kl": 0.05778517574071884,
      "learning_rate": 1e-05,
      "loss": 0.0003,
      "reward": 0.5234375,
      "reward_std": 0.41479695960879326,
      "rewards/correctness_reward_func": 0.5234375,
      "step": 44
    },
    {
      "completion_length": 379.49609375,
      "epoch": 1.448,
      "grad_norm": 0.443359375,
      "kl": 0.052183745545335114,
      "learning_rate": 1e-05,
      "loss": 0.0003,
      "reward": 0.4609375,
      "reward_std": 0.3377937823534012,
      "rewards/correctness_reward_func": 0.4609375,
      "step": 45
    },
    {
      "completion_length": 453.609375,
      "epoch": 1.48,
      "grad_norm": 0.37890625,
      "kl": 0.05021012085489929,
      "learning_rate": 1e-05,
      "loss": 0.0003,
      "reward": 0.484375,
      "reward_std": 0.39433756470680237,
      "rewards/correctness_reward_func": 0.484375,
      "step": 46
    },
    {
      "completion_length": 425.3515625,
      "epoch": 1.512,
      "grad_norm": 0.423828125,
      "kl": 0.0551034901291132,
      "learning_rate": 1e-05,
      "loss": 0.0003,
      "reward": 0.46875,
      "reward_std": 0.2777109779417515,
      "rewards/correctness_reward_func": 0.46875,
      "step": 47
    },
    {
      "completion_length": 436.1796875,
      "epoch": 1.544,
      "grad_norm": 0.373046875,
      "kl": 0.05335203884169459,
      "learning_rate": 1e-05,
      "loss": 0.0003,
      "reward": 0.4921875,
      "reward_std": 0.3245859779417515,
      "rewards/correctness_reward_func": 0.4921875,
      "step": 48
    },
    {
      "completion_length": 552.60546875,
      "epoch": 1.576,
      "grad_norm": 0.43359375,
      "kl": 0.05089649045839906,
      "learning_rate": 1e-05,
      "loss": 0.0003,
      "reward": 0.4765625,
      "reward_std": 0.3426281735301018,
      "rewards/correctness_reward_func": 0.4765625,
      "step": 49
    },
    {
      "completion_length": 509.71484375,
      "epoch": 1.608,
      "grad_norm": 0.474609375,
      "kl": 0.05733613413758576,
      "learning_rate": 1e-05,
      "loss": 0.0003,
      "reward": 0.4375,
      "reward_std": 0.43525634706020355,
      "rewards/correctness_reward_func": 0.4375,
      "step": 50
    },
    {
      "completion_length": 480.94921875,
      "epoch": 1.6400000000000001,
      "grad_norm": 0.375,
      "kl": 0.06065811449661851,
      "learning_rate": 1e-05,
      "loss": 0.0003,
      "reward": 0.34375,
      "reward_std": 0.3270031735301018,
      "rewards/correctness_reward_func": 0.34375,
      "step": 51
    },
    {
      "completion_length": 491.1640625,
      "epoch": 1.6720000000000002,
      "grad_norm": 0.390625,
      "kl": 0.060098053654655814,
      "learning_rate": 1e-05,
      "loss": 0.0003,
      "reward": 0.4375,
      "reward_std": 0.3846687786281109,
      "rewards/correctness_reward_func": 0.4375,
      "step": 52
    },
    {
      "completion_length": 464.2265625,
      "epoch": 1.704,
      "grad_norm": 0.435546875,
      "kl": 0.062199660344049335,
      "learning_rate": 1e-05,
      "loss": 0.0003,
      "reward": 0.4765625,
      "reward_std": 0.35713134706020355,
      "rewards/correctness_reward_func": 0.4765625,
      "step": 53
    },
    {
      "completion_length": 429.90625,
      "epoch": 1.736,
      "grad_norm": 12.625,
      "kl": 0.22791615827009082,
      "learning_rate": 1e-05,
      "loss": 0.0011,
      "reward": 0.734375,
      "reward_std": 0.3402109779417515,
      "rewards/correctness_reward_func": 0.734375,
      "step": 54
    },
    {
      "completion_length": 512.62109375,
      "epoch": 1.768,
      "grad_norm": 0.388671875,
      "kl": 0.059834773652255535,
      "learning_rate": 1e-05,
      "loss": 0.0003,
      "reward": 0.7265625,
      "reward_std": 0.24404378235340118,
      "rewards/correctness_reward_func": 0.7265625,
      "step": 55
    },
    {
      "completion_length": 471.5859375,
      "epoch": 1.8,
      "grad_norm": 0.388671875,
      "kl": 0.07468604180030525,
      "learning_rate": 1e-05,
      "loss": 0.0004,
      "reward": 0.859375,
      "reward_std": 0.2055421955883503,
      "rewards/correctness_reward_func": 0.859375,
      "step": 56
    },
    {
      "completion_length": 446.046875,
      "epoch": 1.8319999999999999,
      "grad_norm": 0.443359375,
      "kl": 0.06021555629558861,
      "learning_rate": 1e-05,
      "loss": 0.0003,
      "reward": 0.890625,
      "reward_std": 0.3221687823534012,
      "rewards/correctness_reward_func": 0.890625,
      "step": 57
    },
    {
      "completion_length": 536.68359375,
      "epoch": 1.8639999999999999,
      "grad_norm": 0.333984375,
      "kl": 0.060477497056126595,
      "learning_rate": 1e-05,
      "loss": 0.0003,
      "reward": 0.609375,
      "reward_std": 0.24646097794175148,
      "rewards/correctness_reward_func": 0.609375,
      "step": 58
    },
    {
      "completion_length": 411.69140625,
      "epoch": 1.896,
      "grad_norm": 0.453125,
      "kl": 0.0779561479575932,
      "learning_rate": 1e-05,
      "loss": 0.0004,
      "reward": 0.9609375,
      "reward_std": 0.2620859779417515,
      "rewards/correctness_reward_func": 0.9609375,
      "step": 59
    },
    {
      "completion_length": 544.3671875,
      "epoch": 1.928,
      "grad_norm": 0.81640625,
      "kl": 0.10888050403445959,
      "learning_rate": 1e-05,
      "loss": 0.0005,
      "reward": 0.5859375,
      "reward_std": 0.24404378235340118,
      "rewards/correctness_reward_func": 0.5859375,
      "step": 60
    },
    {
      "completion_length": 397.546875,
      "epoch": 1.96,
      "grad_norm": 0.6953125,
      "kl": 0.10245927749201655,
      "learning_rate": 1e-05,
      "loss": 0.0005,
      "reward": 0.9375,
      "reward_std": 0.30058756470680237,
      "rewards/correctness_reward_func": 0.9375,
      "step": 61
    },
    {
      "completion_length": 533.109375,
      "epoch": 1.992,
      "grad_norm": 0.65234375,
      "kl": 0.09279383928515017,
      "learning_rate": 1e-05,
      "loss": 0.0005,
      "reward": 0.640625,
      "reward_std": 0.28254536911845207,
      "rewards/correctness_reward_func": 0.640625,
      "step": 62
    },
    {
      "completion_length": 607.078125,
      "epoch": 2.0,
      "grad_norm": 0.2421875,
      "kl": 0.059856025502085686,
      "learning_rate": 1e-05,
      "loss": 0.0001,
      "reward": 0.6875,
      "reward_std": 0.4471687823534012,
      "rewards/correctness_reward_func": 0.6875,
      "step": 63
    },
    {
      "completion_length": 706.515625,
      "epoch": 2.032,
      "grad_norm": 2.9375,
      "kl": 0.23451336496509612,
      "learning_rate": 1e-05,
      "loss": 0.0012,
      "reward": 0.21875,
      "reward_std": 0.19716878235340118,
      "rewards/correctness_reward_func": 0.21875,
      "step": 64
    },
    {
      "completion_length": 661.4453125,
      "epoch": 2.064,
      "grad_norm": 0.455078125,
      "kl": 0.08936465694569051,
      "learning_rate": 1e-05,
      "loss": 0.0004,
      "reward": 0.1796875,
      "reward_std": 0.0961671955883503,
      "rewards/correctness_reward_func": 0.1796875,
      "step": 65
    },
    {
      "completion_length": 644.8046875,
      "epoch": 2.096,
      "grad_norm": 0.287109375,
      "kl": 0.0694590169005096,
      "learning_rate": 1e-05,
      "loss": 0.0003,
      "reward": 0.2109375,
      "reward_std": 0.1454593911767006,
      "rewards/correctness_reward_func": 0.2109375,
      "step": 66
    },
    {
      "completion_length": 331.421875,
      "epoch": 2.128,
      "grad_norm": 0.5625,
      "kl": 0.10819920105859637,
      "learning_rate": 1e-05,
      "loss": 0.0005,
      "reward": 0.7109375,
      "reward_std": 0.39192036911845207,
      "rewards/correctness_reward_func": 0.7109375,
      "step": 67
    },
    {
      "completion_length": 454.83984375,
      "epoch": 2.16,
      "grad_norm": 1.171875,
      "kl": 0.14896808750927448,
      "learning_rate": 1e-05,
      "loss": 0.0007,
      "reward": 0.4453125,
      "reward_std": 0.2885015867650509,
      "rewards/correctness_reward_func": 0.4453125,
      "step": 68
    },
    {
      "completion_length": 474.54296875,
      "epoch": 2.192,
      "grad_norm": 0.56640625,
      "kl": 0.06995820580050349,
      "learning_rate": 1e-05,
      "loss": 0.0003,
      "reward": 0.6015625,
      "reward_std": 0.39192036911845207,
      "rewards/correctness_reward_func": 0.6015625,
      "step": 69
    },
    {
      "completion_length": 440.05859375,
      "epoch": 2.224,
      "grad_norm": 0.50390625,
      "kl": 0.0910127570386976,
      "learning_rate": 1e-05,
      "loss": 0.0005,
      "reward": 0.46875,
      "reward_std": 0.39433756470680237,
      "rewards/correctness_reward_func": 0.46875,
      "step": 70
    },
    {
      "completion_length": 478.2890625,
      "epoch": 2.2560000000000002,
      "grad_norm": 0.419921875,
      "kl": 0.07039178418926895,
      "learning_rate": 1e-05,
      "loss": 0.0004,
      "reward": 0.515625,
      "reward_std": 0.3534187823534012,
      "rewards/correctness_reward_func": 0.515625,
      "step": 71
    },
    {
      "completion_length": 539.33203125,
      "epoch": 2.288,
      "grad_norm": 0.35546875,
      "kl": 0.0686434528324753,
      "learning_rate": 1e-05,
      "loss": 0.0003,
      "reward": 0.4453125,
      "reward_std": 0.2933359779417515,
      "rewards/correctness_reward_func": 0.4453125,
      "step": 72
    },
    {
      "completion_length": 480.2578125,
      "epoch": 2.32,
      "grad_norm": 0.326171875,
      "kl": 0.06706948089413345,
      "learning_rate": 1e-05,
      "loss": 0.0003,
      "reward": 0.40625,
      "reward_std": 0.3582531735301018,
      "rewards/correctness_reward_func": 0.40625,
      "step": 73
    },
    {
      "completion_length": 497.828125,
      "epoch": 2.352,
      "grad_norm": 0.486328125,
      "kl": 0.07771567534655333,
      "learning_rate": 1e-05,
      "loss": 0.0004,
      "reward": 0.4609375,
      "reward_std": 0.45925475656986237,
      "rewards/correctness_reward_func": 0.4609375,
      "step": 74
    },
    {
      "completion_length": 540.3671875,
      "epoch": 2.384,
      "grad_norm": 0.330078125,
      "kl": 0.07498506712727249,
      "learning_rate": 1e-05,
      "loss": 0.0004,
      "reward": 0.359375,
      "reward_std": 0.2909187823534012,
      "rewards/correctness_reward_func": 0.359375,
      "step": 75
    },
    {
      "completion_length": 472.80859375,
      "epoch": 2.416,
      "grad_norm": 0.431640625,
      "kl": 0.06656011822633445,
      "learning_rate": 1e-05,
      "loss": 0.0003,
      "reward": 0.5,
      "reward_std": 0.3534187823534012,
      "rewards/correctness_reward_func": 0.5,
      "step": 76
    },
    {
      "completion_length": 517.7421875,
      "epoch": 2.448,
      "grad_norm": 0.64453125,
      "kl": 0.07896088412962854,
      "learning_rate": 1e-05,
      "loss": 0.0004,
      "reward": 0.375,
      "reward_std": 0.34504536911845207,
      "rewards/correctness_reward_func": 0.375,
      "step": 77
    },
    {
      "completion_length": 589.140625,
      "epoch": 2.48,
      "grad_norm": 1.1875,
      "kl": 0.12798133888281882,
      "learning_rate": 1e-05,
      "loss": 0.0006,
      "reward": 0.3203125,
      "reward_std": 0.34746256470680237,
      "rewards/correctness_reward_func": 0.3203125,
      "step": 78
    },
    {
      "completion_length": 506.98828125,
      "epoch": 2.512,
      "grad_norm": 0.302734375,
      "kl": 0.06648093881085515,
      "learning_rate": 1e-05,
      "loss": 0.0003,
      "reward": 0.2890625,
      "reward_std": 0.29817036911845207,
      "rewards/correctness_reward_func": 0.2890625,
      "step": 79
    },
    {
      "completion_length": 634.9140625,
      "epoch": 2.544,
      "grad_norm": 0.296875,
      "kl": 0.08733222424052656,
      "learning_rate": 1e-05,
      "loss": 0.0004,
      "reward": 0.265625,
      "reward_std": 0.2596687823534012,
      "rewards/correctness_reward_func": 0.265625,
      "step": 80
    },
    {
      "completion_length": 561.2734375,
      "epoch": 2.576,
      "grad_norm": 0.451171875,
      "kl": 0.08239166042767465,
      "learning_rate": 1e-05,
      "loss": 0.0004,
      "reward": 0.5078125,
      "reward_std": 0.45442036911845207,
      "rewards/correctness_reward_func": 0.5078125,
      "step": 81
    },
    {
      "completion_length": 535.79296875,
      "epoch": 2.608,
      "grad_norm": 0.99609375,
      "kl": 0.22772114095278084,
      "learning_rate": 1e-05,
      "loss": 0.0011,
      "reward": 0.3359375,
      "reward_std": 0.36550476029515266,
      "rewards/correctness_reward_func": 0.3359375,
      "step": 82
    },
    {
      "completion_length": 516.10546875,
      "epoch": 2.64,
      "grad_norm": 0.4765625,
      "kl": 0.12832795921713114,
      "learning_rate": 1e-05,
      "loss": 0.0006,
      "reward": 0.390625,
      "reward_std": 0.2728765867650509,
      "rewards/correctness_reward_func": 0.390625,
      "step": 83
    },
    {
      "completion_length": 486.69140625,
      "epoch": 2.672,
      "grad_norm": 0.453125,
      "kl": 0.08598195551894605,
      "learning_rate": 1e-05,
      "loss": 0.0004,
      "reward": 0.4296875,
      "reward_std": 0.3377937823534012,
      "rewards/correctness_reward_func": 0.4296875,
      "step": 84
    },
    {
      "completion_length": 595.859375,
      "epoch": 2.7039999999999997,
      "grad_norm": 0.337890625,
      "kl": 0.08345729601569474,
      "learning_rate": 1e-05,
      "loss": 0.0004,
      "reward": 0.390625,
      "reward_std": 0.3089609779417515,
      "rewards/correctness_reward_func": 0.390625,
      "step": 85
    },
    {
      "completion_length": 613.12109375,
      "epoch": 2.7359999999999998,
      "grad_norm": 0.291015625,
      "kl": 0.0912326027173549,
      "learning_rate": 1e-05,
      "loss": 0.0005,
      "reward": 0.2890625,
      "reward_std": 0.19958597794175148,
      "rewards/correctness_reward_func": 0.2890625,
      "step": 86
    },
    {
      "completion_length": 586.3984375,
      "epoch": 2.768,
      "grad_norm": 0.361328125,
      "kl": 0.08986557251773775,
      "learning_rate": 1e-05,
      "loss": 0.0004,
      "reward": 0.4609375,
      "reward_std": 0.3245859779417515,
      "rewards/correctness_reward_func": 0.4609375,
      "step": 87
    },
    {
      "completion_length": 580.55859375,
      "epoch": 2.8,
      "grad_norm": 0.4140625,
      "kl": 0.08739295578561723,
      "learning_rate": 1e-05,
      "loss": 0.0004,
      "reward": 0.3671875,
      "reward_std": 0.2933359779417515,
      "rewards/correctness_reward_func": 0.3671875,
      "step": 88
    },
    {
      "completion_length": 545.07421875,
      "epoch": 2.832,
      "grad_norm": 0.30078125,
      "kl": 0.0942155783995986,
      "learning_rate": 1e-05,
      "loss": 0.0005,
      "reward": 0.328125,
      "reward_std": 0.2512953653931618,
      "rewards/correctness_reward_func": 0.328125,
      "step": 89
    },
    {
      "completion_length": 507.109375,
      "epoch": 2.864,
      "grad_norm": 0.37109375,
      "kl": 0.08285854570567608,
      "learning_rate": 1e-05,
      "loss": 0.0004,
      "reward": 0.4140625,
      "reward_std": 0.31621256098151207,
      "rewards/correctness_reward_func": 0.4140625,
      "step": 90
    },
    {
      "completion_length": 548.953125,
      "epoch": 2.896,
      "grad_norm": 1.2109375,
      "kl": 0.17426068475469947,
      "learning_rate": 1e-05,
      "loss": 0.0009,
      "reward": 0.5390625,
      "reward_std": 0.2752937823534012,
      "rewards/correctness_reward_func": 0.5390625,
      "step": 91
    },
    {
      "completion_length": 487.5078125,
      "epoch": 2.928,
      "grad_norm": 0.3671875,
      "kl": 0.08832010440528393,
      "learning_rate": 1e-05,
      "loss": 0.0004,
      "reward": 0.75,
      "reward_std": 0.3270031735301018,
      "rewards/correctness_reward_func": 0.75,
      "step": 92
    },
    {
      "completion_length": 464.93359375,
      "epoch": 2.96,
      "grad_norm": 0.625,
      "kl": 0.12189564970321953,
      "learning_rate": 1e-05,
      "loss": 0.0006,
      "reward": 0.5625,
      "reward_std": 0.2645031735301018,
      "rewards/correctness_reward_func": 0.5625,
      "step": 93
    },
    {
      "completion_length": 474.10546875,
      "epoch": 2.992,
      "grad_norm": 0.30078125,
      "kl": 0.08207035693340003,
      "learning_rate": 1e-05,
      "loss": 0.0004,
      "reward": 0.75,
      "reward_std": 0.2596687823534012,
      "rewards/correctness_reward_func": 0.75,
      "step": 94
    },
    {
      "completion_length": 495.15625,
      "epoch": 3.0,
      "grad_norm": 0.107421875,
      "kl": 0.09875499829649925,
      "learning_rate": 1e-05,
      "loss": 0.0001,
      "reward": 0.5,
      "reward_std": 0.25,
      "rewards/correctness_reward_func": 0.5,
      "step": 95
    },
    {
      "completion_length": 693.6640625,
      "epoch": 3.032,
      "grad_norm": 0.287109375,
      "kl": 0.13659441424533725,
      "learning_rate": 1e-05,
      "loss": 0.0007,
      "reward": 0.1953125,
      "reward_std": 0.21279378235340118,
      "rewards/correctness_reward_func": 0.1953125,
      "step": 96
    },
    {
      "completion_length": 681.03125,
      "epoch": 3.064,
      "grad_norm": 0.255859375,
      "kl": 0.13894613785669208,
      "learning_rate": 1e-05,
      "loss": 0.0007,
      "reward": 0.2578125,
      "reward_std": 0.1947515867650509,
      "rewards/correctness_reward_func": 0.2578125,
      "step": 97
    },
    {
      "completion_length": 722.29296875,
      "epoch": 3.096,
      "grad_norm": 0.330078125,
      "kl": 0.1400570566765964,
      "learning_rate": 1e-05,
      "loss": 0.0007,
      "reward": 0.2109375,
      "reward_std": 0.2079593911767006,
      "rewards/correctness_reward_func": 0.2109375,
      "step": 98
    },
    {
      "completion_length": 634.5546875,
      "epoch": 3.128,
      "grad_norm": 0.287109375,
      "kl": 0.13448539143428206,
      "learning_rate": 1e-05,
      "loss": 0.0007,
      "reward": 0.203125,
      "reward_std": 0.1923343911767006,
      "rewards/correctness_reward_func": 0.203125,
      "step": 99
    },
    {
      "completion_length": 406.96875,
      "epoch": 3.16,
      "grad_norm": 0.7109375,
      "kl": 0.13877984089776874,
      "learning_rate": 1e-05,
      "loss": 0.0007,
      "reward": 0.4609375,
      "reward_std": 0.3558359779417515,
      "rewards/correctness_reward_func": 0.4609375,
      "step": 100
    }
  ],
  "logging_steps": 1,
  "max_steps": 125800,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 200,
  "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": 4,
  "trial_name": null,
  "trial_params": null
}