{
  "best_global_step": null,
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 0.009894622272794735,
  "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": 6.0,
      "epoch": 9.894622272794736e-05,
      "grad_norm": 22.976070404052734,
      "kl": 0.0,
      "learning_rate": 1e-06,
      "loss": -0.0,
      "reward": 0.125,
      "reward_std": 0.25,
      "rewards/accuracy_reward": 0.125,
      "rewards/format_reward": 0.0,
      "step": 1
    },
    {
      "completion_length": 6.5,
      "epoch": 0.00019789244545589471,
      "grad_norm": 32.8896369934082,
      "kl": 0.0035400390625,
      "learning_rate": 9.989999999999999e-07,
      "loss": 0.0001,
      "reward": 0.3125,
      "reward_std": 0.5193375647068024,
      "rewards/accuracy_reward": 0.3125,
      "rewards/format_reward": 0.0,
      "step": 2
    },
    {
      "completion_length": 5.0625,
      "epoch": 0.0002968386681838421,
      "grad_norm": 45.44774627685547,
      "kl": 0.00732421875,
      "learning_rate": 9.98e-07,
      "loss": 0.0003,
      "reward": 0.3125,
      "reward_std": 0.26933756470680237,
      "rewards/accuracy_reward": 0.3125,
      "rewards/format_reward": 0.0,
      "step": 3
    },
    {
      "completion_length": 6.25,
      "epoch": 0.00039578489091178943,
      "grad_norm": 35.005924224853516,
      "kl": 0.003936767578125,
      "learning_rate": 9.97e-07,
      "loss": 0.0002,
      "reward": 0.375,
      "reward_std": 0.39433756470680237,
      "rewards/accuracy_reward": 0.375,
      "rewards/format_reward": 0.0,
      "step": 4
    },
    {
      "completion_length": 4.5625,
      "epoch": 0.0004947311136397368,
      "grad_norm": 39.57672882080078,
      "kl": 0.00762176513671875,
      "learning_rate": 9.959999999999999e-07,
      "loss": 0.0003,
      "reward": 0.625,
      "reward_std": 0.39433756470680237,
      "rewards/accuracy_reward": 0.625,
      "rewards/format_reward": 0.0,
      "step": 5
    },
    {
      "completion_length": 4.125,
      "epoch": 0.0005936773363676842,
      "grad_norm": 23.2183837890625,
      "kl": 0.03729248046875,
      "learning_rate": 9.95e-07,
      "loss": 0.0015,
      "reward": 0.5625,
      "reward_std": 0.375,
      "rewards/accuracy_reward": 0.5625,
      "rewards/format_reward": 0.0,
      "step": 6
    },
    {
      "completion_length": 3.3125,
      "epoch": 0.0006926235590956315,
      "grad_norm": 47.74429702758789,
      "kl": 0.06170654296875,
      "learning_rate": 9.94e-07,
      "loss": 0.0025,
      "reward": 0.4375,
      "reward_std": 0.41367512941360474,
      "rewards/accuracy_reward": 0.4375,
      "rewards/format_reward": 0.0,
      "step": 7
    },
    {
      "completion_length": 3.8125,
      "epoch": 0.0007915697818235789,
      "grad_norm": 49.124691009521484,
      "kl": 0.09912109375,
      "learning_rate": 9.929999999999999e-07,
      "loss": 0.004,
      "reward": 0.3125,
      "reward_std": 0.375,
      "rewards/accuracy_reward": 0.3125,
      "rewards/format_reward": 0.0,
      "step": 8
    },
    {
      "completion_length": 3.5625,
      "epoch": 0.0008905160045515262,
      "grad_norm": 43.116676330566406,
      "kl": 0.02130126953125,
      "learning_rate": 9.92e-07,
      "loss": 0.0009,
      "reward": 0.4375,
      "reward_std": 0.5193375647068024,
      "rewards/accuracy_reward": 0.4375,
      "rewards/format_reward": 0.0,
      "step": 9
    },
    {
      "completion_length": 3.5625,
      "epoch": 0.0009894622272794737,
      "grad_norm": 42.24452209472656,
      "kl": 0.030517578125,
      "learning_rate": 9.91e-07,
      "loss": 0.0012,
      "reward": 0.4375,
      "reward_std": 0.26933756470680237,
      "rewards/accuracy_reward": 0.4375,
      "rewards/format_reward": 0.0,
      "step": 10
    },
    {
      "completion_length": 4.5625,
      "epoch": 0.001088408450007421,
      "grad_norm": 9.293623924255371,
      "kl": 0.386474609375,
      "learning_rate": 9.9e-07,
      "loss": 0.0155,
      "reward": 0.75,
      "reward_std": 0.25,
      "rewards/accuracy_reward": 0.75,
      "rewards/format_reward": 0.0,
      "step": 11
    },
    {
      "completion_length": 2.75,
      "epoch": 0.0011873546727353683,
      "grad_norm": 27.51582908630371,
      "kl": 0.119140625,
      "learning_rate": 9.89e-07,
      "loss": 0.0048,
      "reward": 0.9375,
      "reward_std": 0.125,
      "rewards/accuracy_reward": 0.9375,
      "rewards/format_reward": 0.0,
      "step": 12
    },
    {
      "completion_length": 4.125,
      "epoch": 0.0012863008954633157,
      "grad_norm": 42.183692932128906,
      "kl": 1.232391357421875,
      "learning_rate": 9.88e-07,
      "loss": 0.0494,
      "reward": 0.375,
      "reward_std": 0.39433756470680237,
      "rewards/accuracy_reward": 0.375,
      "rewards/format_reward": 0.0,
      "step": 13
    },
    {
      "completion_length": 3.75,
      "epoch": 0.001385247118191263,
      "grad_norm": 16.871009826660156,
      "kl": 0.02880859375,
      "learning_rate": 9.87e-07,
      "loss": 0.0012,
      "reward": 0.8125,
      "reward_std": 0.125,
      "rewards/accuracy_reward": 0.8125,
      "rewards/format_reward": 0.0,
      "step": 14
    },
    {
      "completion_length": 3.1875,
      "epoch": 0.0014841933409192104,
      "grad_norm": 9.070819854736328,
      "kl": 0.31671142578125,
      "learning_rate": 9.86e-07,
      "loss": 0.0127,
      "reward": 0.5,
      "reward_std": 0.0,
      "rewards/accuracy_reward": 0.5,
      "rewards/format_reward": 0.0,
      "step": 15
    },
    {
      "completion_length": 3.875,
      "epoch": 0.0015831395636471577,
      "grad_norm": 34.1285285949707,
      "kl": 0.1123046875,
      "learning_rate": 9.849999999999999e-07,
      "loss": 0.0045,
      "reward": 0.8125,
      "reward_std": 0.26933756470680237,
      "rewards/accuracy_reward": 0.8125,
      "rewards/format_reward": 0.0,
      "step": 16
    },
    {
      "completion_length": 4.5,
      "epoch": 0.001682085786375105,
      "grad_norm": 28.27692985534668,
      "kl": 0.06103515625,
      "learning_rate": 9.84e-07,
      "loss": 0.0024,
      "reward": 0.375,
      "reward_std": 0.14433756470680237,
      "rewards/accuracy_reward": 0.375,
      "rewards/format_reward": 0.0,
      "step": 17
    },
    {
      "completion_length": 3.625,
      "epoch": 0.0017810320091030524,
      "grad_norm": 43.93234634399414,
      "kl": 0.04833984375,
      "learning_rate": 9.83e-07,
      "loss": 0.0019,
      "reward": 0.4375,
      "reward_std": 0.375,
      "rewards/accuracy_reward": 0.4375,
      "rewards/format_reward": 0.0,
      "step": 18
    },
    {
      "completion_length": 4.875,
      "epoch": 0.001879978231831,
      "grad_norm": 35.52457046508789,
      "kl": 0.1285400390625,
      "learning_rate": 9.819999999999999e-07,
      "loss": 0.0052,
      "reward": 0.25,
      "reward_std": 0.25,
      "rewards/accuracy_reward": 0.25,
      "rewards/format_reward": 0.0,
      "step": 19
    },
    {
      "completion_length": 4.6875,
      "epoch": 0.0019789244545589473,
      "grad_norm": 5.439634799957275,
      "kl": 0.283935546875,
      "learning_rate": 9.81e-07,
      "loss": 0.0114,
      "reward": 0.9375,
      "reward_std": 0.125,
      "rewards/accuracy_reward": 0.9375,
      "rewards/format_reward": 0.0,
      "step": 20
    },
    {
      "completion_length": 2.875,
      "epoch": 0.0020778706772868944,
      "grad_norm": 35.29671859741211,
      "kl": 0.076904296875,
      "learning_rate": 9.8e-07,
      "loss": 0.0031,
      "reward": 0.75,
      "reward_std": 0.25,
      "rewards/accuracy_reward": 0.75,
      "rewards/format_reward": 0.0,
      "step": 21
    },
    {
      "completion_length": 3.1875,
      "epoch": 0.002176816900014842,
      "grad_norm": 0.33881574869155884,
      "kl": 0.03955078125,
      "learning_rate": 9.789999999999999e-07,
      "loss": 0.0016,
      "reward": 0.5,
      "reward_std": 0.0,
      "rewards/accuracy_reward": 0.5,
      "rewards/format_reward": 0.0,
      "step": 22
    },
    {
      "completion_length": 3.875,
      "epoch": 0.002275763122742789,
      "grad_norm": 18.542280197143555,
      "kl": 0.13916015625,
      "learning_rate": 9.78e-07,
      "loss": 0.0056,
      "reward": 0.875,
      "reward_std": 0.14433756470680237,
      "rewards/accuracy_reward": 0.875,
      "rewards/format_reward": 0.0,
      "step": 23
    },
    {
      "completion_length": 3.9375,
      "epoch": 0.0023747093454707367,
      "grad_norm": 22.113807678222656,
      "kl": 0.46295166015625,
      "learning_rate": 9.77e-07,
      "loss": 0.0185,
      "reward": 0.5625,
      "reward_std": 0.125,
      "rewards/accuracy_reward": 0.5625,
      "rewards/format_reward": 0.0,
      "step": 24
    },
    {
      "completion_length": 2.875,
      "epoch": 0.002473655568198684,
      "grad_norm": 28.871366500854492,
      "kl": 0.0540771484375,
      "learning_rate": 9.759999999999998e-07,
      "loss": 0.0022,
      "reward": 0.75,
      "reward_std": 0.25,
      "rewards/accuracy_reward": 0.75,
      "rewards/format_reward": 0.0,
      "step": 25
    },
    {
      "completion_length": 2.75,
      "epoch": 0.0025726017909266314,
      "grad_norm": 13.429852485656738,
      "kl": 0.0830078125,
      "learning_rate": 9.75e-07,
      "loss": 0.0033,
      "reward": 0.9375,
      "reward_std": 0.125,
      "rewards/accuracy_reward": 0.9375,
      "rewards/format_reward": 0.0,
      "step": 26
    },
    {
      "completion_length": 3.3125,
      "epoch": 0.002671548013654579,
      "grad_norm": 25.259496688842773,
      "kl": 0.08380126953125,
      "learning_rate": 9.74e-07,
      "loss": 0.0034,
      "reward": 0.9375,
      "reward_std": 0.125,
      "rewards/accuracy_reward": 0.9375,
      "rewards/format_reward": 0.0,
      "step": 27
    },
    {
      "completion_length": 4.6875,
      "epoch": 0.002770494236382526,
      "grad_norm": 35.60582733154297,
      "kl": 0.017974853515625,
      "learning_rate": 9.729999999999998e-07,
      "loss": 0.0007,
      "reward": 0.875,
      "reward_std": 0.14433756470680237,
      "rewards/accuracy_reward": 0.875,
      "rewards/format_reward": 0.0,
      "step": 28
    },
    {
      "completion_length": 2.625,
      "epoch": 0.0028694404591104736,
      "grad_norm": 35.21329116821289,
      "kl": 0.107421875,
      "learning_rate": 9.72e-07,
      "loss": 0.0043,
      "reward": 0.625,
      "reward_std": 0.25,
      "rewards/accuracy_reward": 0.625,
      "rewards/format_reward": 0.0,
      "step": 29
    },
    {
      "completion_length": 3.4375,
      "epoch": 0.0029683866818384207,
      "grad_norm": 41.33916091918945,
      "kl": 0.14288330078125,
      "learning_rate": 9.709999999999999e-07,
      "loss": 0.0057,
      "reward": 0.8125,
      "reward_std": 0.26933756470680237,
      "rewards/accuracy_reward": 0.8125,
      "rewards/format_reward": 0.0,
      "step": 30
    },
    {
      "completion_length": 3.1875,
      "epoch": 0.0030673329045663683,
      "grad_norm": 35.675777435302734,
      "kl": 0.1474609375,
      "learning_rate": 9.7e-07,
      "loss": 0.0059,
      "reward": 0.625,
      "reward_std": 0.25,
      "rewards/accuracy_reward": 0.625,
      "rewards/format_reward": 0.0,
      "step": 31
    },
    {
      "completion_length": 4.0,
      "epoch": 0.0031662791272943154,
      "grad_norm": 18.264066696166992,
      "kl": 0.087646484375,
      "learning_rate": 9.69e-07,
      "loss": 0.0035,
      "reward": 0.9375,
      "reward_std": 0.125,
      "rewards/accuracy_reward": 0.9375,
      "rewards/format_reward": 0.0,
      "step": 32
    },
    {
      "completion_length": 3.5,
      "epoch": 0.003265225350022263,
      "grad_norm": 0.011650191619992256,
      "kl": 0.0262451171875,
      "learning_rate": 9.679999999999999e-07,
      "loss": 0.001,
      "reward": 1.0,
      "reward_std": 0.0,
      "rewards/accuracy_reward": 1.0,
      "rewards/format_reward": 0.0,
      "step": 33
    },
    {
      "completion_length": 3.5,
      "epoch": 0.00336417157275021,
      "grad_norm": 52.62285614013672,
      "kl": 0.115966796875,
      "learning_rate": 9.67e-07,
      "loss": 0.0046,
      "reward": 0.5625,
      "reward_std": 0.125,
      "rewards/accuracy_reward": 0.5625,
      "rewards/format_reward": 0.0,
      "step": 34
    },
    {
      "completion_length": 3.625,
      "epoch": 0.0034631177954781577,
      "grad_norm": 82.99117279052734,
      "kl": 3.158203125,
      "learning_rate": 9.66e-07,
      "loss": 0.1264,
      "reward": 0.4375,
      "reward_std": 0.125,
      "rewards/accuracy_reward": 0.4375,
      "rewards/format_reward": 0.0,
      "step": 35
    },
    {
      "completion_length": 2.4375,
      "epoch": 0.003562064018206105,
      "grad_norm": 33.630123138427734,
      "kl": 0.109375,
      "learning_rate": 9.649999999999999e-07,
      "loss": 0.0044,
      "reward": 0.9375,
      "reward_std": 0.125,
      "rewards/accuracy_reward": 0.9375,
      "rewards/format_reward": 0.0,
      "step": 36
    },
    {
      "completion_length": 3.1875,
      "epoch": 0.0036610102409340524,
      "grad_norm": 31.628652572631836,
      "kl": 0.61328125,
      "learning_rate": 9.64e-07,
      "loss": 0.0246,
      "reward": 0.625,
      "reward_std": 0.14433756470680237,
      "rewards/accuracy_reward": 0.625,
      "rewards/format_reward": 0.0,
      "step": 37
    },
    {
      "completion_length": 4.1875,
      "epoch": 0.003759956463662,
      "grad_norm": 19.04951286315918,
      "kl": 0.2353515625,
      "learning_rate": 9.63e-07,
      "loss": 0.0095,
      "reward": 0.8125,
      "reward_std": 0.26933756470680237,
      "rewards/accuracy_reward": 0.8125,
      "rewards/format_reward": 0.0,
      "step": 38
    },
    {
      "completion_length": 4.3125,
      "epoch": 0.003858902686389947,
      "grad_norm": 26.543506622314453,
      "kl": 0.0684814453125,
      "learning_rate": 9.619999999999999e-07,
      "loss": 0.0027,
      "reward": 0.875,
      "reward_std": 0.25,
      "rewards/accuracy_reward": 0.875,
      "rewards/format_reward": 0.0,
      "step": 39
    },
    {
      "completion_length": 4.0,
      "epoch": 0.003957848909117895,
      "grad_norm": 13.690553665161133,
      "kl": 0.13409423828125,
      "learning_rate": 9.61e-07,
      "loss": 0.0054,
      "reward": 0.5625,
      "reward_std": 0.125,
      "rewards/accuracy_reward": 0.5625,
      "rewards/format_reward": 0.0,
      "step": 40
    },
    {
      "completion_length": 3.6875,
      "epoch": 0.004056795131845842,
      "grad_norm": 13.923758506774902,
      "kl": 0.0458984375,
      "learning_rate": 9.6e-07,
      "loss": 0.0018,
      "reward": 0.9375,
      "reward_std": 0.125,
      "rewards/accuracy_reward": 0.9375,
      "rewards/format_reward": 0.0,
      "step": 41
    },
    {
      "completion_length": 5.875,
      "epoch": 0.004155741354573789,
      "grad_norm": 18.20030403137207,
      "kl": 0.115478515625,
      "learning_rate": 9.589999999999998e-07,
      "loss": 0.0046,
      "reward": 0.625,
      "reward_std": 0.14433756470680237,
      "rewards/accuracy_reward": 0.625,
      "rewards/format_reward": 0.0,
      "step": 42
    },
    {
      "completion_length": 2.75,
      "epoch": 0.004254687577301737,
      "grad_norm": 43.371185302734375,
      "kl": 0.097412109375,
      "learning_rate": 9.58e-07,
      "loss": 0.0039,
      "reward": 0.625,
      "reward_std": 0.14433756470680237,
      "rewards/accuracy_reward": 0.625,
      "rewards/format_reward": 0.0,
      "step": 43
    },
    {
      "completion_length": 3.1875,
      "epoch": 0.004353633800029684,
      "grad_norm": 69.18988037109375,
      "kl": 0.15380859375,
      "learning_rate": 9.57e-07,
      "loss": 0.0062,
      "reward": 0.5625,
      "reward_std": 0.26933756470680237,
      "rewards/accuracy_reward": 0.5625,
      "rewards/format_reward": 0.0,
      "step": 44
    },
    {
      "completion_length": 3.5,
      "epoch": 0.004452580022757631,
      "grad_norm": 42.821083068847656,
      "kl": 0.29052734375,
      "learning_rate": 9.559999999999998e-07,
      "loss": 0.0116,
      "reward": 0.8125,
      "reward_std": 0.26933756470680237,
      "rewards/accuracy_reward": 0.8125,
      "rewards/format_reward": 0.0,
      "step": 45
    },
    {
      "completion_length": 2.75,
      "epoch": 0.004551526245485578,
      "grad_norm": 0.00253496621735394,
      "kl": 0.042724609375,
      "learning_rate": 9.55e-07,
      "loss": 0.0017,
      "reward": 0.75,
      "reward_std": 0.0,
      "rewards/accuracy_reward": 0.75,
      "rewards/format_reward": 0.0,
      "step": 46
    },
    {
      "completion_length": 2.75,
      "epoch": 0.004650472468213526,
      "grad_norm": 28.470979690551758,
      "kl": 0.1875,
      "learning_rate": 9.539999999999999e-07,
      "loss": 0.0075,
      "reward": 0.6875,
      "reward_std": 0.125,
      "rewards/accuracy_reward": 0.6875,
      "rewards/format_reward": 0.0,
      "step": 47
    },
    {
      "completion_length": 3.125,
      "epoch": 0.004749418690941473,
      "grad_norm": 78.60997772216797,
      "kl": 0.44921875,
      "learning_rate": 9.529999999999999e-07,
      "loss": 0.018,
      "reward": 0.8125,
      "reward_std": 0.26933756470680237,
      "rewards/accuracy_reward": 0.8125,
      "rewards/format_reward": 0.0,
      "step": 48
    },
    {
      "completion_length": 3.5,
      "epoch": 0.0048483649136694205,
      "grad_norm": 0.15050888061523438,
      "kl": 0.028778076171875,
      "learning_rate": 9.52e-07,
      "loss": 0.0011,
      "reward": 1.0,
      "reward_std": 0.0,
      "rewards/accuracy_reward": 1.0,
      "rewards/format_reward": 0.0,
      "step": 49
    },
    {
      "completion_length": 3.6875,
      "epoch": 0.004947311136397368,
      "grad_norm": 8.970772743225098,
      "kl": 0.3662109375,
      "learning_rate": 9.509999999999999e-07,
      "loss": 0.0146,
      "reward": 0.4375,
      "reward_std": 0.125,
      "rewards/accuracy_reward": 0.4375,
      "rewards/format_reward": 0.0,
      "step": 50
    },
    {
      "completion_length": 3.5,
      "epoch": 0.005046257359125316,
      "grad_norm": 42.177669525146484,
      "kl": 0.20263671875,
      "learning_rate": 9.499999999999999e-07,
      "loss": 0.0081,
      "reward": 0.875,
      "reward_std": 0.14433756470680237,
      "rewards/accuracy_reward": 0.875,
      "rewards/format_reward": 0.0,
      "step": 51
    },
    {
      "completion_length": 3.125,
      "epoch": 0.005145203581853263,
      "grad_norm": 11.261039733886719,
      "kl": 0.14239501953125,
      "learning_rate": 9.489999999999999e-07,
      "loss": 0.0057,
      "reward": 0.6875,
      "reward_std": 0.125,
      "rewards/accuracy_reward": 0.6875,
      "rewards/format_reward": 0.0,
      "step": 52
    },
    {
      "completion_length": 3.25,
      "epoch": 0.00524414980458121,
      "grad_norm": 0.3226803243160248,
      "kl": 0.0335235595703125,
      "learning_rate": 9.479999999999999e-07,
      "loss": 0.0013,
      "reward": 0.75,
      "reward_std": 0.0,
      "rewards/accuracy_reward": 0.75,
      "rewards/format_reward": 0.0,
      "step": 53
    },
    {
      "completion_length": 2.9375,
      "epoch": 0.005343096027309158,
      "grad_norm": 27.19357681274414,
      "kl": 0.1173553466796875,
      "learning_rate": 9.469999999999999e-07,
      "loss": 0.0047,
      "reward": 0.9375,
      "reward_std": 0.125,
      "rewards/accuracy_reward": 0.9375,
      "rewards/format_reward": 0.0,
      "step": 54
    },
    {
      "completion_length": 2.25,
      "epoch": 0.005442042250037105,
      "grad_norm": 0.14267660677433014,
      "kl": 0.0958251953125,
      "learning_rate": 9.459999999999999e-07,
      "loss": 0.0038,
      "reward": 0.75,
      "reward_std": 0.0,
      "rewards/accuracy_reward": 0.75,
      "rewards/format_reward": 0.0,
      "step": 55
    },
    {
      "completion_length": 3.125,
      "epoch": 0.005540988472765052,
      "grad_norm": 1.3845109939575195,
      "kl": 0.070556640625,
      "learning_rate": 9.45e-07,
      "loss": 0.0028,
      "reward": 0.75,
      "reward_std": 0.0,
      "rewards/accuracy_reward": 0.75,
      "rewards/format_reward": 0.0,
      "step": 56
    },
    {
      "completion_length": 3.5,
      "epoch": 0.005639934695492999,
      "grad_norm": 0.3318254351615906,
      "kl": 0.07275390625,
      "learning_rate": 9.439999999999999e-07,
      "loss": 0.0029,
      "reward": 0.75,
      "reward_std": 0.0,
      "rewards/accuracy_reward": 0.75,
      "rewards/format_reward": 0.0,
      "step": 57
    },
    {
      "completion_length": 3.5,
      "epoch": 0.005738880918220947,
      "grad_norm": 79.56658172607422,
      "kl": 0.147705078125,
      "learning_rate": 9.429999999999999e-07,
      "loss": 0.0059,
      "reward": 0.625,
      "reward_std": 0.14433756470680237,
      "rewards/accuracy_reward": 0.625,
      "rewards/format_reward": 0.0,
      "step": 58
    },
    {
      "completion_length": 3.25,
      "epoch": 0.005837827140948894,
      "grad_norm": 29.595632553100586,
      "kl": 0.08984375,
      "learning_rate": 9.419999999999999e-07,
      "loss": 0.0036,
      "reward": 0.9375,
      "reward_std": 0.125,
      "rewards/accuracy_reward": 0.9375,
      "rewards/format_reward": 0.0,
      "step": 59
    },
    {
      "completion_length": 3.375,
      "epoch": 0.0059367733636768415,
      "grad_norm": 1.569680094718933,
      "kl": 0.1719970703125,
      "learning_rate": 9.409999999999999e-07,
      "loss": 0.0069,
      "reward": 0.5,
      "reward_std": 0.0,
      "rewards/accuracy_reward": 0.5,
      "rewards/format_reward": 0.0,
      "step": 60
    },
    {
      "completion_length": 3.4375,
      "epoch": 0.006035719586404789,
      "grad_norm": 20.28252601623535,
      "kl": 0.01580810546875,
      "learning_rate": 9.399999999999999e-07,
      "loss": 0.0006,
      "reward": 0.6875,
      "reward_std": 0.125,
      "rewards/accuracy_reward": 0.6875,
      "rewards/format_reward": 0.0,
      "step": 61
    },
    {
      "completion_length": 4.0,
      "epoch": 0.006134665809132737,
      "grad_norm": 0.009604735299944878,
      "kl": 0.016845703125,
      "learning_rate": 9.389999999999999e-07,
      "loss": 0.0007,
      "reward": 1.0,
      "reward_std": 0.0,
      "rewards/accuracy_reward": 1.0,
      "rewards/format_reward": 0.0,
      "step": 62
    },
    {
      "completion_length": 3.0,
      "epoch": 0.006233612031860684,
      "grad_norm": 0.054051682353019714,
      "kl": 0.1220703125,
      "learning_rate": 9.379999999999998e-07,
      "loss": 0.0049,
      "reward": 1.0,
      "reward_std": 0.0,
      "rewards/accuracy_reward": 1.0,
      "rewards/format_reward": 0.0,
      "step": 63
    },
    {
      "completion_length": 3.25,
      "epoch": 0.006332558254588631,
      "grad_norm": 43.524681091308594,
      "kl": 0.41552734375,
      "learning_rate": 9.37e-07,
      "loss": 0.0166,
      "reward": 0.5625,
      "reward_std": 0.125,
      "rewards/accuracy_reward": 0.5625,
      "rewards/format_reward": 0.0,
      "step": 64
    },
    {
      "completion_length": 3.3125,
      "epoch": 0.006431504477316579,
      "grad_norm": 19.583276748657227,
      "kl": 0.4501953125,
      "learning_rate": 9.36e-07,
      "loss": 0.018,
      "reward": 0.6875,
      "reward_std": 0.125,
      "rewards/accuracy_reward": 0.6875,
      "rewards/format_reward": 0.0,
      "step": 65
    },
    {
      "completion_length": 3.5,
      "epoch": 0.006530450700044526,
      "grad_norm": 29.74109649658203,
      "kl": 0.42333984375,
      "learning_rate": 9.35e-07,
      "loss": 0.0169,
      "reward": 0.625,
      "reward_std": 0.25,
      "rewards/accuracy_reward": 0.625,
      "rewards/format_reward": 0.0,
      "step": 66
    },
    {
      "completion_length": 3.5625,
      "epoch": 0.006629396922772473,
      "grad_norm": 51.01466369628906,
      "kl": 0.45574951171875,
      "learning_rate": 9.34e-07,
      "loss": 0.0182,
      "reward": 0.9375,
      "reward_std": 0.125,
      "rewards/accuracy_reward": 0.9375,
      "rewards/format_reward": 0.0,
      "step": 67
    },
    {
      "completion_length": 2.875,
      "epoch": 0.00672834314550042,
      "grad_norm": 40.87276840209961,
      "kl": 0.11376953125,
      "learning_rate": 9.33e-07,
      "loss": 0.0046,
      "reward": 0.9375,
      "reward_std": 0.125,
      "rewards/accuracy_reward": 0.9375,
      "rewards/format_reward": 0.0,
      "step": 68
    },
    {
      "completion_length": 4.375,
      "epoch": 0.006827289368228368,
      "grad_norm": 920.1598510742188,
      "kl": 30.021240234375,
      "learning_rate": 9.32e-07,
      "loss": 1.1998,
      "reward": 0.6875,
      "reward_std": 0.125,
      "rewards/accuracy_reward": 0.6875,
      "rewards/format_reward": 0.0,
      "step": 69
    },
    {
      "completion_length": 3.25,
      "epoch": 0.006926235590956315,
      "grad_norm": 0.07895195484161377,
      "kl": 0.0203857421875,
      "learning_rate": 9.31e-07,
      "loss": 0.0008,
      "reward": 0.75,
      "reward_std": 0.0,
      "rewards/accuracy_reward": 0.75,
      "rewards/format_reward": 0.0,
      "step": 70
    },
    {
      "completion_length": 2.9375,
      "epoch": 0.0070251818136842625,
      "grad_norm": 72.71320343017578,
      "kl": 0.9306640625,
      "learning_rate": 9.3e-07,
      "loss": 0.0373,
      "reward": 0.6875,
      "reward_std": 0.125,
      "rewards/accuracy_reward": 0.6875,
      "rewards/format_reward": 0.0,
      "step": 71
    },
    {
      "completion_length": 3.5,
      "epoch": 0.00712412803641221,
      "grad_norm": 0.04532546177506447,
      "kl": 0.07568359375,
      "learning_rate": 9.29e-07,
      "loss": 0.003,
      "reward": 0.75,
      "reward_std": 0.0,
      "rewards/accuracy_reward": 0.75,
      "rewards/format_reward": 0.0,
      "step": 72
    },
    {
      "completion_length": 3.5,
      "epoch": 0.007223074259140158,
      "grad_norm": 59.64799118041992,
      "kl": 0.309326171875,
      "learning_rate": 9.28e-07,
      "loss": 0.0123,
      "reward": 0.625,
      "reward_std": 0.14433756470680237,
      "rewards/accuracy_reward": 0.625,
      "rewards/format_reward": 0.0,
      "step": 73
    },
    {
      "completion_length": 4.25,
      "epoch": 0.007322020481868105,
      "grad_norm": 0.022819090634584427,
      "kl": 0.129150390625,
      "learning_rate": 9.27e-07,
      "loss": 0.0052,
      "reward": 0.75,
      "reward_std": 0.0,
      "rewards/accuracy_reward": 0.75,
      "rewards/format_reward": 0.0,
      "step": 74
    },
    {
      "completion_length": 2.5,
      "epoch": 0.007420966704596052,
      "grad_norm": 0.4349958598613739,
      "kl": 0.1767578125,
      "learning_rate": 9.26e-07,
      "loss": 0.0071,
      "reward": 0.75,
      "reward_std": 0.0,
      "rewards/accuracy_reward": 0.75,
      "rewards/format_reward": 0.0,
      "step": 75
    },
    {
      "completion_length": 3.25,
      "epoch": 0.007519912927324,
      "grad_norm": 0.9703753590583801,
      "kl": 0.1021728515625,
      "learning_rate": 9.25e-07,
      "loss": 0.0041,
      "reward": 1.0,
      "reward_std": 0.0,
      "rewards/accuracy_reward": 1.0,
      "rewards/format_reward": 0.0,
      "step": 76
    },
    {
      "completion_length": 3.6875,
      "epoch": 0.007618859150051947,
      "grad_norm": 0.20156638324260712,
      "kl": 0.0416259765625,
      "learning_rate": 9.24e-07,
      "loss": 0.0017,
      "reward": 0.5,
      "reward_std": 0.0,
      "rewards/accuracy_reward": 0.5,
      "rewards/format_reward": 0.0,
      "step": 77
    },
    {
      "completion_length": 3.0,
      "epoch": 0.007717805372779894,
      "grad_norm": 0.1729861944913864,
      "kl": 0.1273193359375,
      "learning_rate": 9.23e-07,
      "loss": 0.0051,
      "reward": 0.5,
      "reward_std": 0.0,
      "rewards/accuracy_reward": 0.5,
      "rewards/format_reward": 0.0,
      "step": 78
    },
    {
      "completion_length": 2.75,
      "epoch": 0.007816751595507841,
      "grad_norm": 0.019643617793917656,
      "kl": 0.0714111328125,
      "learning_rate": 9.22e-07,
      "loss": 0.0029,
      "reward": 0.75,
      "reward_std": 0.0,
      "rewards/accuracy_reward": 0.75,
      "rewards/format_reward": 0.0,
      "step": 79
    },
    {
      "completion_length": 3.25,
      "epoch": 0.00791569781823579,
      "grad_norm": 0.017483701929450035,
      "kl": 0.23956298828125,
      "learning_rate": 9.21e-07,
      "loss": 0.0096,
      "reward": 1.0,
      "reward_std": 0.0,
      "rewards/accuracy_reward": 1.0,
      "rewards/format_reward": 0.0,
      "step": 80
    },
    {
      "completion_length": 3.75,
      "epoch": 0.008014644040963735,
      "grad_norm": 0.02477215975522995,
      "kl": 0.056396484375,
      "learning_rate": 9.2e-07,
      "loss": 0.0023,
      "reward": 1.0,
      "reward_std": 0.0,
      "rewards/accuracy_reward": 1.0,
      "rewards/format_reward": 0.0,
      "step": 81
    },
    {
      "completion_length": 2.25,
      "epoch": 0.008113590263691683,
      "grad_norm": 57.296180725097656,
      "kl": 0.0313720703125,
      "learning_rate": 9.19e-07,
      "loss": 0.0013,
      "reward": 0.875,
      "reward_std": 0.14433756470680237,
      "rewards/accuracy_reward": 0.875,
      "rewards/format_reward": 0.0,
      "step": 82
    },
    {
      "completion_length": 2.75,
      "epoch": 0.008212536486419631,
      "grad_norm": 0.0005820893566124141,
      "kl": 0.030029296875,
      "learning_rate": 9.18e-07,
      "loss": 0.0012,
      "reward": 0.75,
      "reward_std": 0.0,
      "rewards/accuracy_reward": 0.75,
      "rewards/format_reward": 0.0,
      "step": 83
    },
    {
      "completion_length": 2.75,
      "epoch": 0.008311482709147578,
      "grad_norm": 0.007898775860667229,
      "kl": 0.0718994140625,
      "learning_rate": 9.17e-07,
      "loss": 0.0029,
      "reward": 1.0,
      "reward_std": 0.0,
      "rewards/accuracy_reward": 1.0,
      "rewards/format_reward": 0.0,
      "step": 84
    },
    {
      "completion_length": 5.25,
      "epoch": 0.008410428931875526,
      "grad_norm": 17.927873611450195,
      "kl": 0.056640625,
      "learning_rate": 9.16e-07,
      "loss": 0.0023,
      "reward": 0.8125,
      "reward_std": 0.125,
      "rewards/accuracy_reward": 0.8125,
      "rewards/format_reward": 0.0,
      "step": 85
    },
    {
      "completion_length": 2.9375,
      "epoch": 0.008509375154603474,
      "grad_norm": 0.6610168814659119,
      "kl": 0.05419921875,
      "learning_rate": 9.15e-07,
      "loss": 0.0022,
      "reward": 1.0,
      "reward_std": 0.0,
      "rewards/accuracy_reward": 1.0,
      "rewards/format_reward": 0.0,
      "step": 86
    },
    {
      "completion_length": 6.0625,
      "epoch": 0.00860832137733142,
      "grad_norm": 35.296607971191406,
      "kl": 1.230712890625,
      "learning_rate": 9.14e-07,
      "loss": 0.0497,
      "reward": 0.625,
      "reward_std": 0.25,
      "rewards/accuracy_reward": 0.625,
      "rewards/format_reward": 0.0,
      "step": 87
    },
    {
      "completion_length": 2.5,
      "epoch": 0.008707267600059368,
      "grad_norm": 0.06378056854009628,
      "kl": 0.06463623046875,
      "learning_rate": 9.13e-07,
      "loss": 0.0026,
      "reward": 1.0,
      "reward_std": 0.0,
      "rewards/accuracy_reward": 1.0,
      "rewards/format_reward": 0.0,
      "step": 88
    },
    {
      "completion_length": 3.1875,
      "epoch": 0.008806213822787316,
      "grad_norm": 227.54002380371094,
      "kl": 0.4443359375,
      "learning_rate": 9.12e-07,
      "loss": 0.0178,
      "reward": 0.4375,
      "reward_std": 0.125,
      "rewards/accuracy_reward": 0.4375,
      "rewards/format_reward": 0.0,
      "step": 89
    },
    {
      "completion_length": 2.75,
      "epoch": 0.008905160045515262,
      "grad_norm": 0.0024500866420567036,
      "kl": 0.0645751953125,
      "learning_rate": 9.109999999999999e-07,
      "loss": 0.0026,
      "reward": 0.75,
      "reward_std": 0.0,
      "rewards/accuracy_reward": 0.75,
      "rewards/format_reward": 0.0,
      "step": 90
    },
    {
      "completion_length": 3.0,
      "epoch": 0.00900410626824321,
      "grad_norm": 33.444313049316406,
      "kl": 0.10693359375,
      "learning_rate": 9.1e-07,
      "loss": 0.0043,
      "reward": 0.9375,
      "reward_std": 0.125,
      "rewards/accuracy_reward": 0.9375,
      "rewards/format_reward": 0.0,
      "step": 91
    },
    {
      "completion_length": 3.6875,
      "epoch": 0.009103052490971156,
      "grad_norm": 19.867097854614258,
      "kl": 0.14208984375,
      "learning_rate": 9.09e-07,
      "loss": 0.0057,
      "reward": 0.6875,
      "reward_std": 0.125,
      "rewards/accuracy_reward": 0.6875,
      "rewards/format_reward": 0.0,
      "step": 92
    },
    {
      "completion_length": 3.625,
      "epoch": 0.009201998713699104,
      "grad_norm": 3.6420788764953613,
      "kl": 0.177490234375,
      "learning_rate": 9.08e-07,
      "loss": 0.0071,
      "reward": 0.75,
      "reward_std": 0.0,
      "rewards/accuracy_reward": 0.75,
      "rewards/format_reward": 0.0,
      "step": 93
    },
    {
      "completion_length": 2.0,
      "epoch": 0.009300944936427052,
      "grad_norm": 0.008027822710573673,
      "kl": 0.13671875,
      "learning_rate": 9.07e-07,
      "loss": 0.0055,
      "reward": 1.0,
      "reward_std": 0.0,
      "rewards/accuracy_reward": 1.0,
      "rewards/format_reward": 0.0,
      "step": 94
    },
    {
      "completion_length": 3.625,
      "epoch": 0.009399891159154999,
      "grad_norm": 51.053592681884766,
      "kl": 0.162353515625,
      "learning_rate": 9.06e-07,
      "loss": 0.0065,
      "reward": 0.8125,
      "reward_std": 0.125,
      "rewards/accuracy_reward": 0.8125,
      "rewards/format_reward": 0.0,
      "step": 95
    },
    {
      "completion_length": 2.5,
      "epoch": 0.009498837381882947,
      "grad_norm": 0.0023729894310235977,
      "kl": 0.025390625,
      "learning_rate": 9.05e-07,
      "loss": 0.001,
      "reward": 1.0,
      "reward_std": 0.0,
      "rewards/accuracy_reward": 1.0,
      "rewards/format_reward": 0.0,
      "step": 96
    },
    {
      "completion_length": 3.0,
      "epoch": 0.009597783604610895,
      "grad_norm": 0.1433296650648117,
      "kl": 0.1759033203125,
      "learning_rate": 9.039999999999999e-07,
      "loss": 0.0071,
      "reward": 0.75,
      "reward_std": 0.0,
      "rewards/accuracy_reward": 0.75,
      "rewards/format_reward": 0.0,
      "step": 97
    },
    {
      "completion_length": 3.25,
      "epoch": 0.009696729827338841,
      "grad_norm": 46.357540130615234,
      "kl": 0.344024658203125,
      "learning_rate": 9.03e-07,
      "loss": 0.0137,
      "reward": 0.9375,
      "reward_std": 0.125,
      "rewards/accuracy_reward": 0.9375,
      "rewards/format_reward": 0.0,
      "step": 98
    },
    {
      "completion_length": 4.5625,
      "epoch": 0.009795676050066789,
      "grad_norm": 25.5620059967041,
      "kl": 0.1181640625,
      "learning_rate": 9.02e-07,
      "loss": 0.0047,
      "reward": 0.625,
      "reward_std": 0.25,
      "rewards/accuracy_reward": 0.625,
      "rewards/format_reward": 0.0,
      "step": 99
    },
    {
      "completion_length": 3.8125,
      "epoch": 0.009894622272794735,
      "grad_norm": 31.894678115844727,
      "kl": 0.300445556640625,
      "learning_rate": 9.01e-07,
      "loss": 0.012,
      "reward": 0.3125,
      "reward_std": 0.125,
      "rewards/accuracy_reward": 0.3125,
      "rewards/format_reward": 0.0,
      "step": 100
    }
  ],
  "logging_steps": 1,
  "max_steps": 1000,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 1,
  "save_steps": 100,
  "stateful_callbacks": {
    "TrainerControl": {
      "args": {
        "should_epoch_stop": false,
        "should_evaluate": false,
        "should_log": false,
        "should_save": true,
        "should_training_stop": false
      },
      "attributes": {}
    }
  },
  "total_flos": 0.0,
  "train_batch_size": 1,
  "trial_name": null,
  "trial_params": null
}