{
  "best_global_step": null,
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 0.8,
  "eval_steps": 500,
  "global_step": 1000,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "completion_length": 46.553125,
      "epoch": 0.008,
      "grad_norm": 0.05134107172489166,
      "kl": 0.012939453125,
      "learning_rate": 9.95e-07,
      "loss": 0.0001,
      "reward": 2.703125,
      "reward_std": 0.11205126643180847,
      "rewards/accuracy_reward": 1.7125,
      "rewards/format_reward": 0.990625,
      "step": 10
    },
    {
      "completion_length": 49.34375,
      "epoch": 0.016,
      "grad_norm": 0.06966069340705872,
      "kl": 0.01898193359375,
      "learning_rate": 9.9e-07,
      "loss": 0.0002,
      "reward": 2.775,
      "reward_std": 0.05,
      "rewards/accuracy_reward": 1.78125,
      "rewards/format_reward": 0.99375,
      "step": 20
    },
    {
      "completion_length": 44.63125,
      "epoch": 0.024,
      "grad_norm": 5.11226749420166,
      "kl": 0.0212158203125,
      "learning_rate": 9.849999999999999e-07,
      "loss": 0.0002,
      "reward": 2.546875,
      "reward_std": 0.09568375647068024,
      "rewards/accuracy_reward": 1.55625,
      "rewards/format_reward": 0.990625,
      "step": 30
    },
    {
      "completion_length": 42.775,
      "epoch": 0.032,
      "grad_norm": 0.0820818841457367,
      "kl": 0.042626953125,
      "learning_rate": 9.8e-07,
      "loss": 0.0004,
      "reward": 2.775,
      "reward_std": 0.03943375647068024,
      "rewards/accuracy_reward": 1.775,
      "rewards/format_reward": 1.0,
      "step": 40
    },
    {
      "completion_length": 44.275,
      "epoch": 0.04,
      "grad_norm": 0.06030546873807907,
      "kl": 0.03828125,
      "learning_rate": 9.75e-07,
      "loss": 0.0004,
      "reward": 2.74375,
      "reward_std": 0.026933756470680238,
      "rewards/accuracy_reward": 1.74375,
      "rewards/format_reward": 1.0,
      "step": 50
    },
    {
      "completion_length": 50.990625,
      "epoch": 0.048,
      "grad_norm": 0.10005596280097961,
      "kl": 0.03060302734375,
      "learning_rate": 9.7e-07,
      "loss": 0.0003,
      "reward": 2.60625,
      "reward_std": 0.10193375647068023,
      "rewards/accuracy_reward": 1.60625,
      "rewards/format_reward": 1.0,
      "step": 60
    },
    {
      "completion_length": 54.375,
      "epoch": 0.056,
      "grad_norm": 4.453707695007324,
      "kl": 0.0556640625,
      "learning_rate": 9.649999999999999e-07,
      "loss": 0.0006,
      "reward": 2.590625,
      "reward_std": 0.08318375647068024,
      "rewards/accuracy_reward": 1.59375,
      "rewards/format_reward": 0.996875,
      "step": 70
    },
    {
      "completion_length": 48.815625,
      "epoch": 0.064,
      "grad_norm": 2.5629329681396484,
      "kl": 0.040283203125,
      "learning_rate": 9.6e-07,
      "loss": 0.0004,
      "reward": 2.765625,
      "reward_std": 0.058183756470680234,
      "rewards/accuracy_reward": 1.76875,
      "rewards/format_reward": 0.996875,
      "step": 80
    },
    {
      "completion_length": 47.53125,
      "epoch": 0.072,
      "grad_norm": 0.08292120695114136,
      "kl": 0.0712646484375,
      "learning_rate": 9.55e-07,
      "loss": 0.0007,
      "reward": 2.825,
      "reward_std": 0.05,
      "rewards/accuracy_reward": 1.825,
      "rewards/format_reward": 1.0,
      "step": 90
    },
    {
      "completion_length": 46.703125,
      "epoch": 0.08,
      "grad_norm": 2.7465286254882812,
      "kl": 0.05367431640625,
      "learning_rate": 9.499999999999999e-07,
      "loss": 0.0005,
      "reward": 2.71875,
      "reward_std": 0.07693375647068024,
      "rewards/accuracy_reward": 1.71875,
      "rewards/format_reward": 1.0,
      "step": 100
    },
    {
      "completion_length": 46.225,
      "epoch": 0.088,
      "grad_norm": 2.1839213371276855,
      "kl": 0.0655517578125,
      "learning_rate": 9.45e-07,
      "loss": 0.0007,
      "reward": 2.609375,
      "reward_std": 0.03125,
      "rewards/accuracy_reward": 1.61875,
      "rewards/format_reward": 0.990625,
      "step": 110
    },
    {
      "completion_length": 44.096875,
      "epoch": 0.096,
      "grad_norm": 0.07181887328624725,
      "kl": 0.06865234375,
      "learning_rate": 9.399999999999999e-07,
      "loss": 0.0007,
      "reward": 2.71875,
      "reward_std": 0.0125,
      "rewards/accuracy_reward": 1.71875,
      "rewards/format_reward": 1.0,
      "step": 120
    },
    {
      "completion_length": 44.590625,
      "epoch": 0.104,
      "grad_norm": 0.09902142733335495,
      "kl": 0.0936767578125,
      "learning_rate": 9.35e-07,
      "loss": 0.0009,
      "reward": 2.56875,
      "reward_std": 0.0625,
      "rewards/accuracy_reward": 1.575,
      "rewards/format_reward": 0.99375,
      "step": 130
    },
    {
      "completion_length": 43.81875,
      "epoch": 0.112,
      "grad_norm": 2.340815305709839,
      "kl": 0.066015625,
      "learning_rate": 9.3e-07,
      "loss": 0.0007,
      "reward": 2.75,
      "reward_std": 0.025,
      "rewards/accuracy_reward": 1.75,
      "rewards/format_reward": 1.0,
      "step": 140
    },
    {
      "completion_length": 49.06875,
      "epoch": 0.12,
      "grad_norm": 2.58245849609375,
      "kl": 0.0600341796875,
      "learning_rate": 9.25e-07,
      "loss": 0.0006,
      "reward": 2.7125,
      "reward_std": 0.125,
      "rewards/accuracy_reward": 1.7125,
      "rewards/format_reward": 1.0,
      "step": 150
    },
    {
      "completion_length": 52.84375,
      "epoch": 0.128,
      "grad_norm": 0.06839890778064728,
      "kl": 0.0785400390625,
      "learning_rate": 9.2e-07,
      "loss": 0.0008,
      "reward": 2.7,
      "reward_std": 0.03943375647068024,
      "rewards/accuracy_reward": 1.7,
      "rewards/format_reward": 1.0,
      "step": 160
    },
    {
      "completion_length": 47.7,
      "epoch": 0.136,
      "grad_norm": 0.11428700387477875,
      "kl": 0.06865234375,
      "learning_rate": 9.15e-07,
      "loss": 0.0007,
      "reward": 2.75,
      "reward_std": 0.025,
      "rewards/accuracy_reward": 1.75,
      "rewards/format_reward": 1.0,
      "step": 170
    },
    {
      "completion_length": 43.478125,
      "epoch": 0.144,
      "grad_norm": 2.188392400741577,
      "kl": 0.062451171875,
      "learning_rate": 9.1e-07,
      "loss": 0.0006,
      "reward": 2.615625,
      "reward_std": 0.06875,
      "rewards/accuracy_reward": 1.61875,
      "rewards/format_reward": 0.996875,
      "step": 180
    },
    {
      "completion_length": 42.540625,
      "epoch": 0.152,
      "grad_norm": 3.399991512298584,
      "kl": 0.076953125,
      "learning_rate": 9.05e-07,
      "loss": 0.0008,
      "reward": 2.64375,
      "reward_std": 0.09136751294136047,
      "rewards/accuracy_reward": 1.64375,
      "rewards/format_reward": 1.0,
      "step": 190
    },
    {
      "completion_length": 50.21875,
      "epoch": 0.16,
      "grad_norm": 0.10214658826589584,
      "kl": 0.09365234375,
      "learning_rate": 9e-07,
      "loss": 0.0009,
      "reward": 2.784375,
      "reward_std": 0.05625,
      "rewards/accuracy_reward": 1.7875,
      "rewards/format_reward": 0.996875,
      "step": 200
    },
    {
      "completion_length": 54.35,
      "epoch": 0.168,
      "grad_norm": 0.08639144152402878,
      "kl": 0.1749267578125,
      "learning_rate": 8.95e-07,
      "loss": 0.0017,
      "reward": 2.7875,
      "reward_std": 0.014433756470680237,
      "rewards/accuracy_reward": 1.7875,
      "rewards/format_reward": 1.0,
      "step": 210
    },
    {
      "completion_length": 56.29375,
      "epoch": 0.176,
      "grad_norm": 0.06954076141119003,
      "kl": 0.119091796875,
      "learning_rate": 8.9e-07,
      "loss": 0.0012,
      "reward": 2.75,
      "reward_std": 0.025,
      "rewards/accuracy_reward": 1.75625,
      "rewards/format_reward": 0.99375,
      "step": 220
    },
    {
      "completion_length": 48.8875,
      "epoch": 0.184,
      "grad_norm": 0.06490013003349304,
      "kl": 0.12080078125,
      "learning_rate": 8.85e-07,
      "loss": 0.0012,
      "reward": 2.65,
      "reward_std": 0.07886751294136048,
      "rewards/accuracy_reward": 1.65,
      "rewards/format_reward": 1.0,
      "step": 230
    },
    {
      "completion_length": 41.115625,
      "epoch": 0.192,
      "grad_norm": 0.12679292261600494,
      "kl": 0.12470703125,
      "learning_rate": 8.799999999999999e-07,
      "loss": 0.0012,
      "reward": 2.8125,
      "reward_std": 0.07886751294136048,
      "rewards/accuracy_reward": 1.81875,
      "rewards/format_reward": 0.99375,
      "step": 240
    },
    {
      "completion_length": 40.4125,
      "epoch": 0.2,
      "grad_norm": 0.11438746750354767,
      "kl": 30.11142578125,
      "learning_rate": 8.75e-07,
      "loss": 0.3012,
      "reward": 2.725,
      "reward_std": 0.05,
      "rewards/accuracy_reward": 1.73125,
      "rewards/format_reward": 0.99375,
      "step": 250
    },
    {
      "completion_length": 46.790625,
      "epoch": 0.208,
      "grad_norm": 2.282456159591675,
      "kl": 0.10205078125,
      "learning_rate": 8.699999999999999e-07,
      "loss": 0.001,
      "reward": 2.578125,
      "reward_std": 0.06875,
      "rewards/accuracy_reward": 1.58125,
      "rewards/format_reward": 0.996875,
      "step": 260
    },
    {
      "completion_length": 52.253125,
      "epoch": 0.216,
      "grad_norm": 1.9098315238952637,
      "kl": 0.105810546875,
      "learning_rate": 8.65e-07,
      "loss": 0.0011,
      "reward": 2.8,
      "reward_std": 0.07886751294136048,
      "rewards/accuracy_reward": 1.8,
      "rewards/format_reward": 1.0,
      "step": 270
    },
    {
      "completion_length": 49.828125,
      "epoch": 0.224,
      "grad_norm": 0.058336157351732254,
      "kl": 0.0714599609375,
      "learning_rate": 8.599999999999999e-07,
      "loss": 0.0007,
      "reward": 2.73125,
      "reward_std": 0.0375,
      "rewards/accuracy_reward": 1.73125,
      "rewards/format_reward": 1.0,
      "step": 280
    },
    {
      "completion_length": 47.14375,
      "epoch": 0.232,
      "grad_norm": 0.07711385935544968,
      "kl": 0.08037109375,
      "learning_rate": 8.55e-07,
      "loss": 0.0008,
      "reward": 2.875,
      "reward_std": 0.03943375647068024,
      "rewards/accuracy_reward": 1.875,
      "rewards/format_reward": 1.0,
      "step": 290
    },
    {
      "completion_length": 46.759375,
      "epoch": 0.24,
      "grad_norm": 0.059466563165187836,
      "kl": 0.079248046875,
      "learning_rate": 8.499999999999999e-07,
      "loss": 0.0008,
      "reward": 2.70625,
      "reward_std": 0.051933756470680235,
      "rewards/accuracy_reward": 1.70625,
      "rewards/format_reward": 1.0,
      "step": 300
    },
    {
      "completion_length": 48.540625,
      "epoch": 0.248,
      "grad_norm": 3.2264294624328613,
      "kl": 0.0768310546875,
      "learning_rate": 8.45e-07,
      "loss": 0.0008,
      "reward": 2.7375,
      "reward_std": 0.075,
      "rewards/accuracy_reward": 1.7375,
      "rewards/format_reward": 1.0,
      "step": 310
    },
    {
      "completion_length": 46.85,
      "epoch": 0.256,
      "grad_norm": 0.08373435586690903,
      "kl": 0.088037109375,
      "learning_rate": 8.399999999999999e-07,
      "loss": 0.0009,
      "reward": 2.728125,
      "reward_std": 0.08318375647068024,
      "rewards/accuracy_reward": 1.73125,
      "rewards/format_reward": 0.996875,
      "step": 320
    },
    {
      "completion_length": 45.0125,
      "epoch": 0.264,
      "grad_norm": 0.08248328417539597,
      "kl": 0.084375,
      "learning_rate": 8.349999999999999e-07,
      "loss": 0.0008,
      "reward": 2.684375,
      "reward_std": 0.04568375647068024,
      "rewards/accuracy_reward": 1.6875,
      "rewards/format_reward": 0.996875,
      "step": 330
    },
    {
      "completion_length": 47.1,
      "epoch": 0.272,
      "grad_norm": 0.08357389271259308,
      "kl": 0.07880859375,
      "learning_rate": 8.299999999999999e-07,
      "loss": 0.0008,
      "reward": 2.628125,
      "reward_std": 0.03318375647068024,
      "rewards/accuracy_reward": 1.63125,
      "rewards/format_reward": 0.996875,
      "step": 340
    },
    {
      "completion_length": 48.95625,
      "epoch": 0.28,
      "grad_norm": 1.7901896238327026,
      "kl": 0.084033203125,
      "learning_rate": 8.249999999999999e-07,
      "loss": 0.0008,
      "reward": 2.609375,
      "reward_std": 0.03125,
      "rewards/accuracy_reward": 1.6125,
      "rewards/format_reward": 0.996875,
      "step": 350
    },
    {
      "completion_length": 46.15,
      "epoch": 0.288,
      "grad_norm": 0.07559721171855927,
      "kl": 0.14404296875,
      "learning_rate": 8.199999999999999e-07,
      "loss": 0.0014,
      "reward": 2.778125,
      "reward_std": 0.04375,
      "rewards/accuracy_reward": 1.78125,
      "rewards/format_reward": 0.996875,
      "step": 360
    },
    {
      "completion_length": 43.8375,
      "epoch": 0.296,
      "grad_norm": 3.8727450370788574,
      "kl": 0.109521484375,
      "learning_rate": 8.149999999999999e-07,
      "loss": 0.0011,
      "reward": 2.83125,
      "reward_std": 0.0375,
      "rewards/accuracy_reward": 1.83125,
      "rewards/format_reward": 1.0,
      "step": 370
    },
    {
      "completion_length": 45.9625,
      "epoch": 0.304,
      "grad_norm": 0.05233932286500931,
      "kl": 0.0930908203125,
      "learning_rate": 8.1e-07,
      "loss": 0.0009,
      "reward": 2.796875,
      "reward_std": 0.00625,
      "rewards/accuracy_reward": 1.8,
      "rewards/format_reward": 0.996875,
      "step": 380
    },
    {
      "completion_length": 49.55,
      "epoch": 0.312,
      "grad_norm": 4.457919120788574,
      "kl": 0.0723876953125,
      "learning_rate": 8.05e-07,
      "loss": 0.0007,
      "reward": 2.75,
      "reward_std": 0.053867512941360475,
      "rewards/accuracy_reward": 1.75,
      "rewards/format_reward": 1.0,
      "step": 390
    },
    {
      "completion_length": 50.909375,
      "epoch": 0.32,
      "grad_norm": 0.050397127866744995,
      "kl": 0.08388671875,
      "learning_rate": 8e-07,
      "loss": 0.0008,
      "reward": 2.7625,
      "reward_std": 0.025,
      "rewards/accuracy_reward": 1.7625,
      "rewards/format_reward": 1.0,
      "step": 400
    },
    {
      "completion_length": 49.165625,
      "epoch": 0.328,
      "grad_norm": 0.1388678401708603,
      "kl": 0.084033203125,
      "learning_rate": 7.95e-07,
      "loss": 0.0008,
      "reward": 2.6875,
      "reward_std": 0.014433756470680237,
      "rewards/accuracy_reward": 1.6875,
      "rewards/format_reward": 1.0,
      "step": 410
    },
    {
      "completion_length": 48.190625,
      "epoch": 0.336,
      "grad_norm": 2.034395933151245,
      "kl": 0.078125,
      "learning_rate": 7.9e-07,
      "loss": 0.0008,
      "reward": 2.76875,
      "reward_std": 0.0375,
      "rewards/accuracy_reward": 1.76875,
      "rewards/format_reward": 1.0,
      "step": 420
    },
    {
      "completion_length": 49.45,
      "epoch": 0.344,
      "grad_norm": 2.2621846199035645,
      "kl": 0.075048828125,
      "learning_rate": 7.85e-07,
      "loss": 0.0008,
      "reward": 2.634375,
      "reward_std": 0.03125,
      "rewards/accuracy_reward": 1.6375,
      "rewards/format_reward": 0.996875,
      "step": 430
    },
    {
      "completion_length": 52.03125,
      "epoch": 0.352,
      "grad_norm": 2.9660024642944336,
      "kl": 0.0776123046875,
      "learning_rate": 7.799999999999999e-07,
      "loss": 0.0008,
      "reward": 2.7625,
      "reward_std": 0.03943375647068024,
      "rewards/accuracy_reward": 1.7625,
      "rewards/format_reward": 1.0,
      "step": 440
    },
    {
      "completion_length": 52.496875,
      "epoch": 0.36,
      "grad_norm": 0.040182050317525864,
      "kl": 0.0726806640625,
      "learning_rate": 7.75e-07,
      "loss": 0.0007,
      "reward": 2.6875,
      "reward_std": 0.025,
      "rewards/accuracy_reward": 1.6875,
      "rewards/format_reward": 1.0,
      "step": 450
    },
    {
      "completion_length": 51.725,
      "epoch": 0.368,
      "grad_norm": 0.06841447949409485,
      "kl": 0.0802001953125,
      "learning_rate": 7.699999999999999e-07,
      "loss": 0.0008,
      "reward": 2.8,
      "reward_std": 0.0,
      "rewards/accuracy_reward": 1.8,
      "rewards/format_reward": 1.0,
      "step": 460
    },
    {
      "completion_length": 48.14375,
      "epoch": 0.376,
      "grad_norm": 0.04733005911111832,
      "kl": 0.0659912109375,
      "learning_rate": 7.65e-07,
      "loss": 0.0007,
      "reward": 2.61875,
      "reward_std": 0.0125,
      "rewards/accuracy_reward": 1.61875,
      "rewards/format_reward": 1.0,
      "step": 470
    },
    {
      "completion_length": 46.89375,
      "epoch": 0.384,
      "grad_norm": 2.7484917640686035,
      "kl": 0.0697998046875,
      "learning_rate": 7.599999999999999e-07,
      "loss": 0.0007,
      "reward": 2.74375,
      "reward_std": 0.09136751294136047,
      "rewards/accuracy_reward": 1.74375,
      "rewards/format_reward": 1.0,
      "step": 480
    },
    {
      "completion_length": 48.48125,
      "epoch": 0.392,
      "grad_norm": 1.7968782186508179,
      "kl": 0.0580078125,
      "learning_rate": 7.55e-07,
      "loss": 0.0006,
      "reward": 2.7125,
      "reward_std": 0.025,
      "rewards/accuracy_reward": 1.7125,
      "rewards/format_reward": 1.0,
      "step": 490
    },
    {
      "completion_length": 50.034375,
      "epoch": 0.4,
      "grad_norm": 0.08426347374916077,
      "kl": 0.077099609375,
      "learning_rate": 7.5e-07,
      "loss": 0.0008,
      "reward": 2.68125,
      "reward_std": 0.04136751294136047,
      "rewards/accuracy_reward": 1.68125,
      "rewards/format_reward": 1.0,
      "step": 500
    },
    {
      "completion_length": 51.378125,
      "epoch": 0.408,
      "grad_norm": 0.040815118700265884,
      "kl": 0.06416015625,
      "learning_rate": 7.45e-07,
      "loss": 0.0006,
      "reward": 2.73125,
      "reward_std": 0.026933756470680238,
      "rewards/accuracy_reward": 1.73125,
      "rewards/format_reward": 1.0,
      "step": 510
    },
    {
      "completion_length": 49.878125,
      "epoch": 0.416,
      "grad_norm": 0.06027600169181824,
      "kl": 0.0675537109375,
      "learning_rate": 7.4e-07,
      "loss": 0.0007,
      "reward": 2.671875,
      "reward_std": 0.00625,
      "rewards/accuracy_reward": 1.675,
      "rewards/format_reward": 0.996875,
      "step": 520
    },
    {
      "completion_length": 47.921875,
      "epoch": 0.424,
      "grad_norm": 0.06604389101266861,
      "kl": 0.07177734375,
      "learning_rate": 7.35e-07,
      "loss": 0.0007,
      "reward": 2.675,
      "reward_std": 0.08943375647068023,
      "rewards/accuracy_reward": 1.68125,
      "rewards/format_reward": 0.99375,
      "step": 530
    },
    {
      "completion_length": 41.890625,
      "epoch": 0.432,
      "grad_norm": 2.579275608062744,
      "kl": 0.080859375,
      "learning_rate": 7.3e-07,
      "loss": 0.0008,
      "reward": 2.75,
      "reward_std": 0.03943375647068024,
      "rewards/accuracy_reward": 1.75,
      "rewards/format_reward": 1.0,
      "step": 540
    },
    {
      "completion_length": 44.046875,
      "epoch": 0.44,
      "grad_norm": 0.04179125651717186,
      "kl": 0.076025390625,
      "learning_rate": 7.249999999999999e-07,
      "loss": 0.0008,
      "reward": 2.5375,
      "reward_std": 0.025,
      "rewards/accuracy_reward": 1.5375,
      "rewards/format_reward": 1.0,
      "step": 550
    },
    {
      "completion_length": 46.725,
      "epoch": 0.448,
      "grad_norm": 0.04865502566099167,
      "kl": 0.075830078125,
      "learning_rate": 7.2e-07,
      "loss": 0.0008,
      "reward": 2.66875,
      "reward_std": 0.0125,
      "rewards/accuracy_reward": 1.66875,
      "rewards/format_reward": 1.0,
      "step": 560
    },
    {
      "completion_length": 48.0875,
      "epoch": 0.456,
      "grad_norm": 0.1781499981880188,
      "kl": 92.47451171875,
      "learning_rate": 7.149999999999999e-07,
      "loss": 0.9243,
      "reward": 2.8,
      "reward_std": 0.025,
      "rewards/accuracy_reward": 1.8,
      "rewards/format_reward": 1.0,
      "step": 570
    },
    {
      "completion_length": 49.703125,
      "epoch": 0.464,
      "grad_norm": 0.05255131423473358,
      "kl": 0.0656982421875,
      "learning_rate": 7.1e-07,
      "loss": 0.0007,
      "reward": 2.6625,
      "reward_std": 0.025,
      "rewards/accuracy_reward": 1.6625,
      "rewards/format_reward": 1.0,
      "step": 580
    },
    {
      "completion_length": 52.06875,
      "epoch": 0.472,
      "grad_norm": 0.1266418695449829,
      "kl": 0.0781005859375,
      "learning_rate": 7.049999999999999e-07,
      "loss": 0.0008,
      "reward": 2.75,
      "reward_std": 0.025,
      "rewards/accuracy_reward": 1.75,
      "rewards/format_reward": 1.0,
      "step": 590
    },
    {
      "completion_length": 53.475,
      "epoch": 0.48,
      "grad_norm": 0.07561592757701874,
      "kl": 0.0699951171875,
      "learning_rate": 7e-07,
      "loss": 0.0007,
      "reward": 2.6875,
      "reward_std": 0.053867512941360475,
      "rewards/accuracy_reward": 1.6875,
      "rewards/format_reward": 1.0,
      "step": 600
    },
    {
      "completion_length": 52.0625,
      "epoch": 0.488,
      "grad_norm": 0.04883831739425659,
      "kl": 0.0799560546875,
      "learning_rate": 6.949999999999999e-07,
      "loss": 0.0008,
      "reward": 2.65625,
      "reward_std": 0.0125,
      "rewards/accuracy_reward": 1.65625,
      "rewards/format_reward": 1.0,
      "step": 610
    },
    {
      "completion_length": 49.54375,
      "epoch": 0.496,
      "grad_norm": 2.3243064880371094,
      "kl": 0.0752685546875,
      "learning_rate": 6.9e-07,
      "loss": 0.0008,
      "reward": 2.815625,
      "reward_std": 0.058183756470680234,
      "rewards/accuracy_reward": 1.81875,
      "rewards/format_reward": 0.996875,
      "step": 620
    },
    {
      "completion_length": 48.690625,
      "epoch": 0.504,
      "grad_norm": 0.06750122457742691,
      "kl": 0.06513671875,
      "learning_rate": 6.85e-07,
      "loss": 0.0007,
      "reward": 2.84375,
      "reward_std": 0.0375,
      "rewards/accuracy_reward": 1.84375,
      "rewards/format_reward": 1.0,
      "step": 630
    },
    {
      "completion_length": 49.271875,
      "epoch": 0.512,
      "grad_norm": 0.056099992245435715,
      "kl": 0.0666259765625,
      "learning_rate": 6.800000000000001e-07,
      "loss": 0.0007,
      "reward": 2.69375,
      "reward_std": 0.0125,
      "rewards/accuracy_reward": 1.69375,
      "rewards/format_reward": 1.0,
      "step": 640
    },
    {
      "completion_length": 46.4375,
      "epoch": 0.52,
      "grad_norm": 0.0455087348818779,
      "kl": 0.0546630859375,
      "learning_rate": 6.75e-07,
      "loss": 0.0005,
      "reward": 2.75625,
      "reward_std": 0.0125,
      "rewards/accuracy_reward": 1.75625,
      "rewards/format_reward": 1.0,
      "step": 650
    },
    {
      "completion_length": 46.496875,
      "epoch": 0.528,
      "grad_norm": 0.05418640747666359,
      "kl": 0.0645263671875,
      "learning_rate": 6.7e-07,
      "loss": 0.0006,
      "reward": 2.6875,
      "reward_std": 0.014433756470680237,
      "rewards/accuracy_reward": 1.6875,
      "rewards/format_reward": 1.0,
      "step": 660
    },
    {
      "completion_length": 46.328125,
      "epoch": 0.536,
      "grad_norm": 4.0458455085754395,
      "kl": 0.081103515625,
      "learning_rate": 6.65e-07,
      "loss": 0.0008,
      "reward": 2.65625,
      "reward_std": 0.08080126941204072,
      "rewards/accuracy_reward": 1.65625,
      "rewards/format_reward": 1.0,
      "step": 670
    },
    {
      "completion_length": 48.91875,
      "epoch": 0.544,
      "grad_norm": 0.04970540851354599,
      "kl": 0.0717041015625,
      "learning_rate": 6.6e-07,
      "loss": 0.0007,
      "reward": 2.7625,
      "reward_std": 0.04330126941204071,
      "rewards/accuracy_reward": 1.7625,
      "rewards/format_reward": 1.0,
      "step": 680
    },
    {
      "completion_length": 49.01875,
      "epoch": 0.552,
      "grad_norm": 0.1746923178434372,
      "kl": 0.073779296875,
      "learning_rate": 6.55e-07,
      "loss": 0.0007,
      "reward": 2.75,
      "reward_std": 0.025,
      "rewards/accuracy_reward": 1.75,
      "rewards/format_reward": 1.0,
      "step": 690
    },
    {
      "completion_length": 48.3125,
      "epoch": 0.56,
      "grad_norm": 0.051023293286561966,
      "kl": 0.06783447265625,
      "learning_rate": 6.5e-07,
      "loss": 0.0007,
      "reward": 2.7625,
      "reward_std": 0.025,
      "rewards/accuracy_reward": 1.7625,
      "rewards/format_reward": 1.0,
      "step": 700
    },
    {
      "completion_length": 49.11875,
      "epoch": 0.568,
      "grad_norm": 0.07166194915771484,
      "kl": 0.0619384765625,
      "learning_rate": 6.45e-07,
      "loss": 0.0006,
      "reward": 2.7375,
      "reward_std": 0.014433756470680237,
      "rewards/accuracy_reward": 1.7375,
      "rewards/format_reward": 1.0,
      "step": 710
    },
    {
      "completion_length": 51.103125,
      "epoch": 0.576,
      "grad_norm": 0.08520376682281494,
      "kl": 0.0830078125,
      "learning_rate": 6.4e-07,
      "loss": 0.0008,
      "reward": 2.7375,
      "reward_std": 0.014433756470680237,
      "rewards/accuracy_reward": 1.7375,
      "rewards/format_reward": 1.0,
      "step": 720
    },
    {
      "completion_length": 49.615625,
      "epoch": 0.584,
      "grad_norm": 0.10399647802114487,
      "kl": 0.0688232421875,
      "learning_rate": 6.35e-07,
      "loss": 0.0007,
      "reward": 2.69375,
      "reward_std": 0.0125,
      "rewards/accuracy_reward": 1.69375,
      "rewards/format_reward": 1.0,
      "step": 730
    },
    {
      "completion_length": 50.596875,
      "epoch": 0.592,
      "grad_norm": 0.06369677186012268,
      "kl": 0.087890625,
      "learning_rate": 6.3e-07,
      "loss": 0.0009,
      "reward": 2.61875,
      "reward_std": 0.0375,
      "rewards/accuracy_reward": 1.61875,
      "rewards/format_reward": 1.0,
      "step": 740
    },
    {
      "completion_length": 50.56875,
      "epoch": 0.6,
      "grad_norm": 0.07198835164308548,
      "kl": 0.10087890625,
      "learning_rate": 6.249999999999999e-07,
      "loss": 0.001,
      "reward": 2.625,
      "reward_std": 0.06443375647068024,
      "rewards/accuracy_reward": 1.625,
      "rewards/format_reward": 1.0,
      "step": 750
    },
    {
      "completion_length": 50.953125,
      "epoch": 0.608,
      "grad_norm": 0.04980659857392311,
      "kl": 0.101806640625,
      "learning_rate": 6.2e-07,
      "loss": 0.001,
      "reward": 2.721875,
      "reward_std": 0.03125,
      "rewards/accuracy_reward": 1.725,
      "rewards/format_reward": 0.996875,
      "step": 760
    },
    {
      "completion_length": 46.609375,
      "epoch": 0.616,
      "grad_norm": 2.673631191253662,
      "kl": 0.0730224609375,
      "learning_rate": 6.149999999999999e-07,
      "loss": 0.0007,
      "reward": 2.6875,
      "reward_std": 0.03943375647068024,
      "rewards/accuracy_reward": 1.6875,
      "rewards/format_reward": 1.0,
      "step": 770
    },
    {
      "completion_length": 46.16875,
      "epoch": 0.624,
      "grad_norm": 0.07191024720668793,
      "kl": 0.07197265625,
      "learning_rate": 6.1e-07,
      "loss": 0.0007,
      "reward": 2.6125,
      "reward_std": 0.03943375647068024,
      "rewards/accuracy_reward": 1.6125,
      "rewards/format_reward": 1.0,
      "step": 780
    },
    {
      "completion_length": 47.346875,
      "epoch": 0.632,
      "grad_norm": 0.31487828493118286,
      "kl": 0.0890625,
      "learning_rate": 6.049999999999999e-07,
      "loss": 0.0009,
      "reward": 2.7,
      "reward_std": 0.0,
      "rewards/accuracy_reward": 1.7,
      "rewards/format_reward": 1.0,
      "step": 790
    },
    {
      "completion_length": 48.5125,
      "epoch": 0.64,
      "grad_norm": 0.04281134530901909,
      "kl": 0.0651611328125,
      "learning_rate": 6e-07,
      "loss": 0.0007,
      "reward": 2.65,
      "reward_std": 0.0,
      "rewards/accuracy_reward": 1.65,
      "rewards/format_reward": 1.0,
      "step": 800
    },
    {
      "completion_length": 47.9625,
      "epoch": 0.648,
      "grad_norm": 1.7782899141311646,
      "kl": 0.0711669921875,
      "learning_rate": 5.949999999999999e-07,
      "loss": 0.0007,
      "reward": 2.634375,
      "reward_std": 0.04568375647068024,
      "rewards/accuracy_reward": 1.6375,
      "rewards/format_reward": 0.996875,
      "step": 810
    },
    {
      "completion_length": 47.7,
      "epoch": 0.656,
      "grad_norm": 0.9939271211624146,
      "kl": 0.07099609375,
      "learning_rate": 5.9e-07,
      "loss": 0.0007,
      "reward": 2.6625,
      "reward_std": 0.025,
      "rewards/accuracy_reward": 1.6625,
      "rewards/format_reward": 1.0,
      "step": 820
    },
    {
      "completion_length": 45.790625,
      "epoch": 0.664,
      "grad_norm": 0.05890406668186188,
      "kl": 0.0596923828125,
      "learning_rate": 5.849999999999999e-07,
      "loss": 0.0006,
      "reward": 2.696875,
      "reward_std": 0.00625,
      "rewards/accuracy_reward": 1.7,
      "rewards/format_reward": 0.996875,
      "step": 830
    },
    {
      "completion_length": 46.675,
      "epoch": 0.672,
      "grad_norm": 0.062360286712646484,
      "kl": 0.071240234375,
      "learning_rate": 5.8e-07,
      "loss": 0.0007,
      "reward": 2.5875,
      "reward_std": 0.025,
      "rewards/accuracy_reward": 1.5875,
      "rewards/format_reward": 1.0,
      "step": 840
    },
    {
      "completion_length": 47.621875,
      "epoch": 0.68,
      "grad_norm": 2.2732224464416504,
      "kl": 1.6703369140625,
      "learning_rate": 5.749999999999999e-07,
      "loss": 0.0167,
      "reward": 2.64375,
      "reward_std": 0.04136751294136047,
      "rewards/accuracy_reward": 1.64375,
      "rewards/format_reward": 1.0,
      "step": 850
    },
    {
      "completion_length": 50.35,
      "epoch": 0.688,
      "grad_norm": 2.1026487350463867,
      "kl": 0.09111328125,
      "learning_rate": 5.699999999999999e-07,
      "loss": 0.0009,
      "reward": 2.6625,
      "reward_std": 0.053867512941360475,
      "rewards/accuracy_reward": 1.6625,
      "rewards/format_reward": 1.0,
      "step": 860
    },
    {
      "completion_length": 52.321875,
      "epoch": 0.696,
      "grad_norm": 3.0173561573028564,
      "kl": 321.521728515625,
      "learning_rate": 5.649999999999999e-07,
      "loss": 3.2171,
      "reward": 2.46875,
      "reward_std": 0.04136751294136047,
      "rewards/accuracy_reward": 1.46875,
      "rewards/format_reward": 1.0,
      "step": 870
    },
    {
      "completion_length": 49.7625,
      "epoch": 0.704,
      "grad_norm": 0.06468257308006287,
      "kl": 0.39306640625,
      "learning_rate": 5.6e-07,
      "loss": 0.0039,
      "reward": 2.7125,
      "reward_std": 0.06443375647068024,
      "rewards/accuracy_reward": 1.7125,
      "rewards/format_reward": 1.0,
      "step": 880
    },
    {
      "completion_length": 48.240625,
      "epoch": 0.712,
      "grad_norm": 0.07906866073608398,
      "kl": 0.12939453125,
      "learning_rate": 5.55e-07,
      "loss": 0.0013,
      "reward": 2.68125,
      "reward_std": 0.026933756470680238,
      "rewards/accuracy_reward": 1.68125,
      "rewards/format_reward": 1.0,
      "step": 890
    },
    {
      "completion_length": 49.03125,
      "epoch": 0.72,
      "grad_norm": 0.07313551008701324,
      "kl": 0.18505859375,
      "learning_rate": 5.5e-07,
      "loss": 0.0019,
      "reward": 2.69375,
      "reward_std": 0.026933756470680238,
      "rewards/accuracy_reward": 1.69375,
      "rewards/format_reward": 1.0,
      "step": 900
    },
    {
      "completion_length": 49.60625,
      "epoch": 0.728,
      "grad_norm": 4.0763630867004395,
      "kl": 2.13671875,
      "learning_rate": 5.45e-07,
      "loss": 0.0213,
      "reward": 2.753125,
      "reward_std": 0.058183756470680234,
      "rewards/accuracy_reward": 1.75625,
      "rewards/format_reward": 0.996875,
      "step": 910
    },
    {
      "completion_length": 49.83125,
      "epoch": 0.736,
      "grad_norm": 4.245804786682129,
      "kl": 0.094384765625,
      "learning_rate": 5.4e-07,
      "loss": 0.0009,
      "reward": 2.75625,
      "reward_std": 0.04136751294136047,
      "rewards/accuracy_reward": 1.75625,
      "rewards/format_reward": 1.0,
      "step": 920
    },
    {
      "completion_length": 50.996875,
      "epoch": 0.744,
      "grad_norm": 39.86748504638672,
      "kl": 3.122021484375,
      "learning_rate": 5.35e-07,
      "loss": 0.0313,
      "reward": 2.8125,
      "reward_std": 0.03943375647068024,
      "rewards/accuracy_reward": 1.8125,
      "rewards/format_reward": 1.0,
      "step": 930
    },
    {
      "completion_length": 49.59375,
      "epoch": 0.752,
      "grad_norm": 0.05779128894209862,
      "kl": 0.0998046875,
      "learning_rate": 5.3e-07,
      "loss": 0.001,
      "reward": 2.79375,
      "reward_std": 0.051933756470680235,
      "rewards/accuracy_reward": 1.79375,
      "rewards/format_reward": 1.0,
      "step": 940
    },
    {
      "completion_length": 47.74375,
      "epoch": 0.76,
      "grad_norm": 0.056427907198667526,
      "kl": 0.445361328125,
      "learning_rate": 5.25e-07,
      "loss": 0.0045,
      "reward": 2.76875,
      "reward_std": 0.0125,
      "rewards/accuracy_reward": 1.76875,
      "rewards/format_reward": 1.0,
      "step": 950
    },
    {
      "completion_length": 48.50625,
      "epoch": 0.768,
      "grad_norm": 66.90420532226562,
      "kl": 2.232958984375,
      "learning_rate": 5.2e-07,
      "loss": 0.0223,
      "reward": 2.80625,
      "reward_std": 0.0375,
      "rewards/accuracy_reward": 1.80625,
      "rewards/format_reward": 1.0,
      "step": 960
    },
    {
      "completion_length": 51.01875,
      "epoch": 0.776,
      "grad_norm": 0.09945037215948105,
      "kl": 0.09228515625,
      "learning_rate": 5.149999999999999e-07,
      "loss": 0.0009,
      "reward": 2.775,
      "reward_std": 0.0,
      "rewards/accuracy_reward": 1.775,
      "rewards/format_reward": 1.0,
      "step": 970
    },
    {
      "completion_length": 50.090625,
      "epoch": 0.784,
      "grad_norm": 0.07448896020650864,
      "kl": 0.15419921875,
      "learning_rate": 5.1e-07,
      "loss": 0.0015,
      "reward": 2.76875,
      "reward_std": 0.0375,
      "rewards/accuracy_reward": 1.76875,
      "rewards/format_reward": 1.0,
      "step": 980
    },
    {
      "completion_length": 49.23125,
      "epoch": 0.792,
      "grad_norm": 2.0038902759552,
      "kl": 0.098828125,
      "learning_rate": 5.049999999999999e-07,
      "loss": 0.001,
      "reward": 2.64375,
      "reward_std": 0.0125,
      "rewards/accuracy_reward": 1.64375,
      "rewards/format_reward": 1.0,
      "step": 990
    },
    {
      "completion_length": 49.790625,
      "epoch": 0.8,
      "grad_norm": 0.03871207684278488,
      "kl": 0.31083984375,
      "learning_rate": 5e-07,
      "loss": 0.0031,
      "reward": 2.7625,
      "reward_std": 0.025,
      "rewards/accuracy_reward": 1.7625,
      "rewards/format_reward": 1.0,
      "step": 1000
    }
  ],
  "logging_steps": 10,
  "max_steps": 2000,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 2,
  "save_steps": 50,
  "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
}