{
  "best_global_step": null,
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 2.830188679245283,
  "eval_steps": 500,
  "global_step": 150,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "completion_length": 13.109375,
      "epoch": 0.018867924528301886,
      "grad_norm": 113.25696994976008,
      "kl": 0.0,
      "learning_rate": 1e-06,
      "loss": 0.0,
      "reward": 0.53125,
      "reward_std": 0.6636751294136047,
      "rewards/accuracy_reward_v1": 0.140625,
      "rewards/format_reward": 0.390625,
      "step": 1
    },
    {
      "completion_length": 11.4375,
      "epoch": 0.03773584905660377,
      "grad_norm": 17.455156386548044,
      "kl": 0.0010302291484549642,
      "learning_rate": 9.973045822102425e-07,
      "loss": 0.0,
      "reward": 0.921875,
      "reward_std": 0.5112396776676178,
      "rewards/accuracy_reward_v1": 0.28125,
      "rewards/format_reward": 0.640625,
      "step": 2
    },
    {
      "completion_length": 8.625,
      "epoch": 0.05660377358490566,
      "grad_norm": 99.22339823453805,
      "kl": 0.015800674445927143,
      "learning_rate": 9.946091644204851e-07,
      "loss": 0.0006,
      "reward": 0.265625,
      "reward_std": 0.5958134829998016,
      "rewards/accuracy_reward_v1": 0.046875,
      "rewards/format_reward": 0.21875,
      "step": 3
    },
    {
      "completion_length": 7.5,
      "epoch": 0.07547169811320754,
      "grad_norm": 36.344086106102296,
      "kl": 0.06874167174100876,
      "learning_rate": 9.919137466307277e-07,
      "loss": 0.0027,
      "reward": 0.015625,
      "reward_std": 0.3286440968513489,
      "rewards/accuracy_reward_v1": -0.015625,
      "rewards/format_reward": 0.03125,
      "step": 4
    },
    {
      "completion_length": 9.59375,
      "epoch": 0.09433962264150944,
      "grad_norm": 53.778756712320366,
      "kl": 0.03450273536145687,
      "learning_rate": 9.892183288409703e-07,
      "loss": 0.0014,
      "reward": 0.9375,
      "reward_std": 0.6200955510139465,
      "rewards/accuracy_reward_v1": 0.0625,
      "rewards/format_reward": 0.875,
      "step": 5
    },
    {
      "completion_length": 7.609375,
      "epoch": 0.11320754716981132,
      "grad_norm": 38.17063175193828,
      "kl": 0.03428475372493267,
      "learning_rate": 9.865229110512129e-07,
      "loss": 0.0014,
      "reward": 0.734375,
      "reward_std": 0.20728103816509247,
      "rewards/accuracy_reward_v1": 0.09375,
      "rewards/format_reward": 0.640625,
      "step": 6
    },
    {
      "completion_length": 6.734375,
      "epoch": 0.1320754716981132,
      "grad_norm": 28.849631709247998,
      "kl": 0.0493768819142133,
      "learning_rate": 9.838274932614555e-07,
      "loss": 0.002,
      "reward": 0.4375,
      "reward_std": 0.125,
      "rewards/accuracy_reward_v1": -0.015625,
      "rewards/format_reward": 0.453125,
      "step": 7
    },
    {
      "completion_length": 9.34375,
      "epoch": 0.1509433962264151,
      "grad_norm": 47.40102962097629,
      "kl": 0.08044508285820484,
      "learning_rate": 9.81132075471698e-07,
      "loss": 0.0032,
      "reward": 0.453125,
      "reward_std": 0.48587022721767426,
      "rewards/accuracy_reward_v1": 0.09375,
      "rewards/format_reward": 0.359375,
      "step": 8
    },
    {
      "completion_length": 6.640625,
      "epoch": 0.16981132075471697,
      "grad_norm": 18.963736933364622,
      "kl": 0.007860187441110611,
      "learning_rate": 9.784366576819406e-07,
      "loss": 0.0003,
      "reward": 0.5625,
      "reward_std": 0.125,
      "rewards/accuracy_reward_v1": 0.0625,
      "rewards/format_reward": 0.5,
      "step": 9
    },
    {
      "completion_length": 7.859375,
      "epoch": 0.18867924528301888,
      "grad_norm": 61.17171985485836,
      "kl": 0.13775761052966118,
      "learning_rate": 9.757412398921832e-07,
      "loss": 0.0055,
      "reward": 0.8125,
      "reward_std": 0.44716876745224,
      "rewards/accuracy_reward_v1": 0.078125,
      "rewards/format_reward": 0.734375,
      "step": 10
    },
    {
      "completion_length": 6.328125,
      "epoch": 0.20754716981132076,
      "grad_norm": 15.67651479348042,
      "kl": 0.038907756097614765,
      "learning_rate": 9.730458221024258e-07,
      "loss": 0.0016,
      "reward": 1.015625,
      "reward_std": 0.09375,
      "rewards/accuracy_reward_v1": 0.0,
      "rewards/format_reward": 1.015625,
      "step": 11
    },
    {
      "completion_length": 7.25,
      "epoch": 0.22641509433962265,
      "grad_norm": 6.014558113459187,
      "kl": 0.05433193470162223,
      "learning_rate": 9.703504043126684e-07,
      "loss": 0.0022,
      "reward": 0.546875,
      "reward_std": 0.13950317353010178,
      "rewards/accuracy_reward_v1": -0.046875,
      "rewards/format_reward": 0.59375,
      "step": 12
    },
    {
      "completion_length": 7.140625,
      "epoch": 0.24528301886792453,
      "grad_norm": 9.34578555325466,
      "kl": 0.05158466845750809,
      "learning_rate": 9.67654986522911e-07,
      "loss": 0.0021,
      "reward": 0.4375,
      "reward_std": 0.15358919650316238,
      "rewards/accuracy_reward_v1": -0.015625,
      "rewards/format_reward": 0.453125,
      "step": 13
    },
    {
      "completion_length": 5.984375,
      "epoch": 0.2641509433962264,
      "grad_norm": 61.001487031097064,
      "kl": 0.5286545567214489,
      "learning_rate": 9.649595687331536e-07,
      "loss": 0.0211,
      "reward": 0.65625,
      "reward_std": 0.0625,
      "rewards/accuracy_reward_v1": 0.0,
      "rewards/format_reward": 0.65625,
      "step": 14
    },
    {
      "completion_length": 6.40625,
      "epoch": 0.2830188679245283,
      "grad_norm": 1.9479825463836582,
      "kl": 0.016150618554092944,
      "learning_rate": 9.622641509433961e-07,
      "loss": 0.0006,
      "reward": 0.453125,
      "reward_std": 0.09375,
      "rewards/accuracy_reward_v1": 0.0,
      "rewards/format_reward": 0.453125,
      "step": 15
    },
    {
      "completion_length": 6.0,
      "epoch": 0.3018867924528302,
      "grad_norm": 0.8507446421494759,
      "kl": 0.002415000752080232,
      "learning_rate": 9.595687331536387e-07,
      "loss": 0.0001,
      "reward": 0.515625,
      "reward_std": 0.07864411175251007,
      "rewards/accuracy_reward_v1": 0.015625,
      "rewards/format_reward": 0.5,
      "step": 16
    },
    {
      "completion_length": 6.28125,
      "epoch": 0.32075471698113206,
      "grad_norm": 17.10440040176831,
      "kl": 0.07829345297068357,
      "learning_rate": 9.568733153638813e-07,
      "loss": 0.0031,
      "reward": 0.625,
      "reward_std": 0.125,
      "rewards/accuracy_reward_v1": -0.015625,
      "rewards/format_reward": 0.640625,
      "step": 17
    },
    {
      "completion_length": 7.046875,
      "epoch": 0.33962264150943394,
      "grad_norm": 32.02508384489064,
      "kl": 0.19249150529503822,
      "learning_rate": 9.541778975741239e-07,
      "loss": 0.0077,
      "reward": -0.171875,
      "reward_std": 0.27592839300632477,
      "rewards/accuracy_reward_v1": 0.09375,
      "rewards/format_reward": -0.265625,
      "step": 18
    },
    {
      "completion_length": 6.0,
      "epoch": 0.3584905660377358,
      "grad_norm": 0.6767178894314931,
      "kl": 0.005678875371813774,
      "learning_rate": 9.514824797843666e-07,
      "loss": 0.0002,
      "reward": 0.875,
      "reward_std": 0.0,
      "rewards/accuracy_reward_v1": 0.0,
      "rewards/format_reward": 0.875,
      "step": 19
    },
    {
      "completion_length": 11.875,
      "epoch": 0.37735849056603776,
      "grad_norm": 27.69343537359132,
      "kl": 0.20545678702183068,
      "learning_rate": 9.487870619946092e-07,
      "loss": 0.0082,
      "reward": -0.234375,
      "reward_std": 0.07864411175251007,
      "rewards/accuracy_reward_v1": 0.015625,
      "rewards/format_reward": -0.25,
      "step": 20
    },
    {
      "completion_length": 7.09375,
      "epoch": 0.39622641509433965,
      "grad_norm": 1.6376469352682972,
      "kl": 0.03206097148358822,
      "learning_rate": 9.460916442048516e-07,
      "loss": 0.0013,
      "reward": 1.015625,
      "reward_std": 0.09375,
      "rewards/accuracy_reward_v1": 0.0,
      "rewards/format_reward": 1.015625,
      "step": 21
    },
    {
      "completion_length": 6.0,
      "epoch": 0.41509433962264153,
      "grad_norm": 0.751230013203976,
      "kl": 0.001671111851464957,
      "learning_rate": 9.433962264150943e-07,
      "loss": 0.0001,
      "reward": 0.875,
      "reward_std": 0.0,
      "rewards/accuracy_reward_v1": 0.0,
      "rewards/format_reward": 0.875,
      "step": 22
    },
    {
      "completion_length": 6.0,
      "epoch": 0.4339622641509434,
      "grad_norm": 0.024176332343756202,
      "kl": 0.02829874058807036,
      "learning_rate": 9.407008086253368e-07,
      "loss": 0.0011,
      "reward": 0.125,
      "reward_std": 0.0,
      "rewards/accuracy_reward_v1": 0.0,
      "rewards/format_reward": 0.125,
      "step": 23
    },
    {
      "completion_length": 19.125,
      "epoch": 0.4528301886792453,
      "grad_norm": 8.218280531766865,
      "kl": 0.026557361241430044,
      "learning_rate": 9.380053908355795e-07,
      "loss": 0.0011,
      "reward": 0.28125,
      "reward_std": 0.1875,
      "rewards/accuracy_reward_v1": 0.015625,
      "rewards/format_reward": 0.265625,
      "step": 24
    },
    {
      "completion_length": 6.5625,
      "epoch": 0.4716981132075472,
      "grad_norm": 20.864188370379015,
      "kl": 0.025693180970847607,
      "learning_rate": 9.353099730458221e-07,
      "loss": 0.001,
      "reward": 0.171875,
      "reward_std": 0.09375,
      "rewards/accuracy_reward_v1": 0.046875,
      "rewards/format_reward": 0.125,
      "step": 25
    },
    {
      "completion_length": 6.078125,
      "epoch": 0.49056603773584906,
      "grad_norm": 22.968977069358576,
      "kl": 0.018222793005406857,
      "learning_rate": 9.326145552560647e-07,
      "loss": 0.0007,
      "reward": 0.421875,
      "reward_std": 0.1298343911767006,
      "rewards/accuracy_reward_v1": -0.03125,
      "rewards/format_reward": 0.453125,
      "step": 26
    },
    {
      "completion_length": 6.359375,
      "epoch": 0.5094339622641509,
      "grad_norm": 5.285537623328123,
      "kl": 0.009080744697712362,
      "learning_rate": 9.299191374663073e-07,
      "loss": 0.0004,
      "reward": 0.078125,
      "reward_std": 0.09375,
      "rewards/accuracy_reward_v1": 0.0,
      "rewards/format_reward": 0.078125,
      "step": 27
    },
    {
      "completion_length": 6.8125,
      "epoch": 0.5283018867924528,
      "grad_norm": 8.01142088192983,
      "kl": 0.08269636548357084,
      "learning_rate": 9.272237196765497e-07,
      "loss": 0.0033,
      "reward": 0.484375,
      "reward_std": 0.15625,
      "rewards/accuracy_reward_v1": 0.03125,
      "rewards/format_reward": 0.453125,
      "step": 28
    },
    {
      "completion_length": 9.25,
      "epoch": 0.5471698113207547,
      "grad_norm": 44940.43250898536,
      "kl": 429.3509378628805,
      "learning_rate": 9.245283018867924e-07,
      "loss": 17.174,
      "reward": 0.625,
      "reward_std": 0.23489411175251007,
      "rewards/accuracy_reward_v1": 0.015625,
      "rewards/format_reward": 0.609375,
      "step": 29
    },
    {
      "completion_length": 6.328125,
      "epoch": 0.5660377358490566,
      "grad_norm": 5.3595802935312165,
      "kl": 0.015079768985742703,
      "learning_rate": 9.21832884097035e-07,
      "loss": 0.0006,
      "reward": 1.109375,
      "reward_std": 0.10341878235340118,
      "rewards/accuracy_reward_v1": 0.046875,
      "rewards/format_reward": 1.0625,
      "step": 30
    },
    {
      "completion_length": 6.59375,
      "epoch": 0.5849056603773585,
      "grad_norm": 200.52598781535795,
      "kl": 0.16589106805622578,
      "learning_rate": 9.191374663072776e-07,
      "loss": 0.0066,
      "reward": 0.203125,
      "reward_std": 0.10341878235340118,
      "rewards/accuracy_reward_v1": 0.0625,
      "rewards/format_reward": 0.140625,
      "step": 31
    },
    {
      "completion_length": 6.71875,
      "epoch": 0.6037735849056604,
      "grad_norm": 13.708430791873619,
      "kl": 0.04937315918505192,
      "learning_rate": 9.164420485175202e-07,
      "loss": 0.002,
      "reward": 1.109375,
      "reward_std": 0.28125,
      "rewards/accuracy_reward_v1": 0.0,
      "rewards/format_reward": 1.109375,
      "step": 32
    },
    {
      "completion_length": 135.390625,
      "epoch": 0.6226415094339622,
      "grad_norm": 75.39004677783589,
      "kl": 0.0725557878613472,
      "learning_rate": 9.137466307277629e-07,
      "loss": 0.0029,
      "reward": -0.03125,
      "reward_std": 0.10825317353010178,
      "rewards/accuracy_reward_v1": 0.125,
      "rewards/format_reward": -0.15625,
      "step": 33
    },
    {
      "completion_length": 6.21875,
      "epoch": 0.6415094339622641,
      "grad_norm": 22.61841680640308,
      "kl": 0.1283794054761529,
      "learning_rate": 9.110512129380053e-07,
      "loss": 0.0051,
      "reward": 0.453125,
      "reward_std": 0.09375,
      "rewards/accuracy_reward_v1": -0.015625,
      "rewards/format_reward": 0.46875,
      "step": 34
    },
    {
      "completion_length": 6.8125,
      "epoch": 0.660377358490566,
      "grad_norm": 94.5490460045855,
      "kl": 0.09005002037156373,
      "learning_rate": 9.083557951482479e-07,
      "loss": 0.0036,
      "reward": 0.671875,
      "reward_std": 0.15625,
      "rewards/accuracy_reward_v1": 0.03125,
      "rewards/format_reward": 0.640625,
      "step": 35
    },
    {
      "completion_length": 5.984375,
      "epoch": 0.6792452830188679,
      "grad_norm": 15.827914495508187,
      "kl": 0.017108662985265255,
      "learning_rate": 9.056603773584905e-07,
      "loss": 0.0007,
      "reward": 0.890625,
      "reward_std": 0.03125,
      "rewards/accuracy_reward_v1": 0.0,
      "rewards/format_reward": 0.890625,
      "step": 36
    },
    {
      "completion_length": 6.046875,
      "epoch": 0.6981132075471698,
      "grad_norm": 968.8698350317079,
      "kl": 7.435610045678914,
      "learning_rate": 9.029649595687331e-07,
      "loss": 0.2974,
      "reward": 0.46875,
      "reward_std": 0.0625,
      "rewards/accuracy_reward_v1": -0.015625,
      "rewards/format_reward": 0.484375,
      "step": 37
    },
    {
      "completion_length": 6.0,
      "epoch": 0.7169811320754716,
      "grad_norm": 0.019372855645124972,
      "kl": 0.03183715185150504,
      "learning_rate": 9.002695417789758e-07,
      "loss": 0.0013,
      "reward": 0.125,
      "reward_std": 0.0,
      "rewards/accuracy_reward_v1": 0.0,
      "rewards/format_reward": 0.125,
      "step": 38
    },
    {
      "completion_length": 5.984375,
      "epoch": 0.7358490566037735,
      "grad_norm": 9.38601211541122,
      "kl": 0.034928835928440094,
      "learning_rate": 8.975741239892183e-07,
      "loss": 0.0014,
      "reward": 0.84375,
      "reward_std": 0.0625,
      "rewards/accuracy_reward_v1": 0.0,
      "rewards/format_reward": 0.84375,
      "step": 39
    },
    {
      "completion_length": 5.984375,
      "epoch": 0.7547169811320755,
      "grad_norm": 6.751602282234656,
      "kl": 0.015660904347896576,
      "learning_rate": 8.94878706199461e-07,
      "loss": 0.0006,
      "reward": 0.515625,
      "reward_std": 0.03125,
      "rewards/accuracy_reward_v1": 0.0,
      "rewards/format_reward": 0.515625,
      "step": 40
    },
    {
      "completion_length": 5.984375,
      "epoch": 0.7735849056603774,
      "grad_norm": 18.76399906522087,
      "kl": 0.011825291672721505,
      "learning_rate": 8.921832884097034e-07,
      "loss": 0.0005,
      "reward": 1.453125,
      "reward_std": 0.03125,
      "rewards/accuracy_reward_v1": 0.0,
      "rewards/format_reward": 1.453125,
      "step": 41
    },
    {
      "completion_length": 5.96875,
      "epoch": 0.7924528301886793,
      "grad_norm": 11.64564735069018,
      "kl": 0.031408867333084345,
      "learning_rate": 8.89487870619946e-07,
      "loss": 0.0013,
      "reward": 0.625,
      "reward_std": 0.125,
      "rewards/accuracy_reward_v1": 0.0,
      "rewards/format_reward": 0.625,
      "step": 42
    },
    {
      "completion_length": 6.328125,
      "epoch": 0.8113207547169812,
      "grad_norm": 11.576381065477106,
      "kl": 0.025013845995999873,
      "learning_rate": 8.867924528301887e-07,
      "loss": 0.001,
      "reward": 0.28125,
      "reward_std": 0.1875,
      "rewards/accuracy_reward_v1": 0.015625,
      "rewards/format_reward": 0.265625,
      "step": 43
    },
    {
      "completion_length": 6.421875,
      "epoch": 0.8301886792452831,
      "grad_norm": 23.62669653663731,
      "kl": 0.09232653491199017,
      "learning_rate": 8.840970350404312e-07,
      "loss": 0.0037,
      "reward": 0.21875,
      "reward_std": 0.375,
      "rewards/accuracy_reward_v1": 0.03125,
      "rewards/format_reward": 0.1875,
      "step": 44
    },
    {
      "completion_length": 5.953125,
      "epoch": 0.8490566037735849,
      "grad_norm": 37.130279133532085,
      "kl": 0.39919494092464447,
      "learning_rate": 8.814016172506739e-07,
      "loss": 0.016,
      "reward": 1.203125,
      "reward_std": 0.15625,
      "rewards/accuracy_reward_v1": 0.0,
      "rewards/format_reward": 1.203125,
      "step": 45
    },
    {
      "completion_length": 6.859375,
      "epoch": 0.8679245283018868,
      "grad_norm": 32.16252577438579,
      "kl": 0.16001062956638634,
      "learning_rate": 8.787061994609164e-07,
      "loss": 0.0064,
      "reward": 0.28125,
      "reward_std": 0.23489411175251007,
      "rewards/accuracy_reward_v1": 0.09375,
      "rewards/format_reward": 0.1875,
      "step": 46
    },
    {
      "completion_length": 6.0,
      "epoch": 0.8867924528301887,
      "grad_norm": 0.014299505784278984,
      "kl": 0.012281196657568216,
      "learning_rate": 8.760107816711589e-07,
      "loss": 0.0005,
      "reward": -0.0625,
      "reward_std": 0.0,
      "rewards/accuracy_reward_v1": 0.0,
      "rewards/format_reward": -0.0625,
      "step": 47
    },
    {
      "completion_length": 7.453125,
      "epoch": 0.9056603773584906,
      "grad_norm": 62.832801793657524,
      "kl": 0.05191558599472046,
      "learning_rate": 8.733153638814016e-07,
      "loss": 0.0021,
      "reward": 0.34375,
      "reward_std": 0.3125,
      "rewards/accuracy_reward_v1": 0.09375,
      "rewards/format_reward": 0.25,
      "step": 48
    },
    {
      "completion_length": 6.25,
      "epoch": 0.9245283018867925,
      "grad_norm": 68.50200972708288,
      "kl": 0.08319599367678165,
      "learning_rate": 8.706199460916441e-07,
      "loss": 0.0033,
      "reward": 0.40625,
      "reward_std": 0.15358919650316238,
      "rewards/accuracy_reward_v1": 0.015625,
      "rewards/format_reward": 0.390625,
      "step": 49
    },
    {
      "completion_length": 6.03125,
      "epoch": 0.9433962264150944,
      "grad_norm": 73.71745454546195,
      "kl": 0.04624786600470543,
      "learning_rate": 8.679245283018868e-07,
      "loss": 0.0018,
      "reward": 0.515625,
      "reward_std": 0.15625,
      "rewards/accuracy_reward_v1": 0.03125,
      "rewards/format_reward": 0.484375,
      "step": 50
    },
    {
      "completion_length": 7.140625,
      "epoch": 0.9622641509433962,
      "grad_norm": 34.26377220172008,
      "kl": 0.06593625247478485,
      "learning_rate": 8.652291105121293e-07,
      "loss": 0.0026,
      "reward": 0.328125,
      "reward_std": 0.34375,
      "rewards/accuracy_reward_v1": 0.0625,
      "rewards/format_reward": 0.265625,
      "step": 51
    },
    {
      "completion_length": 5.953125,
      "epoch": 0.9811320754716981,
      "grad_norm": 12.841483298963944,
      "kl": 0.027065719477832317,
      "learning_rate": 8.62533692722372e-07,
      "loss": 0.0011,
      "reward": 0.640625,
      "reward_std": 0.15625,
      "rewards/accuracy_reward_v1": 0.0,
      "rewards/format_reward": 0.640625,
      "step": 52
    },
    {
      "completion_length": 5.900000095367432,
      "epoch": 1.0,
      "grad_norm": 14.937803974771874,
      "kl": 0.046957485377788544,
      "learning_rate": 8.598382749326146e-07,
      "loss": 0.0024,
      "reward": 0.8999999985098839,
      "reward_std": 0.20000000298023224,
      "rewards/accuracy_reward_v1": 0.0,
      "rewards/format_reward": 0.8999999985098839,
      "step": 53
    },
    {
      "completion_length": 136.28125,
      "epoch": 1.0188679245283019,
      "grad_norm": 50.81914218967191,
      "kl": 0.4388061761856079,
      "learning_rate": 8.57142857142857e-07,
      "loss": 0.0176,
      "reward": 0.265625,
      "reward_std": 0.28125,
      "rewards/accuracy_reward_v1": 0.109375,
      "rewards/format_reward": 0.15625,
      "step": 54
    },
    {
      "completion_length": 18.46875,
      "epoch": 1.0377358490566038,
      "grad_norm": 14.265621997516613,
      "kl": 0.08800802286714315,
      "learning_rate": 8.544474393530997e-07,
      "loss": 0.0035,
      "reward": 0.3125,
      "reward_std": 0.0625,
      "rewards/accuracy_reward_v1": -0.015625,
      "rewards/format_reward": 0.328125,
      "step": 55
    },
    {
      "completion_length": 6.75,
      "epoch": 1.0566037735849056,
      "grad_norm": 25.880606401711432,
      "kl": 0.18149923533201218,
      "learning_rate": 8.517520215633422e-07,
      "loss": 0.0073,
      "reward": 0.484375,
      "reward_std": 0.34375,
      "rewards/accuracy_reward_v1": 0.046875,
      "rewards/format_reward": 0.4375,
      "step": 56
    },
    {
      "completion_length": 7.390625,
      "epoch": 1.0754716981132075,
      "grad_norm": 24.976627040965077,
      "kl": 0.05544249154627323,
      "learning_rate": 8.490566037735849e-07,
      "loss": 0.0022,
      "reward": -0.171875,
      "reward_std": 0.28125,
      "rewards/accuracy_reward_v1": 0.09375,
      "rewards/format_reward": -0.265625,
      "step": 57
    },
    {
      "completion_length": 6.28125,
      "epoch": 1.0943396226415094,
      "grad_norm": 38.30509050682159,
      "kl": 0.04700027219951153,
      "learning_rate": 8.463611859838275e-07,
      "loss": 0.0019,
      "reward": -0.0625,
      "reward_std": 0.125,
      "rewards/accuracy_reward_v1": -0.03125,
      "rewards/format_reward": -0.03125,
      "step": 58
    },
    {
      "completion_length": 6.296875,
      "epoch": 1.1132075471698113,
      "grad_norm": 29.616594415696703,
      "kl": 0.04546574130654335,
      "learning_rate": 8.436657681940701e-07,
      "loss": 0.0018,
      "reward": 0.96875,
      "reward_std": 0.3125,
      "rewards/accuracy_reward_v1": 0.0,
      "rewards/format_reward": 0.96875,
      "step": 59
    },
    {
      "completion_length": 7.78125,
      "epoch": 1.1320754716981132,
      "grad_norm": 284.9164497975472,
      "kl": 0.06180373951792717,
      "learning_rate": 8.409703504043126e-07,
      "loss": 0.0025,
      "reward": 0.1875,
      "reward_std": 0.3238462805747986,
      "rewards/accuracy_reward_v1": 0.078125,
      "rewards/format_reward": 0.109375,
      "step": 60
    },
    {
      "completion_length": 6.359375,
      "epoch": 1.150943396226415,
      "grad_norm": 8.169614652406057,
      "kl": 0.04489657515659928,
      "learning_rate": 8.382749326145552e-07,
      "loss": 0.0018,
      "reward": 0.671875,
      "reward_std": 0.03125,
      "rewards/accuracy_reward_v1": -0.015625,
      "rewards/format_reward": 0.6875,
      "step": 61
    },
    {
      "completion_length": 11.375,
      "epoch": 1.169811320754717,
      "grad_norm": 48.81511838253674,
      "kl": 0.16956874169409275,
      "learning_rate": 8.355795148247978e-07,
      "loss": 0.0068,
      "reward": 0.453125,
      "reward_std": 0.28125,
      "rewards/accuracy_reward_v1": -0.015625,
      "rewards/format_reward": 0.46875,
      "step": 62
    },
    {
      "completion_length": 6.0625,
      "epoch": 1.1886792452830188,
      "grad_norm": 19.637736140655225,
      "kl": 0.04594237124547362,
      "learning_rate": 8.328840970350404e-07,
      "loss": 0.0018,
      "reward": 0.125,
      "reward_std": 0.1875,
      "rewards/accuracy_reward_v1": -0.015625,
      "rewards/format_reward": 0.140625,
      "step": 63
    },
    {
      "completion_length": 7.796875,
      "epoch": 1.2075471698113207,
      "grad_norm": 14.926898050021375,
      "kl": 0.05223815282806754,
      "learning_rate": 8.30188679245283e-07,
      "loss": 0.0021,
      "reward": 1.15625,
      "reward_std": 0.1875,
      "rewards/accuracy_reward_v1": -0.015625,
      "rewards/format_reward": 1.171875,
      "step": 64
    },
    {
      "completion_length": 6.078125,
      "epoch": 1.2264150943396226,
      "grad_norm": 30.610327141923904,
      "kl": 0.11978641152381897,
      "learning_rate": 8.274932614555256e-07,
      "loss": 0.0048,
      "reward": 0.546875,
      "reward_std": 0.35187220573425293,
      "rewards/accuracy_reward_v1": 0.0625,
      "rewards/format_reward": 0.484375,
      "step": 65
    },
    {
      "completion_length": 7.15625,
      "epoch": 1.2452830188679245,
      "grad_norm": 21.542244248276532,
      "kl": 0.05250057205557823,
      "learning_rate": 8.247978436657683e-07,
      "loss": 0.0021,
      "reward": 0.875,
      "reward_std": 0.3221687823534012,
      "rewards/accuracy_reward_v1": 0.078125,
      "rewards/format_reward": 0.796875,
      "step": 66
    },
    {
      "completion_length": 5.90625,
      "epoch": 1.2641509433962264,
      "grad_norm": 28.650448380954998,
      "kl": 0.10170655325055122,
      "learning_rate": 8.221024258760107e-07,
      "loss": 0.0041,
      "reward": 0.328125,
      "reward_std": 0.22841878235340118,
      "rewards/accuracy_reward_v1": 0.0,
      "rewards/format_reward": 0.328125,
      "step": 67
    },
    {
      "completion_length": 5.90625,
      "epoch": 1.2830188679245282,
      "grad_norm": 64.68612514929318,
      "kl": 0.1377212107181549,
      "learning_rate": 8.194070080862533e-07,
      "loss": 0.0055,
      "reward": 1.15625,
      "reward_std": 0.2596687823534012,
      "rewards/accuracy_reward_v1": 0.0,
      "rewards/format_reward": 1.15625,
      "step": 68
    },
    {
      "completion_length": 5.84375,
      "epoch": 1.3018867924528301,
      "grad_norm": 28.832798485554264,
      "kl": 0.2311520278453827,
      "learning_rate": 8.167115902964959e-07,
      "loss": 0.0092,
      "reward": 0.421875,
      "reward_std": 0.28125,
      "rewards/accuracy_reward_v1": 0.0,
      "rewards/format_reward": 0.421875,
      "step": 69
    },
    {
      "completion_length": 7.671875,
      "epoch": 1.320754716981132,
      "grad_norm": 75.39703628351455,
      "kl": 0.3764437586069107,
      "learning_rate": 8.140161725067385e-07,
      "loss": 0.0151,
      "reward": 0.75,
      "reward_std": 0.5342842042446136,
      "rewards/accuracy_reward_v1": 0.015625,
      "rewards/format_reward": 0.734375,
      "step": 70
    },
    {
      "completion_length": 5.765625,
      "epoch": 1.3396226415094339,
      "grad_norm": 90.34161478588945,
      "kl": 0.46842779219150543,
      "learning_rate": 8.113207547169812e-07,
      "loss": 0.0187,
      "reward": 0.3125,
      "reward_std": 0.44716876745224,
      "rewards/accuracy_reward_v1": 0.0,
      "rewards/format_reward": 0.3125,
      "step": 71
    },
    {
      "completion_length": 5.875,
      "epoch": 1.3584905660377358,
      "grad_norm": 28.477689678793762,
      "kl": 0.2690361887216568,
      "learning_rate": 8.086253369272237e-07,
      "loss": 0.0108,
      "reward": 0.125,
      "reward_std": 0.5311799347400665,
      "rewards/accuracy_reward_v1": 0.046875,
      "rewards/format_reward": 0.078125,
      "step": 72
    },
    {
      "completion_length": 5.90625,
      "epoch": 1.3773584905660377,
      "grad_norm": 30.745817115659335,
      "kl": 0.44600608944892883,
      "learning_rate": 8.059299191374662e-07,
      "loss": 0.0178,
      "reward": 0.078125,
      "reward_std": 0.43525634706020355,
      "rewards/accuracy_reward_v1": 0.0,
      "rewards/format_reward": 0.078125,
      "step": 73
    },
    {
      "completion_length": 5.734375,
      "epoch": 1.3962264150943398,
      "grad_norm": 63.59315485782405,
      "kl": 0.493766725063324,
      "learning_rate": 8.032345013477088e-07,
      "loss": 0.0198,
      "reward": 0.578125,
      "reward_std": 0.5241719335317612,
      "rewards/accuracy_reward_v1": 0.0,
      "rewards/format_reward": 0.578125,
      "step": 74
    },
    {
      "completion_length": 6.25,
      "epoch": 1.4150943396226414,
      "grad_norm": 376.962883789387,
      "kl": 0.6078239977359772,
      "learning_rate": 8.005390835579514e-07,
      "loss": 0.0243,
      "reward": 0.71875,
      "reward_std": 0.6804219484329224,
      "rewards/accuracy_reward_v1": 0.0,
      "rewards/format_reward": 0.71875,
      "step": 75
    },
    {
      "completion_length": 7.421875,
      "epoch": 1.4339622641509435,
      "grad_norm": 54.980705509223874,
      "kl": 0.728535920381546,
      "learning_rate": 7.978436657681941e-07,
      "loss": 0.0291,
      "reward": 0.421875,
      "reward_std": 0.8220986723899841,
      "rewards/accuracy_reward_v1": 0.03125,
      "rewards/format_reward": 0.390625,
      "step": 76
    },
    {
      "completion_length": 6.203125,
      "epoch": 1.4528301886792452,
      "grad_norm": 61.0567889858235,
      "kl": 0.42994847893714905,
      "learning_rate": 7.951482479784366e-07,
      "loss": 0.0172,
      "reward": 0.296875,
      "reward_std": 0.4810095429420471,
      "rewards/accuracy_reward_v1": 0.0,
      "rewards/format_reward": 0.296875,
      "step": 77
    },
    {
      "completion_length": 8.421875,
      "epoch": 1.4716981132075473,
      "grad_norm": 4648.40617287234,
      "kl": 13.131318747997284,
      "learning_rate": 7.924528301886793e-07,
      "loss": 0.5253,
      "reward": 0.21875,
      "reward_std": 0.5936799049377441,
      "rewards/accuracy_reward_v1": 0.03125,
      "rewards/format_reward": 0.1875,
      "step": 78
    },
    {
      "completion_length": 7.21875,
      "epoch": 1.490566037735849,
      "grad_norm": 70.34769464159976,
      "kl": 0.6175012588500977,
      "learning_rate": 7.897574123989218e-07,
      "loss": 0.0247,
      "reward": 0.625,
      "reward_std": 0.7719283103942871,
      "rewards/accuracy_reward_v1": 0.109375,
      "rewards/format_reward": 0.515625,
      "step": 79
    },
    {
      "completion_length": 5.796875,
      "epoch": 1.509433962264151,
      "grad_norm": 39.1599345059281,
      "kl": 0.3216542750597,
      "learning_rate": 7.870619946091643e-07,
      "loss": 0.0129,
      "reward": 0.796875,
      "reward_std": 0.4255875498056412,
      "rewards/accuracy_reward_v1": 0.0,
      "rewards/format_reward": 0.796875,
      "step": 80
    },
    {
      "completion_length": 7.15625,
      "epoch": 1.5283018867924527,
      "grad_norm": 61.46576454255242,
      "kl": 0.57504503428936,
      "learning_rate": 7.84366576819407e-07,
      "loss": 0.023,
      "reward": 0.625,
      "reward_std": 0.5290063470602036,
      "rewards/accuracy_reward_v1": 0.0,
      "rewards/format_reward": 0.625,
      "step": 81
    },
    {
      "completion_length": 6.5,
      "epoch": 1.5471698113207548,
      "grad_norm": 127.06308053172842,
      "kl": 1.0777413994073868,
      "learning_rate": 7.816711590296495e-07,
      "loss": 0.0431,
      "reward": 0.78125,
      "reward_std": 0.6135310530662537,
      "rewards/accuracy_reward_v1": -0.03125,
      "rewards/format_reward": 0.8125,
      "step": 82
    },
    {
      "completion_length": 6.15625,
      "epoch": 1.5660377358490565,
      "grad_norm": 30.75068892911751,
      "kl": 0.18815521150827408,
      "learning_rate": 7.789757412398922e-07,
      "loss": 0.0075,
      "reward": -0.09375,
      "reward_std": 0.3582531660795212,
      "rewards/accuracy_reward_v1": 0.0,
      "rewards/format_reward": -0.09375,
      "step": 83
    },
    {
      "completion_length": 5.828125,
      "epoch": 1.5849056603773586,
      "grad_norm": 47.407406657671885,
      "kl": 0.39250317215919495,
      "learning_rate": 7.762803234501348e-07,
      "loss": 0.0157,
      "reward": 0.8125,
      "reward_std": 0.6275907456874847,
      "rewards/accuracy_reward_v1": 0.0,
      "rewards/format_reward": 0.8125,
      "step": 84
    },
    {
      "completion_length": 6.640625,
      "epoch": 1.6037735849056602,
      "grad_norm": 36.61081127904801,
      "kl": 0.2916853576898575,
      "learning_rate": 7.735849056603774e-07,
      "loss": 0.0117,
      "reward": 0.34375,
      "reward_std": 0.44716876745224,
      "rewards/accuracy_reward_v1": 0.015625,
      "rewards/format_reward": 0.328125,
      "step": 85
    },
    {
      "completion_length": 6.453125,
      "epoch": 1.6226415094339623,
      "grad_norm": 22.44772023409409,
      "kl": 0.5226494371891022,
      "learning_rate": 7.708894878706199e-07,
      "loss": 0.0209,
      "reward": 0.25,
      "reward_std": 0.7088519036769867,
      "rewards/accuracy_reward_v1": 0.015625,
      "rewards/format_reward": 0.234375,
      "step": 86
    },
    {
      "completion_length": 5.84375,
      "epoch": 1.641509433962264,
      "grad_norm": 142.18053012672632,
      "kl": 0.29771409928798676,
      "learning_rate": 7.681940700808624e-07,
      "loss": 0.0119,
      "reward": 0.984375,
      "reward_std": 0.3534187823534012,
      "rewards/accuracy_reward_v1": 0.0,
      "rewards/format_reward": 0.984375,
      "step": 87
    },
    {
      "completion_length": 5.984375,
      "epoch": 1.6603773584905661,
      "grad_norm": 40.679488846208855,
      "kl": 0.40321239829063416,
      "learning_rate": 7.654986522911051e-07,
      "loss": 0.0161,
      "reward": 0.515625,
      "reward_std": 0.6209013164043427,
      "rewards/accuracy_reward_v1": 0.03125,
      "rewards/format_reward": 0.484375,
      "step": 88
    },
    {
      "completion_length": 6.21875,
      "epoch": 1.6792452830188678,
      "grad_norm": 194.5182883849516,
      "kl": 1.242738962173462,
      "learning_rate": 7.628032345013477e-07,
      "loss": 0.0497,
      "reward": 0.65625,
      "reward_std": 0.6540063619613647,
      "rewards/accuracy_reward_v1": 0.0,
      "rewards/format_reward": 0.65625,
      "step": 89
    },
    {
      "completion_length": 6.875,
      "epoch": 1.6981132075471699,
      "grad_norm": 37.05674248642704,
      "kl": 0.9561429917812347,
      "learning_rate": 7.601078167115903e-07,
      "loss": 0.0382,
      "reward": 0.21875,
      "reward_std": 0.5570628941059113,
      "rewards/accuracy_reward_v1": 0.0,
      "rewards/format_reward": 0.21875,
      "step": 90
    },
    {
      "completion_length": 5.703125,
      "epoch": 1.7169811320754715,
      "grad_norm": 50.629114056375165,
      "kl": 0.6461111009120941,
      "learning_rate": 7.574123989218329e-07,
      "loss": 0.0258,
      "reward": 0.40625,
      "reward_std": 0.6900907456874847,
      "rewards/accuracy_reward_v1": 0.0,
      "rewards/format_reward": 0.40625,
      "step": 91
    },
    {
      "completion_length": 6.484375,
      "epoch": 1.7358490566037736,
      "grad_norm": 44.61354623523612,
      "kl": 0.9098754525184631,
      "learning_rate": 7.547169811320754e-07,
      "loss": 0.0364,
      "reward": 0.46875,
      "reward_std": 0.6656894683837891,
      "rewards/accuracy_reward_v1": 0.015625,
      "rewards/format_reward": 0.453125,
      "step": 92
    },
    {
      "completion_length": 5.859375,
      "epoch": 1.7547169811320755,
      "grad_norm": 56.542951767589514,
      "kl": 0.5751659870147705,
      "learning_rate": 7.52021563342318e-07,
      "loss": 0.023,
      "reward": 0.71875,
      "reward_std": 0.5457531809806824,
      "rewards/accuracy_reward_v1": 0.0,
      "rewards/format_reward": 0.71875,
      "step": 93
    },
    {
      "completion_length": 5.59375,
      "epoch": 1.7735849056603774,
      "grad_norm": 52.48764250264164,
      "kl": 0.6279080808162689,
      "learning_rate": 7.493261455525606e-07,
      "loss": 0.0251,
      "reward": 0.578125,
      "reward_std": 0.7935095131397247,
      "rewards/accuracy_reward_v1": 0.0,
      "rewards/format_reward": 0.578125,
      "step": 94
    },
    {
      "completion_length": 5.671875,
      "epoch": 1.7924528301886793,
      "grad_norm": 42.759663388684906,
      "kl": 0.4385665953159332,
      "learning_rate": 7.466307277628032e-07,
      "loss": 0.0175,
      "reward": 0.453125,
      "reward_std": 0.6420939266681671,
      "rewards/accuracy_reward_v1": 0.0,
      "rewards/format_reward": 0.453125,
      "step": 95
    },
    {
      "completion_length": 6.15625,
      "epoch": 1.8113207547169812,
      "grad_norm": 45.15852397883639,
      "kl": 0.3389554023742676,
      "learning_rate": 7.439353099730458e-07,
      "loss": 0.0136,
      "reward": 0.0625,
      "reward_std": 0.5632905662059784,
      "rewards/accuracy_reward_v1": -0.015625,
      "rewards/format_reward": 0.078125,
      "step": 96
    },
    {
      "completion_length": 5.859375,
      "epoch": 1.830188679245283,
      "grad_norm": 30.03959836472858,
      "kl": 0.12645583972334862,
      "learning_rate": 7.412398921832884e-07,
      "loss": 0.0051,
      "reward": 0.96875,
      "reward_std": 0.41108438372612,
      "rewards/accuracy_reward_v1": 0.0,
      "rewards/format_reward": 0.96875,
      "step": 97
    },
    {
      "completion_length": 5.921875,
      "epoch": 1.849056603773585,
      "grad_norm": 22.224165729384023,
      "kl": 0.08302480168640614,
      "learning_rate": 7.38544474393531e-07,
      "loss": 0.0033,
      "reward": -0.171875,
      "reward_std": 0.1298343911767006,
      "rewards/accuracy_reward_v1": 0.0,
      "rewards/format_reward": -0.171875,
      "step": 98
    },
    {
      "completion_length": 5.828125,
      "epoch": 1.8679245283018868,
      "grad_norm": 19.595510648990068,
      "kl": 0.14540733397006989,
      "learning_rate": 7.358490566037735e-07,
      "loss": 0.0058,
      "reward": 0.25,
      "reward_std": 0.46650636196136475,
      "rewards/accuracy_reward_v1": 0.0,
      "rewards/format_reward": 0.25,
      "step": 99
    },
    {
      "completion_length": 6.609375,
      "epoch": 1.8867924528301887,
      "grad_norm": 59.366563044495884,
      "kl": 0.09590592980384827,
      "learning_rate": 7.331536388140161e-07,
      "loss": 0.0038,
      "reward": 0.140625,
      "reward_std": 0.31733438372612,
      "rewards/accuracy_reward_v1": 0.015625,
      "rewards/format_reward": 0.125,
      "step": 100
    },
    {
      "completion_length": 6.09375,
      "epoch": 1.9056603773584906,
      "grad_norm": 10.457267950620983,
      "kl": 0.041879747062921524,
      "learning_rate": 7.304582210242587e-07,
      "loss": 0.0017,
      "reward": 0.515625,
      "reward_std": 0.15625,
      "rewards/accuracy_reward_v1": 0.03125,
      "rewards/format_reward": 0.484375,
      "step": 101
    },
    {
      "completion_length": 6.78125,
      "epoch": 1.9245283018867925,
      "grad_norm": 29.98865362421412,
      "kl": 0.11279736459255219,
      "learning_rate": 7.277628032345014e-07,
      "loss": 0.0045,
      "reward": 0.5625,
      "reward_std": 0.3480571210384369,
      "rewards/accuracy_reward_v1": 0.03125,
      "rewards/format_reward": 0.53125,
      "step": 102
    },
    {
      "completion_length": 6.515625,
      "epoch": 1.9433962264150944,
      "grad_norm": 17.503120522404213,
      "kl": 0.018309888735529967,
      "learning_rate": 7.250673854447439e-07,
      "loss": 0.0007,
      "reward": -0.140625,
      "reward_std": 0.09375,
      "rewards/accuracy_reward_v1": 0.0,
      "rewards/format_reward": -0.140625,
      "step": 103
    },
    {
      "completion_length": 5.984375,
      "epoch": 1.9622641509433962,
      "grad_norm": 18.78967686622246,
      "kl": 0.018351302947849035,
      "learning_rate": 7.223719676549866e-07,
      "loss": 0.0007,
      "reward": 0.46875,
      "reward_std": 0.0625,
      "rewards/accuracy_reward_v1": 0.0,
      "rewards/format_reward": 0.46875,
      "step": 104
    },
    {
      "completion_length": 6.265625,
      "epoch": 1.9811320754716981,
      "grad_norm": 25.54795055152684,
      "kl": 0.11738302558660507,
      "learning_rate": 7.19676549865229e-07,
      "loss": 0.0047,
      "reward": 0.296875,
      "reward_std": 0.21875,
      "rewards/accuracy_reward_v1": 0.046875,
      "rewards/format_reward": 0.25,
      "step": 105
    },
    {
      "completion_length": 6.0,
      "epoch": 2.0,
      "grad_norm": 1.580962944198986,
      "kl": 0.03179383184760809,
      "learning_rate": 7.169811320754716e-07,
      "loss": 0.0008,
      "reward": -0.7000000029802322,
      "reward_std": 0.0,
      "rewards/accuracy_reward_v1": 0.0,
      "rewards/format_reward": -0.7000000029802322,
      "step": 106
    },
    {
      "completion_length": 11.234375,
      "epoch": 2.018867924528302,
      "grad_norm": 6.997437005547989,
      "kl": 0.01108578743878752,
      "learning_rate": 7.142857142857143e-07,
      "loss": 0.0004,
      "reward": 0.46875,
      "reward_std": 0.0625,
      "rewards/accuracy_reward_v1": 0.0,
      "rewards/format_reward": 0.46875,
      "step": 107
    },
    {
      "completion_length": 8.375,
      "epoch": 2.0377358490566038,
      "grad_norm": 17.559739244329524,
      "kl": 0.02117897942662239,
      "learning_rate": 7.115902964959568e-07,
      "loss": 0.0008,
      "reward": 0.625,
      "reward_std": 0.25,
      "rewards/accuracy_reward_v1": 0.03125,
      "rewards/format_reward": 0.59375,
      "step": 108
    },
    {
      "completion_length": 6.0,
      "epoch": 2.056603773584906,
      "grad_norm": 0.011915576527115862,
      "kl": 0.0025631723401602358,
      "learning_rate": 7.088948787061995e-07,
      "loss": 0.0001,
      "reward": 0.125,
      "reward_std": 0.0,
      "rewards/accuracy_reward_v1": 0.0,
      "rewards/format_reward": 0.125,
      "step": 109
    },
    {
      "completion_length": 14.234375,
      "epoch": 2.0754716981132075,
      "grad_norm": 5.869281630758835,
      "kl": 0.0786725077778101,
      "learning_rate": 7.06199460916442e-07,
      "loss": 0.0031,
      "reward": 1.0,
      "reward_std": 0.25,
      "rewards/accuracy_reward_v1": 0.03125,
      "rewards/format_reward": 0.96875,
      "step": 110
    },
    {
      "completion_length": 6.5625,
      "epoch": 2.0943396226415096,
      "grad_norm": 40.12499099754656,
      "kl": 0.06683788076043129,
      "learning_rate": 7.035040431266847e-07,
      "loss": 0.0027,
      "reward": 0.390625,
      "reward_std": 0.28937220573425293,
      "rewards/accuracy_reward_v1": 0.046875,
      "rewards/format_reward": 0.34375,
      "step": 111
    },
    {
      "completion_length": 6.078125,
      "epoch": 2.1132075471698113,
      "grad_norm": 25.245861878154713,
      "kl": 0.03486994560807943,
      "learning_rate": 7.008086253369272e-07,
      "loss": 0.0014,
      "reward": 0.125,
      "reward_std": 0.125,
      "rewards/accuracy_reward_v1": 0.03125,
      "rewards/format_reward": 0.09375,
      "step": 112
    },
    {
      "completion_length": 8.71875,
      "epoch": 2.1320754716981134,
      "grad_norm": 14.871171677520552,
      "kl": 0.15129369497299194,
      "learning_rate": 6.981132075471697e-07,
      "loss": 0.0061,
      "reward": 0.4375,
      "reward_std": 0.25,
      "rewards/accuracy_reward_v1": 0.03125,
      "rewards/format_reward": 0.40625,
      "step": 113
    },
    {
      "completion_length": 6.0,
      "epoch": 2.150943396226415,
      "grad_norm": 0.7014051810495818,
      "kl": 0.011805298272520304,
      "learning_rate": 6.954177897574124e-07,
      "loss": 0.0005,
      "reward": 0.5,
      "reward_std": 0.0,
      "rewards/accuracy_reward_v1": 0.0,
      "rewards/format_reward": 0.5,
      "step": 114
    },
    {
      "completion_length": 18.4375,
      "epoch": 2.169811320754717,
      "grad_norm": 0.07024164686329869,
      "kl": 0.034326742053963244,
      "learning_rate": 6.927223719676549e-07,
      "loss": 0.0014,
      "reward": 0.859375,
      "reward_std": 0.03125,
      "rewards/accuracy_reward_v1": -0.015625,
      "rewards/format_reward": 0.875,
      "step": 115
    },
    {
      "completion_length": 6.5625,
      "epoch": 2.188679245283019,
      "grad_norm": 17.55893171420226,
      "kl": 0.040992474649101496,
      "learning_rate": 6.900269541778976e-07,
      "loss": 0.0016,
      "reward": 1.390625,
      "reward_std": 0.09375,
      "rewards/accuracy_reward_v1": 0.0,
      "rewards/format_reward": 1.390625,
      "step": 116
    },
    {
      "completion_length": 6.453125,
      "epoch": 2.207547169811321,
      "grad_norm": 16.828995213241594,
      "kl": 0.0306085217744112,
      "learning_rate": 6.873315363881402e-07,
      "loss": 0.0012,
      "reward": 0.53125,
      "reward_std": 0.0625,
      "rewards/accuracy_reward_v1": 0.03125,
      "rewards/format_reward": 0.5,
      "step": 117
    },
    {
      "completion_length": 7.15625,
      "epoch": 2.2264150943396226,
      "grad_norm": 31.650829733459336,
      "kl": 0.0922243557870388,
      "learning_rate": 6.846361185983827e-07,
      "loss": 0.0037,
      "reward": 0.78125,
      "reward_std": 0.13466878235340118,
      "rewards/accuracy_reward_v1": 0.09375,
      "rewards/format_reward": 0.6875,
      "step": 118
    },
    {
      "completion_length": 6.0,
      "epoch": 2.2452830188679247,
      "grad_norm": 0.005777622686102371,
      "kl": 0.0019141640368616208,
      "learning_rate": 6.819407008086253e-07,
      "loss": 0.0001,
      "reward": 0.6875,
      "reward_std": 0.0,
      "rewards/accuracy_reward_v1": 0.0,
      "rewards/format_reward": 0.6875,
      "step": 119
    },
    {
      "completion_length": 7.203125,
      "epoch": 2.2641509433962264,
      "grad_norm": 7.316117668482866,
      "kl": 0.04923913301900029,
      "learning_rate": 6.792452830188678e-07,
      "loss": 0.002,
      "reward": 1.15625,
      "reward_std": 0.1875,
      "rewards/accuracy_reward_v1": 0.0,
      "rewards/format_reward": 1.15625,
      "step": 120
    },
    {
      "completion_length": 6.0,
      "epoch": 2.2830188679245285,
      "grad_norm": 0.0473944187186787,
      "kl": 0.003173680364852771,
      "learning_rate": 6.765498652291105e-07,
      "loss": 0.0001,
      "reward": 0.5,
      "reward_std": 0.0,
      "rewards/accuracy_reward_v1": 0.0,
      "rewards/format_reward": 0.5,
      "step": 121
    },
    {
      "completion_length": 6.046875,
      "epoch": 2.30188679245283,
      "grad_norm": 16.730925417032434,
      "kl": 0.04596833884716034,
      "learning_rate": 6.738544474393531e-07,
      "loss": 0.0018,
      "reward": 0.671875,
      "reward_std": 0.15625,
      "rewards/accuracy_reward_v1": 0.03125,
      "rewards/format_reward": 0.640625,
      "step": 122
    },
    {
      "completion_length": 6.0,
      "epoch": 2.3207547169811322,
      "grad_norm": 0.8235924994429733,
      "kl": 0.011163470335304737,
      "learning_rate": 6.711590296495957e-07,
      "loss": 0.0004,
      "reward": 0.875,
      "reward_std": 0.0,
      "rewards/accuracy_reward_v1": 0.0,
      "rewards/format_reward": 0.875,
      "step": 123
    },
    {
      "completion_length": 7.5,
      "epoch": 2.339622641509434,
      "grad_norm": 93.95542031847113,
      "kl": 0.2978769689798355,
      "learning_rate": 6.684636118598382e-07,
      "loss": 0.0119,
      "reward": 0.046875,
      "reward_std": 0.40625,
      "rewards/accuracy_reward_v1": -0.015625,
      "rewards/format_reward": 0.0625,
      "step": 124
    },
    {
      "completion_length": 6.0,
      "epoch": 2.358490566037736,
      "grad_norm": 0.6089558332172572,
      "kl": 0.03021073527634144,
      "learning_rate": 6.657681940700808e-07,
      "loss": 0.0012,
      "reward": 0.484375,
      "reward_std": 0.03125,
      "rewards/accuracy_reward_v1": -0.015625,
      "rewards/format_reward": 0.5,
      "step": 125
    },
    {
      "completion_length": 6.90625,
      "epoch": 2.3773584905660377,
      "grad_norm": 15.180374038204938,
      "kl": 0.0736192874610424,
      "learning_rate": 6.630727762803234e-07,
      "loss": 0.0029,
      "reward": 0.5,
      "reward_std": 0.25,
      "rewards/accuracy_reward_v1": 0.046875,
      "rewards/format_reward": 0.453125,
      "step": 126
    },
    {
      "completion_length": 10.71875,
      "epoch": 2.3962264150943398,
      "grad_norm": 7.880577023330441,
      "kl": 0.08375321701169014,
      "learning_rate": 6.60377358490566e-07,
      "loss": 0.0034,
      "reward": -0.03125,
      "reward_std": 0.24456289410591125,
      "rewards/accuracy_reward_v1": 0.078125,
      "rewards/format_reward": -0.109375,
      "step": 127
    },
    {
      "completion_length": 9.453125,
      "epoch": 2.4150943396226414,
      "grad_norm": 12.939571057618204,
      "kl": 0.013894313480705023,
      "learning_rate": 6.576819407008086e-07,
      "loss": 0.0006,
      "reward": 0.125,
      "reward_std": 0.3221687823534012,
      "rewards/accuracy_reward_v1": 0.09375,
      "rewards/format_reward": 0.03125,
      "step": 128
    },
    {
      "completion_length": 6.0,
      "epoch": 2.4339622641509435,
      "grad_norm": 0.07251639394373288,
      "kl": 0.0011274355347268283,
      "learning_rate": 6.549865229110512e-07,
      "loss": 0.0,
      "reward": 0.46875,
      "reward_std": 0.03608439117670059,
      "rewards/accuracy_reward_v1": -0.03125,
      "rewards/format_reward": 0.5,
      "step": 129
    },
    {
      "completion_length": 6.6875,
      "epoch": 2.452830188679245,
      "grad_norm": 14.012757408408744,
      "kl": 0.013412876753136516,
      "learning_rate": 6.522911051212939e-07,
      "loss": 0.0005,
      "reward": 0.953125,
      "reward_std": 0.21875,
      "rewards/accuracy_reward_v1": -0.015625,
      "rewards/format_reward": 0.96875,
      "step": 130
    },
    {
      "completion_length": 9.390625,
      "epoch": 2.4716981132075473,
      "grad_norm": 9.440315615957966,
      "kl": 0.0242958867456764,
      "learning_rate": 6.495956873315363e-07,
      "loss": 0.001,
      "reward": 0.265625,
      "reward_std": 0.28125,
      "rewards/accuracy_reward_v1": 0.046875,
      "rewards/format_reward": 0.21875,
      "step": 131
    },
    {
      "completion_length": 135.328125,
      "epoch": 2.490566037735849,
      "grad_norm": 5.164945482149966,
      "kl": 0.006374373915605247,
      "learning_rate": 6.469002695417789e-07,
      "loss": 0.0003,
      "reward": 0.75,
      "reward_std": 0.125,
      "rewards/accuracy_reward_v1": 0.109375,
      "rewards/format_reward": 0.640625,
      "step": 132
    },
    {
      "completion_length": 6.78125,
      "epoch": 2.509433962264151,
      "grad_norm": 11.723072585015913,
      "kl": 0.06275491928681731,
      "learning_rate": 6.442048517520215e-07,
      "loss": 0.0025,
      "reward": 0.671875,
      "reward_std": 0.2645031735301018,
      "rewards/accuracy_reward_v1": 0.03125,
      "rewards/format_reward": 0.640625,
      "step": 133
    },
    {
      "completion_length": 8.546875,
      "epoch": 2.5283018867924527,
      "grad_norm": 31.031718596134894,
      "kl": 0.09517993405461311,
      "learning_rate": 6.415094339622641e-07,
      "loss": 0.0038,
      "reward": 0.46875,
      "reward_std": 0.2607824504375458,
      "rewards/accuracy_reward_v1": 0.015625,
      "rewards/format_reward": 0.453125,
      "step": 134
    },
    {
      "completion_length": 6.875,
      "epoch": 2.547169811320755,
      "grad_norm": 2.97226535410033,
      "kl": 0.0015688572020735592,
      "learning_rate": 6.388140161725068e-07,
      "loss": 0.0001,
      "reward": 0.34375,
      "reward_std": 0.0625,
      "rewards/accuracy_reward_v1": 0.03125,
      "rewards/format_reward": 0.3125,
      "step": 135
    },
    {
      "completion_length": 7.46875,
      "epoch": 2.5660377358490565,
      "grad_norm": 30.534802645399633,
      "kl": 0.12897738814353943,
      "learning_rate": 6.361185983827493e-07,
      "loss": 0.0052,
      "reward": 0.4375,
      "reward_std": 0.3221687823534012,
      "rewards/accuracy_reward_v1": 0.046875,
      "rewards/format_reward": 0.390625,
      "step": 136
    },
    {
      "completion_length": 7.1875,
      "epoch": 2.5849056603773586,
      "grad_norm": 22.500646669412244,
      "kl": 0.024923089891672134,
      "learning_rate": 6.33423180592992e-07,
      "loss": 0.001,
      "reward": 0.203125,
      "reward_std": 0.21875,
      "rewards/accuracy_reward_v1": 0.078125,
      "rewards/format_reward": 0.125,
      "step": 137
    },
    {
      "completion_length": 7.4375,
      "epoch": 2.6037735849056602,
      "grad_norm": 22.638209851348872,
      "kl": 0.08623421750962734,
      "learning_rate": 6.307277628032344e-07,
      "loss": 0.0034,
      "reward": 0.609375,
      "reward_std": 0.21875,
      "rewards/accuracy_reward_v1": -0.03125,
      "rewards/format_reward": 0.640625,
      "step": 138
    },
    {
      "completion_length": 10.875,
      "epoch": 2.6226415094339623,
      "grad_norm": 25.20962860924125,
      "kl": 0.1827621478587389,
      "learning_rate": 6.28032345013477e-07,
      "loss": 0.0073,
      "reward": 0.828125,
      "reward_std": 0.33831289410591125,
      "rewards/accuracy_reward_v1": 0.203125,
      "rewards/format_reward": 0.625,
      "step": 139
    },
    {
      "completion_length": 7.015625,
      "epoch": 2.641509433962264,
      "grad_norm": 5.275033636735031,
      "kl": 0.005385308701079339,
      "learning_rate": 6.253369272237197e-07,
      "loss": 0.0002,
      "reward": 0.328125,
      "reward_std": 0.09375,
      "rewards/accuracy_reward_v1": 0.015625,
      "rewards/format_reward": 0.3125,
      "step": 140
    },
    {
      "completion_length": 8.46875,
      "epoch": 2.660377358490566,
      "grad_norm": 22.17583650118601,
      "kl": 0.0279629435390234,
      "learning_rate": 6.226415094339622e-07,
      "loss": 0.0011,
      "reward": 0.875,
      "reward_std": 0.2957531809806824,
      "rewards/accuracy_reward_v1": 0.0,
      "rewards/format_reward": 0.875,
      "step": 141
    },
    {
      "completion_length": 7.375,
      "epoch": 2.6792452830188678,
      "grad_norm": 21.441145926885206,
      "kl": 0.020797467790544033,
      "learning_rate": 6.199460916442049e-07,
      "loss": 0.0008,
      "reward": 0.859375,
      "reward_std": 0.15625,
      "rewards/accuracy_reward_v1": 0.03125,
      "rewards/format_reward": 0.828125,
      "step": 142
    },
    {
      "completion_length": 9.9375,
      "epoch": 2.69811320754717,
      "grad_norm": 9.816206120316384,
      "kl": 0.018046388868242502,
      "learning_rate": 6.172506738544473e-07,
      "loss": 0.0007,
      "reward": 0.390625,
      "reward_std": 0.34375,
      "rewards/accuracy_reward_v1": 0.03125,
      "rewards/format_reward": 0.359375,
      "step": 143
    },
    {
      "completion_length": 22.484375,
      "epoch": 2.7169811320754715,
      "grad_norm": 49.852406732174884,
      "kl": 0.02830031793564558,
      "learning_rate": 6.1455525606469e-07,
      "loss": 0.0011,
      "reward": 0.265625,
      "reward_std": 0.29091876745224,
      "rewards/accuracy_reward_v1": 0.09375,
      "rewards/format_reward": 0.171875,
      "step": 144
    },
    {
      "completion_length": 30.078125,
      "epoch": 2.7358490566037736,
      "grad_norm": 40.8497610426359,
      "kl": 0.07939931750297546,
      "learning_rate": 6.118598382749326e-07,
      "loss": 0.0032,
      "reward": 0.46875,
      "reward_std": 0.5285891890525818,
      "rewards/accuracy_reward_v1": 0.203125,
      "rewards/format_reward": 0.265625,
      "step": 145
    },
    {
      "completion_length": 9.640625,
      "epoch": 2.7547169811320753,
      "grad_norm": 41.85567107039319,
      "kl": 0.03684032987803221,
      "learning_rate": 6.091644204851751e-07,
      "loss": 0.0015,
      "reward": 0.953125,
      "reward_std": 0.38950315117836,
      "rewards/accuracy_reward_v1": 0.03125,
      "rewards/format_reward": 0.921875,
      "step": 146
    },
    {
      "completion_length": 7.0625,
      "epoch": 2.7735849056603774,
      "grad_norm": 5.795460704591,
      "kl": 0.08030329644680023,
      "learning_rate": 6.064690026954178e-07,
      "loss": 0.0032,
      "reward": 1.15625,
      "reward_std": 0.1875,
      "rewards/accuracy_reward_v1": 0.0,
      "rewards/format_reward": 1.15625,
      "step": 147
    },
    {
      "completion_length": 10.140625,
      "epoch": 2.7924528301886795,
      "grad_norm": 35.07840542892151,
      "kl": 0.03356534452177584,
      "learning_rate": 6.037735849056604e-07,
      "loss": 0.0013,
      "reward": 0.71875,
      "reward_std": 0.25,
      "rewards/accuracy_reward_v1": 0.078125,
      "rewards/format_reward": 0.640625,
      "step": 148
    },
    {
      "completion_length": 6.859375,
      "epoch": 2.811320754716981,
      "grad_norm": 3.671903367779326,
      "kl": 0.029136563651263714,
      "learning_rate": 6.01078167115903e-07,
      "loss": 0.0012,
      "reward": -0.09375,
      "reward_std": 0.125,
      "rewards/accuracy_reward_v1": 0.015625,
      "rewards/format_reward": -0.109375,
      "step": 149
    },
    {
      "completion_length": 9.265625,
      "epoch": 2.830188679245283,
      "grad_norm": 204.5378370992803,
      "kl": 0.019195225555449724,
      "learning_rate": 5.983827493261455e-07,
      "loss": 0.0008,
      "reward": 0.578125,
      "reward_std": 0.2909187823534012,
      "rewards/accuracy_reward_v1": 0.125,
      "rewards/format_reward": 0.453125,
      "step": 150
    }
  ],
  "logging_steps": 1.0,
  "max_steps": 371,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 7,
  "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
}