{
  "best_global_step": null,
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 8.0,
  "eval_steps": 500,
  "global_step": 2000,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "clip_ratio": 0.0,
      "completion_length": 240.75,
      "epoch": 0.004,
      "grad_norm": 3.483977794647217,
      "kl": 0.0011882781982421875,
      "learning_rate": 9.996e-07,
      "loss": 0.0,
      "reward": 0.0625,
      "reward_std": 0.1767766922712326,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.0625,
      "step": 1
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 244.59375,
      "epoch": 0.008,
      "grad_norm": 0.001067359815351665,
      "kl": 0.001041412353515625,
      "learning_rate": 9.992e-07,
      "loss": 0.0,
      "reward": 0.0,
      "reward_std": 0.0,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.0,
      "step": 2
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 246.375,
      "epoch": 0.012,
      "grad_norm": 0.003222062485292554,
      "kl": 0.001255035400390625,
      "learning_rate": 9.988e-07,
      "loss": 0.0,
      "reward": 0.0,
      "reward_std": 0.0,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.0,
      "step": 3
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 231.40625,
      "epoch": 0.016,
      "grad_norm": 4.284538745880127,
      "kl": 0.00179290771484375,
      "learning_rate": 9.983999999999998e-07,
      "loss": 0.0,
      "reward": 0.125,
      "reward_std": 0.2925042062997818,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.125,
      "step": 4
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 226.90625,
      "epoch": 0.02,
      "grad_norm": 5.088000774383545,
      "kl": 0.00188446044921875,
      "learning_rate": 9.98e-07,
      "loss": 0.0,
      "reward": 0.125,
      "reward_std": 0.2925042062997818,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.125,
      "step": 5
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 220.71875,
      "epoch": 0.024,
      "grad_norm": 6.958590507507324,
      "kl": 0.002094268798828125,
      "learning_rate": 9.976e-07,
      "loss": 0.0,
      "reward": 0.0625,
      "reward_std": 0.1767766922712326,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.0625,
      "step": 6
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 212.875,
      "epoch": 0.028,
      "grad_norm": 2.5518085956573486,
      "kl": 0.003185272216796875,
      "learning_rate": 9.972e-07,
      "loss": 0.0,
      "reward": 0.125,
      "reward_std": 0.13363061845302582,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.125,
      "step": 7
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 208.375,
      "epoch": 0.032,
      "grad_norm": 0.003275999566540122,
      "kl": 0.0058441162109375,
      "learning_rate": 9.968e-07,
      "loss": 0.0,
      "reward": 0.0,
      "reward_std": 0.0,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.0,
      "step": 8
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 204.65625,
      "epoch": 0.036,
      "grad_norm": 3.262775182723999,
      "kl": 0.007598876953125,
      "learning_rate": 9.964e-07,
      "loss": 0.0,
      "reward": 0.0625,
      "reward_std": 0.1767766922712326,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.0625,
      "step": 9
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 198.375,
      "epoch": 0.04,
      "grad_norm": 0.004171893931925297,
      "kl": 0.0090484619140625,
      "learning_rate": 9.959999999999999e-07,
      "loss": 0.0,
      "reward": 0.0,
      "reward_std": 0.0,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.0,
      "step": 10
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 185.71875,
      "epoch": 0.044,
      "grad_norm": 11.263163566589355,
      "kl": 0.0082550048828125,
      "learning_rate": 9.956e-07,
      "loss": 0.0,
      "reward": 0.40625,
      "reward_std": 0.3061639815568924,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.40625,
      "step": 11
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 221.5,
      "epoch": 0.048,
      "grad_norm": 2.615539312362671,
      "kl": 0.0104522705078125,
      "learning_rate": 9.952e-07,
      "loss": 0.0,
      "reward": 0.09375,
      "reward_std": 0.1293872892856598,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.09375,
      "step": 12
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 182.0,
      "epoch": 0.052,
      "grad_norm": 4.1748127937316895,
      "kl": 0.0160369873046875,
      "learning_rate": 9.948e-07,
      "loss": 0.0,
      "reward": 0.21875,
      "reward_std": 0.3377464786171913,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.21875,
      "step": 13
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 200.09375,
      "epoch": 0.056,
      "grad_norm": 4.833225727081299,
      "kl": 0.018402099609375,
      "learning_rate": 9.944e-07,
      "loss": 0.0,
      "reward": 0.15625,
      "reward_std": 0.24511480331420898,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.15625,
      "step": 14
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 161.21875,
      "epoch": 0.06,
      "grad_norm": 6.787618637084961,
      "kl": 0.03887939453125,
      "learning_rate": 9.94e-07,
      "loss": 0.0,
      "reward": 0.125,
      "reward_std": 0.2314550280570984,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.125,
      "step": 15
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 179.53125,
      "epoch": 0.064,
      "grad_norm": 4.547823905944824,
      "kl": 0.03192138671875,
      "learning_rate": 9.936e-07,
      "loss": 0.0,
      "reward": 0.15625,
      "reward_std": 0.3198433741927147,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.15625,
      "step": 16
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 171.78125,
      "epoch": 0.068,
      "grad_norm": 3.955144166946411,
      "kl": 0.030029296875,
      "learning_rate": 9.931999999999999e-07,
      "loss": 0.0,
      "reward": 0.28125,
      "reward_std": 0.2041158601641655,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.28125,
      "step": 17
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 184.0,
      "epoch": 0.072,
      "grad_norm": 3.1369616985321045,
      "kl": 0.0242919921875,
      "learning_rate": 9.928e-07,
      "loss": 0.0,
      "reward": 0.03125,
      "reward_std": 0.0883883461356163,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.03125,
      "step": 18
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 165.71875,
      "epoch": 0.076,
      "grad_norm": 3.3629579544067383,
      "kl": 0.03326416015625,
      "learning_rate": 9.923999999999998e-07,
      "loss": 0.0,
      "reward": 0.125,
      "reward_std": 0.2314550280570984,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.125,
      "step": 19
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 173.09375,
      "epoch": 0.08,
      "grad_norm": 1.9398032426834106,
      "kl": 0.032012939453125,
      "learning_rate": 9.92e-07,
      "loss": 0.0,
      "reward": 0.03125,
      "reward_std": 0.0883883461356163,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.03125,
      "step": 20
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 165.28125,
      "epoch": 0.084,
      "grad_norm": 3.363475799560547,
      "kl": 0.03326416015625,
      "learning_rate": 9.916e-07,
      "loss": 0.0,
      "reward": 0.03125,
      "reward_std": 0.0883883461356163,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.03125,
      "step": 21
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 178.875,
      "epoch": 0.088,
      "grad_norm": 24.74161720275879,
      "kl": 0.02996826171875,
      "learning_rate": 9.912e-07,
      "loss": 0.0,
      "reward": 0.1875,
      "reward_std": 0.3335031494498253,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.1875,
      "step": 22
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 143.53125,
      "epoch": 0.092,
      "grad_norm": 5.056652069091797,
      "kl": 0.035888671875,
      "learning_rate": 9.908e-07,
      "loss": 0.0,
      "reward": 0.09375,
      "reward_std": 0.2041158601641655,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.09375,
      "step": 23
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 149.125,
      "epoch": 0.096,
      "grad_norm": 2.984828233718872,
      "kl": 0.036224365234375,
      "learning_rate": 9.903999999999999e-07,
      "loss": 0.0,
      "reward": 0.34375,
      "reward_std": 0.24511480331420898,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.34375,
      "step": 24
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 140.375,
      "epoch": 0.1,
      "grad_norm": 11.72681713104248,
      "kl": 0.042724609375,
      "learning_rate": 9.9e-07,
      "loss": 0.0,
      "reward": 0.15625,
      "reward_std": 0.3808925524353981,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.15625,
      "step": 25
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 156.3125,
      "epoch": 0.104,
      "grad_norm": 9.360048294067383,
      "kl": 0.0421142578125,
      "learning_rate": 9.896e-07,
      "loss": 0.0,
      "reward": 0.25,
      "reward_std": 0.4492306634783745,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.25,
      "step": 26
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 156.8125,
      "epoch": 0.108,
      "grad_norm": 5.80740213394165,
      "kl": 0.04278564453125,
      "learning_rate": 9.892e-07,
      "loss": 0.0,
      "reward": 0.15625,
      "reward_std": 0.3198433741927147,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.15625,
      "step": 27
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 152.4375,
      "epoch": 0.112,
      "grad_norm": 5.797030448913574,
      "kl": 0.04522705078125,
      "learning_rate": 9.888e-07,
      "loss": 0.0,
      "reward": 0.1875,
      "reward_std": 0.249358132481575,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.1875,
      "step": 28
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 160.8125,
      "epoch": 0.116,
      "grad_norm": 15.973736763000488,
      "kl": 0.0457763671875,
      "learning_rate": 9.884e-07,
      "loss": 0.0,
      "reward": 0.125,
      "reward_std": 0.2925042062997818,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.125,
      "step": 29
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 144.46875,
      "epoch": 0.12,
      "grad_norm": 52.75318908691406,
      "kl": 0.05126953125,
      "learning_rate": 9.88e-07,
      "loss": 0.0001,
      "reward": 0.25,
      "reward_std": 0.4492306634783745,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.25,
      "step": 30
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 142.625,
      "epoch": 0.124,
      "grad_norm": 19.736919403076172,
      "kl": 0.05572509765625,
      "learning_rate": 9.876e-07,
      "loss": 0.0001,
      "reward": 0.15625,
      "reward_std": 0.3061639815568924,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.15625,
      "step": 31
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 159.875,
      "epoch": 0.128,
      "grad_norm": 5.151854515075684,
      "kl": 0.05242919921875,
      "learning_rate": 9.871999999999998e-07,
      "loss": 0.0001,
      "reward": 0.125,
      "reward_std": 0.2177756354212761,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.125,
      "step": 32
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 163.5,
      "epoch": 0.132,
      "grad_norm": 4.104671001434326,
      "kl": 0.04571533203125,
      "learning_rate": 9.868e-07,
      "loss": 0.0,
      "reward": 0.125,
      "reward_std": 0.2314550280570984,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.125,
      "step": 33
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 151.46875,
      "epoch": 0.136,
      "grad_norm": 11.92469310760498,
      "kl": 0.05950927734375,
      "learning_rate": 9.864e-07,
      "loss": 0.0001,
      "reward": 0.25,
      "reward_std": 0.3514062538743019,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.25,
      "step": 34
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 136.3125,
      "epoch": 0.14,
      "grad_norm": 6.980234622955322,
      "kl": 0.05780029296875,
      "learning_rate": 9.86e-07,
      "loss": 0.0001,
      "reward": 0.375,
      "reward_std": 0.3924051970243454,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.375,
      "step": 35
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 146.34375,
      "epoch": 0.144,
      "grad_norm": 2.9053189754486084,
      "kl": 0.06768798828125,
      "learning_rate": 9.856e-07,
      "loss": 0.0001,
      "reward": 0.03125,
      "reward_std": 0.0883883461356163,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.03125,
      "step": 36
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 132.1875,
      "epoch": 0.148,
      "grad_norm": 10.398985862731934,
      "kl": 0.06573486328125,
      "learning_rate": 9.852e-07,
      "loss": 0.0001,
      "reward": 0.125,
      "reward_std": 0.2314550280570984,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.125,
      "step": 37
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 134.4375,
      "epoch": 0.152,
      "grad_norm": 12.788537979125977,
      "kl": 0.07025146484375,
      "learning_rate": 9.847999999999999e-07,
      "loss": 0.0001,
      "reward": 0.09375,
      "reward_std": 0.2041158601641655,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.09375,
      "step": 38
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 125.6875,
      "epoch": 0.156,
      "grad_norm": 3.894649028778076,
      "kl": 0.068603515625,
      "learning_rate": 9.844e-07,
      "loss": 0.0001,
      "reward": 0.375,
      "reward_std": 0.2314550280570984,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.375,
      "step": 39
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 128.875,
      "epoch": 0.16,
      "grad_norm": 4.223596096038818,
      "kl": 0.07427978515625,
      "learning_rate": 9.84e-07,
      "loss": 0.0001,
      "reward": 0.09375,
      "reward_std": 0.2041158601641655,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.09375,
      "step": 40
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 135.375,
      "epoch": 0.164,
      "grad_norm": 3.229766845703125,
      "kl": 0.08056640625,
      "learning_rate": 9.836e-07,
      "loss": 0.0001,
      "reward": 0.28125,
      "reward_std": 0.2630179077386856,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.28125,
      "step": 41
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 126.0625,
      "epoch": 0.168,
      "grad_norm": 5.142240047454834,
      "kl": 0.076904296875,
      "learning_rate": 9.832e-07,
      "loss": 0.0001,
      "reward": 0.15625,
      "reward_std": 0.3061639815568924,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.15625,
      "step": 42
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 137.46875,
      "epoch": 0.172,
      "grad_norm": 6.515385627746582,
      "kl": 0.08544921875,
      "learning_rate": 9.828e-07,
      "loss": 0.0001,
      "reward": 0.34375,
      "reward_std": 0.4218914955854416,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.34375,
      "step": 43
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 163.40625,
      "epoch": 0.176,
      "grad_norm": 6.223419666290283,
      "kl": 0.06097412109375,
      "learning_rate": 9.824e-07,
      "loss": 0.0001,
      "reward": 0.28125,
      "reward_std": 0.3787454217672348,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.28125,
      "step": 44
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 131.65625,
      "epoch": 0.18,
      "grad_norm": 4.957655429840088,
      "kl": 0.0849609375,
      "learning_rate": 9.819999999999999e-07,
      "loss": 0.0001,
      "reward": 0.34375,
      "reward_std": 0.3198433741927147,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.34375,
      "step": 45
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 132.1875,
      "epoch": 0.184,
      "grad_norm": 2.228003978729248,
      "kl": 0.0794677734375,
      "learning_rate": 9.816e-07,
      "loss": 0.0001,
      "reward": 0.0625,
      "reward_std": 0.1157275140285492,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.0625,
      "step": 46
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 143.875,
      "epoch": 0.188,
      "grad_norm": 11.523205757141113,
      "kl": 0.0767822265625,
      "learning_rate": 9.811999999999998e-07,
      "loss": 0.0001,
      "reward": 0.125,
      "reward_std": 0.2177756354212761,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.125,
      "step": 47
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 156.6875,
      "epoch": 0.192,
      "grad_norm": 7.209992408752441,
      "kl": 0.073486328125,
      "learning_rate": 9.808e-07,
      "loss": 0.0001,
      "reward": 0.25,
      "reward_std": 0.3514062538743019,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.25,
      "step": 48
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 155.40625,
      "epoch": 0.196,
      "grad_norm": 5.14667272567749,
      "kl": 0.1248779296875,
      "learning_rate": 9.804e-07,
      "loss": 0.0001,
      "reward": 0.25,
      "reward_std": 0.4492306634783745,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.25,
      "step": 49
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 146.96875,
      "epoch": 0.2,
      "grad_norm": 4.949409008026123,
      "kl": 0.0828857421875,
      "learning_rate": 9.8e-07,
      "loss": 0.0001,
      "reward": 0.28125,
      "reward_std": 0.4628904387354851,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.28125,
      "step": 50
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 122.375,
      "epoch": 0.204,
      "grad_norm": 13.097493171691895,
      "kl": 0.0899658203125,
      "learning_rate": 9.796e-07,
      "loss": 0.0001,
      "reward": 0.15625,
      "reward_std": 0.3808925524353981,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.15625,
      "step": 51
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 135.15625,
      "epoch": 0.208,
      "grad_norm": 12.397465705871582,
      "kl": 0.08447265625,
      "learning_rate": 9.791999999999999e-07,
      "loss": 0.0001,
      "reward": 0.1875,
      "reward_std": 0.3335031494498253,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.1875,
      "step": 52
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 125.75,
      "epoch": 0.212,
      "grad_norm": 6.712313175201416,
      "kl": 0.0906982421875,
      "learning_rate": 9.788e-07,
      "loss": 0.0001,
      "reward": 0.375,
      "reward_std": 0.408231720328331,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.375,
      "step": 53
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 141.625,
      "epoch": 0.216,
      "grad_norm": 5.286505222320557,
      "kl": 0.0909423828125,
      "learning_rate": 9.784e-07,
      "loss": 0.0001,
      "reward": 0.3125,
      "reward_std": 0.3745020925998688,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.3125,
      "step": 54
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 152.25,
      "epoch": 0.22,
      "grad_norm": 4.269443511962891,
      "kl": 0.08026123046875,
      "learning_rate": 9.78e-07,
      "loss": 0.0001,
      "reward": 0.0625,
      "reward_std": 0.1767766922712326,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.0625,
      "step": 55
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 138.09375,
      "epoch": 0.224,
      "grad_norm": 4.18522310256958,
      "kl": 0.0966796875,
      "learning_rate": 9.776e-07,
      "loss": 0.0001,
      "reward": 0.25,
      "reward_std": 0.2925042062997818,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.25,
      "step": 56
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 140.3125,
      "epoch": 0.228,
      "grad_norm": 7.846221446990967,
      "kl": 0.1097412109375,
      "learning_rate": 9.772e-07,
      "loss": 0.0001,
      "reward": 0.4375,
      "reward_std": 0.2925042062997818,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.4375,
      "step": 57
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 141.59375,
      "epoch": 0.232,
      "grad_norm": 2.465056896209717,
      "kl": 0.0821533203125,
      "learning_rate": 9.768e-07,
      "loss": 0.0001,
      "reward": 0.09375,
      "reward_std": 0.1293872892856598,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.09375,
      "step": 58
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 142.6875,
      "epoch": 0.236,
      "grad_norm": 34.07876968383789,
      "kl": 0.07476806640625,
      "learning_rate": 9.764e-07,
      "loss": 0.0001,
      "reward": 0.3125,
      "reward_std": 0.3745020925998688,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.3125,
      "step": 59
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 136.40625,
      "epoch": 0.24,
      "grad_norm": 4.285093784332275,
      "kl": 0.07568359375,
      "learning_rate": 9.759999999999998e-07,
      "loss": 0.0001,
      "reward": 0.1875,
      "reward_std": 0.3104073107242584,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.1875,
      "step": 60
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 156.5,
      "epoch": 0.244,
      "grad_norm": 4.140926837921143,
      "kl": 0.08203125,
      "learning_rate": 9.756e-07,
      "loss": 0.0001,
      "reward": 0.15625,
      "reward_std": 0.22201896458864212,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.15625,
      "step": 61
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 154.0625,
      "epoch": 0.248,
      "grad_norm": 5.816662311553955,
      "kl": 0.0833740234375,
      "learning_rate": 9.752e-07,
      "loss": 0.0001,
      "reward": 0.125,
      "reward_std": 0.2177756354212761,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.125,
      "step": 62
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 144.75,
      "epoch": 0.252,
      "grad_norm": 5.923285961151123,
      "kl": 0.079345703125,
      "learning_rate": 9.748e-07,
      "loss": 0.0001,
      "reward": 0.5,
      "reward_std": 0.4355512708425522,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.5,
      "step": 63
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 143.6875,
      "epoch": 0.256,
      "grad_norm": 3.0318973064422607,
      "kl": 0.1051025390625,
      "learning_rate": 9.744e-07,
      "loss": 0.0001,
      "reward": 0.0625,
      "reward_std": 0.1157275140285492,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.0625,
      "step": 64
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 137.15625,
      "epoch": 0.26,
      "grad_norm": 9.24459171295166,
      "kl": 0.081298828125,
      "learning_rate": 9.74e-07,
      "loss": 0.0001,
      "reward": 0.34375,
      "reward_std": 0.3608423173427582,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.34375,
      "step": 65
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 139.09375,
      "epoch": 0.264,
      "grad_norm": 2.5574748516082764,
      "kl": 0.0794677734375,
      "learning_rate": 9.735999999999999e-07,
      "loss": 0.0001,
      "reward": 0.21875,
      "reward_std": 0.0883883461356163,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.21875,
      "step": 66
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 147.875,
      "epoch": 0.268,
      "grad_norm": 3.8737809658050537,
      "kl": 0.8548583984375,
      "learning_rate": 9.731999999999998e-07,
      "loss": 0.0009,
      "reward": 0.25,
      "reward_std": 0.1767766922712326,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.25,
      "step": 67
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 133.0,
      "epoch": 0.272,
      "grad_norm": 18.663503646850586,
      "kl": 0.1005859375,
      "learning_rate": 9.728e-07,
      "loss": 0.0001,
      "reward": 0.3125,
      "reward_std": 0.3924051970243454,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.3125,
      "step": 68
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 145.0,
      "epoch": 0.276,
      "grad_norm": 3.6664018630981445,
      "kl": 0.0904541015625,
      "learning_rate": 9.724e-07,
      "loss": 0.0001,
      "reward": 0.21875,
      "reward_std": 0.2041158601641655,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.21875,
      "step": 69
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 134.03125,
      "epoch": 0.28,
      "grad_norm": 4.882191181182861,
      "kl": 0.0919189453125,
      "learning_rate": 9.72e-07,
      "loss": 0.0001,
      "reward": 0.375,
      "reward_std": 0.3745020925998688,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.375,
      "step": 70
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 159.375,
      "epoch": 0.284,
      "grad_norm": 3.3950247764587402,
      "kl": 0.11431884765625,
      "learning_rate": 9.716e-07,
      "loss": 0.0001,
      "reward": 0.21875,
      "reward_std": 0.2041158601641655,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.21875,
      "step": 71
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 128.4375,
      "epoch": 0.288,
      "grad_norm": 3.155332326889038,
      "kl": 0.0821533203125,
      "learning_rate": 9.712e-07,
      "loss": 0.0001,
      "reward": 0.5625,
      "reward_std": 0.1157275140285492,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.5625,
      "step": 72
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 137.1875,
      "epoch": 0.292,
      "grad_norm": 5.511759281158447,
      "kl": 0.095458984375,
      "learning_rate": 9.707999999999999e-07,
      "loss": 0.0001,
      "reward": 0.34375,
      "reward_std": 0.2041158601641655,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.34375,
      "step": 73
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 145.21875,
      "epoch": 0.296,
      "grad_norm": 3.31162428855896,
      "kl": 0.8125,
      "learning_rate": 9.704e-07,
      "loss": 0.0008,
      "reward": 0.28125,
      "reward_std": 0.0883883461356163,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.28125,
      "step": 74
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 145.5625,
      "epoch": 0.3,
      "grad_norm": 3.7722182273864746,
      "kl": 0.0821533203125,
      "learning_rate": 9.7e-07,
      "loss": 0.0001,
      "reward": 0.59375,
      "reward_std": 0.22201896458864212,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.59375,
      "step": 75
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 143.4375,
      "epoch": 0.304,
      "grad_norm": 10.5135498046875,
      "kl": 0.0877685546875,
      "learning_rate": 9.696e-07,
      "loss": 0.0001,
      "reward": 0.28125,
      "reward_std": 0.2041158601641655,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.28125,
      "step": 76
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 139.25,
      "epoch": 0.308,
      "grad_norm": 5.515931606292725,
      "kl": 0.125732421875,
      "learning_rate": 9.692e-07,
      "loss": 0.0001,
      "reward": 0.34375,
      "reward_std": 0.3966485261917114,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.34375,
      "step": 77
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 170.375,
      "epoch": 0.312,
      "grad_norm": 3.9698073863983154,
      "kl": 0.0877685546875,
      "learning_rate": 9.688e-07,
      "loss": 0.0001,
      "reward": 0.03125,
      "reward_std": 0.0883883461356163,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.03125,
      "step": 78
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 165.96875,
      "epoch": 0.316,
      "grad_norm": 0.01992066204547882,
      "kl": 0.1063232421875,
      "learning_rate": 9.684e-07,
      "loss": 0.0001,
      "reward": 0.0,
      "reward_std": 0.0,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.0,
      "step": 79
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 146.0,
      "epoch": 0.32,
      "grad_norm": 2.0439631938934326,
      "kl": 0.08642578125,
      "learning_rate": 9.679999999999999e-07,
      "loss": 0.0001,
      "reward": 0.46875,
      "reward_std": 0.0883883461356163,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.46875,
      "step": 80
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 146.0625,
      "epoch": 0.324,
      "grad_norm": 5.627711772918701,
      "kl": 0.2181396484375,
      "learning_rate": 9.676e-07,
      "loss": 0.0002,
      "reward": 0.0625,
      "reward_std": 0.1767766922712326,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.0625,
      "step": 81
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 149.03125,
      "epoch": 0.328,
      "grad_norm": 3.953463077545166,
      "kl": 0.1114501953125,
      "learning_rate": 9.671999999999998e-07,
      "loss": 0.0001,
      "reward": 0.53125,
      "reward_std": 0.2041158601641655,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.53125,
      "step": 82
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 136.28125,
      "epoch": 0.332,
      "grad_norm": 4.612481117248535,
      "kl": 0.1048583984375,
      "learning_rate": 9.668e-07,
      "loss": 0.0001,
      "reward": 0.40625,
      "reward_std": 0.2041158601641655,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.40625,
      "step": 83
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 137.59375,
      "epoch": 0.336,
      "grad_norm": 6.570859909057617,
      "kl": 0.2337646484375,
      "learning_rate": 9.664e-07,
      "loss": 0.0002,
      "reward": 0.3125,
      "reward_std": 0.3335031494498253,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.3125,
      "step": 84
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 152.6875,
      "epoch": 0.34,
      "grad_norm": 8.282193183898926,
      "kl": 0.0938720703125,
      "learning_rate": 9.66e-07,
      "loss": 0.0001,
      "reward": 0.3125,
      "reward_std": 0.4355512708425522,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.3125,
      "step": 85
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 135.28125,
      "epoch": 0.344,
      "grad_norm": 3.57267165184021,
      "kl": 0.1107177734375,
      "learning_rate": 9.656e-07,
      "loss": 0.0001,
      "reward": 0.28125,
      "reward_std": 0.2041158601641655,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.28125,
      "step": 86
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 153.8125,
      "epoch": 0.348,
      "grad_norm": 9.419325828552246,
      "kl": 0.0953369140625,
      "learning_rate": 9.651999999999999e-07,
      "loss": 0.0001,
      "reward": 0.34375,
      "reward_std": 0.1293872892856598,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.34375,
      "step": 87
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 143.5625,
      "epoch": 0.352,
      "grad_norm": 11.560566902160645,
      "kl": 0.1787109375,
      "learning_rate": 9.647999999999999e-07,
      "loss": 0.0002,
      "reward": 0.34375,
      "reward_std": 0.2041158601641655,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.34375,
      "step": 88
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 145.1875,
      "epoch": 0.356,
      "grad_norm": 8.131706237792969,
      "kl": 0.12060546875,
      "learning_rate": 9.644e-07,
      "loss": 0.0001,
      "reward": 0.03125,
      "reward_std": 0.0883883461356163,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.03125,
      "step": 89
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 139.21875,
      "epoch": 0.36,
      "grad_norm": 11.170744895935059,
      "kl": 0.1241455078125,
      "learning_rate": 9.64e-07,
      "loss": 0.0001,
      "reward": 0.34375,
      "reward_std": 0.3377464786171913,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.34375,
      "step": 90
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 155.6875,
      "epoch": 0.364,
      "grad_norm": 27.866975784301758,
      "kl": 0.1082763671875,
      "learning_rate": 9.636e-07,
      "loss": 0.0001,
      "reward": 0.125,
      "reward_std": 0.2177756354212761,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.125,
      "step": 91
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 151.75,
      "epoch": 0.368,
      "grad_norm": 5.540046215057373,
      "kl": 0.1146240234375,
      "learning_rate": 9.632e-07,
      "loss": 0.0001,
      "reward": 0.21875,
      "reward_std": 0.0883883461356163,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.21875,
      "step": 92
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 136.96875,
      "epoch": 0.372,
      "grad_norm": 4.550509929656982,
      "kl": 0.1181640625,
      "learning_rate": 9.628e-07,
      "loss": 0.0001,
      "reward": 0.625,
      "reward_std": 0.3745020925998688,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.625,
      "step": 93
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 151.28125,
      "epoch": 0.376,
      "grad_norm": 6.27862548828125,
      "kl": 0.1434326171875,
      "learning_rate": 9.624e-07,
      "loss": 0.0001,
      "reward": 0.0625,
      "reward_std": 0.1767766922712326,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.0625,
      "step": 94
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 135.4375,
      "epoch": 0.38,
      "grad_norm": 14.521442413330078,
      "kl": 0.1177978515625,
      "learning_rate": 9.619999999999999e-07,
      "loss": 0.0001,
      "reward": 0.1875,
      "reward_std": 0.2177756354212761,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.1875,
      "step": 95
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 160.3125,
      "epoch": 0.384,
      "grad_norm": 5.92653751373291,
      "kl": 0.1239013671875,
      "learning_rate": 9.616e-07,
      "loss": 0.0001,
      "reward": 0.3125,
      "reward_std": 0.2925042062997818,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.3125,
      "step": 96
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 137.625,
      "epoch": 0.388,
      "grad_norm": 8.451234817504883,
      "kl": 0.134521484375,
      "learning_rate": 9.612e-07,
      "loss": 0.0001,
      "reward": 0.625,
      "reward_std": 0.3650856465101242,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.625,
      "step": 97
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 142.65625,
      "epoch": 0.392,
      "grad_norm": 0.014137900434434414,
      "kl": 0.146728515625,
      "learning_rate": 9.608e-07,
      "loss": 0.0001,
      "reward": 0.5,
      "reward_std": 0.0,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.5,
      "step": 98
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 137.15625,
      "epoch": 0.396,
      "grad_norm": 5.509148120880127,
      "kl": 0.1878662109375,
      "learning_rate": 9.604e-07,
      "loss": 0.0002,
      "reward": 0.25,
      "reward_std": 0.2587745785713196,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.25,
      "step": 99
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 146.71875,
      "epoch": 0.4,
      "grad_norm": 9.132729530334473,
      "kl": 0.1285400390625,
      "learning_rate": 9.6e-07,
      "loss": 0.0001,
      "reward": 0.0625,
      "reward_std": 0.1767766922712326,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.0625,
      "step": 100
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 135.40625,
      "epoch": 0.404,
      "grad_norm": 4.046815872192383,
      "kl": 0.152587890625,
      "learning_rate": 9.595999999999999e-07,
      "loss": 0.0002,
      "reward": 0.125,
      "reward_std": 0.13363061845302582,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.125,
      "step": 101
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 143.25,
      "epoch": 0.408,
      "grad_norm": 4.168700218200684,
      "kl": 0.1993408203125,
      "learning_rate": 9.592e-07,
      "loss": 0.0002,
      "reward": 0.28125,
      "reward_std": 0.0883883461356163,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.28125,
      "step": 102
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 131.03125,
      "epoch": 0.412,
      "grad_norm": 0.014560648240149021,
      "kl": 0.1416015625,
      "learning_rate": 9.588e-07,
      "loss": 0.0001,
      "reward": 0.5,
      "reward_std": 0.0,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.5,
      "step": 103
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 140.71875,
      "epoch": 0.416,
      "grad_norm": 2.6636085510253906,
      "kl": 0.1611328125,
      "learning_rate": 9.584e-07,
      "loss": 0.0002,
      "reward": 0.46875,
      "reward_std": 0.0883883461356163,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.46875,
      "step": 104
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 138.375,
      "epoch": 0.42,
      "grad_norm": 7.546946048736572,
      "kl": 0.1434326171875,
      "learning_rate": 9.58e-07,
      "loss": 0.0001,
      "reward": 0.28125,
      "reward_std": 0.24511480331420898,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.28125,
      "step": 105
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 148.96875,
      "epoch": 0.424,
      "grad_norm": 5.07991361618042,
      "kl": 0.1376953125,
      "learning_rate": 9.576e-07,
      "loss": 0.0001,
      "reward": 0.03125,
      "reward_std": 0.0883883461356163,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.03125,
      "step": 106
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 157.375,
      "epoch": 0.428,
      "grad_norm": 2.303396463394165,
      "kl": 0.1376953125,
      "learning_rate": 9.572e-07,
      "loss": 0.0001,
      "reward": 0.3125,
      "reward_std": 0.1157275140285492,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.3125,
      "step": 107
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 152.40625,
      "epoch": 0.432,
      "grad_norm": 10.513725280761719,
      "kl": 0.217529296875,
      "learning_rate": 9.567999999999999e-07,
      "loss": 0.0002,
      "reward": 0.71875,
      "reward_std": 0.2651650384068489,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.71875,
      "step": 108
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 148.21875,
      "epoch": 0.436,
      "grad_norm": 2.932323932647705,
      "kl": 0.213623046875,
      "learning_rate": 9.564e-07,
      "loss": 0.0002,
      "reward": 0.46875,
      "reward_std": 0.0883883461356163,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.46875,
      "step": 109
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 144.3125,
      "epoch": 0.44,
      "grad_norm": 2.535456895828247,
      "kl": 0.150634765625,
      "learning_rate": 9.559999999999998e-07,
      "loss": 0.0002,
      "reward": 0.21875,
      "reward_std": 0.0883883461356163,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.21875,
      "step": 110
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 151.46875,
      "epoch": 0.444,
      "grad_norm": 14.733839988708496,
      "kl": 0.1661376953125,
      "learning_rate": 9.556e-07,
      "loss": 0.0002,
      "reward": 0.0625,
      "reward_std": 0.1767766922712326,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.0625,
      "step": 111
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 142.34375,
      "epoch": 0.448,
      "grad_norm": 0.015656903386116028,
      "kl": 0.1395263671875,
      "learning_rate": 9.552e-07,
      "loss": 0.0001,
      "reward": 0.25,
      "reward_std": 0.0,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.25,
      "step": 112
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 153.34375,
      "epoch": 0.452,
      "grad_norm": 0.022109676152467728,
      "kl": 0.1488037109375,
      "learning_rate": 9.548e-07,
      "loss": 0.0001,
      "reward": 0.75,
      "reward_std": 0.0,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.75,
      "step": 113
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 156.15625,
      "epoch": 0.456,
      "grad_norm": 2.3757565021514893,
      "kl": 0.2210693359375,
      "learning_rate": 9.544e-07,
      "loss": 0.0002,
      "reward": 0.28125,
      "reward_std": 0.0883883461356163,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.28125,
      "step": 114
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 167.625,
      "epoch": 0.46,
      "grad_norm": 4.812951564788818,
      "kl": 0.1298828125,
      "learning_rate": 9.539999999999999e-07,
      "loss": 0.0001,
      "reward": 0.5625,
      "reward_std": 0.1157275140285492,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.5625,
      "step": 115
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 154.96875,
      "epoch": 0.464,
      "grad_norm": 2.8628149032592773,
      "kl": 0.183349609375,
      "learning_rate": 9.536e-07,
      "loss": 0.0002,
      "reward": 0.0625,
      "reward_std": 0.1767766922712326,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.0625,
      "step": 116
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 168.78125,
      "epoch": 0.468,
      "grad_norm": 0.01167032215744257,
      "kl": 0.123046875,
      "learning_rate": 9.532e-07,
      "loss": 0.0001,
      "reward": 0.5,
      "reward_std": 0.0,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.5,
      "step": 117
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 164.46875,
      "epoch": 0.472,
      "grad_norm": 8.528116226196289,
      "kl": 0.150390625,
      "learning_rate": 9.527999999999999e-07,
      "loss": 0.0002,
      "reward": 0.09375,
      "reward_std": 0.2041158601641655,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.09375,
      "step": 118
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 173.25,
      "epoch": 0.476,
      "grad_norm": 3.113943338394165,
      "kl": 0.1400146484375,
      "learning_rate": 9.524e-07,
      "loss": 0.0001,
      "reward": 0.28125,
      "reward_std": 0.0883883461356163,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.28125,
      "step": 119
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 166.0,
      "epoch": 0.48,
      "grad_norm": 1.5100096464157104,
      "kl": 0.139404296875,
      "learning_rate": 9.52e-07,
      "loss": 0.0001,
      "reward": 0.21875,
      "reward_std": 0.0883883461356163,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.21875,
      "step": 120
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 157.34375,
      "epoch": 0.484,
      "grad_norm": 2.4216766357421875,
      "kl": 0.1324462890625,
      "learning_rate": 9.515999999999999e-07,
      "loss": 0.0001,
      "reward": 0.71875,
      "reward_std": 0.0883883461356163,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.71875,
      "step": 121
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 164.25,
      "epoch": 0.488,
      "grad_norm": 8.379853248596191,
      "kl": 0.146728515625,
      "learning_rate": 9.512e-07,
      "loss": 0.0001,
      "reward": 0.375,
      "reward_std": 0.3650856465101242,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.375,
      "step": 122
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 176.40625,
      "epoch": 0.492,
      "grad_norm": 21.827362060546875,
      "kl": 0.1123046875,
      "learning_rate": 9.508e-07,
      "loss": 0.0001,
      "reward": 0.5,
      "reward_std": 0.1767766922712326,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.5,
      "step": 123
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 165.84375,
      "epoch": 0.496,
      "grad_norm": 3.238415479660034,
      "kl": 0.10888671875,
      "learning_rate": 9.503999999999999e-07,
      "loss": 0.0001,
      "reward": 0.09375,
      "reward_std": 0.1293872892856598,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.09375,
      "step": 124
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 176.3125,
      "epoch": 0.5,
      "grad_norm": 2.2066924571990967,
      "kl": 0.12939453125,
      "learning_rate": 9.499999999999999e-07,
      "loss": 0.0001,
      "reward": 0.1875,
      "reward_std": 0.1157275140285492,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.1875,
      "step": 125
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 163.25,
      "epoch": 0.504,
      "grad_norm": 7.6757307052612305,
      "kl": 0.176025390625,
      "learning_rate": 9.496e-07,
      "loss": 0.0002,
      "reward": 0.15625,
      "reward_std": 0.1293872892856598,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.15625,
      "step": 126
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 170.125,
      "epoch": 0.508,
      "grad_norm": 3.0611886978149414,
      "kl": 0.1181640625,
      "learning_rate": 9.492e-07,
      "loss": 0.0001,
      "reward": 0.375,
      "reward_std": 0.2177756354212761,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.375,
      "step": 127
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 160.15625,
      "epoch": 0.512,
      "grad_norm": 0.009313968941569328,
      "kl": 0.11865234375,
      "learning_rate": 9.487999999999999e-07,
      "loss": 0.0001,
      "reward": 0.25,
      "reward_std": 0.0,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.25,
      "step": 128
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 159.03125,
      "epoch": 0.516,
      "grad_norm": 1.8127228021621704,
      "kl": 0.1759033203125,
      "learning_rate": 9.484e-07,
      "loss": 0.0002,
      "reward": 0.71875,
      "reward_std": 0.0883883461356163,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.71875,
      "step": 129
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 132.25,
      "epoch": 0.52,
      "grad_norm": 2.159052610397339,
      "kl": 0.1309814453125,
      "learning_rate": 9.479999999999999e-07,
      "loss": 0.0001,
      "reward": 0.28125,
      "reward_std": 0.0883883461356163,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.28125,
      "step": 130
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 149.84375,
      "epoch": 0.524,
      "grad_norm": 5.0523552894592285,
      "kl": 0.15087890625,
      "learning_rate": 9.475999999999999e-07,
      "loss": 0.0002,
      "reward": 0.21875,
      "reward_std": 0.24511480331420898,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.21875,
      "step": 131
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 146.65625,
      "epoch": 0.528,
      "grad_norm": 3.5751190185546875,
      "kl": 0.1422119140625,
      "learning_rate": 9.472e-07,
      "loss": 0.0001,
      "reward": 0.03125,
      "reward_std": 0.0883883461356163,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.03125,
      "step": 132
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 138.75,
      "epoch": 0.532,
      "grad_norm": 7.118408203125,
      "kl": 0.1361083984375,
      "learning_rate": 9.468e-07,
      "loss": 0.0001,
      "reward": 0.46875,
      "reward_std": 0.2041158601641655,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.46875,
      "step": 133
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 142.90625,
      "epoch": 0.536,
      "grad_norm": 7.05600118637085,
      "kl": 0.1453857421875,
      "learning_rate": 9.464e-07,
      "loss": 0.0001,
      "reward": 0.03125,
      "reward_std": 0.0883883461356163,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.03125,
      "step": 134
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 146.53125,
      "epoch": 0.54,
      "grad_norm": 14.140984535217285,
      "kl": 0.24853515625,
      "learning_rate": 9.459999999999999e-07,
      "loss": 0.0002,
      "reward": 0.28125,
      "reward_std": 0.24511480331420898,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.28125,
      "step": 135
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 140.25,
      "epoch": 0.544,
      "grad_norm": 10.306221961975098,
      "kl": 0.1552734375,
      "learning_rate": 9.456e-07,
      "loss": 0.0002,
      "reward": 0.25,
      "reward_std": 0.1767766922712326,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.25,
      "step": 136
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 149.5,
      "epoch": 0.548,
      "grad_norm": 10.985379219055176,
      "kl": 0.150634765625,
      "learning_rate": 9.452e-07,
      "loss": 0.0002,
      "reward": 0.25,
      "reward_std": 0.3514062538743019,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.25,
      "step": 137
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 143.25,
      "epoch": 0.552,
      "grad_norm": 5.7292022705078125,
      "kl": 0.164306640625,
      "learning_rate": 9.447999999999999e-07,
      "loss": 0.0002,
      "reward": 0.15625,
      "reward_std": 0.3061639815568924,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.15625,
      "step": 138
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 152.625,
      "epoch": 0.556,
      "grad_norm": 8.641803741455078,
      "kl": 0.117431640625,
      "learning_rate": 9.444e-07,
      "loss": 0.0001,
      "reward": 0.28125,
      "reward_std": 0.24511480331420898,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.28125,
      "step": 139
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 140.40625,
      "epoch": 0.56,
      "grad_norm": 3.8259010314941406,
      "kl": 0.18310546875,
      "learning_rate": 9.439999999999999e-07,
      "loss": 0.0002,
      "reward": 0.34375,
      "reward_std": 0.2041158601641655,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.34375,
      "step": 140
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 142.0,
      "epoch": 0.564,
      "grad_norm": 0.026461126282811165,
      "kl": 0.150634765625,
      "learning_rate": 9.436e-07,
      "loss": 0.0002,
      "reward": 0.25,
      "reward_std": 0.0,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.25,
      "step": 141
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 130.40625,
      "epoch": 0.568,
      "grad_norm": 20.714466094970703,
      "kl": 0.164306640625,
      "learning_rate": 9.432e-07,
      "loss": 0.0002,
      "reward": 0.75,
      "reward_std": 0.3335031494498253,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.75,
      "step": 142
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 146.25,
      "epoch": 0.572,
      "grad_norm": 2.2029976844787598,
      "kl": 0.2403564453125,
      "learning_rate": 9.427999999999999e-07,
      "loss": 0.0002,
      "reward": 0.03125,
      "reward_std": 0.0883883461356163,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.03125,
      "step": 143
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 157.15625,
      "epoch": 0.576,
      "grad_norm": 4.797538757324219,
      "kl": 0.125,
      "learning_rate": 9.424e-07,
      "loss": 0.0001,
      "reward": 0.03125,
      "reward_std": 0.0883883461356163,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.03125,
      "step": 144
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 141.0625,
      "epoch": 0.58,
      "grad_norm": 4.411031723022461,
      "kl": 0.14599609375,
      "learning_rate": 9.419999999999999e-07,
      "loss": 0.0001,
      "reward": 0.21875,
      "reward_std": 0.0883883461356163,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.21875,
      "step": 145
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 149.6875,
      "epoch": 0.584,
      "grad_norm": 13.82177448272705,
      "kl": 0.1361083984375,
      "learning_rate": 9.415999999999999e-07,
      "loss": 0.0001,
      "reward": 0.46875,
      "reward_std": 0.2041158601641655,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.46875,
      "step": 146
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 143.125,
      "epoch": 0.588,
      "grad_norm": 3.6890528202056885,
      "kl": 0.1322021484375,
      "learning_rate": 9.412e-07,
      "loss": 0.0001,
      "reward": 0.0625,
      "reward_std": 0.1157275140285492,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.0625,
      "step": 147
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 156.0,
      "epoch": 0.592,
      "grad_norm": 7.011545181274414,
      "kl": 0.153564453125,
      "learning_rate": 9.408e-07,
      "loss": 0.0002,
      "reward": 0.5,
      "reward_std": 0.26726123690605164,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.5,
      "step": 148
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 153.5625,
      "epoch": 0.596,
      "grad_norm": 7.351828098297119,
      "kl": 0.1778564453125,
      "learning_rate": 9.403999999999999e-07,
      "loss": 0.0002,
      "reward": 0.34375,
      "reward_std": 0.22201896458864212,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.34375,
      "step": 149
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 142.5625,
      "epoch": 0.6,
      "grad_norm": 3.5653135776519775,
      "kl": 0.1507568359375,
      "learning_rate": 9.399999999999999e-07,
      "loss": 0.0002,
      "reward": 0.5,
      "reward_std": 0.1767766922712326,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.5,
      "step": 150
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 135.15625,
      "epoch": 0.604,
      "grad_norm": 2.1044363975524902,
      "kl": 0.168701171875,
      "learning_rate": 9.396e-07,
      "loss": 0.0002,
      "reward": 0.46875,
      "reward_std": 0.0883883461356163,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.46875,
      "step": 151
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 153.0,
      "epoch": 0.608,
      "grad_norm": 5.389544486999512,
      "kl": 0.1319580078125,
      "learning_rate": 9.391999999999999e-07,
      "loss": 0.0001,
      "reward": 0.25,
      "reward_std": 0.2587745785713196,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.25,
      "step": 152
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 143.625,
      "epoch": 0.612,
      "grad_norm": 7.576977729797363,
      "kl": 0.154541015625,
      "learning_rate": 9.387999999999999e-07,
      "loss": 0.0002,
      "reward": 0.21875,
      "reward_std": 0.24511480331420898,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.21875,
      "step": 153
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 154.96875,
      "epoch": 0.616,
      "grad_norm": 17.224998474121094,
      "kl": 0.1458740234375,
      "learning_rate": 9.384e-07,
      "loss": 0.0001,
      "reward": 0.4375,
      "reward_std": 0.1767766922712326,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.4375,
      "step": 154
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 130.90625,
      "epoch": 0.62,
      "grad_norm": 6.7807297706604,
      "kl": 0.1806640625,
      "learning_rate": 9.379999999999998e-07,
      "loss": 0.0002,
      "reward": 0.28125,
      "reward_std": 0.0883883461356163,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.28125,
      "step": 155
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 138.875,
      "epoch": 0.624,
      "grad_norm": 24.668161392211914,
      "kl": 0.158935546875,
      "learning_rate": 9.375999999999999e-07,
      "loss": 0.0002,
      "reward": 0.25,
      "reward_std": 0.3650856465101242,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.25,
      "step": 156
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 154.71875,
      "epoch": 0.628,
      "grad_norm": 3.2327768802642822,
      "kl": 0.153076171875,
      "learning_rate": 9.372e-07,
      "loss": 0.0002,
      "reward": 0.0625,
      "reward_std": 0.1157275140285492,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.0625,
      "step": 157
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 150.53125,
      "epoch": 0.632,
      "grad_norm": 4.513970851898193,
      "kl": 0.17236328125,
      "learning_rate": 9.368e-07,
      "loss": 0.0002,
      "reward": 0.3125,
      "reward_std": 0.1767766922712326,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.3125,
      "step": 158
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 125.9375,
      "epoch": 0.636,
      "grad_norm": 0.025679027661681175,
      "kl": 0.18505859375,
      "learning_rate": 9.363999999999999e-07,
      "loss": 0.0002,
      "reward": 0.25,
      "reward_std": 0.0,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.25,
      "step": 159
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 130.125,
      "epoch": 0.64,
      "grad_norm": 2.8930089473724365,
      "kl": 0.193359375,
      "learning_rate": 9.36e-07,
      "loss": 0.0002,
      "reward": 0.3125,
      "reward_std": 0.1157275140285492,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.3125,
      "step": 160
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 156.65625,
      "epoch": 0.644,
      "grad_norm": 4.403336048126221,
      "kl": 0.16552734375,
      "learning_rate": 9.356e-07,
      "loss": 0.0002,
      "reward": 0.28125,
      "reward_std": 0.2041158601641655,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.28125,
      "step": 161
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 151.75,
      "epoch": 0.648,
      "grad_norm": 9.439102172851562,
      "kl": 0.152587890625,
      "learning_rate": 9.352e-07,
      "loss": 0.0002,
      "reward": 0.4375,
      "reward_std": 0.2925042062997818,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.4375,
      "step": 162
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 128.6875,
      "epoch": 0.652,
      "grad_norm": 0.018322864547371864,
      "kl": 0.18701171875,
      "learning_rate": 9.347999999999999e-07,
      "loss": 0.0002,
      "reward": 0.5,
      "reward_std": 0.0,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.5,
      "step": 163
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 144.375,
      "epoch": 0.656,
      "grad_norm": 3.9991912841796875,
      "kl": 0.1630859375,
      "learning_rate": 9.344e-07,
      "loss": 0.0002,
      "reward": 0.21875,
      "reward_std": 0.2041158601641655,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.21875,
      "step": 164
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 140.125,
      "epoch": 0.66,
      "grad_norm": 9.71346378326416,
      "kl": 0.188720703125,
      "learning_rate": 9.34e-07,
      "loss": 0.0002,
      "reward": 0.5625,
      "reward_std": 0.2177756354212761,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.5625,
      "step": 165
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 158.96875,
      "epoch": 0.664,
      "grad_norm": 4.799767971038818,
      "kl": 0.1461181640625,
      "learning_rate": 9.335999999999999e-07,
      "loss": 0.0001,
      "reward": 0.34375,
      "reward_std": 0.1293872892856598,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.34375,
      "step": 166
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 160.78125,
      "epoch": 0.668,
      "grad_norm": 2.5199387073516846,
      "kl": 0.1396484375,
      "learning_rate": 9.332e-07,
      "loss": 0.0001,
      "reward": 0.46875,
      "reward_std": 0.0883883461356163,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.46875,
      "step": 167
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 158.875,
      "epoch": 0.672,
      "grad_norm": 6.527361869812012,
      "kl": 0.192138671875,
      "learning_rate": 9.327999999999999e-07,
      "loss": 0.0002,
      "reward": 0.125,
      "reward_std": 0.2177756354212761,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.125,
      "step": 168
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 149.96875,
      "epoch": 0.676,
      "grad_norm": 2.801525592803955,
      "kl": 0.2891845703125,
      "learning_rate": 9.324e-07,
      "loss": 0.0003,
      "reward": 0.46875,
      "reward_std": 0.0883883461356163,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.46875,
      "step": 169
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 149.53125,
      "epoch": 0.68,
      "grad_norm": 8.147767066955566,
      "kl": 0.18017578125,
      "learning_rate": 9.32e-07,
      "loss": 0.0002,
      "reward": 0.3125,
      "reward_std": 0.2177756354212761,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.3125,
      "step": 170
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 165.625,
      "epoch": 0.684,
      "grad_norm": 4.073009967803955,
      "kl": 0.20751953125,
      "learning_rate": 9.315999999999999e-07,
      "loss": 0.0002,
      "reward": 0.25,
      "reward_std": 0.3335031494498253,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.25,
      "step": 171
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 165.25,
      "epoch": 0.688,
      "grad_norm": 6.553722858428955,
      "kl": 0.1534423828125,
      "learning_rate": 9.312e-07,
      "loss": 0.0002,
      "reward": 0.3125,
      "reward_std": 0.2177756354212761,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.3125,
      "step": 172
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 174.40625,
      "epoch": 0.692,
      "grad_norm": 4.628172874450684,
      "kl": 0.14404296875,
      "learning_rate": 9.307999999999999e-07,
      "loss": 0.0001,
      "reward": 0.03125,
      "reward_std": 0.0883883461356163,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.03125,
      "step": 173
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 173.96875,
      "epoch": 0.696,
      "grad_norm": 0.014567671343684196,
      "kl": 0.133544921875,
      "learning_rate": 9.303999999999999e-07,
      "loss": 0.0001,
      "reward": 0.0,
      "reward_std": 0.0,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.0,
      "step": 174
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 144.1875,
      "epoch": 0.7,
      "grad_norm": 0.012290208600461483,
      "kl": 0.155517578125,
      "learning_rate": 9.3e-07,
      "loss": 0.0002,
      "reward": 0.75,
      "reward_std": 0.0,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.75,
      "step": 175
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 152.25,
      "epoch": 0.704,
      "grad_norm": 4.443714141845703,
      "kl": 0.1324462890625,
      "learning_rate": 9.296e-07,
      "loss": 0.0001,
      "reward": 0.46875,
      "reward_std": 0.2041158601641655,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.46875,
      "step": 176
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 155.40625,
      "epoch": 0.708,
      "grad_norm": 0.017265308648347855,
      "kl": 0.1400146484375,
      "learning_rate": 9.292e-07,
      "loss": 0.0001,
      "reward": 0.0,
      "reward_std": 0.0,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.0,
      "step": 177
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 150.0625,
      "epoch": 0.712,
      "grad_norm": 2.628682851791382,
      "kl": 0.1719970703125,
      "learning_rate": 9.287999999999999e-07,
      "loss": 0.0002,
      "reward": 0.3125,
      "reward_std": 0.1157275140285492,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.3125,
      "step": 178
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 162.46875,
      "epoch": 0.716,
      "grad_norm": 6.353070259094238,
      "kl": 0.14990234375,
      "learning_rate": 9.284e-07,
      "loss": 0.0001,
      "reward": 0.84375,
      "reward_std": 0.3061639815568924,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.84375,
      "step": 179
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 133.40625,
      "epoch": 0.72,
      "grad_norm": 8.15472412109375,
      "kl": 0.16064453125,
      "learning_rate": 9.28e-07,
      "loss": 0.0002,
      "reward": 0.46875,
      "reward_std": 0.0883883461356163,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.46875,
      "step": 180
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 138.71875,
      "epoch": 0.724,
      "grad_norm": 5.264553070068359,
      "kl": 0.15869140625,
      "learning_rate": 9.275999999999999e-07,
      "loss": 0.0002,
      "reward": 0.1875,
      "reward_std": 0.2177756354212761,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.1875,
      "step": 181
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 150.8125,
      "epoch": 0.728,
      "grad_norm": 7.217442989349365,
      "kl": 0.15283203125,
      "learning_rate": 9.272e-07,
      "loss": 0.0002,
      "reward": 0.0625,
      "reward_std": 0.1157275140285492,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.0625,
      "step": 182
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 157.875,
      "epoch": 0.732,
      "grad_norm": 3.5819404125213623,
      "kl": 0.167236328125,
      "learning_rate": 9.268e-07,
      "loss": 0.0002,
      "reward": 0.53125,
      "reward_std": 0.2041158601641655,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.53125,
      "step": 183
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 152.53125,
      "epoch": 0.736,
      "grad_norm": 20.323453903198242,
      "kl": 0.172119140625,
      "learning_rate": 9.263999999999999e-07,
      "loss": 0.0002,
      "reward": 0.21875,
      "reward_std": 0.0883883461356163,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.21875,
      "step": 184
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 144.90625,
      "epoch": 0.74,
      "grad_norm": 0.01625896245241165,
      "kl": 0.1708984375,
      "learning_rate": 9.26e-07,
      "loss": 0.0002,
      "reward": 0.25,
      "reward_std": 0.0,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.25,
      "step": 185
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 142.6875,
      "epoch": 0.744,
      "grad_norm": 7.118624687194824,
      "kl": 0.15478515625,
      "learning_rate": 9.256e-07,
      "loss": 0.0002,
      "reward": 0.53125,
      "reward_std": 0.0883883461356163,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.53125,
      "step": 186
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 150.0,
      "epoch": 0.748,
      "grad_norm": 0.018267452716827393,
      "kl": 0.157470703125,
      "learning_rate": 9.251999999999999e-07,
      "loss": 0.0002,
      "reward": 0.75,
      "reward_std": 0.0,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.75,
      "step": 187
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 156.25,
      "epoch": 0.752,
      "grad_norm": 4.424862861633301,
      "kl": 0.1611328125,
      "learning_rate": 9.247999999999999e-07,
      "loss": 0.0002,
      "reward": 0.53125,
      "reward_std": 0.0883883461356163,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.53125,
      "step": 188
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 145.34375,
      "epoch": 0.756,
      "grad_norm": 5.989232063293457,
      "kl": 0.29541015625,
      "learning_rate": 9.244e-07,
      "loss": 0.0003,
      "reward": 0.53125,
      "reward_std": 0.2630179077386856,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.53125,
      "step": 189
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 147.0625,
      "epoch": 0.76,
      "grad_norm": 5.551213264465332,
      "kl": 0.17138671875,
      "learning_rate": 9.24e-07,
      "loss": 0.0002,
      "reward": 0.5625,
      "reward_std": 0.2177756354212761,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.5625,
      "step": 190
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 139.84375,
      "epoch": 0.764,
      "grad_norm": 3.024784803390503,
      "kl": 0.162841796875,
      "learning_rate": 9.235999999999999e-07,
      "loss": 0.0002,
      "reward": 0.4375,
      "reward_std": 0.1157275140285492,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.4375,
      "step": 191
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 164.21875,
      "epoch": 0.768,
      "grad_norm": 2.841665744781494,
      "kl": 0.18115234375,
      "learning_rate": 9.232e-07,
      "loss": 0.0002,
      "reward": 0.21875,
      "reward_std": 0.0883883461356163,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.21875,
      "step": 192
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 144.15625,
      "epoch": 0.772,
      "grad_norm": 0.2720145285129547,
      "kl": 0.3896484375,
      "learning_rate": 9.227999999999999e-07,
      "loss": 0.0004,
      "reward": 0.25,
      "reward_std": 0.0,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.25,
      "step": 193
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 134.84375,
      "epoch": 0.776,
      "grad_norm": 13.293619155883789,
      "kl": 0.171630859375,
      "learning_rate": 9.224e-07,
      "loss": 0.0002,
      "reward": 0.375,
      "reward_std": 0.2177756354212761,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.375,
      "step": 194
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 146.0,
      "epoch": 0.78,
      "grad_norm": 4.161295413970947,
      "kl": 0.1806640625,
      "learning_rate": 9.22e-07,
      "loss": 0.0002,
      "reward": 0.0625,
      "reward_std": 0.1767766922712326,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.0625,
      "step": 195
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 126.53125,
      "epoch": 0.784,
      "grad_norm": 2.5194199085235596,
      "kl": 0.1943359375,
      "learning_rate": 9.215999999999999e-07,
      "loss": 0.0002,
      "reward": 0.09375,
      "reward_std": 0.1293872892856598,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.09375,
      "step": 196
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 155.9375,
      "epoch": 0.788,
      "grad_norm": 0.013363498263061047,
      "kl": 0.154052734375,
      "learning_rate": 9.212e-07,
      "loss": 0.0002,
      "reward": 0.0,
      "reward_std": 0.0,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.0,
      "step": 197
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 148.5,
      "epoch": 0.792,
      "grad_norm": 0.011504505760967731,
      "kl": 0.169921875,
      "learning_rate": 9.207999999999999e-07,
      "loss": 0.0002,
      "reward": 0.0,
      "reward_std": 0.0,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.0,
      "step": 198
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 148.34375,
      "epoch": 0.796,
      "grad_norm": 3.968836545944214,
      "kl": 0.1748046875,
      "learning_rate": 9.203999999999999e-07,
      "loss": 0.0002,
      "reward": 0.1875,
      "reward_std": 0.1157275140285492,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.1875,
      "step": 199
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 148.65625,
      "epoch": 0.8,
      "grad_norm": 0.018608927726745605,
      "kl": 0.1884765625,
      "learning_rate": 9.2e-07,
      "loss": 0.0002,
      "reward": 0.0,
      "reward_std": 0.0,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.0,
      "step": 200
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 148.78125,
      "epoch": 0.804,
      "grad_norm": 7.09287166595459,
      "kl": 0.186767578125,
      "learning_rate": 9.196e-07,
      "loss": 0.0002,
      "reward": 0.0625,
      "reward_std": 0.1767766922712326,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.0625,
      "step": 201
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 127.8125,
      "epoch": 0.808,
      "grad_norm": 0.012347673065960407,
      "kl": 0.1826171875,
      "learning_rate": 9.192e-07,
      "loss": 0.0002,
      "reward": 0.0,
      "reward_std": 0.0,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.0,
      "step": 202
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 134.8125,
      "epoch": 0.812,
      "grad_norm": 5.15930700302124,
      "kl": 0.170654296875,
      "learning_rate": 9.187999999999999e-07,
      "loss": 0.0002,
      "reward": 0.1875,
      "reward_std": 0.2177756354212761,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.1875,
      "step": 203
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 131.0,
      "epoch": 0.816,
      "grad_norm": 6.69101619720459,
      "kl": 0.175048828125,
      "learning_rate": 9.184e-07,
      "loss": 0.0002,
      "reward": 0.1875,
      "reward_std": 0.2177756354212761,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.1875,
      "step": 204
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 143.4375,
      "epoch": 0.82,
      "grad_norm": 11.913467407226562,
      "kl": 0.16259765625,
      "learning_rate": 9.18e-07,
      "loss": 0.0002,
      "reward": 0.53125,
      "reward_std": 0.2651650384068489,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.53125,
      "step": 205
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 140.6875,
      "epoch": 0.824,
      "grad_norm": 0.013337607495486736,
      "kl": 0.1787109375,
      "learning_rate": 9.175999999999999e-07,
      "loss": 0.0002,
      "reward": 0.25,
      "reward_std": 0.0,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.25,
      "step": 206
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 138.875,
      "epoch": 0.828,
      "grad_norm": 3.1044652462005615,
      "kl": 0.192138671875,
      "learning_rate": 9.172e-07,
      "loss": 0.0002,
      "reward": 0.03125,
      "reward_std": 0.0883883461356163,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.03125,
      "step": 207
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 136.90625,
      "epoch": 0.832,
      "grad_norm": 4.334849834442139,
      "kl": 0.159912109375,
      "learning_rate": 9.168e-07,
      "loss": 0.0002,
      "reward": 0.5,
      "reward_std": 0.1767766922712326,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.5,
      "step": 208
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 138.84375,
      "epoch": 0.836,
      "grad_norm": 0.02636542171239853,
      "kl": 0.161865234375,
      "learning_rate": 9.163999999999999e-07,
      "loss": 0.0002,
      "reward": 0.25,
      "reward_std": 0.0,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.25,
      "step": 209
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 125.65625,
      "epoch": 0.84,
      "grad_norm": 22.57999038696289,
      "kl": 0.178466796875,
      "learning_rate": 9.16e-07,
      "loss": 0.0002,
      "reward": 0.4375,
      "reward_std": 0.2177756354212761,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.4375,
      "step": 210
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 138.4375,
      "epoch": 0.844,
      "grad_norm": 13.246150016784668,
      "kl": 0.298583984375,
      "learning_rate": 9.156e-07,
      "loss": 0.0003,
      "reward": 0.15625,
      "reward_std": 0.22201896458864212,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.15625,
      "step": 211
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 151.25,
      "epoch": 0.848,
      "grad_norm": 6.394719123840332,
      "kl": 0.1474609375,
      "learning_rate": 9.151999999999999e-07,
      "loss": 0.0001,
      "reward": 0.0625,
      "reward_std": 0.1767766922712326,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.0625,
      "step": 212
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 135.9375,
      "epoch": 0.852,
      "grad_norm": 4.514334201812744,
      "kl": 0.180419921875,
      "learning_rate": 9.147999999999999e-07,
      "loss": 0.0002,
      "reward": 0.5,
      "reward_std": 0.3335031494498253,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.5,
      "step": 213
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 158.84375,
      "epoch": 0.856,
      "grad_norm": 4.87898588180542,
      "kl": 0.2811279296875,
      "learning_rate": 9.144e-07,
      "loss": 0.0003,
      "reward": 0.09375,
      "reward_std": 0.1293872892856598,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.09375,
      "step": 214
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 161.3125,
      "epoch": 0.86,
      "grad_norm": 4.790887355804443,
      "kl": 0.15283203125,
      "learning_rate": 9.14e-07,
      "loss": 0.0002,
      "reward": 0.40625,
      "reward_std": 0.3198433741927147,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.40625,
      "step": 215
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 148.53125,
      "epoch": 0.864,
      "grad_norm": 9.748350143432617,
      "kl": 0.176025390625,
      "learning_rate": 9.135999999999999e-07,
      "loss": 0.0002,
      "reward": 0.40625,
      "reward_std": 0.3198433741927147,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.40625,
      "step": 216
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 148.21875,
      "epoch": 0.868,
      "grad_norm": 3.753589391708374,
      "kl": 0.16259765625,
      "learning_rate": 9.132e-07,
      "loss": 0.0002,
      "reward": 0.28125,
      "reward_std": 0.2630179077386856,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.28125,
      "step": 217
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 132.125,
      "epoch": 0.872,
      "grad_norm": 5.923015594482422,
      "kl": 0.17041015625,
      "learning_rate": 9.127999999999999e-07,
      "loss": 0.0002,
      "reward": 0.375,
      "reward_std": 0.3650856465101242,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.375,
      "step": 218
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 133.28125,
      "epoch": 0.876,
      "grad_norm": 5.67153787612915,
      "kl": 0.1689453125,
      "learning_rate": 9.123999999999999e-07,
      "loss": 0.0002,
      "reward": 0.03125,
      "reward_std": 0.0883883461356163,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.03125,
      "step": 219
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 140.90625,
      "epoch": 0.88,
      "grad_norm": 2.6040894985198975,
      "kl": 0.15185546875,
      "learning_rate": 9.12e-07,
      "loss": 0.0002,
      "reward": 0.1875,
      "reward_std": 0.1157275140285492,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.1875,
      "step": 220
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 134.71875,
      "epoch": 0.884,
      "grad_norm": 4.3682379722595215,
      "kl": 0.16845703125,
      "learning_rate": 9.115999999999999e-07,
      "loss": 0.0002,
      "reward": 0.40625,
      "reward_std": 0.22201896458864212,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.40625,
      "step": 221
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 149.0,
      "epoch": 0.888,
      "grad_norm": 7.323733806610107,
      "kl": 0.158203125,
      "learning_rate": 9.112e-07,
      "loss": 0.0002,
      "reward": 0.5625,
      "reward_std": 0.4355512708425522,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.5625,
      "step": 222
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 133.3125,
      "epoch": 0.892,
      "grad_norm": 8.492609977722168,
      "kl": 0.1689453125,
      "learning_rate": 9.108e-07,
      "loss": 0.0002,
      "reward": 0.5,
      "reward_std": 0.3745020925998688,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.5,
      "step": 223
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 124.625,
      "epoch": 0.896,
      "grad_norm": 4.2470622062683105,
      "kl": 0.300537109375,
      "learning_rate": 9.103999999999999e-07,
      "loss": 0.0003,
      "reward": 0.3125,
      "reward_std": 0.1157275140285492,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.3125,
      "step": 224
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 117.75,
      "epoch": 0.9,
      "grad_norm": 2.365232467651367,
      "kl": 0.175048828125,
      "learning_rate": 9.1e-07,
      "loss": 0.0002,
      "reward": 0.3125,
      "reward_std": 0.1157275140285492,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.3125,
      "step": 225
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 126.9375,
      "epoch": 0.904,
      "grad_norm": 5.005716323852539,
      "kl": 0.184814453125,
      "learning_rate": 9.095999999999999e-07,
      "loss": 0.0002,
      "reward": 0.15625,
      "reward_std": 0.22201896458864212,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.15625,
      "step": 226
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 145.34375,
      "epoch": 0.908,
      "grad_norm": 3.4398090839385986,
      "kl": 0.403076171875,
      "learning_rate": 9.092e-07,
      "loss": 0.0004,
      "reward": 0.28125,
      "reward_std": 0.0883883461356163,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.28125,
      "step": 227
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 126.3125,
      "epoch": 0.912,
      "grad_norm": 5.781132698059082,
      "kl": 0.185546875,
      "learning_rate": 9.088e-07,
      "loss": 0.0002,
      "reward": 0.1875,
      "reward_std": 0.3335031494498253,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.1875,
      "step": 228
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 121.375,
      "epoch": 0.916,
      "grad_norm": 8.078886032104492,
      "kl": 0.184326171875,
      "learning_rate": 9.084e-07,
      "loss": 0.0002,
      "reward": 0.53125,
      "reward_std": 0.3471629247069359,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.53125,
      "step": 229
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 112.1875,
      "epoch": 0.92,
      "grad_norm": 17.53875732421875,
      "kl": 0.181640625,
      "learning_rate": 9.08e-07,
      "loss": 0.0002,
      "reward": 0.5,
      "reward_std": 0.2925042062997818,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.5,
      "step": 230
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 145.3125,
      "epoch": 0.924,
      "grad_norm": 6.732425212860107,
      "kl": 0.180419921875,
      "learning_rate": 9.075999999999999e-07,
      "loss": 0.0002,
      "reward": 0.21875,
      "reward_std": 0.3377464786171913,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.21875,
      "step": 231
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 131.15625,
      "epoch": 0.928,
      "grad_norm": 5.378574371337891,
      "kl": 0.183349609375,
      "learning_rate": 9.072e-07,
      "loss": 0.0002,
      "reward": 0.40625,
      "reward_std": 0.22201896458864212,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.40625,
      "step": 232
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 126.34375,
      "epoch": 0.932,
      "grad_norm": 8.546733856201172,
      "kl": 0.246337890625,
      "learning_rate": 9.068e-07,
      "loss": 0.0002,
      "reward": 0.40625,
      "reward_std": 0.5217924863100052,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.40625,
      "step": 233
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 119.28125,
      "epoch": 0.936,
      "grad_norm": 4.824804782867432,
      "kl": 0.21923828125,
      "learning_rate": 9.063999999999999e-07,
      "loss": 0.0002,
      "reward": 0.375,
      "reward_std": 0.2177756354212761,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.375,
      "step": 234
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 129.71875,
      "epoch": 0.94,
      "grad_norm": 4.122939109802246,
      "kl": 0.18359375,
      "learning_rate": 9.06e-07,
      "loss": 0.0002,
      "reward": 0.09375,
      "reward_std": 0.2041158601641655,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.09375,
      "step": 235
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 135.875,
      "epoch": 0.944,
      "grad_norm": 2.8131566047668457,
      "kl": 0.26416015625,
      "learning_rate": 9.056e-07,
      "loss": 0.0003,
      "reward": 0.03125,
      "reward_std": 0.0883883461356163,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.03125,
      "step": 236
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 139.0,
      "epoch": 0.948,
      "grad_norm": 2.190013885498047,
      "kl": 0.18359375,
      "learning_rate": 9.051999999999999e-07,
      "loss": 0.0002,
      "reward": 0.46875,
      "reward_std": 0.0883883461356163,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.46875,
      "step": 237
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 157.4375,
      "epoch": 0.952,
      "grad_norm": 3.7834203243255615,
      "kl": 0.1572265625,
      "learning_rate": 9.048e-07,
      "loss": 0.0002,
      "reward": 0.0625,
      "reward_std": 0.1157275140285492,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.0625,
      "step": 238
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 133.78125,
      "epoch": 0.956,
      "grad_norm": 32.57776641845703,
      "kl": 0.179443359375,
      "learning_rate": 9.044e-07,
      "loss": 0.0002,
      "reward": 0.40625,
      "reward_std": 0.5123760402202606,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.40625,
      "step": 239
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 119.53125,
      "epoch": 0.96,
      "grad_norm": 3.308706045150757,
      "kl": 0.210693359375,
      "learning_rate": 9.039999999999999e-07,
      "loss": 0.0002,
      "reward": 0.09375,
      "reward_std": 0.1293872892856598,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.09375,
      "step": 240
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 120.59375,
      "epoch": 0.964,
      "grad_norm": 6.456974506378174,
      "kl": 0.187255859375,
      "learning_rate": 9.035999999999999e-07,
      "loss": 0.0002,
      "reward": 0.40625,
      "reward_std": 0.3377464786171913,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.40625,
      "step": 241
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 140.21875,
      "epoch": 0.968,
      "grad_norm": 13.71239185333252,
      "kl": 0.162353515625,
      "learning_rate": 9.032e-07,
      "loss": 0.0002,
      "reward": 0.1875,
      "reward_std": 0.3104073107242584,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.1875,
      "step": 242
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 154.0625,
      "epoch": 0.972,
      "grad_norm": 10.391451835632324,
      "kl": 0.160888671875,
      "learning_rate": 9.028e-07,
      "loss": 0.0002,
      "reward": 0.0625,
      "reward_std": 0.1767766922712326,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.0625,
      "step": 243
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 133.5,
      "epoch": 0.976,
      "grad_norm": 5.7491044998168945,
      "kl": 0.192138671875,
      "learning_rate": 9.023999999999999e-07,
      "loss": 0.0002,
      "reward": 0.28125,
      "reward_std": 0.2651650384068489,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.28125,
      "step": 244
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 144.59375,
      "epoch": 0.98,
      "grad_norm": 6.993992328643799,
      "kl": 0.16796875,
      "learning_rate": 9.02e-07,
      "loss": 0.0002,
      "reward": 0.03125,
      "reward_std": 0.0883883461356163,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.03125,
      "step": 245
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 140.78125,
      "epoch": 0.984,
      "grad_norm": 11.08362102508545,
      "kl": 0.192626953125,
      "learning_rate": 9.015999999999999e-07,
      "loss": 0.0002,
      "reward": 0.40625,
      "reward_std": 0.22201896458864212,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.40625,
      "step": 246
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 149.75,
      "epoch": 0.988,
      "grad_norm": 35.28593826293945,
      "kl": 0.166015625,
      "learning_rate": 9.011999999999999e-07,
      "loss": 0.0002,
      "reward": 0.375,
      "reward_std": 0.4492306634783745,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.375,
      "step": 247
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 142.125,
      "epoch": 0.992,
      "grad_norm": 8.120216369628906,
      "kl": 0.1826171875,
      "learning_rate": 9.008e-07,
      "loss": 0.0002,
      "reward": 0.21875,
      "reward_std": 0.24511480331420898,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.21875,
      "step": 248
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 129.09375,
      "epoch": 0.996,
      "grad_norm": 7.45528507232666,
      "kl": 0.2109375,
      "learning_rate": 9.004e-07,
      "loss": 0.0002,
      "reward": 0.03125,
      "reward_std": 0.0883883461356163,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.03125,
      "step": 249
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 150.21875,
      "epoch": 1.0,
      "grad_norm": 12.991154670715332,
      "kl": 0.177978515625,
      "learning_rate": 9e-07,
      "loss": 0.0002,
      "reward": 0.125,
      "reward_std": 0.2925042062997818,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.125,
      "step": 250
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 135.71875,
      "epoch": 1.004,
      "grad_norm": 10.966651916503906,
      "kl": 0.2197265625,
      "learning_rate": 8.995999999999999e-07,
      "loss": 0.0002,
      "reward": 0.125,
      "reward_std": 0.2925042062997818,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.125,
      "step": 251
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 141.90625,
      "epoch": 1.008,
      "grad_norm": 6.208240509033203,
      "kl": 0.205078125,
      "learning_rate": 8.992e-07,
      "loss": 0.0002,
      "reward": 0.3125,
      "reward_std": 0.3514062538743019,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.3125,
      "step": 252
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 151.46875,
      "epoch": 1.012,
      "grad_norm": 6.70610237121582,
      "kl": 1.283935546875,
      "learning_rate": 8.988e-07,
      "loss": 0.0013,
      "reward": 0.40625,
      "reward_std": 0.3608423173427582,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.40625,
      "step": 253
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 130.21875,
      "epoch": 1.016,
      "grad_norm": 3.005783796310425,
      "kl": 0.173095703125,
      "learning_rate": 8.983999999999999e-07,
      "loss": 0.0002,
      "reward": 0.53125,
      "reward_std": 0.0883883461356163,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.53125,
      "step": 254
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 146.21875,
      "epoch": 1.02,
      "grad_norm": 5.299511432647705,
      "kl": 0.185302734375,
      "learning_rate": 8.98e-07,
      "loss": 0.0002,
      "reward": 0.34375,
      "reward_std": 0.3061639815568924,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.34375,
      "step": 255
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 145.0625,
      "epoch": 1.024,
      "grad_norm": 7.629411220550537,
      "kl": 0.26318359375,
      "learning_rate": 8.975999999999999e-07,
      "loss": 0.0003,
      "reward": 0.46875,
      "reward_std": 0.3471629247069359,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.46875,
      "step": 256
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 135.5625,
      "epoch": 1.028,
      "grad_norm": 15.968307495117188,
      "kl": 0.17822265625,
      "learning_rate": 8.972e-07,
      "loss": 0.0002,
      "reward": 0.5625,
      "reward_std": 0.2587745785713196,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.5625,
      "step": 257
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 140.8125,
      "epoch": 1.032,
      "grad_norm": 9.633905410766602,
      "kl": 0.265380859375,
      "learning_rate": 8.968e-07,
      "loss": 0.0003,
      "reward": 0.3125,
      "reward_std": 0.3514062538743019,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.3125,
      "step": 258
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 167.53125,
      "epoch": 1.036,
      "grad_norm": 5.413832187652588,
      "kl": 0.187255859375,
      "learning_rate": 8.963999999999999e-07,
      "loss": 0.0002,
      "reward": 0.15625,
      "reward_std": 0.24511480331420898,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.15625,
      "step": 259
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 145.28125,
      "epoch": 1.04,
      "grad_norm": 42.84470748901367,
      "kl": 0.178466796875,
      "learning_rate": 8.96e-07,
      "loss": 0.0002,
      "reward": 0.40625,
      "reward_std": 0.4218914955854416,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.40625,
      "step": 260
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 139.1875,
      "epoch": 1.044,
      "grad_norm": 5.803390979766846,
      "kl": 0.201904296875,
      "learning_rate": 8.955999999999999e-07,
      "loss": 0.0002,
      "reward": 0.34375,
      "reward_std": 0.4807935431599617,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.34375,
      "step": 261
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 157.59375,
      "epoch": 1.048,
      "grad_norm": 9.147906303405762,
      "kl": 0.20263671875,
      "learning_rate": 8.951999999999999e-07,
      "loss": 0.0002,
      "reward": 0.21875,
      "reward_std": 0.3987956568598747,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.21875,
      "step": 262
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 153.8125,
      "epoch": 1.052,
      "grad_norm": 3.2289392948150635,
      "kl": 0.22314453125,
      "learning_rate": 8.948e-07,
      "loss": 0.0002,
      "reward": 0.625,
      "reward_std": 0.2314550280570984,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.625,
      "step": 263
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 145.0,
      "epoch": 1.056,
      "grad_norm": 16.819181442260742,
      "kl": 0.2177734375,
      "learning_rate": 8.944e-07,
      "loss": 0.0002,
      "reward": 0.21875,
      "reward_std": 0.3061639815568924,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.21875,
      "step": 264
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 149.71875,
      "epoch": 1.06,
      "grad_norm": 14.53752326965332,
      "kl": 0.1953125,
      "learning_rate": 8.939999999999999e-07,
      "loss": 0.0002,
      "reward": 0.5,
      "reward_std": 0.2587745785713196,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.5,
      "step": 265
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 142.96875,
      "epoch": 1.064,
      "grad_norm": 3.053736925125122,
      "kl": 0.21142578125,
      "learning_rate": 8.935999999999999e-07,
      "loss": 0.0002,
      "reward": 0.34375,
      "reward_std": 0.1293872892856598,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.34375,
      "step": 266
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 158.25,
      "epoch": 1.068,
      "grad_norm": 4.489502906799316,
      "kl": 0.180908203125,
      "learning_rate": 8.932e-07,
      "loss": 0.0002,
      "reward": 0.5,
      "reward_std": 0.2925042062997818,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.5,
      "step": 267
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 140.0625,
      "epoch": 1.072,
      "grad_norm": 7.516378402709961,
      "kl": 0.201416015625,
      "learning_rate": 8.928e-07,
      "loss": 0.0002,
      "reward": 0.28125,
      "reward_std": 0.0883883461356163,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.28125,
      "step": 268
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 143.28125,
      "epoch": 1.076,
      "grad_norm": 7.0981645584106445,
      "kl": 0.21875,
      "learning_rate": 8.923999999999999e-07,
      "loss": 0.0002,
      "reward": 0.1875,
      "reward_std": 0.3335031494498253,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.1875,
      "step": 269
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 133.59375,
      "epoch": 1.08,
      "grad_norm": 5.433140754699707,
      "kl": 0.222900390625,
      "learning_rate": 8.92e-07,
      "loss": 0.0002,
      "reward": 0.3125,
      "reward_std": 0.1767766922712326,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.3125,
      "step": 270
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 141.125,
      "epoch": 1.084,
      "grad_norm": 11.345158576965332,
      "kl": 0.216552734375,
      "learning_rate": 8.915999999999999e-07,
      "loss": 0.0002,
      "reward": 0.4375,
      "reward_std": 0.4492306634783745,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.4375,
      "step": 271
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 146.71875,
      "epoch": 1.088,
      "grad_norm": 7.191331386566162,
      "kl": 0.253173828125,
      "learning_rate": 8.911999999999999e-07,
      "loss": 0.0003,
      "reward": 0.53125,
      "reward_std": 0.3787454217672348,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.53125,
      "step": 272
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 122.96875,
      "epoch": 1.092,
      "grad_norm": 5.7541985511779785,
      "kl": 0.203125,
      "learning_rate": 8.908e-07,
      "loss": 0.0002,
      "reward": 0.34375,
      "reward_std": 0.2041158601641655,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.34375,
      "step": 273
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 144.6875,
      "epoch": 1.096,
      "grad_norm": 6.448830604553223,
      "kl": 0.213623046875,
      "learning_rate": 8.904e-07,
      "loss": 0.0002,
      "reward": 0.34375,
      "reward_std": 0.3808925524353981,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.34375,
      "step": 274
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 142.125,
      "epoch": 1.1,
      "grad_norm": 3.787655830383301,
      "kl": 0.217529296875,
      "learning_rate": 8.9e-07,
      "loss": 0.0002,
      "reward": 0.1875,
      "reward_std": 0.2177756354212761,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.1875,
      "step": 275
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 145.21875,
      "epoch": 1.104,
      "grad_norm": 3.446657180786133,
      "kl": 0.217529296875,
      "learning_rate": 8.895999999999999e-07,
      "loss": 0.0002,
      "reward": 0.6875,
      "reward_std": 0.249358132481575,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.6875,
      "step": 276
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 129.0625,
      "epoch": 1.108,
      "grad_norm": 54.69266128540039,
      "kl": 0.211669921875,
      "learning_rate": 8.892e-07,
      "loss": 0.0002,
      "reward": 0.34375,
      "reward_std": 0.3377464786171913,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.34375,
      "step": 277
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 126.71875,
      "epoch": 1.112,
      "grad_norm": 32.58818817138672,
      "kl": 0.248779296875,
      "learning_rate": 8.888e-07,
      "loss": 0.0002,
      "reward": 0.28125,
      "reward_std": 0.2630179077386856,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.28125,
      "step": 278
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 127.875,
      "epoch": 1.116,
      "grad_norm": 0.0929594337940216,
      "kl": 0.3173828125,
      "learning_rate": 8.883999999999999e-07,
      "loss": 0.0003,
      "reward": 0.75,
      "reward_std": 0.0,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.75,
      "step": 279
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 121.375,
      "epoch": 1.12,
      "grad_norm": 3.5213842391967773,
      "kl": 0.2275390625,
      "learning_rate": 8.88e-07,
      "loss": 0.0002,
      "reward": 0.3125,
      "reward_std": 0.1157275140285492,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.3125,
      "step": 280
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 132.9375,
      "epoch": 1.124,
      "grad_norm": 5.264272212982178,
      "kl": 0.28369140625,
      "learning_rate": 8.875999999999999e-07,
      "loss": 0.0003,
      "reward": 0.375,
      "reward_std": 0.3104073107242584,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.375,
      "step": 281
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 116.53125,
      "epoch": 1.1280000000000001,
      "grad_norm": 4.806576728820801,
      "kl": 0.26025390625,
      "learning_rate": 8.872e-07,
      "loss": 0.0003,
      "reward": 0.34375,
      "reward_std": 0.22201896458864212,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.34375,
      "step": 282
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 144.34375,
      "epoch": 1.1320000000000001,
      "grad_norm": 16.849353790283203,
      "kl": 0.2763671875,
      "learning_rate": 8.868e-07,
      "loss": 0.0003,
      "reward": 0.5625,
      "reward_std": 0.1157275140285492,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.5625,
      "step": 283
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 119.5,
      "epoch": 1.1360000000000001,
      "grad_norm": 0.024166446179151535,
      "kl": 0.277587890625,
      "learning_rate": 8.863999999999999e-07,
      "loss": 0.0003,
      "reward": 0.25,
      "reward_std": 0.0,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.25,
      "step": 284
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 125.375,
      "epoch": 1.1400000000000001,
      "grad_norm": 3.680453300476074,
      "kl": 0.27392578125,
      "learning_rate": 8.86e-07,
      "loss": 0.0003,
      "reward": 0.53125,
      "reward_std": 0.2041158601641655,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.53125,
      "step": 285
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 115.375,
      "epoch": 1.144,
      "grad_norm": 3.488722562789917,
      "kl": 0.292724609375,
      "learning_rate": 8.856e-07,
      "loss": 0.0003,
      "reward": 0.3125,
      "reward_std": 0.1157275140285492,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.3125,
      "step": 286
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 125.96875,
      "epoch": 1.148,
      "grad_norm": 4.979753494262695,
      "kl": 0.269287109375,
      "learning_rate": 8.851999999999999e-07,
      "loss": 0.0003,
      "reward": 0.4375,
      "reward_std": 0.3514062538743019,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.4375,
      "step": 287
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 138.21875,
      "epoch": 1.152,
      "grad_norm": 6.298620223999023,
      "kl": 0.27734375,
      "learning_rate": 8.848e-07,
      "loss": 0.0003,
      "reward": 0.1875,
      "reward_std": 0.2587745785713196,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.1875,
      "step": 288
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 127.78125,
      "epoch": 1.156,
      "grad_norm": 5.3922882080078125,
      "kl": 0.296875,
      "learning_rate": 8.844e-07,
      "loss": 0.0003,
      "reward": 0.1875,
      "reward_std": 0.249358132481575,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.1875,
      "step": 289
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 136.90625,
      "epoch": 1.16,
      "grad_norm": 5.702232360839844,
      "kl": 0.27685546875,
      "learning_rate": 8.839999999999999e-07,
      "loss": 0.0003,
      "reward": 0.4375,
      "reward_std": 0.2177756354212761,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.4375,
      "step": 290
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 131.28125,
      "epoch": 1.164,
      "grad_norm": 0.024661913514137268,
      "kl": 0.2802734375,
      "learning_rate": 8.836e-07,
      "loss": 0.0003,
      "reward": 0.25,
      "reward_std": 0.0,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.25,
      "step": 291
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 137.59375,
      "epoch": 1.168,
      "grad_norm": 32.11168670654297,
      "kl": 0.240966796875,
      "learning_rate": 8.832e-07,
      "loss": 0.0002,
      "reward": 0.46875,
      "reward_std": 0.24511480331420898,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.46875,
      "step": 292
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 149.03125,
      "epoch": 1.172,
      "grad_norm": 36.96295166015625,
      "kl": 0.34814453125,
      "learning_rate": 8.827999999999999e-07,
      "loss": 0.0003,
      "reward": 0.09375,
      "reward_std": 0.2041158601641655,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.09375,
      "step": 293
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 122.3125,
      "epoch": 1.176,
      "grad_norm": 3.696056604385376,
      "kl": 0.2783203125,
      "learning_rate": 8.823999999999999e-07,
      "loss": 0.0003,
      "reward": 0.28125,
      "reward_std": 0.0883883461356163,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.28125,
      "step": 294
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 138.875,
      "epoch": 1.18,
      "grad_norm": 2.5873639583587646,
      "kl": 0.35009765625,
      "learning_rate": 8.82e-07,
      "loss": 0.0003,
      "reward": 0.71875,
      "reward_std": 0.0883883461356163,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.71875,
      "step": 295
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 133.125,
      "epoch": 1.184,
      "grad_norm": 5.720484733581543,
      "kl": 0.4599609375,
      "learning_rate": 8.816000000000001e-07,
      "loss": 0.0005,
      "reward": 0.375,
      "reward_std": 0.3650856465101242,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.375,
      "step": 296
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 133.3125,
      "epoch": 1.188,
      "grad_norm": 10.18944263458252,
      "kl": 0.31298828125,
      "learning_rate": 8.811999999999999e-07,
      "loss": 0.0003,
      "reward": 0.46875,
      "reward_std": 0.4765698313713074,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.46875,
      "step": 297
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 152.625,
      "epoch": 1.192,
      "grad_norm": 5.152041435241699,
      "kl": 0.27001953125,
      "learning_rate": 8.808e-07,
      "loss": 0.0003,
      "reward": 0.03125,
      "reward_std": 0.0883883461356163,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.03125,
      "step": 298
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 151.46875,
      "epoch": 1.196,
      "grad_norm": 4.417140007019043,
      "kl": 0.5234375,
      "learning_rate": 8.804e-07,
      "loss": 0.0005,
      "reward": 0.125,
      "reward_std": 0.2925042062997818,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.125,
      "step": 299
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 146.8125,
      "epoch": 1.2,
      "grad_norm": 6.4717230796813965,
      "kl": 0.294677734375,
      "learning_rate": 8.799999999999999e-07,
      "loss": 0.0003,
      "reward": 0.375,
      "reward_std": 0.3945523276925087,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.375,
      "step": 300
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 142.75,
      "epoch": 1.204,
      "grad_norm": 5.155836582183838,
      "kl": 0.26806640625,
      "learning_rate": 8.796e-07,
      "loss": 0.0003,
      "reward": 0.6875,
      "reward_std": 0.3335031494498253,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.6875,
      "step": 301
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 130.65625,
      "epoch": 1.208,
      "grad_norm": 3.839174270629883,
      "kl": 0.27783203125,
      "learning_rate": 8.792e-07,
      "loss": 0.0003,
      "reward": 0.53125,
      "reward_std": 0.2651650384068489,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.53125,
      "step": 302
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 141.34375,
      "epoch": 1.212,
      "grad_norm": 4.404873371124268,
      "kl": 0.29296875,
      "learning_rate": 8.788e-07,
      "loss": 0.0003,
      "reward": 0.28125,
      "reward_std": 0.2630179077386856,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.28125,
      "step": 303
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 140.65625,
      "epoch": 1.216,
      "grad_norm": 9.610031127929688,
      "kl": 0.310546875,
      "learning_rate": 8.783999999999999e-07,
      "loss": 0.0003,
      "reward": 0.65625,
      "reward_std": 0.2041158601641655,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.65625,
      "step": 304
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 146.75,
      "epoch": 1.22,
      "grad_norm": 5.232041835784912,
      "kl": 0.3271484375,
      "learning_rate": 8.78e-07,
      "loss": 0.0003,
      "reward": 0.65625,
      "reward_std": 0.2041158601641655,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.65625,
      "step": 305
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 155.65625,
      "epoch": 1.224,
      "grad_norm": 5.603926181793213,
      "kl": 0.4677734375,
      "learning_rate": 8.776e-07,
      "loss": 0.0005,
      "reward": 0.59375,
      "reward_std": 0.4218914955854416,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.59375,
      "step": 306
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 145.84375,
      "epoch": 1.228,
      "grad_norm": 6.439726829528809,
      "kl": 0.31396484375,
      "learning_rate": 8.771999999999999e-07,
      "loss": 0.0003,
      "reward": 0.40625,
      "reward_std": 0.3198433741927147,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.40625,
      "step": 307
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 163.09375,
      "epoch": 1.232,
      "grad_norm": 4.445047855377197,
      "kl": 0.32958984375,
      "learning_rate": 8.768e-07,
      "loss": 0.0003,
      "reward": 0.25,
      "reward_std": 0.1767766922712326,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.25,
      "step": 308
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 151.9375,
      "epoch": 1.236,
      "grad_norm": 9.926246643066406,
      "kl": 0.313720703125,
      "learning_rate": 8.763999999999999e-07,
      "loss": 0.0003,
      "reward": 0.4375,
      "reward_std": 0.3945523276925087,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.4375,
      "step": 309
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 160.65625,
      "epoch": 1.24,
      "grad_norm": 4.820209503173828,
      "kl": 0.3916015625,
      "learning_rate": 8.76e-07,
      "loss": 0.0004,
      "reward": 0.3125,
      "reward_std": 0.2177756354212761,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.3125,
      "step": 310
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 160.875,
      "epoch": 1.244,
      "grad_norm": 4.64197301864624,
      "kl": 0.28564453125,
      "learning_rate": 8.756e-07,
      "loss": 0.0003,
      "reward": 0.40625,
      "reward_std": 0.24511480331420898,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.40625,
      "step": 311
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 171.34375,
      "epoch": 1.248,
      "grad_norm": 6.238957405090332,
      "kl": 0.269775390625,
      "learning_rate": 8.751999999999999e-07,
      "loss": 0.0003,
      "reward": 0.46875,
      "reward_std": 0.2041158601641655,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.46875,
      "step": 312
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 163.0625,
      "epoch": 1.252,
      "grad_norm": 16.75998878479004,
      "kl": 0.29443359375,
      "learning_rate": 8.748e-07,
      "loss": 0.0003,
      "reward": 0.21875,
      "reward_std": 0.4355708882212639,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.21875,
      "step": 313
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 180.09375,
      "epoch": 1.256,
      "grad_norm": 1.9939095973968506,
      "kl": 0.248291015625,
      "learning_rate": 8.743999999999999e-07,
      "loss": 0.0002,
      "reward": 0.03125,
      "reward_std": 0.0883883461356163,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.03125,
      "step": 314
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 174.625,
      "epoch": 1.26,
      "grad_norm": 11.04178524017334,
      "kl": 0.2802734375,
      "learning_rate": 8.739999999999999e-07,
      "loss": 0.0003,
      "reward": 0.53125,
      "reward_std": 0.3061639815568924,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.53125,
      "step": 315
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 172.40625,
      "epoch": 1.264,
      "grad_norm": 4.063099384307861,
      "kl": 0.29736328125,
      "learning_rate": 8.736e-07,
      "loss": 0.0003,
      "reward": 0.6875,
      "reward_std": 0.2925042062997818,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.6875,
      "step": 316
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 149.1875,
      "epoch": 1.268,
      "grad_norm": 3.432239055633545,
      "kl": 0.260009765625,
      "learning_rate": 8.732e-07,
      "loss": 0.0003,
      "reward": 0.3125,
      "reward_std": 0.1157275140285492,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.3125,
      "step": 317
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 169.625,
      "epoch": 1.272,
      "grad_norm": 1.9211883544921875,
      "kl": 0.4375,
      "learning_rate": 8.728e-07,
      "loss": 0.0004,
      "reward": 0.21875,
      "reward_std": 0.0883883461356163,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.21875,
      "step": 318
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 160.90625,
      "epoch": 1.276,
      "grad_norm": 2.502659559249878,
      "kl": 0.31591796875,
      "learning_rate": 8.723999999999999e-07,
      "loss": 0.0003,
      "reward": 0.28125,
      "reward_std": 0.0883883461356163,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.28125,
      "step": 319
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 148.875,
      "epoch": 1.28,
      "grad_norm": 3.1834092140197754,
      "kl": 0.2861328125,
      "learning_rate": 8.72e-07,
      "loss": 0.0003,
      "reward": 0.3125,
      "reward_std": 0.1157275140285492,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.3125,
      "step": 320
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 166.03125,
      "epoch": 1.284,
      "grad_norm": 37.28458786010742,
      "kl": 0.27490234375,
      "learning_rate": 8.716e-07,
      "loss": 0.0003,
      "reward": 0.28125,
      "reward_std": 0.0883883461356163,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.28125,
      "step": 321
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 175.5625,
      "epoch": 1.288,
      "grad_norm": 3.5250649452209473,
      "kl": 0.42919921875,
      "learning_rate": 8.711999999999999e-07,
      "loss": 0.0004,
      "reward": 0.28125,
      "reward_std": 0.2041158601641655,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.28125,
      "step": 322
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 168.46875,
      "epoch": 1.292,
      "grad_norm": 6.868240833282471,
      "kl": 0.3134765625,
      "learning_rate": 8.708e-07,
      "loss": 0.0003,
      "reward": 0.15625,
      "reward_std": 0.22201896458864212,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.15625,
      "step": 323
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 159.75,
      "epoch": 1.296,
      "grad_norm": 6.121273040771484,
      "kl": 0.34033203125,
      "learning_rate": 8.704e-07,
      "loss": 0.0003,
      "reward": 0.40625,
      "reward_std": 0.3377464786171913,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.40625,
      "step": 324
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 157.71875,
      "epoch": 1.3,
      "grad_norm": 3.7920565605163574,
      "kl": 0.298828125,
      "learning_rate": 8.699999999999999e-07,
      "loss": 0.0003,
      "reward": 0.46875,
      "reward_std": 0.0883883461356163,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.46875,
      "step": 325
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 159.71875,
      "epoch": 1.304,
      "grad_norm": 3.7230024337768555,
      "kl": 0.3349609375,
      "learning_rate": 8.696e-07,
      "loss": 0.0003,
      "reward": 0.0625,
      "reward_std": 0.1157275140285492,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.0625,
      "step": 326
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 162.6875,
      "epoch": 1.308,
      "grad_norm": 3.6794321537017822,
      "kl": 0.302978515625,
      "learning_rate": 8.692e-07,
      "loss": 0.0003,
      "reward": 0.34375,
      "reward_std": 0.1293872892856598,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.34375,
      "step": 327
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 160.8125,
      "epoch": 1.312,
      "grad_norm": 2.1902525424957275,
      "kl": 0.279052734375,
      "learning_rate": 8.687999999999999e-07,
      "loss": 0.0003,
      "reward": 0.4375,
      "reward_std": 0.1157275140285492,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.4375,
      "step": 328
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 165.96875,
      "epoch": 1.316,
      "grad_norm": 3.8480546474456787,
      "kl": 0.38330078125,
      "learning_rate": 8.683999999999999e-07,
      "loss": 0.0004,
      "reward": 0.71875,
      "reward_std": 0.2630179077386856,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.71875,
      "step": 329
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 170.0625,
      "epoch": 1.32,
      "grad_norm": 3.8732831478118896,
      "kl": 0.333984375,
      "learning_rate": 8.68e-07,
      "loss": 0.0003,
      "reward": 0.28125,
      "reward_std": 0.0883883461356163,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.28125,
      "step": 330
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 168.4375,
      "epoch": 1.324,
      "grad_norm": 6.415884971618652,
      "kl": 0.28662109375,
      "learning_rate": 8.676e-07,
      "loss": 0.0003,
      "reward": 0.1875,
      "reward_std": 0.249358132481575,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.1875,
      "step": 331
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 180.34375,
      "epoch": 1.328,
      "grad_norm": 2.7691843509674072,
      "kl": 0.59912109375,
      "learning_rate": 8.671999999999999e-07,
      "loss": 0.0006,
      "reward": 0.15625,
      "reward_std": 0.1293872892856598,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.15625,
      "step": 332
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 182.03125,
      "epoch": 1.332,
      "grad_norm": 8.122827529907227,
      "kl": 0.31787109375,
      "learning_rate": 8.668e-07,
      "loss": 0.0003,
      "reward": 0.6875,
      "reward_std": 0.1767766922712326,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.6875,
      "step": 333
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 163.28125,
      "epoch": 1.336,
      "grad_norm": 11.689245223999023,
      "kl": 0.43017578125,
      "learning_rate": 8.663999999999999e-07,
      "loss": 0.0004,
      "reward": 0.53125,
      "reward_std": 0.35564958304166794,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.53125,
      "step": 334
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 163.5625,
      "epoch": 1.34,
      "grad_norm": 0.040142327547073364,
      "kl": 0.3466796875,
      "learning_rate": 8.659999999999999e-07,
      "loss": 0.0003,
      "reward": 0.0,
      "reward_std": 0.0,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.0,
      "step": 335
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 168.15625,
      "epoch": 1.3439999999999999,
      "grad_norm": 7.134670257568359,
      "kl": 0.31787109375,
      "learning_rate": 8.656e-07,
      "loss": 0.0003,
      "reward": 0.625,
      "reward_std": 0.2177756354212761,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.625,
      "step": 336
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 160.8125,
      "epoch": 1.3479999999999999,
      "grad_norm": 6.3668599128723145,
      "kl": 0.35205078125,
      "learning_rate": 8.651999999999999e-07,
      "loss": 0.0004,
      "reward": 0.3125,
      "reward_std": 0.1157275140285492,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.3125,
      "step": 337
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 174.4375,
      "epoch": 1.3519999999999999,
      "grad_norm": 3.3299219608306885,
      "kl": 0.37939453125,
      "learning_rate": 8.648e-07,
      "loss": 0.0004,
      "reward": 0.46875,
      "reward_std": 0.0883883461356163,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.46875,
      "step": 338
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 172.84375,
      "epoch": 1.3559999999999999,
      "grad_norm": 3.8048033714294434,
      "kl": 0.35791015625,
      "learning_rate": 8.643999999999999e-07,
      "loss": 0.0004,
      "reward": 0.21875,
      "reward_std": 0.0883883461356163,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.21875,
      "step": 339
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 175.4375,
      "epoch": 1.3599999999999999,
      "grad_norm": 0.019212258979678154,
      "kl": 0.3486328125,
      "learning_rate": 8.639999999999999e-07,
      "loss": 0.0003,
      "reward": 0.25,
      "reward_std": 0.0,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.25,
      "step": 340
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 172.0625,
      "epoch": 1.3639999999999999,
      "grad_norm": 4.146425724029541,
      "kl": 0.365234375,
      "learning_rate": 8.636e-07,
      "loss": 0.0004,
      "reward": 0.1875,
      "reward_std": 0.1157275140285492,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.1875,
      "step": 341
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 163.9375,
      "epoch": 1.3679999999999999,
      "grad_norm": 5.75056266784668,
      "kl": 0.36328125,
      "learning_rate": 8.632e-07,
      "loss": 0.0004,
      "reward": 0.25,
      "reward_std": 0.2925042062997818,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.25,
      "step": 342
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 160.375,
      "epoch": 1.3719999999999999,
      "grad_norm": 5.63866662979126,
      "kl": 0.330078125,
      "learning_rate": 8.628e-07,
      "loss": 0.0003,
      "reward": 0.4375,
      "reward_std": 0.3335031494498253,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.4375,
      "step": 343
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 149.59375,
      "epoch": 1.376,
      "grad_norm": 2.1933727264404297,
      "kl": 0.52587890625,
      "learning_rate": 8.624e-07,
      "loss": 0.0005,
      "reward": 0.5625,
      "reward_std": 0.1157275140285492,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.5625,
      "step": 344
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 148.125,
      "epoch": 1.38,
      "grad_norm": 3.293926954269409,
      "kl": 0.51171875,
      "learning_rate": 8.62e-07,
      "loss": 0.0005,
      "reward": 0.28125,
      "reward_std": 0.0883883461356163,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.28125,
      "step": 345
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 145.4375,
      "epoch": 1.384,
      "grad_norm": 27.214008331298828,
      "kl": 0.3837890625,
      "learning_rate": 8.616e-07,
      "loss": 0.0004,
      "reward": 0.3125,
      "reward_std": 0.1767766922712326,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.3125,
      "step": 346
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 155.65625,
      "epoch": 1.388,
      "grad_norm": 8.472949981689453,
      "kl": 0.4326171875,
      "learning_rate": 8.611999999999999e-07,
      "loss": 0.0004,
      "reward": 0.4375,
      "reward_std": 0.249358132481575,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.4375,
      "step": 347
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 147.125,
      "epoch": 1.392,
      "grad_norm": 2.3920438289642334,
      "kl": 0.369140625,
      "learning_rate": 8.608e-07,
      "loss": 0.0004,
      "reward": 0.09375,
      "reward_std": 0.1293872892856598,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.09375,
      "step": 348
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 140.3125,
      "epoch": 1.396,
      "grad_norm": 4.007767677307129,
      "kl": 0.5234375,
      "learning_rate": 8.604000000000001e-07,
      "loss": 0.0005,
      "reward": 0.46875,
      "reward_std": 0.24511480331420898,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.46875,
      "step": 349
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 153.3125,
      "epoch": 1.4,
      "grad_norm": 4.542917728424072,
      "kl": 0.4287109375,
      "learning_rate": 8.599999999999999e-07,
      "loss": 0.0004,
      "reward": 0.4375,
      "reward_std": 0.3335031494498253,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.4375,
      "step": 350
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 146.375,
      "epoch": 1.404,
      "grad_norm": 5.640623092651367,
      "kl": 0.365234375,
      "learning_rate": 8.596e-07,
      "loss": 0.0004,
      "reward": 0.15625,
      "reward_std": 0.24511480331420898,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.15625,
      "step": 351
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 149.40625,
      "epoch": 1.408,
      "grad_norm": 2.198486328125,
      "kl": 0.419921875,
      "learning_rate": 8.592e-07,
      "loss": 0.0004,
      "reward": 0.03125,
      "reward_std": 0.0883883461356163,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.03125,
      "step": 352
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 152.3125,
      "epoch": 1.412,
      "grad_norm": 5.3141984939575195,
      "kl": 0.44580078125,
      "learning_rate": 8.587999999999999e-07,
      "loss": 0.0004,
      "reward": 0.125,
      "reward_std": 0.2314550280570984,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.125,
      "step": 353
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 144.875,
      "epoch": 1.416,
      "grad_norm": 5.0913777351379395,
      "kl": 0.3828125,
      "learning_rate": 8.584e-07,
      "loss": 0.0004,
      "reward": 0.625,
      "reward_std": 0.2314550280570984,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.625,
      "step": 354
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 145.65625,
      "epoch": 1.42,
      "grad_norm": 6.999328136444092,
      "kl": 0.3427734375,
      "learning_rate": 8.58e-07,
      "loss": 0.0003,
      "reward": 0.65625,
      "reward_std": 0.3377464786171913,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.65625,
      "step": 355
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 158.3125,
      "epoch": 1.424,
      "grad_norm": 4.412006855010986,
      "kl": 0.35302734375,
      "learning_rate": 8.576e-07,
      "loss": 0.0004,
      "reward": 0.15625,
      "reward_std": 0.3061639815568924,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.15625,
      "step": 356
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 156.96875,
      "epoch": 1.428,
      "grad_norm": 6.386446952819824,
      "kl": 0.380859375,
      "learning_rate": 8.571999999999999e-07,
      "loss": 0.0004,
      "reward": 0.28125,
      "reward_std": 0.24511480331420898,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.28125,
      "step": 357
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 155.8125,
      "epoch": 1.432,
      "grad_norm": 0.12800700962543488,
      "kl": 0.380859375,
      "learning_rate": 8.568e-07,
      "loss": 0.0004,
      "reward": 0.0,
      "reward_std": 0.0,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.0,
      "step": 358
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 145.75,
      "epoch": 1.436,
      "grad_norm": 3.816267490386963,
      "kl": 0.63037109375,
      "learning_rate": 8.564e-07,
      "loss": 0.0006,
      "reward": 0.6875,
      "reward_std": 0.1767766922712326,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.6875,
      "step": 359
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 153.46875,
      "epoch": 1.44,
      "grad_norm": 3.641408681869507,
      "kl": 0.49169921875,
      "learning_rate": 8.559999999999999e-07,
      "loss": 0.0005,
      "reward": 0.40625,
      "reward_std": 0.1293872892856598,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.40625,
      "step": 360
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 152.78125,
      "epoch": 1.444,
      "grad_norm": 6.496709823608398,
      "kl": 0.43115234375,
      "learning_rate": 8.556e-07,
      "loss": 0.0004,
      "reward": 0.28125,
      "reward_std": 0.4218914955854416,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.28125,
      "step": 361
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 158.8125,
      "epoch": 1.448,
      "grad_norm": 6.215940952301025,
      "kl": 0.40771484375,
      "learning_rate": 8.551999999999999e-07,
      "loss": 0.0004,
      "reward": 0.375,
      "reward_std": 0.3945523276925087,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.375,
      "step": 362
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 161.875,
      "epoch": 1.452,
      "grad_norm": 3.668593645095825,
      "kl": 0.5849609375,
      "learning_rate": 8.548e-07,
      "loss": 0.0006,
      "reward": 0.46875,
      "reward_std": 0.0883883461356163,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.46875,
      "step": 363
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 150.78125,
      "epoch": 1.456,
      "grad_norm": 4.504969120025635,
      "kl": 0.37060546875,
      "learning_rate": 8.544e-07,
      "loss": 0.0004,
      "reward": 0.65625,
      "reward_std": 0.1293872892856598,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.65625,
      "step": 364
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 162.0,
      "epoch": 1.46,
      "grad_norm": 2.0712342262268066,
      "kl": 0.3798828125,
      "learning_rate": 8.539999999999999e-07,
      "loss": 0.0004,
      "reward": 0.53125,
      "reward_std": 0.0883883461356163,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.53125,
      "step": 365
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 162.0625,
      "epoch": 1.464,
      "grad_norm": 6.902523994445801,
      "kl": 0.7861328125,
      "learning_rate": 8.536e-07,
      "loss": 0.0008,
      "reward": 0.1875,
      "reward_std": 0.3104073107242584,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.1875,
      "step": 366
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 157.09375,
      "epoch": 1.468,
      "grad_norm": 2.4611382484436035,
      "kl": 0.37109375,
      "learning_rate": 8.531999999999999e-07,
      "loss": 0.0004,
      "reward": 0.375,
      "reward_std": 0.13363061845302582,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.375,
      "step": 367
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 149.59375,
      "epoch": 1.472,
      "grad_norm": 3.2094199657440186,
      "kl": 0.40185546875,
      "learning_rate": 8.528e-07,
      "loss": 0.0004,
      "reward": 0.4375,
      "reward_std": 0.1157275140285492,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.4375,
      "step": 368
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 158.90625,
      "epoch": 1.476,
      "grad_norm": 2.43998122215271,
      "kl": 0.380859375,
      "learning_rate": 8.524e-07,
      "loss": 0.0004,
      "reward": 0.0625,
      "reward_std": 0.1157275140285492,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.0625,
      "step": 369
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 166.625,
      "epoch": 1.48,
      "grad_norm": 4.880451202392578,
      "kl": 0.4033203125,
      "learning_rate": 8.52e-07,
      "loss": 0.0004,
      "reward": 0.5625,
      "reward_std": 0.1767766922712326,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.5625,
      "step": 370
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 163.125,
      "epoch": 1.484,
      "grad_norm": 2.5401484966278076,
      "kl": 0.4384765625,
      "learning_rate": 8.516e-07,
      "loss": 0.0004,
      "reward": 0.4375,
      "reward_std": 0.1157275140285492,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.4375,
      "step": 371
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 169.3125,
      "epoch": 1.488,
      "grad_norm": 3.9937148094177246,
      "kl": 0.41357421875,
      "learning_rate": 8.511999999999999e-07,
      "loss": 0.0004,
      "reward": 0.46875,
      "reward_std": 0.2041158601641655,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.46875,
      "step": 372
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 166.71875,
      "epoch": 1.492,
      "grad_norm": 19.11022186279297,
      "kl": 0.39599609375,
      "learning_rate": 8.508e-07,
      "loss": 0.0004,
      "reward": 0.28125,
      "reward_std": 0.0883883461356163,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.28125,
      "step": 373
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 165.375,
      "epoch": 1.496,
      "grad_norm": 6.305721759796143,
      "kl": 0.36376953125,
      "learning_rate": 8.504e-07,
      "loss": 0.0004,
      "reward": 0.34375,
      "reward_std": 0.3377464786171913,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.34375,
      "step": 374
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 159.75,
      "epoch": 1.5,
      "grad_norm": 4.335789203643799,
      "kl": 0.404296875,
      "learning_rate": 8.499999999999999e-07,
      "loss": 0.0004,
      "reward": 0.25,
      "reward_std": 0.1767766922712326,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.25,
      "step": 375
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 163.6875,
      "epoch": 1.504,
      "grad_norm": 0.017256610095500946,
      "kl": 0.39697265625,
      "learning_rate": 8.496e-07,
      "loss": 0.0004,
      "reward": 0.25,
      "reward_std": 0.0,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.25,
      "step": 376
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 161.96875,
      "epoch": 1.508,
      "grad_norm": 5.186189651489258,
      "kl": 0.40478515625,
      "learning_rate": 8.492e-07,
      "loss": 0.0004,
      "reward": 0.125,
      "reward_std": 0.2177756354212761,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.125,
      "step": 377
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 174.40625,
      "epoch": 1.512,
      "grad_norm": 8.06170654296875,
      "kl": 0.39599609375,
      "learning_rate": 8.487999999999999e-07,
      "loss": 0.0004,
      "reward": 0.125,
      "reward_std": 0.2177756354212761,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.125,
      "step": 378
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 173.75,
      "epoch": 1.516,
      "grad_norm": 4.342880725860596,
      "kl": 0.33984375,
      "learning_rate": 8.484e-07,
      "loss": 0.0003,
      "reward": 0.25,
      "reward_std": 0.1767766922712326,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.25,
      "step": 379
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 175.375,
      "epoch": 1.52,
      "grad_norm": 2.14447021484375,
      "kl": 0.37890625,
      "learning_rate": 8.48e-07,
      "loss": 0.0004,
      "reward": 0.34375,
      "reward_std": 0.1293872892856598,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.34375,
      "step": 380
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 166.96875,
      "epoch": 1.524,
      "grad_norm": 3.2591371536254883,
      "kl": 0.44140625,
      "learning_rate": 8.475999999999999e-07,
      "loss": 0.0004,
      "reward": 0.59375,
      "reward_std": 0.24511480331420898,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.59375,
      "step": 381
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 172.21875,
      "epoch": 1.528,
      "grad_norm": 7.309144020080566,
      "kl": 0.32958984375,
      "learning_rate": 8.471999999999999e-07,
      "loss": 0.0003,
      "reward": 0.53125,
      "reward_std": 0.24511480331420898,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.53125,
      "step": 382
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 178.9375,
      "epoch": 1.532,
      "grad_norm": 5.914097309112549,
      "kl": 0.33544921875,
      "learning_rate": 8.468e-07,
      "loss": 0.0003,
      "reward": 0.4375,
      "reward_std": 0.3745020925998688,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.4375,
      "step": 383
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 174.03125,
      "epoch": 1.536,
      "grad_norm": 5.55394172668457,
      "kl": 0.3583984375,
      "learning_rate": 8.464e-07,
      "loss": 0.0004,
      "reward": 0.53125,
      "reward_std": 0.2041158601641655,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.53125,
      "step": 384
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 182.0,
      "epoch": 1.54,
      "grad_norm": 4.632668972015381,
      "kl": 0.35986328125,
      "learning_rate": 8.459999999999999e-07,
      "loss": 0.0004,
      "reward": 0.46875,
      "reward_std": 0.0883883461356163,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.46875,
      "step": 385
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 173.8125,
      "epoch": 1.544,
      "grad_norm": 6.332529067993164,
      "kl": 0.3095703125,
      "learning_rate": 8.456e-07,
      "loss": 0.0003,
      "reward": 0.5625,
      "reward_std": 0.2177756354212761,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.5625,
      "step": 386
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 183.71875,
      "epoch": 1.548,
      "grad_norm": 5.1340012550354,
      "kl": 0.41455078125,
      "learning_rate": 8.451999999999999e-07,
      "loss": 0.0004,
      "reward": 0.3125,
      "reward_std": 0.2587745785713196,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.3125,
      "step": 387
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 181.65625,
      "epoch": 1.552,
      "grad_norm": 1.8199496269226074,
      "kl": 0.4033203125,
      "learning_rate": 8.447999999999999e-07,
      "loss": 0.0004,
      "reward": 0.46875,
      "reward_std": 0.0883883461356163,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.46875,
      "step": 388
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 180.21875,
      "epoch": 1.556,
      "grad_norm": 3.395925760269165,
      "kl": 0.37939453125,
      "learning_rate": 8.444e-07,
      "loss": 0.0004,
      "reward": 0.40625,
      "reward_std": 0.1293872892856598,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.40625,
      "step": 389
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 184.15625,
      "epoch": 1.56,
      "grad_norm": 2.5946571826934814,
      "kl": 0.38134765625,
      "learning_rate": 8.439999999999999e-07,
      "loss": 0.0004,
      "reward": 0.03125,
      "reward_std": 0.0883883461356163,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.03125,
      "step": 390
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 172.375,
      "epoch": 1.564,
      "grad_norm": 3.0371081829071045,
      "kl": 0.40673828125,
      "learning_rate": 8.436e-07,
      "loss": 0.0004,
      "reward": 0.59375,
      "reward_std": 0.1293872892856598,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.59375,
      "step": 391
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 182.78125,
      "epoch": 1.568,
      "grad_norm": 0.05896129459142685,
      "kl": 0.36083984375,
      "learning_rate": 8.431999999999999e-07,
      "loss": 0.0004,
      "reward": 0.25,
      "reward_std": 0.0,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.25,
      "step": 392
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 160.625,
      "epoch": 1.572,
      "grad_norm": 0.7617104649543762,
      "kl": 0.40087890625,
      "learning_rate": 8.428e-07,
      "loss": 0.0004,
      "reward": 0.75,
      "reward_std": 0.0,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.75,
      "step": 393
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 176.0625,
      "epoch": 1.576,
      "grad_norm": 11.398433685302734,
      "kl": 0.34228515625,
      "learning_rate": 8.424e-07,
      "loss": 0.0003,
      "reward": 0.6875,
      "reward_std": 0.1767766922712326,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.6875,
      "step": 394
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 167.125,
      "epoch": 1.58,
      "grad_norm": 0.6610611081123352,
      "kl": 0.62158203125,
      "learning_rate": 8.419999999999999e-07,
      "loss": 0.0006,
      "reward": 0.75,
      "reward_std": 0.0,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.75,
      "step": 395
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 171.0625,
      "epoch": 1.584,
      "grad_norm": 1.8818594217300415,
      "kl": 0.39404296875,
      "learning_rate": 8.416e-07,
      "loss": 0.0004,
      "reward": 0.53125,
      "reward_std": 0.0883883461356163,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.53125,
      "step": 396
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 181.1875,
      "epoch": 1.588,
      "grad_norm": 2.491208791732788,
      "kl": 0.3828125,
      "learning_rate": 8.411999999999999e-07,
      "loss": 0.0004,
      "reward": 0.46875,
      "reward_std": 0.0883883461356163,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.46875,
      "step": 397
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 173.8125,
      "epoch": 1.592,
      "grad_norm": 0.01651681773364544,
      "kl": 0.34423828125,
      "learning_rate": 8.408e-07,
      "loss": 0.0003,
      "reward": 0.75,
      "reward_std": 0.0,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.75,
      "step": 398
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 170.90625,
      "epoch": 1.596,
      "grad_norm": 12.821555137634277,
      "kl": 0.37255859375,
      "learning_rate": 8.404e-07,
      "loss": 0.0004,
      "reward": 0.28125,
      "reward_std": 0.0883883461356163,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.28125,
      "step": 399
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 173.46875,
      "epoch": 1.6,
      "grad_norm": 2.659421920776367,
      "kl": 0.4072265625,
      "learning_rate": 8.399999999999999e-07,
      "loss": 0.0004,
      "reward": 0.28125,
      "reward_std": 0.0883883461356163,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.28125,
      "step": 400
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 171.875,
      "epoch": 1.604,
      "grad_norm": 3.0120511054992676,
      "kl": 0.3759765625,
      "learning_rate": 8.396e-07,
      "loss": 0.0004,
      "reward": 0.65625,
      "reward_std": 0.1293872892856598,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.65625,
      "step": 401
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 170.09375,
      "epoch": 1.608,
      "grad_norm": 0.057813845574855804,
      "kl": 0.474609375,
      "learning_rate": 8.391999999999999e-07,
      "loss": 0.0005,
      "reward": 0.5,
      "reward_std": 0.0,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.5,
      "step": 402
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 168.46875,
      "epoch": 1.612,
      "grad_norm": 0.05991186574101448,
      "kl": 0.474609375,
      "learning_rate": 8.387999999999999e-07,
      "loss": 0.0005,
      "reward": 0.25,
      "reward_std": 0.0,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.25,
      "step": 403
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 163.09375,
      "epoch": 1.616,
      "grad_norm": 12.902780532836914,
      "kl": 0.3994140625,
      "learning_rate": 8.384e-07,
      "loss": 0.0004,
      "reward": 0.65625,
      "reward_std": 0.3061639815568924,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.65625,
      "step": 404
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 173.46875,
      "epoch": 1.62,
      "grad_norm": 0.03287380188703537,
      "kl": 0.39794921875,
      "learning_rate": 8.38e-07,
      "loss": 0.0004,
      "reward": 0.25,
      "reward_std": 0.0,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.25,
      "step": 405
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 173.625,
      "epoch": 1.624,
      "grad_norm": 0.035295650362968445,
      "kl": 0.408203125,
      "learning_rate": 8.375999999999999e-07,
      "loss": 0.0004,
      "reward": 0.0,
      "reward_std": 0.0,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.0,
      "step": 406
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 173.4375,
      "epoch": 1.6280000000000001,
      "grad_norm": 6.105829238891602,
      "kl": 0.3671875,
      "learning_rate": 8.372e-07,
      "loss": 0.0004,
      "reward": 0.84375,
      "reward_std": 0.1293872892856598,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.84375,
      "step": 407
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 155.75,
      "epoch": 1.6320000000000001,
      "grad_norm": 0.024823768064379692,
      "kl": 0.4033203125,
      "learning_rate": 8.368e-07,
      "loss": 0.0004,
      "reward": 0.5,
      "reward_std": 0.0,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.5,
      "step": 408
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 180.59375,
      "epoch": 1.6360000000000001,
      "grad_norm": 3.6022651195526123,
      "kl": 0.423828125,
      "learning_rate": 8.363999999999999e-07,
      "loss": 0.0004,
      "reward": 0.625,
      "reward_std": 0.2314550280570984,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.625,
      "step": 409
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 168.5625,
      "epoch": 1.6400000000000001,
      "grad_norm": 34.663330078125,
      "kl": 0.4638671875,
      "learning_rate": 8.359999999999999e-07,
      "loss": 0.0005,
      "reward": 0.65625,
      "reward_std": 0.1293872892856598,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.65625,
      "step": 410
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 174.6875,
      "epoch": 1.6440000000000001,
      "grad_norm": 0.057538438588380814,
      "kl": 0.41748046875,
      "learning_rate": 8.356e-07,
      "loss": 0.0004,
      "reward": 0.25,
      "reward_std": 0.0,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.25,
      "step": 411
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 165.0,
      "epoch": 1.6480000000000001,
      "grad_norm": 4.462127208709717,
      "kl": 0.4384765625,
      "learning_rate": 8.352000000000001e-07,
      "loss": 0.0004,
      "reward": 0.0625,
      "reward_std": 0.1157275140285492,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.0625,
      "step": 412
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 157.65625,
      "epoch": 1.6520000000000001,
      "grad_norm": 5.903091907501221,
      "kl": 0.44580078125,
      "learning_rate": 8.347999999999999e-07,
      "loss": 0.0004,
      "reward": 0.5625,
      "reward_std": 0.1157275140285492,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.5625,
      "step": 413
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 164.71875,
      "epoch": 1.6560000000000001,
      "grad_norm": 3.8491811752319336,
      "kl": 0.4384765625,
      "learning_rate": 8.344e-07,
      "loss": 0.0004,
      "reward": 0.59375,
      "reward_std": 0.22201896458864212,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.59375,
      "step": 414
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 167.71875,
      "epoch": 1.6600000000000001,
      "grad_norm": 3.3095595836639404,
      "kl": 0.54345703125,
      "learning_rate": 8.34e-07,
      "loss": 0.0005,
      "reward": 0.09375,
      "reward_std": 0.1293872892856598,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.09375,
      "step": 415
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 164.65625,
      "epoch": 1.6640000000000001,
      "grad_norm": 0.035175856202840805,
      "kl": 0.41015625,
      "learning_rate": 8.335999999999999e-07,
      "loss": 0.0004,
      "reward": 0.75,
      "reward_std": 0.0,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.75,
      "step": 416
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 165.71875,
      "epoch": 1.6680000000000001,
      "grad_norm": 0.03981241583824158,
      "kl": 0.40087890625,
      "learning_rate": 8.332e-07,
      "loss": 0.0004,
      "reward": 0.75,
      "reward_std": 0.0,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.75,
      "step": 417
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 169.75,
      "epoch": 1.6720000000000002,
      "grad_norm": 2.5092711448669434,
      "kl": 0.42333984375,
      "learning_rate": 8.328e-07,
      "loss": 0.0004,
      "reward": 0.375,
      "reward_std": 0.13363061845302582,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.375,
      "step": 418
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 163.78125,
      "epoch": 1.6760000000000002,
      "grad_norm": 5.321307182312012,
      "kl": 0.45068359375,
      "learning_rate": 8.324e-07,
      "loss": 0.0005,
      "reward": 0.5625,
      "reward_std": 0.2177756354212761,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.5625,
      "step": 419
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 163.4375,
      "epoch": 1.6800000000000002,
      "grad_norm": 0.02313672937452793,
      "kl": 0.42529296875,
      "learning_rate": 8.319999999999999e-07,
      "loss": 0.0004,
      "reward": 1.0,
      "reward_std": 0.0,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 1.0,
      "step": 420
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 170.375,
      "epoch": 1.6840000000000002,
      "grad_norm": 4.892299175262451,
      "kl": 0.38623046875,
      "learning_rate": 8.316e-07,
      "loss": 0.0004,
      "reward": 0.5625,
      "reward_std": 0.2177756354212761,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.5625,
      "step": 421
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 171.0,
      "epoch": 1.688,
      "grad_norm": 4.429207801818848,
      "kl": 0.3623046875,
      "learning_rate": 8.312e-07,
      "loss": 0.0004,
      "reward": 0.65625,
      "reward_std": 0.22201896458864212,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.65625,
      "step": 422
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 163.34375,
      "epoch": 1.692,
      "grad_norm": 5.539219379425049,
      "kl": 0.41748046875,
      "learning_rate": 8.308e-07,
      "loss": 0.0004,
      "reward": 0.5,
      "reward_std": 0.1767766922712326,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.5,
      "step": 423
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 164.5625,
      "epoch": 1.696,
      "grad_norm": 15.078441619873047,
      "kl": 0.41650390625,
      "learning_rate": 8.304e-07,
      "loss": 0.0004,
      "reward": 0.21875,
      "reward_std": 0.0883883461356163,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.21875,
      "step": 424
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 160.84375,
      "epoch": 1.7,
      "grad_norm": 2.2660272121429443,
      "kl": 0.375,
      "learning_rate": 8.299999999999999e-07,
      "loss": 0.0004,
      "reward": 0.125,
      "reward_std": 0.13363061845302582,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.125,
      "step": 425
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 158.46875,
      "epoch": 1.704,
      "grad_norm": 2.217654228210449,
      "kl": 0.3955078125,
      "learning_rate": 8.296e-07,
      "loss": 0.0004,
      "reward": 0.78125,
      "reward_std": 0.0883883461356163,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.78125,
      "step": 426
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 161.90625,
      "epoch": 1.708,
      "grad_norm": 16.062503814697266,
      "kl": 0.49462890625,
      "learning_rate": 8.292e-07,
      "loss": 0.0005,
      "reward": 0.15625,
      "reward_std": 0.22201896458864212,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.15625,
      "step": 427
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 155.15625,
      "epoch": 1.712,
      "grad_norm": 2.677586317062378,
      "kl": 0.43212890625,
      "learning_rate": 8.287999999999999e-07,
      "loss": 0.0004,
      "reward": 0.28125,
      "reward_std": 0.0883883461356163,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.28125,
      "step": 428
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 167.125,
      "epoch": 1.716,
      "grad_norm": 14.95690631866455,
      "kl": 0.42626953125,
      "learning_rate": 8.284e-07,
      "loss": 0.0004,
      "reward": 0.5,
      "reward_std": 0.1767766922712326,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.5,
      "step": 429
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 162.0,
      "epoch": 1.72,
      "grad_norm": 0.026368090882897377,
      "kl": 0.42333984375,
      "learning_rate": 8.28e-07,
      "loss": 0.0004,
      "reward": 0.5,
      "reward_std": 0.0,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.5,
      "step": 430
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 163.75,
      "epoch": 1.724,
      "grad_norm": 2.9461612701416016,
      "kl": 0.45849609375,
      "learning_rate": 8.275999999999999e-07,
      "loss": 0.0005,
      "reward": 0.28125,
      "reward_std": 0.0883883461356163,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.28125,
      "step": 431
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 167.15625,
      "epoch": 1.728,
      "grad_norm": 0.02435384877026081,
      "kl": 0.41357421875,
      "learning_rate": 8.272e-07,
      "loss": 0.0004,
      "reward": 0.25,
      "reward_std": 0.0,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.25,
      "step": 432
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 160.8125,
      "epoch": 1.732,
      "grad_norm": 0.02652944251894951,
      "kl": 0.41015625,
      "learning_rate": 8.268e-07,
      "loss": 0.0004,
      "reward": 0.25,
      "reward_std": 0.0,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.25,
      "step": 433
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 154.90625,
      "epoch": 1.736,
      "grad_norm": 6.1886115074157715,
      "kl": 0.4853515625,
      "learning_rate": 8.263999999999999e-07,
      "loss": 0.0005,
      "reward": 0.21875,
      "reward_std": 0.0883883461356163,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.21875,
      "step": 434
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 157.875,
      "epoch": 1.74,
      "grad_norm": 13.261157989501953,
      "kl": 0.45166015625,
      "learning_rate": 8.259999999999999e-07,
      "loss": 0.0005,
      "reward": 0.3125,
      "reward_std": 0.3514062538743019,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.3125,
      "step": 435
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 160.25,
      "epoch": 1.744,
      "grad_norm": 1.860460877418518,
      "kl": 0.40966796875,
      "learning_rate": 8.256e-07,
      "loss": 0.0004,
      "reward": 0.15625,
      "reward_std": 0.1293872892856598,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.15625,
      "step": 436
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 158.125,
      "epoch": 1.748,
      "grad_norm": 0.02234129048883915,
      "kl": 0.4365234375,
      "learning_rate": 8.252000000000001e-07,
      "loss": 0.0004,
      "reward": 0.25,
      "reward_std": 0.0,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.25,
      "step": 437
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 150.09375,
      "epoch": 1.752,
      "grad_norm": 0.08076012879610062,
      "kl": 0.4140625,
      "learning_rate": 8.247999999999999e-07,
      "loss": 0.0004,
      "reward": 0.0,
      "reward_std": 0.0,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.0,
      "step": 438
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 167.28125,
      "epoch": 1.756,
      "grad_norm": 2.5461416244506836,
      "kl": 0.35107421875,
      "learning_rate": 8.244e-07,
      "loss": 0.0004,
      "reward": 0.3125,
      "reward_std": 0.1157275140285492,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.3125,
      "step": 439
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 162.25,
      "epoch": 1.76,
      "grad_norm": 4.5673346519470215,
      "kl": 0.41455078125,
      "learning_rate": 8.24e-07,
      "loss": 0.0004,
      "reward": 0.46875,
      "reward_std": 0.0883883461356163,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.46875,
      "step": 440
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 164.71875,
      "epoch": 1.764,
      "grad_norm": 10.48403549194336,
      "kl": 0.37548828125,
      "learning_rate": 8.235999999999999e-07,
      "loss": 0.0004,
      "reward": 0.40625,
      "reward_std": 0.1293872892856598,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.40625,
      "step": 441
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 170.4375,
      "epoch": 1.768,
      "grad_norm": 0.018004318699240685,
      "kl": 0.376953125,
      "learning_rate": 8.232e-07,
      "loss": 0.0004,
      "reward": 0.25,
      "reward_std": 0.0,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.25,
      "step": 442
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 163.53125,
      "epoch": 1.772,
      "grad_norm": 0.020243454724550247,
      "kl": 0.4091796875,
      "learning_rate": 8.228e-07,
      "loss": 0.0004,
      "reward": 0.5,
      "reward_std": 0.0,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.5,
      "step": 443
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 163.1875,
      "epoch": 1.776,
      "grad_norm": 0.02688312903046608,
      "kl": 0.42138671875,
      "learning_rate": 8.224e-07,
      "loss": 0.0004,
      "reward": 0.75,
      "reward_std": 0.0,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.75,
      "step": 444
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 170.4375,
      "epoch": 1.78,
      "grad_norm": 8.352977752685547,
      "kl": 0.396484375,
      "learning_rate": 8.219999999999999e-07,
      "loss": 0.0004,
      "reward": 0.34375,
      "reward_std": 0.1293872892856598,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.34375,
      "step": 445
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 164.84375,
      "epoch": 1.784,
      "grad_norm": 0.057273704558610916,
      "kl": 0.40283203125,
      "learning_rate": 8.216e-07,
      "loss": 0.0004,
      "reward": 1.0,
      "reward_std": 0.0,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 1.0,
      "step": 446
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 166.84375,
      "epoch": 1.788,
      "grad_norm": 1.9381979703903198,
      "kl": 0.3701171875,
      "learning_rate": 8.212e-07,
      "loss": 0.0004,
      "reward": 0.4375,
      "reward_std": 0.1157275140285492,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.4375,
      "step": 447
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 159.6875,
      "epoch": 1.792,
      "grad_norm": 7.398300647735596,
      "kl": 0.37939453125,
      "learning_rate": 8.207999999999999e-07,
      "loss": 0.0004,
      "reward": 0.34375,
      "reward_std": 0.24511480331420898,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.34375,
      "step": 448
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 170.59375,
      "epoch": 1.796,
      "grad_norm": 5.741900444030762,
      "kl": 0.4267578125,
      "learning_rate": 8.204e-07,
      "loss": 0.0004,
      "reward": 0.46875,
      "reward_std": 0.2041158601641655,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.46875,
      "step": 449
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 178.9375,
      "epoch": 1.8,
      "grad_norm": 0.04533645510673523,
      "kl": 0.3720703125,
      "learning_rate": 8.199999999999999e-07,
      "loss": 0.0004,
      "reward": 0.5,
      "reward_std": 0.0,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.5,
      "step": 450
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 183.4375,
      "epoch": 1.804,
      "grad_norm": 2.4087061882019043,
      "kl": 0.58642578125,
      "learning_rate": 8.196e-07,
      "loss": 0.0006,
      "reward": 0.03125,
      "reward_std": 0.0883883461356163,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.03125,
      "step": 451
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 176.4375,
      "epoch": 1.808,
      "grad_norm": 5.901292324066162,
      "kl": 0.3544921875,
      "learning_rate": 8.192e-07,
      "loss": 0.0004,
      "reward": 0.96875,
      "reward_std": 0.0883883461356163,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.96875,
      "step": 452
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 172.03125,
      "epoch": 1.812,
      "grad_norm": 0.013622630387544632,
      "kl": 0.3583984375,
      "learning_rate": 8.187999999999999e-07,
      "loss": 0.0004,
      "reward": 0.5,
      "reward_std": 0.0,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.5,
      "step": 453
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 185.15625,
      "epoch": 1.8159999999999998,
      "grad_norm": 13.24351692199707,
      "kl": 0.35791015625,
      "learning_rate": 8.184e-07,
      "loss": 0.0004,
      "reward": 0.46875,
      "reward_std": 0.3198433741927147,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.46875,
      "step": 454
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 180.03125,
      "epoch": 1.8199999999999998,
      "grad_norm": 9.442585945129395,
      "kl": 0.3681640625,
      "learning_rate": 8.179999999999999e-07,
      "loss": 0.0004,
      "reward": 0.6875,
      "reward_std": 0.2925042062997818,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.6875,
      "step": 455
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 182.25,
      "epoch": 1.8239999999999998,
      "grad_norm": 19.553857803344727,
      "kl": 0.32421875,
      "learning_rate": 8.175999999999999e-07,
      "loss": 0.0003,
      "reward": 0.375,
      "reward_std": 0.2177756354212761,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.375,
      "step": 456
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 165.625,
      "epoch": 1.8279999999999998,
      "grad_norm": 3.0148234367370605,
      "kl": 0.3466796875,
      "learning_rate": 8.172e-07,
      "loss": 0.0003,
      "reward": 0.3125,
      "reward_std": 0.2177756354212761,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.3125,
      "step": 457
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 169.5,
      "epoch": 1.8319999999999999,
      "grad_norm": 3.1484971046447754,
      "kl": 0.36279296875,
      "learning_rate": 8.168e-07,
      "loss": 0.0004,
      "reward": 0.625,
      "reward_std": 0.13363061845302582,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.625,
      "step": 458
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 190.34375,
      "epoch": 1.8359999999999999,
      "grad_norm": 2.9252376556396484,
      "kl": 0.37353515625,
      "learning_rate": 8.163999999999999e-07,
      "loss": 0.0004,
      "reward": 0.65625,
      "reward_std": 0.2041158601641655,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.65625,
      "step": 459
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 191.3125,
      "epoch": 1.8399999999999999,
      "grad_norm": 4.039405345916748,
      "kl": 0.30712890625,
      "learning_rate": 8.159999999999999e-07,
      "loss": 0.0003,
      "reward": 0.40625,
      "reward_std": 0.1293872892856598,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.40625,
      "step": 460
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 182.84375,
      "epoch": 1.8439999999999999,
      "grad_norm": 3.5302557945251465,
      "kl": 0.34228515625,
      "learning_rate": 8.156e-07,
      "loss": 0.0003,
      "reward": 0.5625,
      "reward_std": 0.2177756354212761,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.5625,
      "step": 461
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 178.625,
      "epoch": 1.8479999999999999,
      "grad_norm": 2.647287607192993,
      "kl": 0.38916015625,
      "learning_rate": 8.152e-07,
      "loss": 0.0004,
      "reward": 0.40625,
      "reward_std": 0.1293872892856598,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.40625,
      "step": 462
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 177.34375,
      "epoch": 1.8519999999999999,
      "grad_norm": 3.616725206375122,
      "kl": 0.41162109375,
      "learning_rate": 8.147999999999999e-07,
      "loss": 0.0004,
      "reward": 0.71875,
      "reward_std": 0.0883883461356163,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.71875,
      "step": 463
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 191.34375,
      "epoch": 1.8559999999999999,
      "grad_norm": 26.745315551757812,
      "kl": 0.359375,
      "learning_rate": 8.144e-07,
      "loss": 0.0004,
      "reward": 0.46875,
      "reward_std": 0.2041158601641655,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.46875,
      "step": 464
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 180.84375,
      "epoch": 1.8599999999999999,
      "grad_norm": 7.421332359313965,
      "kl": 0.3837890625,
      "learning_rate": 8.14e-07,
      "loss": 0.0004,
      "reward": 0.375,
      "reward_std": 0.2177756354212761,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.375,
      "step": 465
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 170.96875,
      "epoch": 1.8639999999999999,
      "grad_norm": 5.073638439178467,
      "kl": 0.365234375,
      "learning_rate": 8.135999999999999e-07,
      "loss": 0.0004,
      "reward": 0.59375,
      "reward_std": 0.3061639815568924,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.59375,
      "step": 466
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 174.09375,
      "epoch": 1.8679999999999999,
      "grad_norm": 4.215259552001953,
      "kl": 0.341796875,
      "learning_rate": 8.132e-07,
      "loss": 0.0003,
      "reward": 0.59375,
      "reward_std": 0.22201896458864212,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.59375,
      "step": 467
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 183.46875,
      "epoch": 1.8719999999999999,
      "grad_norm": 30.839479446411133,
      "kl": 0.345703125,
      "learning_rate": 8.128e-07,
      "loss": 0.0003,
      "reward": 0.21875,
      "reward_std": 0.0883883461356163,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.21875,
      "step": 468
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 163.125,
      "epoch": 1.876,
      "grad_norm": 3.3285722732543945,
      "kl": 0.38720703125,
      "learning_rate": 8.123999999999999e-07,
      "loss": 0.0004,
      "reward": 0.65625,
      "reward_std": 0.1293872892856598,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.65625,
      "step": 469
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 158.03125,
      "epoch": 1.88,
      "grad_norm": 0.04166668653488159,
      "kl": 0.458984375,
      "learning_rate": 8.12e-07,
      "loss": 0.0005,
      "reward": 0.25,
      "reward_std": 0.0,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.25,
      "step": 470
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 164.84375,
      "epoch": 1.884,
      "grad_norm": 6.062572479248047,
      "kl": 0.40771484375,
      "learning_rate": 8.116e-07,
      "loss": 0.0004,
      "reward": 0.09375,
      "reward_std": 0.2041158601641655,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.09375,
      "step": 471
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 156.3125,
      "epoch": 1.888,
      "grad_norm": 9.836076736450195,
      "kl": 0.36474609375,
      "learning_rate": 8.112e-07,
      "loss": 0.0004,
      "reward": 0.6875,
      "reward_std": 0.1767766922712326,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.6875,
      "step": 472
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 165.4375,
      "epoch": 1.892,
      "grad_norm": 2.3117265701293945,
      "kl": 0.38330078125,
      "learning_rate": 8.107999999999999e-07,
      "loss": 0.0004,
      "reward": 0.34375,
      "reward_std": 0.1293872892856598,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.34375,
      "step": 473
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 165.1875,
      "epoch": 1.896,
      "grad_norm": 0.01977263018488884,
      "kl": 0.39892578125,
      "learning_rate": 8.104e-07,
      "loss": 0.0004,
      "reward": 0.25,
      "reward_std": 0.0,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.25,
      "step": 474
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 150.375,
      "epoch": 1.9,
      "grad_norm": 7.866539001464844,
      "kl": 0.48486328125,
      "learning_rate": 8.1e-07,
      "loss": 0.0005,
      "reward": 0.46875,
      "reward_std": 0.0883883461356163,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.46875,
      "step": 475
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 164.0625,
      "epoch": 1.904,
      "grad_norm": 3.914083242416382,
      "kl": 0.3837890625,
      "learning_rate": 8.095999999999999e-07,
      "loss": 0.0004,
      "reward": 0.28125,
      "reward_std": 0.2041158601641655,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.28125,
      "step": 476
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 151.4375,
      "epoch": 1.908,
      "grad_norm": 3.30718994140625,
      "kl": 0.537109375,
      "learning_rate": 8.092e-07,
      "loss": 0.0005,
      "reward": 0.28125,
      "reward_std": 0.0883883461356163,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.28125,
      "step": 477
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 165.25,
      "epoch": 1.912,
      "grad_norm": 1.9141907691955566,
      "kl": 0.37841796875,
      "learning_rate": 8.087999999999999e-07,
      "loss": 0.0004,
      "reward": 0.71875,
      "reward_std": 0.0883883461356163,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.71875,
      "step": 478
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 152.0625,
      "epoch": 1.916,
      "grad_norm": 3.471940279006958,
      "kl": 0.42333984375,
      "learning_rate": 8.084e-07,
      "loss": 0.0004,
      "reward": 0.53125,
      "reward_std": 0.0883883461356163,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.53125,
      "step": 479
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 166.78125,
      "epoch": 1.92,
      "grad_norm": 0.02734890766441822,
      "kl": 0.3984375,
      "learning_rate": 8.08e-07,
      "loss": 0.0004,
      "reward": 0.25,
      "reward_std": 0.0,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.25,
      "step": 480
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 155.5625,
      "epoch": 1.924,
      "grad_norm": 2.8940510749816895,
      "kl": 0.408203125,
      "learning_rate": 8.075999999999999e-07,
      "loss": 0.0004,
      "reward": 0.28125,
      "reward_std": 0.0883883461356163,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.28125,
      "step": 481
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 152.875,
      "epoch": 1.928,
      "grad_norm": 6.591331481933594,
      "kl": 0.36474609375,
      "learning_rate": 8.072e-07,
      "loss": 0.0004,
      "reward": 0.34375,
      "reward_std": 0.2041158601641655,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.34375,
      "step": 482
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 163.28125,
      "epoch": 1.932,
      "grad_norm": 0.044952016323804855,
      "kl": 0.36328125,
      "learning_rate": 8.067999999999999e-07,
      "loss": 0.0004,
      "reward": 0.25,
      "reward_std": 0.0,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.25,
      "step": 483
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 160.0,
      "epoch": 1.936,
      "grad_norm": 2.8192989826202393,
      "kl": 0.423828125,
      "learning_rate": 8.064e-07,
      "loss": 0.0004,
      "reward": 0.96875,
      "reward_std": 0.0883883461356163,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.96875,
      "step": 484
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 145.90625,
      "epoch": 1.94,
      "grad_norm": 0.05787898600101471,
      "kl": 0.421875,
      "learning_rate": 8.06e-07,
      "loss": 0.0004,
      "reward": 0.25,
      "reward_std": 0.0,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.25,
      "step": 485
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 149.84375,
      "epoch": 1.944,
      "grad_norm": 6.145750045776367,
      "kl": 0.50146484375,
      "learning_rate": 8.056e-07,
      "loss": 0.0005,
      "reward": 0.4375,
      "reward_std": 0.2177756354212761,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.4375,
      "step": 486
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 154.90625,
      "epoch": 1.948,
      "grad_norm": 2.458923578262329,
      "kl": 0.37451171875,
      "learning_rate": 8.052e-07,
      "loss": 0.0004,
      "reward": 0.34375,
      "reward_std": 0.1293872892856598,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.34375,
      "step": 487
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 165.46875,
      "epoch": 1.952,
      "grad_norm": 2.168976306915283,
      "kl": 0.390625,
      "learning_rate": 8.047999999999999e-07,
      "loss": 0.0004,
      "reward": 0.40625,
      "reward_std": 0.1293872892856598,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.40625,
      "step": 488
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 167.3125,
      "epoch": 1.956,
      "grad_norm": 5.698536396026611,
      "kl": 0.408203125,
      "learning_rate": 8.044e-07,
      "loss": 0.0004,
      "reward": 0.65625,
      "reward_std": 0.2041158601641655,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.65625,
      "step": 489
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 161.0,
      "epoch": 1.96,
      "grad_norm": 0.03852026164531708,
      "kl": 0.44873046875,
      "learning_rate": 8.04e-07,
      "loss": 0.0004,
      "reward": 0.5,
      "reward_std": 0.0,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.5,
      "step": 490
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 158.40625,
      "epoch": 1.964,
      "grad_norm": 0.02888845093548298,
      "kl": 0.4013671875,
      "learning_rate": 8.035999999999999e-07,
      "loss": 0.0004,
      "reward": 0.0,
      "reward_std": 0.0,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.0,
      "step": 491
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 180.625,
      "epoch": 1.968,
      "grad_norm": 2.9419054985046387,
      "kl": 0.53662109375,
      "learning_rate": 8.032e-07,
      "loss": 0.0005,
      "reward": 0.4375,
      "reward_std": 0.1157275140285492,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.4375,
      "step": 492
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 161.75,
      "epoch": 1.972,
      "grad_norm": 2.9988346099853516,
      "kl": 0.3994140625,
      "learning_rate": 8.028e-07,
      "loss": 0.0004,
      "reward": 0.46875,
      "reward_std": 0.0883883461356163,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.46875,
      "step": 493
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 171.875,
      "epoch": 1.976,
      "grad_norm": 0.023221736773848534,
      "kl": 0.4091796875,
      "learning_rate": 8.023999999999999e-07,
      "loss": 0.0004,
      "reward": 0.0,
      "reward_std": 0.0,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.0,
      "step": 494
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 175.125,
      "epoch": 1.98,
      "grad_norm": 0.026831550523638725,
      "kl": 0.3896484375,
      "learning_rate": 8.02e-07,
      "loss": 0.0004,
      "reward": 0.25,
      "reward_std": 0.0,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.25,
      "step": 495
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 176.75,
      "epoch": 1.984,
      "grad_norm": 1.9323009252548218,
      "kl": 0.39208984375,
      "learning_rate": 8.016e-07,
      "loss": 0.0004,
      "reward": 0.46875,
      "reward_std": 0.0883883461356163,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.46875,
      "step": 496
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 179.9375,
      "epoch": 1.988,
      "grad_norm": 2.6626555919647217,
      "kl": 0.37744140625,
      "learning_rate": 8.012e-07,
      "loss": 0.0004,
      "reward": 0.4375,
      "reward_std": 0.1157275140285492,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.4375,
      "step": 497
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 168.875,
      "epoch": 1.992,
      "grad_norm": 5.306944847106934,
      "kl": 0.34814453125,
      "learning_rate": 8.007999999999999e-07,
      "loss": 0.0003,
      "reward": 0.59375,
      "reward_std": 0.1293872892856598,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.59375,
      "step": 498
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 183.96875,
      "epoch": 1.996,
      "grad_norm": 0.02336149290204048,
      "kl": 0.41064453125,
      "learning_rate": 8.004e-07,
      "loss": 0.0004,
      "reward": 0.25,
      "reward_std": 0.0,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.25,
      "step": 499
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 175.15625,
      "epoch": 2.0,
      "grad_norm": 5.964815616607666,
      "kl": 0.4072265625,
      "learning_rate": 8e-07,
      "loss": 0.0004,
      "reward": 0.4375,
      "reward_std": 0.1767766922712326,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.4375,
      "step": 500
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 180.6875,
      "epoch": 2.004,
      "grad_norm": 3.620058298110962,
      "kl": 0.40673828125,
      "learning_rate": 7.995999999999999e-07,
      "loss": 0.0004,
      "reward": 0.71875,
      "reward_std": 0.0883883461356163,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.71875,
      "step": 501
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 162.90625,
      "epoch": 2.008,
      "grad_norm": 3.434760093688965,
      "kl": 0.3681640625,
      "learning_rate": 7.992e-07,
      "loss": 0.0004,
      "reward": 0.375,
      "reward_std": 0.2177756354212761,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.375,
      "step": 502
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 177.78125,
      "epoch": 2.012,
      "grad_norm": 1.9072635173797607,
      "kl": 0.3291015625,
      "learning_rate": 7.987999999999999e-07,
      "loss": 0.0003,
      "reward": 0.3125,
      "reward_std": 0.1157275140285492,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.3125,
      "step": 503
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 168.0625,
      "epoch": 2.016,
      "grad_norm": 5.137371063232422,
      "kl": 0.408203125,
      "learning_rate": 7.984e-07,
      "loss": 0.0004,
      "reward": 0.6875,
      "reward_std": 0.1157275140285492,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.6875,
      "step": 504
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 163.9375,
      "epoch": 2.02,
      "grad_norm": 7.093776226043701,
      "kl": 0.41357421875,
      "learning_rate": 7.98e-07,
      "loss": 0.0004,
      "reward": 0.65625,
      "reward_std": 0.3787454217672348,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.65625,
      "step": 505
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 175.0625,
      "epoch": 2.024,
      "grad_norm": 2.0226447582244873,
      "kl": 0.4013671875,
      "learning_rate": 7.975999999999999e-07,
      "loss": 0.0004,
      "reward": 0.21875,
      "reward_std": 0.0883883461356163,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.21875,
      "step": 506
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 166.28125,
      "epoch": 2.028,
      "grad_norm": 3.9577460289001465,
      "kl": 0.39599609375,
      "learning_rate": 7.972e-07,
      "loss": 0.0004,
      "reward": 0.53125,
      "reward_std": 0.0883883461356163,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.53125,
      "step": 507
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 179.09375,
      "epoch": 2.032,
      "grad_norm": 0.051929961889982224,
      "kl": 0.4296875,
      "learning_rate": 7.967999999999999e-07,
      "loss": 0.0004,
      "reward": 0.25,
      "reward_std": 0.0,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.25,
      "step": 508
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 162.875,
      "epoch": 2.036,
      "grad_norm": 5.923417568206787,
      "kl": 0.39111328125,
      "learning_rate": 7.964e-07,
      "loss": 0.0004,
      "reward": 0.3125,
      "reward_std": 0.249358132481575,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.3125,
      "step": 509
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 157.9375,
      "epoch": 2.04,
      "grad_norm": 10.605561256408691,
      "kl": 0.3994140625,
      "learning_rate": 7.96e-07,
      "loss": 0.0004,
      "reward": 0.4375,
      "reward_std": 0.2925042062997818,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.4375,
      "step": 510
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 168.875,
      "epoch": 2.044,
      "grad_norm": 0.1380130797624588,
      "kl": 0.42333984375,
      "learning_rate": 7.956e-07,
      "loss": 0.0004,
      "reward": 0.75,
      "reward_std": 0.0,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.75,
      "step": 511
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 185.1875,
      "epoch": 2.048,
      "grad_norm": 6.401873588562012,
      "kl": 0.46826171875,
      "learning_rate": 7.952e-07,
      "loss": 0.0005,
      "reward": 0.46875,
      "reward_std": 0.0883883461356163,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.46875,
      "step": 512
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 169.96875,
      "epoch": 2.052,
      "grad_norm": 6.9412713050842285,
      "kl": 0.458984375,
      "learning_rate": 7.947999999999999e-07,
      "loss": 0.0005,
      "reward": 0.3125,
      "reward_std": 0.1157275140285492,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.3125,
      "step": 513
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 160.40625,
      "epoch": 2.056,
      "grad_norm": 2.1416547298431396,
      "kl": 0.42041015625,
      "learning_rate": 7.944e-07,
      "loss": 0.0004,
      "reward": 0.53125,
      "reward_std": 0.0883883461356163,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.53125,
      "step": 514
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 166.5625,
      "epoch": 2.06,
      "grad_norm": 0.03822876140475273,
      "kl": 0.3828125,
      "learning_rate": 7.94e-07,
      "loss": 0.0004,
      "reward": 0.5,
      "reward_std": 0.0,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.5,
      "step": 515
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 162.25,
      "epoch": 2.064,
      "grad_norm": 0.15952487289905548,
      "kl": 0.51953125,
      "learning_rate": 7.935999999999999e-07,
      "loss": 0.0005,
      "reward": 0.25,
      "reward_std": 0.0,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.25,
      "step": 516
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 169.5,
      "epoch": 2.068,
      "grad_norm": 5.158685207366943,
      "kl": 0.40087890625,
      "learning_rate": 7.932e-07,
      "loss": 0.0004,
      "reward": 0.4375,
      "reward_std": 0.249358132481575,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.4375,
      "step": 517
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 176.15625,
      "epoch": 2.072,
      "grad_norm": 4.253637313842773,
      "kl": 0.40673828125,
      "learning_rate": 7.928e-07,
      "loss": 0.0004,
      "reward": 0.4375,
      "reward_std": 0.249358132481575,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.4375,
      "step": 518
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 168.375,
      "epoch": 2.076,
      "grad_norm": 5.6791791915893555,
      "kl": 0.46435546875,
      "learning_rate": 7.923999999999999e-07,
      "loss": 0.0005,
      "reward": 0.46875,
      "reward_std": 0.0883883461356163,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.46875,
      "step": 519
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 174.71875,
      "epoch": 2.08,
      "grad_norm": 2.477609634399414,
      "kl": 0.4140625,
      "learning_rate": 7.92e-07,
      "loss": 0.0004,
      "reward": 0.46875,
      "reward_std": 0.0883883461356163,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.46875,
      "step": 520
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 170.25,
      "epoch": 2.084,
      "grad_norm": 0.023585015907883644,
      "kl": 0.39892578125,
      "learning_rate": 7.916e-07,
      "loss": 0.0004,
      "reward": 0.75,
      "reward_std": 0.0,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.75,
      "step": 521
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 185.21875,
      "epoch": 2.088,
      "grad_norm": 0.03890880569815636,
      "kl": 0.412109375,
      "learning_rate": 7.911999999999999e-07,
      "loss": 0.0004,
      "reward": 0.25,
      "reward_std": 0.0,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.25,
      "step": 522
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 171.0,
      "epoch": 2.092,
      "grad_norm": 0.06995756179094315,
      "kl": 0.46923828125,
      "learning_rate": 7.907999999999999e-07,
      "loss": 0.0005,
      "reward": 0.75,
      "reward_std": 0.0,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.75,
      "step": 523
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 180.03125,
      "epoch": 2.096,
      "grad_norm": 2.6263272762298584,
      "kl": 0.39501953125,
      "learning_rate": 7.904e-07,
      "loss": 0.0004,
      "reward": 0.28125,
      "reward_std": 0.0883883461356163,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.28125,
      "step": 524
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 178.46875,
      "epoch": 2.1,
      "grad_norm": 4.536235332489014,
      "kl": 0.4228515625,
      "learning_rate": 7.9e-07,
      "loss": 0.0004,
      "reward": 0.78125,
      "reward_std": 0.0883883461356163,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.78125,
      "step": 525
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 166.71875,
      "epoch": 2.104,
      "grad_norm": 0.024363024160265923,
      "kl": 0.41064453125,
      "learning_rate": 7.895999999999999e-07,
      "loss": 0.0004,
      "reward": 0.5,
      "reward_std": 0.0,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.5,
      "step": 526
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 166.5,
      "epoch": 2.108,
      "grad_norm": 2.463157892227173,
      "kl": 0.37353515625,
      "learning_rate": 7.892e-07,
      "loss": 0.0004,
      "reward": 0.5625,
      "reward_std": 0.1157275140285492,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.5625,
      "step": 527
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 177.75,
      "epoch": 2.112,
      "grad_norm": 0.023945072665810585,
      "kl": 0.44921875,
      "learning_rate": 7.887999999999999e-07,
      "loss": 0.0004,
      "reward": 0.25,
      "reward_std": 0.0,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.25,
      "step": 528
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 168.625,
      "epoch": 2.116,
      "grad_norm": 13.871514320373535,
      "kl": 0.41943359375,
      "learning_rate": 7.883999999999999e-07,
      "loss": 0.0004,
      "reward": 0.28125,
      "reward_std": 0.0883883461356163,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.28125,
      "step": 529
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 170.65625,
      "epoch": 2.12,
      "grad_norm": 6.096253395080566,
      "kl": 0.4150390625,
      "learning_rate": 7.88e-07,
      "loss": 0.0004,
      "reward": 0.625,
      "reward_std": 0.13363061845302582,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.625,
      "step": 530
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 172.65625,
      "epoch": 2.124,
      "grad_norm": 4.070682525634766,
      "kl": 0.40283203125,
      "learning_rate": 7.875999999999999e-07,
      "loss": 0.0004,
      "reward": 0.5625,
      "reward_std": 0.2177756354212761,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.5625,
      "step": 531
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 151.375,
      "epoch": 2.128,
      "grad_norm": 0.03539380058646202,
      "kl": 0.40869140625,
      "learning_rate": 7.872e-07,
      "loss": 0.0004,
      "reward": 0.5,
      "reward_std": 0.0,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.5,
      "step": 532
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 168.03125,
      "epoch": 2.132,
      "grad_norm": 4.45925235748291,
      "kl": 0.51171875,
      "learning_rate": 7.868e-07,
      "loss": 0.0005,
      "reward": 0.3125,
      "reward_std": 0.1157275140285492,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.3125,
      "step": 533
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 167.6875,
      "epoch": 2.136,
      "grad_norm": 0.03435595706105232,
      "kl": 0.4697265625,
      "learning_rate": 7.864e-07,
      "loss": 0.0005,
      "reward": 0.25,
      "reward_std": 0.0,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.25,
      "step": 534
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 157.34375,
      "epoch": 2.14,
      "grad_norm": 3.7830166816711426,
      "kl": 0.39306640625,
      "learning_rate": 7.86e-07,
      "loss": 0.0004,
      "reward": 0.625,
      "reward_std": 0.2177756354212761,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.625,
      "step": 535
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 146.75,
      "epoch": 2.144,
      "grad_norm": 0.031088387593626976,
      "kl": 0.4296875,
      "learning_rate": 7.855999999999999e-07,
      "loss": 0.0004,
      "reward": 0.0,
      "reward_std": 0.0,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.0,
      "step": 536
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 141.625,
      "epoch": 2.148,
      "grad_norm": 0.019815722480416298,
      "kl": 0.3515625,
      "learning_rate": 7.852e-07,
      "loss": 0.0004,
      "reward": 0.5,
      "reward_std": 0.0,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.5,
      "step": 537
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 160.75,
      "epoch": 2.152,
      "grad_norm": 0.025819187983870506,
      "kl": 0.40283203125,
      "learning_rate": 7.848e-07,
      "loss": 0.0004,
      "reward": 0.0,
      "reward_std": 0.0,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.0,
      "step": 538
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 169.9375,
      "epoch": 2.156,
      "grad_norm": 2.3410308361053467,
      "kl": 0.435546875,
      "learning_rate": 7.844e-07,
      "loss": 0.0004,
      "reward": 0.09375,
      "reward_std": 0.1293872892856598,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.09375,
      "step": 539
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 166.78125,
      "epoch": 2.16,
      "grad_norm": 0.019718121737241745,
      "kl": 0.38037109375,
      "learning_rate": 7.84e-07,
      "loss": 0.0004,
      "reward": 0.5,
      "reward_std": 0.0,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.5,
      "step": 540
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 157.0625,
      "epoch": 2.164,
      "grad_norm": 7.418305397033691,
      "kl": 1.94970703125,
      "learning_rate": 7.835999999999999e-07,
      "loss": 0.002,
      "reward": 0.71875,
      "reward_std": 0.0883883461356163,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.71875,
      "step": 541
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 158.03125,
      "epoch": 2.168,
      "grad_norm": 0.02193170040845871,
      "kl": 0.41650390625,
      "learning_rate": 7.832e-07,
      "loss": 0.0004,
      "reward": 0.5,
      "reward_std": 0.0,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.5,
      "step": 542
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 149.21875,
      "epoch": 2.172,
      "grad_norm": 3.54630446434021,
      "kl": 0.44091796875,
      "learning_rate": 7.828e-07,
      "loss": 0.0004,
      "reward": 0.03125,
      "reward_std": 0.0883883461356163,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.03125,
      "step": 543
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 152.03125,
      "epoch": 2.176,
      "grad_norm": 2.7670624256134033,
      "kl": 0.40234375,
      "learning_rate": 7.823999999999999e-07,
      "loss": 0.0004,
      "reward": 0.5625,
      "reward_std": 0.1157275140285492,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.5625,
      "step": 544
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 153.25,
      "epoch": 2.18,
      "grad_norm": 20.923410415649414,
      "kl": 0.41845703125,
      "learning_rate": 7.82e-07,
      "loss": 0.0004,
      "reward": 0.5,
      "reward_std": 0.3650856465101242,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.5,
      "step": 545
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 151.71875,
      "epoch": 2.184,
      "grad_norm": 0.06470020115375519,
      "kl": 0.392578125,
      "learning_rate": 7.816e-07,
      "loss": 0.0004,
      "reward": 0.75,
      "reward_std": 0.0,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.75,
      "step": 546
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 159.5625,
      "epoch": 2.188,
      "grad_norm": 2.351090908050537,
      "kl": 0.56787109375,
      "learning_rate": 7.811999999999999e-07,
      "loss": 0.0006,
      "reward": 0.28125,
      "reward_std": 0.0883883461356163,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.28125,
      "step": 547
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 148.6875,
      "epoch": 2.192,
      "grad_norm": 2.147427797317505,
      "kl": 0.4541015625,
      "learning_rate": 7.808e-07,
      "loss": 0.0005,
      "reward": 0.6875,
      "reward_std": 0.1157275140285492,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.6875,
      "step": 548
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 140.25,
      "epoch": 2.196,
      "grad_norm": 7.3476457595825195,
      "kl": 0.3994140625,
      "learning_rate": 7.804e-07,
      "loss": 0.0004,
      "reward": 0.21875,
      "reward_std": 0.2041158601641655,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.21875,
      "step": 549
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 152.4375,
      "epoch": 2.2,
      "grad_norm": 3.0128211975097656,
      "kl": 0.5068359375,
      "learning_rate": 7.799999999999999e-07,
      "loss": 0.0005,
      "reward": 0.28125,
      "reward_std": 0.0883883461356163,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.28125,
      "step": 550
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 158.34375,
      "epoch": 2.204,
      "grad_norm": 0.03397554159164429,
      "kl": 0.46142578125,
      "learning_rate": 7.795999999999999e-07,
      "loss": 0.0005,
      "reward": 0.25,
      "reward_std": 0.0,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.25,
      "step": 551
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 150.0625,
      "epoch": 2.208,
      "grad_norm": 0.02539452351629734,
      "kl": 0.45654296875,
      "learning_rate": 7.792e-07,
      "loss": 0.0005,
      "reward": 0.75,
      "reward_std": 0.0,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.75,
      "step": 552
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 153.375,
      "epoch": 2.212,
      "grad_norm": 24.509769439697266,
      "kl": 0.4794921875,
      "learning_rate": 7.788000000000001e-07,
      "loss": 0.0005,
      "reward": 0.65625,
      "reward_std": 0.1293872892856598,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.65625,
      "step": 553
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 156.96875,
      "epoch": 2.216,
      "grad_norm": 3.392266035079956,
      "kl": 0.45068359375,
      "learning_rate": 7.783999999999999e-07,
      "loss": 0.0005,
      "reward": 0.59375,
      "reward_std": 0.22201896458864212,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.59375,
      "step": 554
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 169.09375,
      "epoch": 2.22,
      "grad_norm": 3.5731542110443115,
      "kl": 0.7041015625,
      "learning_rate": 7.78e-07,
      "loss": 0.0007,
      "reward": 0.34375,
      "reward_std": 0.22201896458864212,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.34375,
      "step": 555
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 152.28125,
      "epoch": 2.224,
      "grad_norm": 7.183278560638428,
      "kl": 0.474609375,
      "learning_rate": 7.776e-07,
      "loss": 0.0005,
      "reward": 0.5625,
      "reward_std": 0.2177756354212761,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.5625,
      "step": 556
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 151.53125,
      "epoch": 2.228,
      "grad_norm": 55.40394973754883,
      "kl": 0.41796875,
      "learning_rate": 7.771999999999999e-07,
      "loss": 0.0004,
      "reward": 0.4375,
      "reward_std": 0.1157275140285492,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.4375,
      "step": 557
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 157.34375,
      "epoch": 2.232,
      "grad_norm": 0.03296584263443947,
      "kl": 0.39501953125,
      "learning_rate": 7.768e-07,
      "loss": 0.0004,
      "reward": 0.5,
      "reward_std": 0.0,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.5,
      "step": 558
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 154.9375,
      "epoch": 2.2359999999999998,
      "grad_norm": 2.9076058864593506,
      "kl": 0.45556640625,
      "learning_rate": 7.764e-07,
      "loss": 0.0005,
      "reward": 0.34375,
      "reward_std": 0.2041158601641655,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.34375,
      "step": 559
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 162.03125,
      "epoch": 2.24,
      "grad_norm": 2.430602550506592,
      "kl": 0.43896484375,
      "learning_rate": 7.76e-07,
      "loss": 0.0004,
      "reward": 0.9375,
      "reward_std": 0.1157275140285492,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.9375,
      "step": 560
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 162.34375,
      "epoch": 2.2439999999999998,
      "grad_norm": 3.7969930171966553,
      "kl": 0.4169921875,
      "learning_rate": 7.755999999999999e-07,
      "loss": 0.0004,
      "reward": 0.53125,
      "reward_std": 0.0883883461356163,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.53125,
      "step": 561
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 163.25,
      "epoch": 2.248,
      "grad_norm": 0.0564698651432991,
      "kl": 0.43017578125,
      "learning_rate": 7.752e-07,
      "loss": 0.0004,
      "reward": 0.25,
      "reward_std": 0.0,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.25,
      "step": 562
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 162.0,
      "epoch": 2.252,
      "grad_norm": 3.547658681869507,
      "kl": 0.43701171875,
      "learning_rate": 7.748e-07,
      "loss": 0.0004,
      "reward": 0.375,
      "reward_std": 0.13363061845302582,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.375,
      "step": 563
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 158.25,
      "epoch": 2.2560000000000002,
      "grad_norm": 3.529600143432617,
      "kl": 0.443359375,
      "learning_rate": 7.743999999999999e-07,
      "loss": 0.0004,
      "reward": 0.6875,
      "reward_std": 0.1157275140285492,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.6875,
      "step": 564
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 168.5,
      "epoch": 2.26,
      "grad_norm": 8.721165657043457,
      "kl": 0.44873046875,
      "learning_rate": 7.74e-07,
      "loss": 0.0004,
      "reward": 0.1875,
      "reward_std": 0.2177756354212761,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.1875,
      "step": 565
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 168.53125,
      "epoch": 2.2640000000000002,
      "grad_norm": 3.0871663093566895,
      "kl": 0.39697265625,
      "learning_rate": 7.735999999999999e-07,
      "loss": 0.0004,
      "reward": 0.28125,
      "reward_std": 0.0883883461356163,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.28125,
      "step": 566
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 167.46875,
      "epoch": 2.268,
      "grad_norm": 3.3945159912109375,
      "kl": 0.513671875,
      "learning_rate": 7.732e-07,
      "loss": 0.0005,
      "reward": 0.8125,
      "reward_std": 0.1157275140285492,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.8125,
      "step": 567
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 171.21875,
      "epoch": 2.2720000000000002,
      "grad_norm": 4.4708709716796875,
      "kl": 0.44970703125,
      "learning_rate": 7.728e-07,
      "loss": 0.0005,
      "reward": 0.59375,
      "reward_std": 0.22201896458864212,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.59375,
      "step": 568
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 163.34375,
      "epoch": 2.276,
      "grad_norm": 2.917856454849243,
      "kl": 0.45263671875,
      "learning_rate": 7.723999999999999e-07,
      "loss": 0.0005,
      "reward": 0.875,
      "reward_std": 0.13363061845302582,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.875,
      "step": 569
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 169.0625,
      "epoch": 2.2800000000000002,
      "grad_norm": 17.440567016601562,
      "kl": 0.48779296875,
      "learning_rate": 7.72e-07,
      "loss": 0.0005,
      "reward": 0.125,
      "reward_std": 0.2177756354212761,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.125,
      "step": 570
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 164.75,
      "epoch": 2.284,
      "grad_norm": 5.56218147277832,
      "kl": 0.48388671875,
      "learning_rate": 7.716e-07,
      "loss": 0.0005,
      "reward": 0.5,
      "reward_std": 0.2587745785713196,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.5,
      "step": 571
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 156.53125,
      "epoch": 2.288,
      "grad_norm": 2.8152010440826416,
      "kl": 0.416015625,
      "learning_rate": 7.711999999999999e-07,
      "loss": 0.0004,
      "reward": 0.53125,
      "reward_std": 0.0883883461356163,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.53125,
      "step": 572
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 160.75,
      "epoch": 2.292,
      "grad_norm": 3.292689085006714,
      "kl": 0.498046875,
      "learning_rate": 7.708e-07,
      "loss": 0.0005,
      "reward": 0.71875,
      "reward_std": 0.0883883461356163,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.71875,
      "step": 573
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 157.15625,
      "epoch": 2.296,
      "grad_norm": 8.015246391296387,
      "kl": 0.5244140625,
      "learning_rate": 7.704e-07,
      "loss": 0.0005,
      "reward": 0.6875,
      "reward_std": 0.1157275140285492,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.6875,
      "step": 574
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 168.5625,
      "epoch": 2.3,
      "grad_norm": 2.3412256240844727,
      "kl": 0.42919921875,
      "learning_rate": 7.699999999999999e-07,
      "loss": 0.0004,
      "reward": 0.96875,
      "reward_std": 0.0883883461356163,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.96875,
      "step": 575
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 162.84375,
      "epoch": 2.304,
      "grad_norm": 0.01675044745206833,
      "kl": 0.43212890625,
      "learning_rate": 7.695999999999999e-07,
      "loss": 0.0004,
      "reward": 0.75,
      "reward_std": 0.0,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.75,
      "step": 576
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 159.9375,
      "epoch": 2.308,
      "grad_norm": 4.7579193115234375,
      "kl": 0.38623046875,
      "learning_rate": 7.692e-07,
      "loss": 0.0004,
      "reward": 0.34375,
      "reward_std": 0.1293872892856598,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.34375,
      "step": 577
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 173.59375,
      "epoch": 2.312,
      "grad_norm": 5.626319885253906,
      "kl": 0.4697265625,
      "learning_rate": 7.688000000000001e-07,
      "loss": 0.0005,
      "reward": 0.21875,
      "reward_std": 0.3377464786171913,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.21875,
      "step": 578
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 158.8125,
      "epoch": 2.316,
      "grad_norm": 0.03451302647590637,
      "kl": 0.4462890625,
      "learning_rate": 7.683999999999999e-07,
      "loss": 0.0004,
      "reward": 0.25,
      "reward_std": 0.0,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.25,
      "step": 579
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 161.875,
      "epoch": 2.32,
      "grad_norm": 3.034019947052002,
      "kl": 0.45849609375,
      "learning_rate": 7.68e-07,
      "loss": 0.0005,
      "reward": 0.15625,
      "reward_std": 0.1293872892856598,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.15625,
      "step": 580
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 167.53125,
      "epoch": 2.324,
      "grad_norm": 0.054354228079319,
      "kl": 0.4072265625,
      "learning_rate": 7.676e-07,
      "loss": 0.0004,
      "reward": 0.5,
      "reward_std": 0.0,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.5,
      "step": 581
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 167.6875,
      "epoch": 2.328,
      "grad_norm": 0.024297289550304413,
      "kl": 0.43115234375,
      "learning_rate": 7.671999999999999e-07,
      "loss": 0.0004,
      "reward": 0.75,
      "reward_std": 0.0,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.75,
      "step": 582
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 150.75,
      "epoch": 2.332,
      "grad_norm": 0.015270624309778214,
      "kl": 0.3720703125,
      "learning_rate": 7.668e-07,
      "loss": 0.0004,
      "reward": 0.25,
      "reward_std": 0.0,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.25,
      "step": 583
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 186.34375,
      "epoch": 2.336,
      "grad_norm": 2.0129873752593994,
      "kl": 0.42626953125,
      "learning_rate": 7.664e-07,
      "loss": 0.0004,
      "reward": 0.71875,
      "reward_std": 0.0883883461356163,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.71875,
      "step": 584
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 151.625,
      "epoch": 2.34,
      "grad_norm": 3.7725656032562256,
      "kl": 0.3935546875,
      "learning_rate": 7.66e-07,
      "loss": 0.0004,
      "reward": 0.65625,
      "reward_std": 0.1293872892856598,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.65625,
      "step": 585
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 163.59375,
      "epoch": 2.344,
      "grad_norm": 0.020550815388560295,
      "kl": 0.419921875,
      "learning_rate": 7.655999999999999e-07,
      "loss": 0.0004,
      "reward": 0.5,
      "reward_std": 0.0,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.5,
      "step": 586
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 156.03125,
      "epoch": 2.348,
      "grad_norm": 0.029289547353982925,
      "kl": 0.427734375,
      "learning_rate": 7.652e-07,
      "loss": 0.0004,
      "reward": 0.0,
      "reward_std": 0.0,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.0,
      "step": 587
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 177.78125,
      "epoch": 2.352,
      "grad_norm": 0.029086973518133163,
      "kl": 0.443359375,
      "learning_rate": 7.648e-07,
      "loss": 0.0004,
      "reward": 0.5,
      "reward_std": 0.0,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.5,
      "step": 588
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 165.75,
      "epoch": 2.356,
      "grad_norm": 2.1231865882873535,
      "kl": 0.4501953125,
      "learning_rate": 7.643999999999999e-07,
      "loss": 0.0005,
      "reward": 0.34375,
      "reward_std": 0.1293872892856598,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.34375,
      "step": 589
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 166.625,
      "epoch": 2.36,
      "grad_norm": 3.6606342792510986,
      "kl": 0.41796875,
      "learning_rate": 7.64e-07,
      "loss": 0.0004,
      "reward": 0.15625,
      "reward_std": 0.22201896458864212,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.15625,
      "step": 590
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 146.4375,
      "epoch": 2.364,
      "grad_norm": 0.03885011747479439,
      "kl": 0.4072265625,
      "learning_rate": 7.635999999999999e-07,
      "loss": 0.0004,
      "reward": 0.75,
      "reward_std": 0.0,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.75,
      "step": 591
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 165.8125,
      "epoch": 2.368,
      "grad_norm": 2.6665732860565186,
      "kl": 0.3623046875,
      "learning_rate": 7.632e-07,
      "loss": 0.0004,
      "reward": 0.1875,
      "reward_std": 0.1157275140285492,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.1875,
      "step": 592
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 164.25,
      "epoch": 2.372,
      "grad_norm": 14.737106323242188,
      "kl": 0.3984375,
      "learning_rate": 7.628e-07,
      "loss": 0.0004,
      "reward": 0.375,
      "reward_std": 0.3335031494498253,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.375,
      "step": 593
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 155.78125,
      "epoch": 2.376,
      "grad_norm": 2.2816948890686035,
      "kl": 0.47998046875,
      "learning_rate": 7.623999999999999e-07,
      "loss": 0.0005,
      "reward": 0.1875,
      "reward_std": 0.1157275140285492,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.1875,
      "step": 594
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 160.625,
      "epoch": 2.38,
      "grad_norm": 4.556805610656738,
      "kl": 0.40478515625,
      "learning_rate": 7.62e-07,
      "loss": 0.0004,
      "reward": 0.625,
      "reward_std": 0.2314550280570984,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.625,
      "step": 595
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 176.125,
      "epoch": 2.384,
      "grad_norm": 2.1612496376037598,
      "kl": 0.4345703125,
      "learning_rate": 7.616e-07,
      "loss": 0.0004,
      "reward": 0.40625,
      "reward_std": 0.1293872892856598,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.40625,
      "step": 596
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 173.875,
      "epoch": 2.388,
      "grad_norm": 0.02121126651763916,
      "kl": 0.43798828125,
      "learning_rate": 7.611999999999999e-07,
      "loss": 0.0004,
      "reward": 0.25,
      "reward_std": 0.0,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.25,
      "step": 597
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 157.21875,
      "epoch": 2.392,
      "grad_norm": 7.606374263763428,
      "kl": 0.44970703125,
      "learning_rate": 7.608e-07,
      "loss": 0.0004,
      "reward": 0.8125,
      "reward_std": 0.1157275140285492,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.8125,
      "step": 598
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 149.84375,
      "epoch": 2.396,
      "grad_norm": 0.020341958850622177,
      "kl": 0.4169921875,
      "learning_rate": 7.604e-07,
      "loss": 0.0004,
      "reward": 1.0,
      "reward_std": 0.0,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 1.0,
      "step": 599
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 171.875,
      "epoch": 2.4,
      "grad_norm": 0.022393569350242615,
      "kl": 0.373046875,
      "learning_rate": 7.599999999999999e-07,
      "loss": 0.0004,
      "reward": 1.0,
      "reward_std": 0.0,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 1.0,
      "step": 600
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 148.78125,
      "epoch": 2.404,
      "grad_norm": 0.0184276532381773,
      "kl": 0.35009765625,
      "learning_rate": 7.596e-07,
      "loss": 0.0004,
      "reward": 0.5,
      "reward_std": 0.0,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.5,
      "step": 601
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 158.0625,
      "epoch": 2.408,
      "grad_norm": 1.964552640914917,
      "kl": 0.41845703125,
      "learning_rate": 7.592e-07,
      "loss": 0.0004,
      "reward": 0.3125,
      "reward_std": 0.1157275140285492,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.3125,
      "step": 602
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 162.4375,
      "epoch": 2.412,
      "grad_norm": 5.75462007522583,
      "kl": 0.40478515625,
      "learning_rate": 7.588e-07,
      "loss": 0.0004,
      "reward": 0.8125,
      "reward_std": 0.3104073107242584,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.8125,
      "step": 603
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 168.4375,
      "epoch": 2.416,
      "grad_norm": 2.917724847793579,
      "kl": 0.38232421875,
      "learning_rate": 7.583999999999999e-07,
      "loss": 0.0004,
      "reward": 0.625,
      "reward_std": 0.13363061845302582,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.625,
      "step": 604
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 175.0,
      "epoch": 2.42,
      "grad_norm": 2.8104546070098877,
      "kl": 0.37646484375,
      "learning_rate": 7.58e-07,
      "loss": 0.0004,
      "reward": 0.59375,
      "reward_std": 0.1293872892856598,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.59375,
      "step": 605
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 162.96875,
      "epoch": 2.424,
      "grad_norm": 9.585522651672363,
      "kl": 0.3671875,
      "learning_rate": 7.576000000000001e-07,
      "loss": 0.0004,
      "reward": 0.6875,
      "reward_std": 0.1157275140285492,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.6875,
      "step": 606
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 157.15625,
      "epoch": 2.428,
      "grad_norm": 0.014563312754034996,
      "kl": 0.39306640625,
      "learning_rate": 7.571999999999999e-07,
      "loss": 0.0004,
      "reward": 0.5,
      "reward_std": 0.0,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.5,
      "step": 607
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 159.71875,
      "epoch": 2.432,
      "grad_norm": 3.35884428024292,
      "kl": 0.384765625,
      "learning_rate": 7.568e-07,
      "loss": 0.0004,
      "reward": 0.5625,
      "reward_std": 0.249358132481575,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.5625,
      "step": 608
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 165.09375,
      "epoch": 2.436,
      "grad_norm": 6.7689337730407715,
      "kl": 0.46826171875,
      "learning_rate": 7.564e-07,
      "loss": 0.0005,
      "reward": 0.0625,
      "reward_std": 0.1157275140285492,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.0625,
      "step": 609
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 163.25,
      "epoch": 2.44,
      "grad_norm": 3.2988665103912354,
      "kl": 0.369140625,
      "learning_rate": 7.559999999999999e-07,
      "loss": 0.0004,
      "reward": 0.09375,
      "reward_std": 0.1293872892856598,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.09375,
      "step": 610
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 155.03125,
      "epoch": 2.444,
      "grad_norm": 3.5551517009735107,
      "kl": 0.38916015625,
      "learning_rate": 7.556e-07,
      "loss": 0.0004,
      "reward": 0.03125,
      "reward_std": 0.0883883461356163,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.03125,
      "step": 611
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 171.65625,
      "epoch": 2.448,
      "grad_norm": 6.909402847290039,
      "kl": 0.39892578125,
      "learning_rate": 7.552e-07,
      "loss": 0.0004,
      "reward": 0.375,
      "reward_std": 0.13363061845302582,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.375,
      "step": 612
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 150.65625,
      "epoch": 2.452,
      "grad_norm": 4.217846870422363,
      "kl": 0.376953125,
      "learning_rate": 7.548e-07,
      "loss": 0.0004,
      "reward": 0.71875,
      "reward_std": 0.24511480331420898,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.71875,
      "step": 613
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 163.65625,
      "epoch": 2.456,
      "grad_norm": 0.039842791855335236,
      "kl": 0.4345703125,
      "learning_rate": 7.543999999999999e-07,
      "loss": 0.0004,
      "reward": 0.0,
      "reward_std": 0.0,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.0,
      "step": 614
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 149.96875,
      "epoch": 2.46,
      "grad_norm": 0.06290771812200546,
      "kl": 0.47998046875,
      "learning_rate": 7.54e-07,
      "loss": 0.0005,
      "reward": 0.0,
      "reward_std": 0.0,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.0,
      "step": 615
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 144.1875,
      "epoch": 2.464,
      "grad_norm": 10.857304573059082,
      "kl": 0.48046875,
      "learning_rate": 7.536e-07,
      "loss": 0.0005,
      "reward": 0.53125,
      "reward_std": 0.24511480331420898,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.53125,
      "step": 616
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 149.78125,
      "epoch": 2.468,
      "grad_norm": 5.542413711547852,
      "kl": 0.41162109375,
      "learning_rate": 7.531999999999999e-07,
      "loss": 0.0004,
      "reward": 0.46875,
      "reward_std": 0.0883883461356163,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.46875,
      "step": 617
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 148.4375,
      "epoch": 2.472,
      "grad_norm": 6.406118869781494,
      "kl": 0.39208984375,
      "learning_rate": 7.528e-07,
      "loss": 0.0004,
      "reward": 0.28125,
      "reward_std": 0.0883883461356163,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.28125,
      "step": 618
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 137.65625,
      "epoch": 2.476,
      "grad_norm": 3.404924154281616,
      "kl": 0.3837890625,
      "learning_rate": 7.523999999999999e-07,
      "loss": 0.0004,
      "reward": 0.75,
      "reward_std": 0.2314550280570984,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.75,
      "step": 619
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 146.46875,
      "epoch": 2.48,
      "grad_norm": 0.05681978166103363,
      "kl": 0.46435546875,
      "learning_rate": 7.52e-07,
      "loss": 0.0005,
      "reward": 0.25,
      "reward_std": 0.0,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.25,
      "step": 620
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 149.5625,
      "epoch": 2.484,
      "grad_norm": 4.457178592681885,
      "kl": 0.43359375,
      "learning_rate": 7.516e-07,
      "loss": 0.0004,
      "reward": 0.59375,
      "reward_std": 0.3608423173427582,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.59375,
      "step": 621
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 147.4375,
      "epoch": 2.488,
      "grad_norm": 2.3256123065948486,
      "kl": 0.43408203125,
      "learning_rate": 7.511999999999999e-07,
      "loss": 0.0004,
      "reward": 0.5625,
      "reward_std": 0.1157275140285492,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.5625,
      "step": 622
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 154.6875,
      "epoch": 2.492,
      "grad_norm": 32.79850387573242,
      "kl": 0.4013671875,
      "learning_rate": 7.508e-07,
      "loss": 0.0004,
      "reward": 0.5,
      "reward_std": 0.2314550280570984,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.5,
      "step": 623
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 146.03125,
      "epoch": 2.496,
      "grad_norm": 0.02143729291856289,
      "kl": 0.45068359375,
      "learning_rate": 7.503999999999999e-07,
      "loss": 0.0004,
      "reward": 0.0,
      "reward_std": 0.0,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.0,
      "step": 624
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 140.28125,
      "epoch": 2.5,
      "grad_norm": 6.017480373382568,
      "kl": 0.3935546875,
      "learning_rate": 7.5e-07,
      "loss": 0.0004,
      "reward": 0.625,
      "reward_std": 0.13363061845302582,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.625,
      "step": 625
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 136.25,
      "epoch": 2.504,
      "grad_norm": 3.216672897338867,
      "kl": 0.453125,
      "learning_rate": 7.496e-07,
      "loss": 0.0005,
      "reward": 0.53125,
      "reward_std": 0.0883883461356163,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.53125,
      "step": 626
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 136.1875,
      "epoch": 2.508,
      "grad_norm": 3.319479465484619,
      "kl": 0.40380859375,
      "learning_rate": 7.492e-07,
      "loss": 0.0004,
      "reward": 0.375,
      "reward_std": 0.13363061845302582,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.375,
      "step": 627
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 149.40625,
      "epoch": 2.512,
      "grad_norm": 33.24382019042969,
      "kl": 0.4365234375,
      "learning_rate": 7.488e-07,
      "loss": 0.0004,
      "reward": 0.53125,
      "reward_std": 0.2041158601641655,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.53125,
      "step": 628
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 143.375,
      "epoch": 2.516,
      "grad_norm": 4.061643123626709,
      "kl": 1.07275390625,
      "learning_rate": 7.483999999999999e-07,
      "loss": 0.0011,
      "reward": 0.59375,
      "reward_std": 0.24511480331420898,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.59375,
      "step": 629
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 144.9375,
      "epoch": 2.52,
      "grad_norm": 2.824666738510132,
      "kl": 0.43701171875,
      "learning_rate": 7.48e-07,
      "loss": 0.0004,
      "reward": 0.40625,
      "reward_std": 0.1293872892856598,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.40625,
      "step": 630
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 148.46875,
      "epoch": 2.524,
      "grad_norm": 0.018655454739928246,
      "kl": 0.455078125,
      "learning_rate": 7.476e-07,
      "loss": 0.0005,
      "reward": 0.5,
      "reward_std": 0.0,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.5,
      "step": 631
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 156.78125,
      "epoch": 2.528,
      "grad_norm": 0.08967535942792892,
      "kl": 0.47998046875,
      "learning_rate": 7.471999999999999e-07,
      "loss": 0.0005,
      "reward": 0.25,
      "reward_std": 0.0,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.25,
      "step": 632
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 144.34375,
      "epoch": 2.532,
      "grad_norm": 15.28121566772461,
      "kl": 0.52685546875,
      "learning_rate": 7.468e-07,
      "loss": 0.0005,
      "reward": 0.71875,
      "reward_std": 0.0883883461356163,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.71875,
      "step": 633
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 150.1875,
      "epoch": 2.536,
      "grad_norm": 3.1732590198516846,
      "kl": 0.4677734375,
      "learning_rate": 7.464e-07,
      "loss": 0.0005,
      "reward": 0.5625,
      "reward_std": 0.1157275140285492,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.5625,
      "step": 634
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 152.5,
      "epoch": 2.54,
      "grad_norm": 3.1518759727478027,
      "kl": 3.7236328125,
      "learning_rate": 7.459999999999999e-07,
      "loss": 0.0037,
      "reward": 0.21875,
      "reward_std": 0.0883883461356163,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.21875,
      "step": 635
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 158.84375,
      "epoch": 2.544,
      "grad_norm": 0.018218638375401497,
      "kl": 0.40576171875,
      "learning_rate": 7.456e-07,
      "loss": 0.0004,
      "reward": 0.5,
      "reward_std": 0.0,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.5,
      "step": 636
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 162.34375,
      "epoch": 2.548,
      "grad_norm": 9.241228103637695,
      "kl": 0.41650390625,
      "learning_rate": 7.452e-07,
      "loss": 0.0004,
      "reward": 0.59375,
      "reward_std": 0.22201896458864212,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.59375,
      "step": 637
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 146.375,
      "epoch": 2.552,
      "grad_norm": 10.745528221130371,
      "kl": 0.4189453125,
      "learning_rate": 7.447999999999999e-07,
      "loss": 0.0004,
      "reward": 0.59375,
      "reward_std": 0.1293872892856598,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.59375,
      "step": 638
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 158.03125,
      "epoch": 2.556,
      "grad_norm": 4.1044182777404785,
      "kl": 1.0205078125,
      "learning_rate": 7.443999999999999e-07,
      "loss": 0.001,
      "reward": 0.8125,
      "reward_std": 0.2177756354212761,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.8125,
      "step": 639
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 156.125,
      "epoch": 2.56,
      "grad_norm": 0.12027113139629364,
      "kl": 0.58447265625,
      "learning_rate": 7.44e-07,
      "loss": 0.0006,
      "reward": 0.5,
      "reward_std": 0.0,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.5,
      "step": 640
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 157.03125,
      "epoch": 2.564,
      "grad_norm": 7.116557598114014,
      "kl": 0.39501953125,
      "learning_rate": 7.436e-07,
      "loss": 0.0004,
      "reward": 0.375,
      "reward_std": 0.2177756354212761,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.375,
      "step": 641
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 151.5625,
      "epoch": 2.568,
      "grad_norm": 1.8864772319793701,
      "kl": 0.4033203125,
      "learning_rate": 7.431999999999999e-07,
      "loss": 0.0004,
      "reward": 0.28125,
      "reward_std": 0.0883883461356163,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.28125,
      "step": 642
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 149.1875,
      "epoch": 2.572,
      "grad_norm": 5.301823616027832,
      "kl": 0.42236328125,
      "learning_rate": 7.428e-07,
      "loss": 0.0004,
      "reward": 0.75,
      "reward_std": 0.1767766922712326,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.75,
      "step": 643
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 147.28125,
      "epoch": 2.576,
      "grad_norm": 3.431767702102661,
      "kl": 0.42919921875,
      "learning_rate": 7.423999999999999e-07,
      "loss": 0.0004,
      "reward": 0.28125,
      "reward_std": 0.0883883461356163,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.28125,
      "step": 644
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 153.125,
      "epoch": 2.58,
      "grad_norm": 0.016840370371937752,
      "kl": 0.3896484375,
      "learning_rate": 7.42e-07,
      "loss": 0.0004,
      "reward": 0.25,
      "reward_std": 0.0,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.25,
      "step": 645
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 159.625,
      "epoch": 2.584,
      "grad_norm": 2.3346869945526123,
      "kl": 0.39892578125,
      "learning_rate": 7.416e-07,
      "loss": 0.0004,
      "reward": 0.71875,
      "reward_std": 0.0883883461356163,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.71875,
      "step": 646
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 148.75,
      "epoch": 2.588,
      "grad_norm": 5.551220417022705,
      "kl": 0.37890625,
      "learning_rate": 7.411999999999999e-07,
      "loss": 0.0004,
      "reward": 0.40625,
      "reward_std": 0.1293872892856598,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.40625,
      "step": 647
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 141.03125,
      "epoch": 2.592,
      "grad_norm": 0.03564630076289177,
      "kl": 0.44189453125,
      "learning_rate": 7.408e-07,
      "loss": 0.0004,
      "reward": 0.5,
      "reward_std": 0.0,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.5,
      "step": 648
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 146.59375,
      "epoch": 2.596,
      "grad_norm": 9.534584999084473,
      "kl": 0.39697265625,
      "learning_rate": 7.403999999999999e-07,
      "loss": 0.0004,
      "reward": 0.53125,
      "reward_std": 0.2041158601641655,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.53125,
      "step": 649
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 148.5625,
      "epoch": 2.6,
      "grad_norm": 21.58906364440918,
      "kl": 0.39208984375,
      "learning_rate": 7.4e-07,
      "loss": 0.0004,
      "reward": 0.34375,
      "reward_std": 0.1293872892856598,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.34375,
      "step": 650
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 150.84375,
      "epoch": 2.604,
      "grad_norm": 3.9090545177459717,
      "kl": 0.4091796875,
      "learning_rate": 7.396e-07,
      "loss": 0.0004,
      "reward": 0.3125,
      "reward_std": 0.1157275140285492,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.3125,
      "step": 651
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 148.0,
      "epoch": 2.608,
      "grad_norm": 0.02115592174232006,
      "kl": 0.3798828125,
      "learning_rate": 7.392e-07,
      "loss": 0.0004,
      "reward": 0.5,
      "reward_std": 0.0,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.5,
      "step": 652
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 143.96875,
      "epoch": 2.612,
      "grad_norm": 0.0582285113632679,
      "kl": 0.48876953125,
      "learning_rate": 7.388e-07,
      "loss": 0.0005,
      "reward": 0.0,
      "reward_std": 0.0,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.0,
      "step": 653
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 157.90625,
      "epoch": 2.616,
      "grad_norm": 0.050456397235393524,
      "kl": 0.42529296875,
      "learning_rate": 7.383999999999999e-07,
      "loss": 0.0004,
      "reward": 0.75,
      "reward_std": 0.0,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.75,
      "step": 654
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 153.9375,
      "epoch": 2.62,
      "grad_norm": 0.01952575333416462,
      "kl": 0.3984375,
      "learning_rate": 7.38e-07,
      "loss": 0.0004,
      "reward": 1.0,
      "reward_std": 0.0,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 1.0,
      "step": 655
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 149.21875,
      "epoch": 2.624,
      "grad_norm": 3.644275426864624,
      "kl": 0.38720703125,
      "learning_rate": 7.376e-07,
      "loss": 0.0004,
      "reward": 0.21875,
      "reward_std": 0.0883883461356163,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.21875,
      "step": 656
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 145.09375,
      "epoch": 2.628,
      "grad_norm": 3.330442190170288,
      "kl": 0.43017578125,
      "learning_rate": 7.371999999999999e-07,
      "loss": 0.0004,
      "reward": 0.9375,
      "reward_std": 0.1157275140285492,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.9375,
      "step": 657
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 159.3125,
      "epoch": 2.632,
      "grad_norm": 0.025550298392772675,
      "kl": 0.37060546875,
      "learning_rate": 7.368e-07,
      "loss": 0.0004,
      "reward": 0.75,
      "reward_std": 0.0,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.75,
      "step": 658
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 145.09375,
      "epoch": 2.636,
      "grad_norm": 0.03615197539329529,
      "kl": 0.38916015625,
      "learning_rate": 7.364000000000001e-07,
      "loss": 0.0004,
      "reward": 0.25,
      "reward_std": 0.0,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.25,
      "step": 659
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 144.53125,
      "epoch": 2.64,
      "grad_norm": 0.11624439060688019,
      "kl": 0.65771484375,
      "learning_rate": 7.359999999999999e-07,
      "loss": 0.0007,
      "reward": 0.5,
      "reward_std": 0.0,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.5,
      "step": 660
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 132.03125,
      "epoch": 2.644,
      "grad_norm": 0.038560472428798676,
      "kl": 0.39404296875,
      "learning_rate": 7.356e-07,
      "loss": 0.0004,
      "reward": 0.0,
      "reward_std": 0.0,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.0,
      "step": 661
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 144.15625,
      "epoch": 2.648,
      "grad_norm": 3.804847478866577,
      "kl": 0.396484375,
      "learning_rate": 7.352e-07,
      "loss": 0.0004,
      "reward": 0.46875,
      "reward_std": 0.2041158601641655,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.46875,
      "step": 662
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 139.46875,
      "epoch": 2.652,
      "grad_norm": 0.013049962930381298,
      "kl": 0.3662109375,
      "learning_rate": 7.347999999999999e-07,
      "loss": 0.0004,
      "reward": 0.25,
      "reward_std": 0.0,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.25,
      "step": 663
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 148.28125,
      "epoch": 2.656,
      "grad_norm": 1.7591242790222168,
      "kl": 0.3662109375,
      "learning_rate": 7.344e-07,
      "loss": 0.0004,
      "reward": 0.28125,
      "reward_std": 0.0883883461356163,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.28125,
      "step": 664
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 145.03125,
      "epoch": 2.66,
      "grad_norm": 2.448904514312744,
      "kl": 0.38330078125,
      "learning_rate": 7.34e-07,
      "loss": 0.0004,
      "reward": 0.6875,
      "reward_std": 0.1157275140285492,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.6875,
      "step": 665
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 137.71875,
      "epoch": 2.664,
      "grad_norm": 2.5101022720336914,
      "kl": 0.40966796875,
      "learning_rate": 7.336e-07,
      "loss": 0.0004,
      "reward": 0.0625,
      "reward_std": 0.1157275140285492,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.0625,
      "step": 666
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 147.75,
      "epoch": 2.668,
      "grad_norm": 3.5953409671783447,
      "kl": 0.40771484375,
      "learning_rate": 7.331999999999999e-07,
      "loss": 0.0004,
      "reward": 0.375,
      "reward_std": 0.13363061845302582,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.375,
      "step": 667
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 142.59375,
      "epoch": 2.672,
      "grad_norm": 5.663808822631836,
      "kl": 0.44482421875,
      "learning_rate": 7.328e-07,
      "loss": 0.0004,
      "reward": 0.46875,
      "reward_std": 0.0883883461356163,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.46875,
      "step": 668
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 147.4375,
      "epoch": 2.676,
      "grad_norm": 20.628957748413086,
      "kl": 0.37646484375,
      "learning_rate": 7.324e-07,
      "loss": 0.0004,
      "reward": 0.625,
      "reward_std": 0.2177756354212761,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.625,
      "step": 669
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 159.1875,
      "epoch": 2.68,
      "grad_norm": 2.000304937362671,
      "kl": 0.36669921875,
      "learning_rate": 7.319999999999999e-07,
      "loss": 0.0004,
      "reward": 0.125,
      "reward_std": 0.13363061845302582,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.125,
      "step": 670
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 152.03125,
      "epoch": 2.684,
      "grad_norm": 3.6491291522979736,
      "kl": 0.3828125,
      "learning_rate": 7.316e-07,
      "loss": 0.0004,
      "reward": 0.03125,
      "reward_std": 0.0883883461356163,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.03125,
      "step": 671
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 151.84375,
      "epoch": 2.6879999999999997,
      "grad_norm": 8.330862045288086,
      "kl": 0.41796875,
      "learning_rate": 7.311999999999999e-07,
      "loss": 0.0004,
      "reward": 0.6875,
      "reward_std": 0.249358132481575,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.6875,
      "step": 672
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 153.1875,
      "epoch": 2.692,
      "grad_norm": 3.8399322032928467,
      "kl": 0.4296875,
      "learning_rate": 7.308e-07,
      "loss": 0.0004,
      "reward": 0.46875,
      "reward_std": 0.0883883461356163,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.46875,
      "step": 673
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 158.71875,
      "epoch": 2.6959999999999997,
      "grad_norm": 0.01773027703166008,
      "kl": 0.365234375,
      "learning_rate": 7.304e-07,
      "loss": 0.0004,
      "reward": 0.0,
      "reward_std": 0.0,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.0,
      "step": 674
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 134.8125,
      "epoch": 2.7,
      "grad_norm": 3.9474873542785645,
      "kl": 0.33642578125,
      "learning_rate": 7.3e-07,
      "loss": 0.0003,
      "reward": 0.53125,
      "reward_std": 0.0883883461356163,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.53125,
      "step": 675
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 155.90625,
      "epoch": 2.7039999999999997,
      "grad_norm": 2.9829750061035156,
      "kl": 0.5849609375,
      "learning_rate": 7.296e-07,
      "loss": 0.0006,
      "reward": 0.03125,
      "reward_std": 0.0883883461356163,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.03125,
      "step": 676
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 148.53125,
      "epoch": 2.708,
      "grad_norm": 4.705109119415283,
      "kl": 0.39794921875,
      "learning_rate": 7.291999999999999e-07,
      "loss": 0.0004,
      "reward": 0.21875,
      "reward_std": 0.0883883461356163,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.21875,
      "step": 677
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 144.09375,
      "epoch": 2.7119999999999997,
      "grad_norm": 0.01359782088547945,
      "kl": 0.35498046875,
      "learning_rate": 7.288e-07,
      "loss": 0.0004,
      "reward": 0.25,
      "reward_std": 0.0,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.25,
      "step": 678
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 145.375,
      "epoch": 2.716,
      "grad_norm": 0.01723480224609375,
      "kl": 0.37548828125,
      "learning_rate": 7.284e-07,
      "loss": 0.0004,
      "reward": 0.5,
      "reward_std": 0.0,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.5,
      "step": 679
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 142.59375,
      "epoch": 2.7199999999999998,
      "grad_norm": 0.016932839527726173,
      "kl": 0.39794921875,
      "learning_rate": 7.28e-07,
      "loss": 0.0004,
      "reward": 0.25,
      "reward_std": 0.0,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.25,
      "step": 680
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 151.125,
      "epoch": 2.724,
      "grad_norm": 0.03179251030087471,
      "kl": 0.36376953125,
      "learning_rate": 7.276e-07,
      "loss": 0.0004,
      "reward": 0.5,
      "reward_std": 0.0,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.5,
      "step": 681
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 148.5625,
      "epoch": 2.7279999999999998,
      "grad_norm": 2.521477222442627,
      "kl": 0.41845703125,
      "learning_rate": 7.271999999999999e-07,
      "loss": 0.0004,
      "reward": 0.375,
      "reward_std": 0.13363061845302582,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.375,
      "step": 682
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 148.875,
      "epoch": 2.732,
      "grad_norm": 4.156079292297363,
      "kl": 0.35546875,
      "learning_rate": 7.268e-07,
      "loss": 0.0004,
      "reward": 0.53125,
      "reward_std": 0.0883883461356163,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.53125,
      "step": 683
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 173.5,
      "epoch": 2.7359999999999998,
      "grad_norm": 0.015807529911398888,
      "kl": 0.34814453125,
      "learning_rate": 7.264e-07,
      "loss": 0.0003,
      "reward": 0.25,
      "reward_std": 0.0,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.25,
      "step": 684
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 155.15625,
      "epoch": 2.74,
      "grad_norm": 0.017792237922549248,
      "kl": 0.3857421875,
      "learning_rate": 7.259999999999999e-07,
      "loss": 0.0004,
      "reward": 0.5,
      "reward_std": 0.0,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.5,
      "step": 685
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 161.625,
      "epoch": 2.7439999999999998,
      "grad_norm": 3.6412100791931152,
      "kl": 0.39453125,
      "learning_rate": 7.256e-07,
      "loss": 0.0004,
      "reward": 0.1875,
      "reward_std": 0.1157275140285492,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.1875,
      "step": 686
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 159.78125,
      "epoch": 2.748,
      "grad_norm": 3.08406138420105,
      "kl": 0.38818359375,
      "learning_rate": 7.252e-07,
      "loss": 0.0004,
      "reward": 0.28125,
      "reward_std": 0.0883883461356163,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.28125,
      "step": 687
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 155.15625,
      "epoch": 2.752,
      "grad_norm": 3.5069499015808105,
      "kl": 0.37109375,
      "learning_rate": 7.247999999999999e-07,
      "loss": 0.0004,
      "reward": 0.0625,
      "reward_std": 0.1157275140285492,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.0625,
      "step": 688
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 166.875,
      "epoch": 2.7560000000000002,
      "grad_norm": 2.569697380065918,
      "kl": 0.36083984375,
      "learning_rate": 7.244e-07,
      "loss": 0.0004,
      "reward": 0.40625,
      "reward_std": 0.1293872892856598,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.40625,
      "step": 689
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 167.40625,
      "epoch": 2.76,
      "grad_norm": 0.019567696377635002,
      "kl": 0.3740234375,
      "learning_rate": 7.24e-07,
      "loss": 0.0004,
      "reward": 0.25,
      "reward_std": 0.0,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.25,
      "step": 690
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 146.53125,
      "epoch": 2.7640000000000002,
      "grad_norm": 0.03392166271805763,
      "kl": 0.384765625,
      "learning_rate": 7.235999999999999e-07,
      "loss": 0.0004,
      "reward": 0.25,
      "reward_std": 0.0,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.25,
      "step": 691
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 162.15625,
      "epoch": 2.768,
      "grad_norm": 3.425001859664917,
      "kl": 0.3642578125,
      "learning_rate": 7.231999999999999e-07,
      "loss": 0.0004,
      "reward": 0.59375,
      "reward_std": 0.24511480331420898,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.59375,
      "step": 692
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 152.125,
      "epoch": 2.7720000000000002,
      "grad_norm": 0.021558105945587158,
      "kl": 0.3681640625,
      "learning_rate": 7.228e-07,
      "loss": 0.0004,
      "reward": 0.0,
      "reward_std": 0.0,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.0,
      "step": 693
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 171.3125,
      "epoch": 2.776,
      "grad_norm": 9.331954002380371,
      "kl": 0.3466796875,
      "learning_rate": 7.224e-07,
      "loss": 0.0003,
      "reward": 0.0625,
      "reward_std": 0.1157275140285492,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.0625,
      "step": 694
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 151.3125,
      "epoch": 2.7800000000000002,
      "grad_norm": 3.3615882396698,
      "kl": 0.3916015625,
      "learning_rate": 7.219999999999999e-07,
      "loss": 0.0004,
      "reward": 0.3125,
      "reward_std": 0.1157275140285492,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.3125,
      "step": 695
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 168.09375,
      "epoch": 2.784,
      "grad_norm": 0.11704137176275253,
      "kl": 0.353515625,
      "learning_rate": 7.216e-07,
      "loss": 0.0004,
      "reward": 0.25,
      "reward_std": 0.0,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.25,
      "step": 696
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 170.40625,
      "epoch": 2.7880000000000003,
      "grad_norm": 0.018887672573328018,
      "kl": 0.37646484375,
      "learning_rate": 7.211999999999999e-07,
      "loss": 0.0004,
      "reward": 0.0,
      "reward_std": 0.0,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.0,
      "step": 697
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 155.71875,
      "epoch": 2.792,
      "grad_norm": 4.669219017028809,
      "kl": 0.359375,
      "learning_rate": 7.207999999999999e-07,
      "loss": 0.0004,
      "reward": 0.65625,
      "reward_std": 0.2041158601641655,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.65625,
      "step": 698
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 175.5625,
      "epoch": 2.7960000000000003,
      "grad_norm": 0.025426549836993217,
      "kl": 0.3515625,
      "learning_rate": 7.204e-07,
      "loss": 0.0004,
      "reward": 0.25,
      "reward_std": 0.0,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.25,
      "step": 699
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 184.625,
      "epoch": 2.8,
      "grad_norm": 3.630786418914795,
      "kl": 0.45947265625,
      "learning_rate": 7.2e-07,
      "loss": 0.0005,
      "reward": 0.5625,
      "reward_std": 0.1157275140285492,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.5625,
      "step": 700
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 168.5,
      "epoch": 2.8040000000000003,
      "grad_norm": 7.253315448760986,
      "kl": 0.35546875,
      "learning_rate": 7.196e-07,
      "loss": 0.0004,
      "reward": 0.28125,
      "reward_std": 0.2041158601641655,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.28125,
      "step": 701
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 178.125,
      "epoch": 2.808,
      "grad_norm": 6.582237243652344,
      "kl": 0.32470703125,
      "learning_rate": 7.191999999999999e-07,
      "loss": 0.0003,
      "reward": 0.4375,
      "reward_std": 0.1767766922712326,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.4375,
      "step": 702
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 169.5625,
      "epoch": 2.8120000000000003,
      "grad_norm": 0.07108395546674728,
      "kl": 0.443359375,
      "learning_rate": 7.188e-07,
      "loss": 0.0004,
      "reward": 0.0,
      "reward_std": 0.0,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.0,
      "step": 703
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 186.40625,
      "epoch": 2.816,
      "grad_norm": 4.2947773933410645,
      "kl": 0.318359375,
      "learning_rate": 7.184e-07,
      "loss": 0.0003,
      "reward": 0.34375,
      "reward_std": 0.1293872892856598,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.34375,
      "step": 704
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 151.75,
      "epoch": 2.82,
      "grad_norm": 17.98407554626465,
      "kl": 0.34716796875,
      "learning_rate": 7.179999999999999e-07,
      "loss": 0.0003,
      "reward": 0.34375,
      "reward_std": 0.2041158601641655,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.34375,
      "step": 705
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 165.46875,
      "epoch": 2.824,
      "grad_norm": 0.012600379064679146,
      "kl": 0.3330078125,
      "learning_rate": 7.176e-07,
      "loss": 0.0003,
      "reward": 0.5,
      "reward_std": 0.0,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.5,
      "step": 706
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 165.375,
      "epoch": 2.828,
      "grad_norm": 2.7540090084075928,
      "kl": 0.36865234375,
      "learning_rate": 7.171999999999999e-07,
      "loss": 0.0004,
      "reward": 0.28125,
      "reward_std": 0.0883883461356163,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.28125,
      "step": 707
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 158.28125,
      "epoch": 2.832,
      "grad_norm": 2.8922812938690186,
      "kl": 0.36083984375,
      "learning_rate": 7.168e-07,
      "loss": 0.0004,
      "reward": 0.0625,
      "reward_std": 0.1157275140285492,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.0625,
      "step": 708
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 151.65625,
      "epoch": 2.836,
      "grad_norm": 2.8475775718688965,
      "kl": 0.359375,
      "learning_rate": 7.164e-07,
      "loss": 0.0004,
      "reward": 0.1875,
      "reward_std": 0.1157275140285492,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.1875,
      "step": 709
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 169.9375,
      "epoch": 2.84,
      "grad_norm": 4.290299892425537,
      "kl": 0.35400390625,
      "learning_rate": 7.159999999999999e-07,
      "loss": 0.0004,
      "reward": 0.28125,
      "reward_std": 0.2041158601641655,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.28125,
      "step": 710
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 156.78125,
      "epoch": 2.844,
      "grad_norm": 3.043875217437744,
      "kl": 0.35595703125,
      "learning_rate": 7.156e-07,
      "loss": 0.0004,
      "reward": 0.6875,
      "reward_std": 0.1157275140285492,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.6875,
      "step": 711
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 171.9375,
      "epoch": 2.848,
      "grad_norm": 6.214578151702881,
      "kl": 0.41552734375,
      "learning_rate": 7.151999999999999e-07,
      "loss": 0.0004,
      "reward": 0.40625,
      "reward_std": 0.3987956568598747,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.40625,
      "step": 712
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 150.46875,
      "epoch": 2.852,
      "grad_norm": 0.013492027297616005,
      "kl": 0.328125,
      "learning_rate": 7.147999999999999e-07,
      "loss": 0.0003,
      "reward": 0.5,
      "reward_std": 0.0,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.5,
      "step": 713
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 160.40625,
      "epoch": 2.856,
      "grad_norm": 0.043924566358327866,
      "kl": 0.3505859375,
      "learning_rate": 7.144e-07,
      "loss": 0.0004,
      "reward": 0.75,
      "reward_std": 0.0,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.75,
      "step": 714
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 163.28125,
      "epoch": 2.86,
      "grad_norm": 2.5488409996032715,
      "kl": 0.35107421875,
      "learning_rate": 7.14e-07,
      "loss": 0.0004,
      "reward": 0.53125,
      "reward_std": 0.0883883461356163,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.53125,
      "step": 715
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 155.625,
      "epoch": 2.864,
      "grad_norm": 3.5106072425842285,
      "kl": 0.42822265625,
      "learning_rate": 7.135999999999999e-07,
      "loss": 0.0004,
      "reward": 0.84375,
      "reward_std": 0.1293872892856598,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.84375,
      "step": 716
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 157.25,
      "epoch": 2.868,
      "grad_norm": 2.403311252593994,
      "kl": 0.41748046875,
      "learning_rate": 7.131999999999999e-07,
      "loss": 0.0004,
      "reward": 0.5625,
      "reward_std": 0.1157275140285492,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.5625,
      "step": 717
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 160.59375,
      "epoch": 2.872,
      "grad_norm": 2.681488275527954,
      "kl": 0.4287109375,
      "learning_rate": 7.128e-07,
      "loss": 0.0004,
      "reward": 0.625,
      "reward_std": 0.13363061845302582,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.625,
      "step": 718
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 154.3125,
      "epoch": 2.876,
      "grad_norm": 1.7415363788604736,
      "kl": 0.3798828125,
      "learning_rate": 7.124e-07,
      "loss": 0.0004,
      "reward": 0.21875,
      "reward_std": 0.0883883461356163,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.21875,
      "step": 719
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 150.5,
      "epoch": 2.88,
      "grad_norm": 0.035293303430080414,
      "kl": 0.40966796875,
      "learning_rate": 7.119999999999999e-07,
      "loss": 0.0004,
      "reward": 0.25,
      "reward_std": 0.0,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.25,
      "step": 720
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 152.5625,
      "epoch": 2.884,
      "grad_norm": 2.288606643676758,
      "kl": 0.36083984375,
      "learning_rate": 7.116e-07,
      "loss": 0.0004,
      "reward": 0.5625,
      "reward_std": 0.1157275140285492,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.5625,
      "step": 721
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 128.8125,
      "epoch": 2.888,
      "grad_norm": 0.0395616739988327,
      "kl": 0.396484375,
      "learning_rate": 7.112000000000001e-07,
      "loss": 0.0004,
      "reward": 1.0,
      "reward_std": 0.0,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 1.0,
      "step": 722
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 154.625,
      "epoch": 2.892,
      "grad_norm": 4.201551914215088,
      "kl": 0.38330078125,
      "learning_rate": 7.107999999999999e-07,
      "loss": 0.0004,
      "reward": 0.28125,
      "reward_std": 0.0883883461356163,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.28125,
      "step": 723
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 143.0625,
      "epoch": 2.896,
      "grad_norm": 8.179118156433105,
      "kl": 0.36669921875,
      "learning_rate": 7.104e-07,
      "loss": 0.0004,
      "reward": 0.46875,
      "reward_std": 0.0883883461356163,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.46875,
      "step": 724
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 128.34375,
      "epoch": 2.9,
      "grad_norm": 0.016004957258701324,
      "kl": 0.38427734375,
      "learning_rate": 7.1e-07,
      "loss": 0.0004,
      "reward": 0.25,
      "reward_std": 0.0,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.25,
      "step": 725
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 145.4375,
      "epoch": 2.904,
      "grad_norm": 4.614955902099609,
      "kl": 0.3623046875,
      "learning_rate": 7.096e-07,
      "loss": 0.0004,
      "reward": 0.5625,
      "reward_std": 0.3335031494498253,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.5625,
      "step": 726
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 145.8125,
      "epoch": 2.908,
      "grad_norm": 9.6151123046875,
      "kl": 0.39697265625,
      "learning_rate": 7.092e-07,
      "loss": 0.0004,
      "reward": 0.34375,
      "reward_std": 0.3808925524353981,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.34375,
      "step": 727
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 148.59375,
      "epoch": 2.912,
      "grad_norm": 5.9742021560668945,
      "kl": 0.41015625,
      "learning_rate": 7.088e-07,
      "loss": 0.0004,
      "reward": 0.1875,
      "reward_std": 0.1157275140285492,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.1875,
      "step": 728
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 134.96875,
      "epoch": 2.916,
      "grad_norm": 6.691777229309082,
      "kl": 0.4296875,
      "learning_rate": 7.084e-07,
      "loss": 0.0004,
      "reward": 0.21875,
      "reward_std": 0.0883883461356163,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.21875,
      "step": 729
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 148.15625,
      "epoch": 2.92,
      "grad_norm": 7.736252784729004,
      "kl": 0.4423828125,
      "learning_rate": 7.079999999999999e-07,
      "loss": 0.0004,
      "reward": 0.46875,
      "reward_std": 0.0883883461356163,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.46875,
      "step": 730
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 124.75,
      "epoch": 2.924,
      "grad_norm": 0.12620128691196442,
      "kl": 0.4228515625,
      "learning_rate": 7.076e-07,
      "loss": 0.0004,
      "reward": 0.75,
      "reward_std": 0.0,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.75,
      "step": 731
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 116.5,
      "epoch": 2.928,
      "grad_norm": 0.026465388014912605,
      "kl": 0.443359375,
      "learning_rate": 7.072e-07,
      "loss": 0.0004,
      "reward": 0.25,
      "reward_std": 0.0,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.25,
      "step": 732
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 130.625,
      "epoch": 2.932,
      "grad_norm": 3.161349058151245,
      "kl": 0.47607421875,
      "learning_rate": 7.068e-07,
      "loss": 0.0005,
      "reward": 0.71875,
      "reward_std": 0.0883883461356163,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.71875,
      "step": 733
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 128.96875,
      "epoch": 2.936,
      "grad_norm": 6.915829658508301,
      "kl": 0.41015625,
      "learning_rate": 7.064e-07,
      "loss": 0.0004,
      "reward": 0.28125,
      "reward_std": 0.0883883461356163,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.28125,
      "step": 734
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 126.6875,
      "epoch": 2.94,
      "grad_norm": 10.095046043395996,
      "kl": 0.42626953125,
      "learning_rate": 7.059999999999999e-07,
      "loss": 0.0004,
      "reward": 0.15625,
      "reward_std": 0.22201896458864212,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.15625,
      "step": 735
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 128.28125,
      "epoch": 2.944,
      "grad_norm": 6.878199100494385,
      "kl": 0.48876953125,
      "learning_rate": 7.056e-07,
      "loss": 0.0005,
      "reward": 0.28125,
      "reward_std": 0.0883883461356163,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.28125,
      "step": 736
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 125.25,
      "epoch": 2.948,
      "grad_norm": 14.080079078674316,
      "kl": 0.40234375,
      "learning_rate": 7.052e-07,
      "loss": 0.0004,
      "reward": 0.25,
      "reward_std": 0.1767766922712326,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.25,
      "step": 737
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 133.84375,
      "epoch": 2.952,
      "grad_norm": 0.01706613413989544,
      "kl": 0.40673828125,
      "learning_rate": 7.047999999999999e-07,
      "loss": 0.0004,
      "reward": 0.0,
      "reward_std": 0.0,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.0,
      "step": 738
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 140.875,
      "epoch": 2.956,
      "grad_norm": 4.372480869293213,
      "kl": 0.412109375,
      "learning_rate": 7.044e-07,
      "loss": 0.0004,
      "reward": 0.5625,
      "reward_std": 0.1157275140285492,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.5625,
      "step": 739
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 125.0,
      "epoch": 2.96,
      "grad_norm": 0.021647008135914803,
      "kl": 0.46142578125,
      "learning_rate": 7.04e-07,
      "loss": 0.0005,
      "reward": 1.0,
      "reward_std": 0.0,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 1.0,
      "step": 740
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 121.0625,
      "epoch": 2.964,
      "grad_norm": 2.4064769744873047,
      "kl": 0.47705078125,
      "learning_rate": 7.035999999999999e-07,
      "loss": 0.0005,
      "reward": 0.28125,
      "reward_std": 0.0883883461356163,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.28125,
      "step": 741
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 125.125,
      "epoch": 2.968,
      "grad_norm": 0.015492529608309269,
      "kl": 0.40625,
      "learning_rate": 7.032e-07,
      "loss": 0.0004,
      "reward": 0.5,
      "reward_std": 0.0,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.5,
      "step": 742
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 136.84375,
      "epoch": 2.972,
      "grad_norm": 4.725335597991943,
      "kl": 0.4384765625,
      "learning_rate": 7.028e-07,
      "loss": 0.0004,
      "reward": 0.59375,
      "reward_std": 0.1293872892856598,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.59375,
      "step": 743
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 141.15625,
      "epoch": 2.976,
      "grad_norm": 0.01998346671462059,
      "kl": 0.4453125,
      "learning_rate": 7.024e-07,
      "loss": 0.0004,
      "reward": 0.5,
      "reward_std": 0.0,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.5,
      "step": 744
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 136.1875,
      "epoch": 2.98,
      "grad_norm": 25.889923095703125,
      "kl": 0.404296875,
      "learning_rate": 7.019999999999999e-07,
      "loss": 0.0004,
      "reward": 0.46875,
      "reward_std": 0.0883883461356163,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.46875,
      "step": 745
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 124.28125,
      "epoch": 2.984,
      "grad_norm": 2.9660818576812744,
      "kl": 0.44580078125,
      "learning_rate": 7.016e-07,
      "loss": 0.0004,
      "reward": 0.21875,
      "reward_std": 0.0883883461356163,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.21875,
      "step": 746
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 121.4375,
      "epoch": 2.988,
      "grad_norm": 0.07212622463703156,
      "kl": 0.52978515625,
      "learning_rate": 7.012000000000001e-07,
      "loss": 0.0005,
      "reward": 1.0,
      "reward_std": 0.0,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 1.0,
      "step": 747
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 135.28125,
      "epoch": 2.992,
      "grad_norm": 0.018393320962786674,
      "kl": 0.4306640625,
      "learning_rate": 7.007999999999999e-07,
      "loss": 0.0004,
      "reward": 0.5,
      "reward_std": 0.0,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.5,
      "step": 748
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 137.375,
      "epoch": 2.996,
      "grad_norm": 0.0674157664179802,
      "kl": 0.408203125,
      "learning_rate": 7.004e-07,
      "loss": 0.0004,
      "reward": 0.0,
      "reward_std": 0.0,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.0,
      "step": 749
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 129.84375,
      "epoch": 3.0,
      "grad_norm": 5.797313690185547,
      "kl": 0.4384765625,
      "learning_rate": 7e-07,
      "loss": 0.0004,
      "reward": 0.59375,
      "reward_std": 0.22201896458864212,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.59375,
      "step": 750
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 127.0,
      "epoch": 3.004,
      "grad_norm": 3.3457190990448,
      "kl": 0.4287109375,
      "learning_rate": 6.995999999999999e-07,
      "loss": 0.0004,
      "reward": 0.28125,
      "reward_std": 0.0883883461356163,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.28125,
      "step": 751
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 135.0625,
      "epoch": 3.008,
      "grad_norm": 5.22096061706543,
      "kl": 0.46923828125,
      "learning_rate": 6.992e-07,
      "loss": 0.0005,
      "reward": 0.375,
      "reward_std": 0.2314550280570984,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.375,
      "step": 752
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 146.34375,
      "epoch": 3.012,
      "grad_norm": 0.015305626206099987,
      "kl": 0.37451171875,
      "learning_rate": 6.988e-07,
      "loss": 0.0004,
      "reward": 0.0,
      "reward_std": 0.0,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.0,
      "step": 753
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 133.46875,
      "epoch": 3.016,
      "grad_norm": 4.47354793548584,
      "kl": 0.5,
      "learning_rate": 6.984e-07,
      "loss": 0.0005,
      "reward": 0.34375,
      "reward_std": 0.1293872892856598,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.34375,
      "step": 754
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 141.71875,
      "epoch": 3.02,
      "grad_norm": 0.039025239646434784,
      "kl": 0.416015625,
      "learning_rate": 6.979999999999999e-07,
      "loss": 0.0004,
      "reward": 0.5,
      "reward_std": 0.0,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.5,
      "step": 755
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 136.6875,
      "epoch": 3.024,
      "grad_norm": 0.028519855812191963,
      "kl": 0.41552734375,
      "learning_rate": 6.976e-07,
      "loss": 0.0004,
      "reward": 0.5,
      "reward_std": 0.0,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.5,
      "step": 756
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 128.3125,
      "epoch": 3.028,
      "grad_norm": 3.1949472427368164,
      "kl": 0.5087890625,
      "learning_rate": 6.972e-07,
      "loss": 0.0005,
      "reward": 0.6875,
      "reward_std": 0.1157275140285492,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.6875,
      "step": 757
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 129.75,
      "epoch": 3.032,
      "grad_norm": 7.704174518585205,
      "kl": 0.4091796875,
      "learning_rate": 6.967999999999999e-07,
      "loss": 0.0004,
      "reward": 0.53125,
      "reward_std": 0.0883883461356163,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.53125,
      "step": 758
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 144.625,
      "epoch": 3.036,
      "grad_norm": 0.01879843696951866,
      "kl": 0.39306640625,
      "learning_rate": 6.964e-07,
      "loss": 0.0004,
      "reward": 0.5,
      "reward_std": 0.0,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.5,
      "step": 759
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 122.46875,
      "epoch": 3.04,
      "grad_norm": 0.02637748420238495,
      "kl": 0.46142578125,
      "learning_rate": 6.959999999999999e-07,
      "loss": 0.0005,
      "reward": 0.25,
      "reward_std": 0.0,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.25,
      "step": 760
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 155.59375,
      "epoch": 3.044,
      "grad_norm": 1.9660699367523193,
      "kl": 0.37841796875,
      "learning_rate": 6.956e-07,
      "loss": 0.0004,
      "reward": 0.78125,
      "reward_std": 0.0883883461356163,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.78125,
      "step": 761
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 132.5,
      "epoch": 3.048,
      "grad_norm": 4.0345001220703125,
      "kl": 0.423828125,
      "learning_rate": 6.952e-07,
      "loss": 0.0004,
      "reward": 0.6875,
      "reward_std": 0.1767766922712326,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.6875,
      "step": 762
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 145.25,
      "epoch": 3.052,
      "grad_norm": 2.4527487754821777,
      "kl": 0.37744140625,
      "learning_rate": 6.947999999999999e-07,
      "loss": 0.0004,
      "reward": 0.59375,
      "reward_std": 0.1293872892856598,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.59375,
      "step": 763
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 152.21875,
      "epoch": 3.056,
      "grad_norm": 8.350884437561035,
      "kl": 0.36328125,
      "learning_rate": 6.944e-07,
      "loss": 0.0004,
      "reward": 0.375,
      "reward_std": 0.13363061845302582,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.375,
      "step": 764
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 135.96875,
      "epoch": 3.06,
      "grad_norm": 4.205017566680908,
      "kl": 0.4365234375,
      "learning_rate": 6.939999999999999e-07,
      "loss": 0.0004,
      "reward": 0.34375,
      "reward_std": 0.2041158601641655,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.34375,
      "step": 765
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 136.34375,
      "epoch": 3.064,
      "grad_norm": 25.4359073638916,
      "kl": 0.4267578125,
      "learning_rate": 6.935999999999999e-07,
      "loss": 0.0004,
      "reward": 0.65625,
      "reward_std": 0.2041158601641655,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.65625,
      "step": 766
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 153.4375,
      "epoch": 3.068,
      "grad_norm": 3.0910513401031494,
      "kl": 0.384765625,
      "learning_rate": 6.932e-07,
      "loss": 0.0004,
      "reward": 0.125,
      "reward_std": 0.2177756354212761,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.125,
      "step": 767
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 136.28125,
      "epoch": 3.072,
      "grad_norm": 0.019646013155579567,
      "kl": 0.3759765625,
      "learning_rate": 6.928e-07,
      "loss": 0.0004,
      "reward": 0.25,
      "reward_std": 0.0,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.25,
      "step": 768
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 129.0,
      "epoch": 3.076,
      "grad_norm": 5.3454179763793945,
      "kl": 0.43359375,
      "learning_rate": 6.924e-07,
      "loss": 0.0004,
      "reward": 0.46875,
      "reward_std": 0.24511480331420898,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.46875,
      "step": 769
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 138.0,
      "epoch": 3.08,
      "grad_norm": 3.4114677906036377,
      "kl": 0.37890625,
      "learning_rate": 6.919999999999999e-07,
      "loss": 0.0004,
      "reward": 0.53125,
      "reward_std": 0.0883883461356163,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.53125,
      "step": 770
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 134.71875,
      "epoch": 3.084,
      "grad_norm": 0.07146432250738144,
      "kl": 0.466796875,
      "learning_rate": 6.916e-07,
      "loss": 0.0005,
      "reward": 0.25,
      "reward_std": 0.0,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.25,
      "step": 771
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 137.6875,
      "epoch": 3.088,
      "grad_norm": 17.032373428344727,
      "kl": 0.369140625,
      "learning_rate": 6.912e-07,
      "loss": 0.0004,
      "reward": 0.5,
      "reward_std": 0.2314550280570984,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.5,
      "step": 772
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 131.59375,
      "epoch": 3.092,
      "grad_norm": 7.009899616241455,
      "kl": 0.40087890625,
      "learning_rate": 6.907999999999999e-07,
      "loss": 0.0004,
      "reward": 0.53125,
      "reward_std": 0.2041158601641655,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.53125,
      "step": 773
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 155.65625,
      "epoch": 3.096,
      "grad_norm": 0.02873370051383972,
      "kl": 0.361328125,
      "learning_rate": 6.904e-07,
      "loss": 0.0004,
      "reward": 0.75,
      "reward_std": 0.0,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.75,
      "step": 774
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 140.625,
      "epoch": 3.1,
      "grad_norm": 3.3769822120666504,
      "kl": 0.39404296875,
      "learning_rate": 6.9e-07,
      "loss": 0.0004,
      "reward": 0.21875,
      "reward_std": 0.0883883461356163,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.21875,
      "step": 775
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 152.3125,
      "epoch": 3.104,
      "grad_norm": 0.05919076129794121,
      "kl": 0.41259765625,
      "learning_rate": 6.895999999999999e-07,
      "loss": 0.0004,
      "reward": 0.5,
      "reward_std": 0.0,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.5,
      "step": 776
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 136.6875,
      "epoch": 3.108,
      "grad_norm": 5.29287052154541,
      "kl": 0.40869140625,
      "learning_rate": 6.892e-07,
      "loss": 0.0004,
      "reward": 0.40625,
      "reward_std": 0.1293872892856598,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.40625,
      "step": 777
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 152.0625,
      "epoch": 3.112,
      "grad_norm": 0.015367278829216957,
      "kl": 0.36767578125,
      "learning_rate": 6.888e-07,
      "loss": 0.0004,
      "reward": 0.25,
      "reward_std": 0.0,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.25,
      "step": 778
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 126.4375,
      "epoch": 3.116,
      "grad_norm": 5.413784027099609,
      "kl": 0.4296875,
      "learning_rate": 6.883999999999999e-07,
      "loss": 0.0004,
      "reward": 0.28125,
      "reward_std": 0.0883883461356163,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.28125,
      "step": 779
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 156.96875,
      "epoch": 3.12,
      "grad_norm": 0.01983475312590599,
      "kl": 0.36279296875,
      "learning_rate": 6.879999999999999e-07,
      "loss": 0.0004,
      "reward": 0.25,
      "reward_std": 0.0,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.25,
      "step": 780
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 141.875,
      "epoch": 3.124,
      "grad_norm": 3.2898848056793213,
      "kl": 0.4111328125,
      "learning_rate": 6.876e-07,
      "loss": 0.0004,
      "reward": 0.53125,
      "reward_std": 0.2041158601641655,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.53125,
      "step": 781
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 141.96875,
      "epoch": 3.128,
      "grad_norm": 2.971181869506836,
      "kl": 0.376953125,
      "learning_rate": 6.872e-07,
      "loss": 0.0004,
      "reward": 0.59375,
      "reward_std": 0.1293872892856598,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.59375,
      "step": 782
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 154.0625,
      "epoch": 3.132,
      "grad_norm": 14.039602279663086,
      "kl": 0.34619140625,
      "learning_rate": 6.867999999999999e-07,
      "loss": 0.0003,
      "reward": 0.71875,
      "reward_std": 0.0883883461356163,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.71875,
      "step": 783
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 142.625,
      "epoch": 3.136,
      "grad_norm": 0.014979549683630466,
      "kl": 0.349609375,
      "learning_rate": 6.864e-07,
      "loss": 0.0003,
      "reward": 0.75,
      "reward_std": 0.0,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.75,
      "step": 784
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 138.53125,
      "epoch": 3.14,
      "grad_norm": 6.834115505218506,
      "kl": 0.3701171875,
      "learning_rate": 6.86e-07,
      "loss": 0.0004,
      "reward": 0.03125,
      "reward_std": 0.0883883461356163,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.03125,
      "step": 785
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 145.875,
      "epoch": 3.144,
      "grad_norm": 2.302032232284546,
      "kl": 0.388671875,
      "learning_rate": 6.855999999999999e-07,
      "loss": 0.0004,
      "reward": 0.53125,
      "reward_std": 0.0883883461356163,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.53125,
      "step": 786
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 132.46875,
      "epoch": 3.148,
      "grad_norm": 0.01389025803655386,
      "kl": 0.34912109375,
      "learning_rate": 6.852e-07,
      "loss": 0.0003,
      "reward": 0.75,
      "reward_std": 0.0,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.75,
      "step": 787
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 151.84375,
      "epoch": 3.152,
      "grad_norm": 4.1845879554748535,
      "kl": 0.353515625,
      "learning_rate": 6.847999999999999e-07,
      "loss": 0.0004,
      "reward": 0.625,
      "reward_std": 0.2177756354212761,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.625,
      "step": 788
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 115.0,
      "epoch": 3.156,
      "grad_norm": 7.745528697967529,
      "kl": 0.375,
      "learning_rate": 6.844e-07,
      "loss": 0.0004,
      "reward": 0.28125,
      "reward_std": 0.0883883461356163,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.28125,
      "step": 789
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 122.03125,
      "epoch": 3.16,
      "grad_norm": 5.739799499511719,
      "kl": 0.36279296875,
      "learning_rate": 6.84e-07,
      "loss": 0.0004,
      "reward": 0.40625,
      "reward_std": 0.3061639815568924,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.40625,
      "step": 790
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 158.34375,
      "epoch": 3.164,
      "grad_norm": 0.02073458395898342,
      "kl": 0.35546875,
      "learning_rate": 6.836e-07,
      "loss": 0.0004,
      "reward": 0.25,
      "reward_std": 0.0,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.25,
      "step": 791
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 143.78125,
      "epoch": 3.168,
      "grad_norm": 7.826498985290527,
      "kl": 0.3896484375,
      "learning_rate": 6.832e-07,
      "loss": 0.0004,
      "reward": 0.21875,
      "reward_std": 0.0883883461356163,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.21875,
      "step": 792
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 110.5625,
      "epoch": 3.172,
      "grad_norm": 0.026096664369106293,
      "kl": 0.396484375,
      "learning_rate": 6.827999999999999e-07,
      "loss": 0.0004,
      "reward": 0.25,
      "reward_std": 0.0,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.25,
      "step": 793
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 131.75,
      "epoch": 3.176,
      "grad_norm": 2.7475388050079346,
      "kl": 0.47021484375,
      "learning_rate": 6.824e-07,
      "loss": 0.0005,
      "reward": 0.5625,
      "reward_std": 0.1157275140285492,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.5625,
      "step": 794
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 129.3125,
      "epoch": 3.18,
      "grad_norm": 5.418735027313232,
      "kl": 0.40771484375,
      "learning_rate": 6.82e-07,
      "loss": 0.0004,
      "reward": 0.59375,
      "reward_std": 0.22201896458864212,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.59375,
      "step": 795
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 136.71875,
      "epoch": 3.184,
      "grad_norm": 4.583237648010254,
      "kl": 0.419921875,
      "learning_rate": 6.816e-07,
      "loss": 0.0004,
      "reward": 0.65625,
      "reward_std": 0.3061639815568924,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.65625,
      "step": 796
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 143.28125,
      "epoch": 3.188,
      "grad_norm": 8.272915840148926,
      "kl": 0.802734375,
      "learning_rate": 6.812e-07,
      "loss": 0.0008,
      "reward": 0.1875,
      "reward_std": 0.1157275140285492,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.1875,
      "step": 797
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 138.125,
      "epoch": 3.192,
      "grad_norm": 5.474944114685059,
      "kl": 0.41650390625,
      "learning_rate": 6.807999999999999e-07,
      "loss": 0.0004,
      "reward": 0.5,
      "reward_std": 0.26726123690605164,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.5,
      "step": 798
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 130.90625,
      "epoch": 3.196,
      "grad_norm": 2.3101980686187744,
      "kl": 0.43310546875,
      "learning_rate": 6.804e-07,
      "loss": 0.0004,
      "reward": 0.53125,
      "reward_std": 0.0883883461356163,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.53125,
      "step": 799
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 146.625,
      "epoch": 3.2,
      "grad_norm": 6.07039213180542,
      "kl": 0.4140625,
      "learning_rate": 6.800000000000001e-07,
      "loss": 0.0004,
      "reward": 0.65625,
      "reward_std": 0.1293872892856598,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.65625,
      "step": 800
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 132.65625,
      "epoch": 3.204,
      "grad_norm": 0.0351131372153759,
      "kl": 0.38916015625,
      "learning_rate": 6.795999999999999e-07,
      "loss": 0.0004,
      "reward": 0.5,
      "reward_std": 0.0,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.5,
      "step": 801
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 123.5625,
      "epoch": 3.208,
      "grad_norm": 10.315834999084473,
      "kl": 0.396484375,
      "learning_rate": 6.792e-07,
      "loss": 0.0004,
      "reward": 0.46875,
      "reward_std": 0.2041158601641655,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.46875,
      "step": 802
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 137.28125,
      "epoch": 3.212,
      "grad_norm": 0.01975666731595993,
      "kl": 0.40185546875,
      "learning_rate": 6.788e-07,
      "loss": 0.0004,
      "reward": 0.75,
      "reward_std": 0.0,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.75,
      "step": 803
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 147.1875,
      "epoch": 3.216,
      "grad_norm": 9.307049751281738,
      "kl": 0.41015625,
      "learning_rate": 6.783999999999999e-07,
      "loss": 0.0004,
      "reward": 0.3125,
      "reward_std": 0.3514062538743019,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.3125,
      "step": 804
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 137.4375,
      "epoch": 3.22,
      "grad_norm": 5.075401306152344,
      "kl": 0.3857421875,
      "learning_rate": 6.78e-07,
      "loss": 0.0004,
      "reward": 0.65625,
      "reward_std": 0.1293872892856598,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.65625,
      "step": 805
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 158.28125,
      "epoch": 3.224,
      "grad_norm": 0.01594218611717224,
      "kl": 0.36669921875,
      "learning_rate": 6.776e-07,
      "loss": 0.0004,
      "reward": 0.5,
      "reward_std": 0.0,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.5,
      "step": 806
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 121.28125,
      "epoch": 3.228,
      "grad_norm": 4.991201400756836,
      "kl": 0.4228515625,
      "learning_rate": 6.772e-07,
      "loss": 0.0004,
      "reward": 0.375,
      "reward_std": 0.13363061845302582,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.375,
      "step": 807
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 139.28125,
      "epoch": 3.232,
      "grad_norm": 7.386273384094238,
      "kl": 0.44677734375,
      "learning_rate": 6.767999999999999e-07,
      "loss": 0.0004,
      "reward": 0.5,
      "reward_std": 0.1767766922712326,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.5,
      "step": 808
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 139.25,
      "epoch": 3.2359999999999998,
      "grad_norm": 0.2657659649848938,
      "kl": 0.75,
      "learning_rate": 6.764e-07,
      "loss": 0.0008,
      "reward": 0.75,
      "reward_std": 0.0,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.75,
      "step": 809
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 139.34375,
      "epoch": 3.24,
      "grad_norm": 0.1089489683508873,
      "kl": 0.509765625,
      "learning_rate": 6.76e-07,
      "loss": 0.0005,
      "reward": 0.75,
      "reward_std": 0.0,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.75,
      "step": 810
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 121.28125,
      "epoch": 3.2439999999999998,
      "grad_norm": 2.7961416244506836,
      "kl": 0.4296875,
      "learning_rate": 6.755999999999999e-07,
      "loss": 0.0004,
      "reward": 0.40625,
      "reward_std": 0.1293872892856598,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.40625,
      "step": 811
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 140.96875,
      "epoch": 3.248,
      "grad_norm": 3.71555233001709,
      "kl": 0.380859375,
      "learning_rate": 6.752e-07,
      "loss": 0.0004,
      "reward": 0.625,
      "reward_std": 0.13363061845302582,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.625,
      "step": 812
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 142.84375,
      "epoch": 3.252,
      "grad_norm": 3.440845012664795,
      "kl": 0.45556640625,
      "learning_rate": 6.747999999999999e-07,
      "loss": 0.0005,
      "reward": 0.53125,
      "reward_std": 0.0883883461356163,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.53125,
      "step": 813
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 148.59375,
      "epoch": 3.2560000000000002,
      "grad_norm": 2.4388344287872314,
      "kl": 0.4111328125,
      "learning_rate": 6.744e-07,
      "loss": 0.0004,
      "reward": 0.46875,
      "reward_std": 0.0883883461356163,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.46875,
      "step": 814
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 152.375,
      "epoch": 3.26,
      "grad_norm": 0.03461618721485138,
      "kl": 0.3935546875,
      "learning_rate": 6.74e-07,
      "loss": 0.0004,
      "reward": 0.25,
      "reward_std": 0.0,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.25,
      "step": 815
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 134.5625,
      "epoch": 3.2640000000000002,
      "grad_norm": 2.8926398754119873,
      "kl": 0.40869140625,
      "learning_rate": 6.736e-07,
      "loss": 0.0004,
      "reward": 0.3125,
      "reward_std": 0.1157275140285492,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.3125,
      "step": 816
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 154.21875,
      "epoch": 3.268,
      "grad_norm": 4.236978054046631,
      "kl": 0.39794921875,
      "learning_rate": 6.732e-07,
      "loss": 0.0004,
      "reward": 0.65625,
      "reward_std": 0.1293872892856598,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.65625,
      "step": 817
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 152.4375,
      "epoch": 3.2720000000000002,
      "grad_norm": 5.333902835845947,
      "kl": 0.4140625,
      "learning_rate": 6.727999999999999e-07,
      "loss": 0.0004,
      "reward": 0.5625,
      "reward_std": 0.1157275140285492,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.5625,
      "step": 818
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 145.1875,
      "epoch": 3.276,
      "grad_norm": 0.011696533299982548,
      "kl": 0.3828125,
      "learning_rate": 6.724e-07,
      "loss": 0.0004,
      "reward": 0.75,
      "reward_std": 0.0,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.75,
      "step": 819
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 138.125,
      "epoch": 3.2800000000000002,
      "grad_norm": 0.036399614065885544,
      "kl": 0.4296875,
      "learning_rate": 6.72e-07,
      "loss": 0.0004,
      "reward": 0.5,
      "reward_std": 0.0,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.5,
      "step": 820
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 154.03125,
      "epoch": 3.284,
      "grad_norm": 2.6736812591552734,
      "kl": 0.3701171875,
      "learning_rate": 6.716e-07,
      "loss": 0.0004,
      "reward": 0.375,
      "reward_std": 0.13363061845302582,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.375,
      "step": 821
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 152.28125,
      "epoch": 3.288,
      "grad_norm": 4.77061128616333,
      "kl": 0.36767578125,
      "learning_rate": 6.712e-07,
      "loss": 0.0004,
      "reward": 0.46875,
      "reward_std": 0.0883883461356163,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.46875,
      "step": 822
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 154.90625,
      "epoch": 3.292,
      "grad_norm": 0.02860257215797901,
      "kl": 0.38916015625,
      "learning_rate": 6.707999999999999e-07,
      "loss": 0.0004,
      "reward": 0.25,
      "reward_std": 0.0,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.25,
      "step": 823
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 136.875,
      "epoch": 3.296,
      "grad_norm": 3.1526553630828857,
      "kl": 0.37353515625,
      "learning_rate": 6.704e-07,
      "loss": 0.0004,
      "reward": 0.3125,
      "reward_std": 0.1157275140285492,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.3125,
      "step": 824
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 155.71875,
      "epoch": 3.3,
      "grad_norm": 0.18805958330631256,
      "kl": 0.3837890625,
      "learning_rate": 6.7e-07,
      "loss": 0.0004,
      "reward": 0.0,
      "reward_std": 0.0,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.0,
      "step": 825
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 153.5625,
      "epoch": 3.304,
      "grad_norm": 0.014158704318106174,
      "kl": 0.388671875,
      "learning_rate": 6.695999999999999e-07,
      "loss": 0.0004,
      "reward": 0.0,
      "reward_std": 0.0,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.0,
      "step": 826
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 144.5625,
      "epoch": 3.308,
      "grad_norm": 0.016404077410697937,
      "kl": 0.42529296875,
      "learning_rate": 6.692e-07,
      "loss": 0.0004,
      "reward": 0.25,
      "reward_std": 0.0,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.25,
      "step": 827
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 154.5,
      "epoch": 3.312,
      "grad_norm": 0.0162848848849535,
      "kl": 0.33740234375,
      "learning_rate": 6.688e-07,
      "loss": 0.0003,
      "reward": 0.5,
      "reward_std": 0.0,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.5,
      "step": 828
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 181.0625,
      "epoch": 3.316,
      "grad_norm": 0.013668426312506199,
      "kl": 0.35302734375,
      "learning_rate": 6.683999999999999e-07,
      "loss": 0.0004,
      "reward": 0.75,
      "reward_std": 0.0,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.75,
      "step": 829
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 178.71875,
      "epoch": 3.32,
      "grad_norm": 6.193622589111328,
      "kl": 0.341796875,
      "learning_rate": 6.68e-07,
      "loss": 0.0003,
      "reward": 0.4375,
      "reward_std": 0.2177756354212761,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.4375,
      "step": 830
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 154.0625,
      "epoch": 3.324,
      "grad_norm": 2.833545207977295,
      "kl": 0.3212890625,
      "learning_rate": 6.676e-07,
      "loss": 0.0003,
      "reward": 0.625,
      "reward_std": 0.13363061845302582,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.625,
      "step": 831
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 146.625,
      "epoch": 3.328,
      "grad_norm": 3.596762180328369,
      "kl": 0.345703125,
      "learning_rate": 6.671999999999999e-07,
      "loss": 0.0003,
      "reward": 0.1875,
      "reward_std": 0.1157275140285492,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.1875,
      "step": 832
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 160.59375,
      "epoch": 3.332,
      "grad_norm": 40.31398391723633,
      "kl": 0.34326171875,
      "learning_rate": 6.667999999999999e-07,
      "loss": 0.0003,
      "reward": 0.0625,
      "reward_std": 0.1157275140285492,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.0625,
      "step": 833
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 151.59375,
      "epoch": 3.336,
      "grad_norm": 3.4847445487976074,
      "kl": 0.36181640625,
      "learning_rate": 6.664e-07,
      "loss": 0.0004,
      "reward": 0.40625,
      "reward_std": 0.2041158601641655,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.40625,
      "step": 834
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 178.34375,
      "epoch": 3.34,
      "grad_norm": 6.8640289306640625,
      "kl": 0.39453125,
      "learning_rate": 6.66e-07,
      "loss": 0.0004,
      "reward": 0.40625,
      "reward_std": 0.2041158601641655,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.40625,
      "step": 835
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 151.625,
      "epoch": 3.344,
      "grad_norm": 5.504528045654297,
      "kl": 0.3740234375,
      "learning_rate": 6.655999999999999e-07,
      "loss": 0.0004,
      "reward": 0.875,
      "reward_std": 0.2314550280570984,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.875,
      "step": 836
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 169.3125,
      "epoch": 3.348,
      "grad_norm": 0.012134571559727192,
      "kl": 0.357421875,
      "learning_rate": 6.652e-07,
      "loss": 0.0004,
      "reward": 0.5,
      "reward_std": 0.0,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.5,
      "step": 837
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 150.5,
      "epoch": 3.352,
      "grad_norm": 2.0329971313476562,
      "kl": 0.38671875,
      "learning_rate": 6.647999999999999e-07,
      "loss": 0.0004,
      "reward": 0.28125,
      "reward_std": 0.0883883461356163,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.28125,
      "step": 838
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 188.90625,
      "epoch": 3.356,
      "grad_norm": 3.4732301235198975,
      "kl": 0.32861328125,
      "learning_rate": 6.643999999999999e-07,
      "loss": 0.0003,
      "reward": 0.5625,
      "reward_std": 0.2177756354212761,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.5625,
      "step": 839
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 170.75,
      "epoch": 3.36,
      "grad_norm": 7.612156867980957,
      "kl": 0.33056640625,
      "learning_rate": 6.64e-07,
      "loss": 0.0003,
      "reward": 0.5625,
      "reward_std": 0.3104073107242584,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.5625,
      "step": 840
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 159.5625,
      "epoch": 3.364,
      "grad_norm": 0.01553240604698658,
      "kl": 0.35986328125,
      "learning_rate": 6.636e-07,
      "loss": 0.0004,
      "reward": 0.25,
      "reward_std": 0.0,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.25,
      "step": 841
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 173.09375,
      "epoch": 3.368,
      "grad_norm": 8.351917266845703,
      "kl": 0.30712890625,
      "learning_rate": 6.632e-07,
      "loss": 0.0003,
      "reward": 0.59375,
      "reward_std": 0.3608423173427582,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.59375,
      "step": 842
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 179.03125,
      "epoch": 3.372,
      "grad_norm": 0.024146631360054016,
      "kl": 0.31396484375,
      "learning_rate": 6.627999999999999e-07,
      "loss": 0.0003,
      "reward": 0.25,
      "reward_std": 0.0,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.25,
      "step": 843
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 159.03125,
      "epoch": 3.376,
      "grad_norm": 3.2570314407348633,
      "kl": 0.3349609375,
      "learning_rate": 6.624e-07,
      "loss": 0.0003,
      "reward": 0.6875,
      "reward_std": 0.1767766922712326,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.6875,
      "step": 844
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 173.8125,
      "epoch": 3.38,
      "grad_norm": 4.933349132537842,
      "kl": 0.313232421875,
      "learning_rate": 6.62e-07,
      "loss": 0.0003,
      "reward": 0.96875,
      "reward_std": 0.0883883461356163,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.96875,
      "step": 845
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 170.5,
      "epoch": 3.384,
      "grad_norm": 2.968602180480957,
      "kl": 0.31103515625,
      "learning_rate": 6.615999999999999e-07,
      "loss": 0.0003,
      "reward": 0.6875,
      "reward_std": 0.1767766922712326,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.6875,
      "step": 846
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 149.46875,
      "epoch": 3.388,
      "grad_norm": 4.907530784606934,
      "kl": 0.357421875,
      "learning_rate": 6.612e-07,
      "loss": 0.0004,
      "reward": 0.375,
      "reward_std": 0.3945523276925087,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.375,
      "step": 847
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 169.5,
      "epoch": 3.392,
      "grad_norm": 7.897972583770752,
      "kl": 0.34375,
      "learning_rate": 6.608e-07,
      "loss": 0.0003,
      "reward": 0.125,
      "reward_std": 0.2177756354212761,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.125,
      "step": 848
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 176.46875,
      "epoch": 3.396,
      "grad_norm": 5.373469829559326,
      "kl": 0.28955078125,
      "learning_rate": 6.604e-07,
      "loss": 0.0003,
      "reward": 0.5,
      "reward_std": 0.1767766922712326,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.5,
      "step": 849
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 154.09375,
      "epoch": 3.4,
      "grad_norm": 1.930970311164856,
      "kl": 0.41357421875,
      "learning_rate": 6.6e-07,
      "loss": 0.0004,
      "reward": 0.71875,
      "reward_std": 0.0883883461356163,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.71875,
      "step": 850
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 160.96875,
      "epoch": 3.404,
      "grad_norm": 0.46500715613365173,
      "kl": 0.36767578125,
      "learning_rate": 6.595999999999999e-07,
      "loss": 0.0004,
      "reward": 1.0,
      "reward_std": 0.0,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 1.0,
      "step": 851
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 167.03125,
      "epoch": 3.408,
      "grad_norm": 4.348776817321777,
      "kl": 0.3359375,
      "learning_rate": 6.592e-07,
      "loss": 0.0003,
      "reward": 0.3125,
      "reward_std": 0.1767766922712326,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.3125,
      "step": 852
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 129.4375,
      "epoch": 3.412,
      "grad_norm": 6.376671314239502,
      "kl": 0.390625,
      "learning_rate": 6.588e-07,
      "loss": 0.0004,
      "reward": 0.4375,
      "reward_std": 0.1767766922712326,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.4375,
      "step": 853
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 144.3125,
      "epoch": 3.416,
      "grad_norm": 5.63402795791626,
      "kl": 0.38818359375,
      "learning_rate": 6.583999999999999e-07,
      "loss": 0.0004,
      "reward": 0.5,
      "reward_std": 0.2925042062997818,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.5,
      "step": 854
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 155.3125,
      "epoch": 3.42,
      "grad_norm": 0.014279457740485668,
      "kl": 0.3291015625,
      "learning_rate": 6.58e-07,
      "loss": 0.0003,
      "reward": 0.75,
      "reward_std": 0.0,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.75,
      "step": 855
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 152.5625,
      "epoch": 3.424,
      "grad_norm": 3.837803602218628,
      "kl": 0.3671875,
      "learning_rate": 6.576e-07,
      "loss": 0.0004,
      "reward": 0.5625,
      "reward_std": 0.2177756354212761,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.5625,
      "step": 856
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 116.71875,
      "epoch": 3.428,
      "grad_norm": 0.028354397043585777,
      "kl": 0.36376953125,
      "learning_rate": 6.571999999999999e-07,
      "loss": 0.0004,
      "reward": 1.0,
      "reward_std": 0.0,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 1.0,
      "step": 857
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 156.375,
      "epoch": 3.432,
      "grad_norm": 2.4925522804260254,
      "kl": 0.33935546875,
      "learning_rate": 6.568e-07,
      "loss": 0.0003,
      "reward": 0.53125,
      "reward_std": 0.0883883461356163,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.53125,
      "step": 858
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 143.09375,
      "epoch": 3.436,
      "grad_norm": 8.009958267211914,
      "kl": 0.3701171875,
      "learning_rate": 6.564e-07,
      "loss": 0.0004,
      "reward": 0.625,
      "reward_std": 0.3650856465101242,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.625,
      "step": 859
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 159.40625,
      "epoch": 3.44,
      "grad_norm": 2.3914809226989746,
      "kl": 0.34130859375,
      "learning_rate": 6.56e-07,
      "loss": 0.0003,
      "reward": 0.40625,
      "reward_std": 0.1293872892856598,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.40625,
      "step": 860
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 142.46875,
      "epoch": 3.444,
      "grad_norm": 0.014350131154060364,
      "kl": 0.34716796875,
      "learning_rate": 6.555999999999999e-07,
      "loss": 0.0003,
      "reward": 0.25,
      "reward_std": 0.0,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.25,
      "step": 861
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 155.6875,
      "epoch": 3.448,
      "grad_norm": 5.6997389793396,
      "kl": 0.3251953125,
      "learning_rate": 6.552e-07,
      "loss": 0.0003,
      "reward": 0.8125,
      "reward_std": 0.1157275140285492,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.8125,
      "step": 862
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 145.78125,
      "epoch": 3.452,
      "grad_norm": 0.019816750660538673,
      "kl": 0.31494140625,
      "learning_rate": 6.548000000000001e-07,
      "loss": 0.0003,
      "reward": 0.75,
      "reward_std": 0.0,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.75,
      "step": 863
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 129.78125,
      "epoch": 3.456,
      "grad_norm": 9.636545181274414,
      "kl": 0.32421875,
      "learning_rate": 6.543999999999999e-07,
      "loss": 0.0003,
      "reward": 0.6875,
      "reward_std": 0.3514062538743019,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.6875,
      "step": 864
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 149.96875,
      "epoch": 3.46,
      "grad_norm": 0.026093220338225365,
      "kl": 0.38037109375,
      "learning_rate": 6.54e-07,
      "loss": 0.0004,
      "reward": 0.5,
      "reward_std": 0.0,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.5,
      "step": 865
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 153.375,
      "epoch": 3.464,
      "grad_norm": 0.07297931611537933,
      "kl": 0.3173828125,
      "learning_rate": 6.536e-07,
      "loss": 0.0003,
      "reward": 0.75,
      "reward_std": 0.0,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.75,
      "step": 866
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 125.625,
      "epoch": 3.468,
      "grad_norm": 0.02715158648788929,
      "kl": 0.3544921875,
      "learning_rate": 6.531999999999999e-07,
      "loss": 0.0004,
      "reward": 0.75,
      "reward_std": 0.0,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.75,
      "step": 867
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 130.71875,
      "epoch": 3.472,
      "grad_norm": 3.483520030975342,
      "kl": 0.35302734375,
      "learning_rate": 6.528e-07,
      "loss": 0.0004,
      "reward": 0.5625,
      "reward_std": 0.2177756354212761,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.5625,
      "step": 868
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 119.96875,
      "epoch": 3.476,
      "grad_norm": 4.439292907714844,
      "kl": 0.39404296875,
      "learning_rate": 6.524e-07,
      "loss": 0.0004,
      "reward": 0.5625,
      "reward_std": 0.1157275140285492,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.5625,
      "step": 869
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 134.875,
      "epoch": 3.48,
      "grad_norm": 0.03019680641591549,
      "kl": 0.35400390625,
      "learning_rate": 6.52e-07,
      "loss": 0.0004,
      "reward": 0.5,
      "reward_std": 0.0,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.5,
      "step": 870
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 144.09375,
      "epoch": 3.484,
      "grad_norm": 0.019718941301107407,
      "kl": 0.30908203125,
      "learning_rate": 6.515999999999999e-07,
      "loss": 0.0003,
      "reward": 0.5,
      "reward_std": 0.0,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.5,
      "step": 871
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 127.71875,
      "epoch": 3.488,
      "grad_norm": 5.099468231201172,
      "kl": 0.40283203125,
      "learning_rate": 6.512e-07,
      "loss": 0.0004,
      "reward": 0.3125,
      "reward_std": 0.1767766922712326,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.3125,
      "step": 872
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 134.9375,
      "epoch": 3.492,
      "grad_norm": 0.01486902590841055,
      "kl": 0.353515625,
      "learning_rate": 6.508e-07,
      "loss": 0.0004,
      "reward": 0.75,
      "reward_std": 0.0,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.75,
      "step": 873
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 123.625,
      "epoch": 3.496,
      "grad_norm": 0.023112261667847633,
      "kl": 0.349609375,
      "learning_rate": 6.504e-07,
      "loss": 0.0003,
      "reward": 0.75,
      "reward_std": 0.0,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.75,
      "step": 874
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 130.34375,
      "epoch": 3.5,
      "grad_norm": 0.01583489030599594,
      "kl": 0.3779296875,
      "learning_rate": 6.5e-07,
      "loss": 0.0004,
      "reward": 1.0,
      "reward_std": 0.0,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 1.0,
      "step": 875
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 126.6875,
      "epoch": 3.504,
      "grad_norm": 0.020490597933530807,
      "kl": 0.33837890625,
      "learning_rate": 6.495999999999999e-07,
      "loss": 0.0003,
      "reward": 1.0,
      "reward_std": 0.0,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 1.0,
      "step": 876
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 138.1875,
      "epoch": 3.508,
      "grad_norm": 16.952529907226562,
      "kl": 0.33740234375,
      "learning_rate": 6.492e-07,
      "loss": 0.0003,
      "reward": 0.8125,
      "reward_std": 0.1157275140285492,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.8125,
      "step": 877
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 135.90625,
      "epoch": 3.512,
      "grad_norm": 3.115140438079834,
      "kl": 0.38037109375,
      "learning_rate": 6.488e-07,
      "loss": 0.0004,
      "reward": 0.6875,
      "reward_std": 0.1157275140285492,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.6875,
      "step": 878
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 125.21875,
      "epoch": 3.516,
      "grad_norm": 0.015883751213550568,
      "kl": 0.39306640625,
      "learning_rate": 6.483999999999999e-07,
      "loss": 0.0004,
      "reward": 0.5,
      "reward_std": 0.0,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.5,
      "step": 879
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 123.4375,
      "epoch": 3.52,
      "grad_norm": 0.020492278039455414,
      "kl": 0.390625,
      "learning_rate": 6.48e-07,
      "loss": 0.0004,
      "reward": 0.0,
      "reward_std": 0.0,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.0,
      "step": 880
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 104.34375,
      "epoch": 3.524,
      "grad_norm": 4.5268754959106445,
      "kl": 0.46533203125,
      "learning_rate": 6.476e-07,
      "loss": 0.0005,
      "reward": 0.40625,
      "reward_std": 0.1293872892856598,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.40625,
      "step": 881
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 119.25,
      "epoch": 3.528,
      "grad_norm": 5.353838920593262,
      "kl": 0.43017578125,
      "learning_rate": 6.471999999999999e-07,
      "loss": 0.0004,
      "reward": 0.65625,
      "reward_std": 0.2041158601641655,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.65625,
      "step": 882
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 127.4375,
      "epoch": 3.532,
      "grad_norm": 0.014445086009800434,
      "kl": 0.3349609375,
      "learning_rate": 6.468e-07,
      "loss": 0.0003,
      "reward": 0.5,
      "reward_std": 0.0,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.5,
      "step": 883
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 120.71875,
      "epoch": 3.536,
      "grad_norm": 0.057173583656549454,
      "kl": 0.380859375,
      "learning_rate": 6.464e-07,
      "loss": 0.0004,
      "reward": 0.75,
      "reward_std": 0.0,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.75,
      "step": 884
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 92.40625,
      "epoch": 3.54,
      "grad_norm": 4.019151210784912,
      "kl": 0.4072265625,
      "learning_rate": 6.46e-07,
      "loss": 0.0004,
      "reward": 0.6875,
      "reward_std": 0.1157275140285492,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.6875,
      "step": 885
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 121.09375,
      "epoch": 3.544,
      "grad_norm": 4.002956390380859,
      "kl": 0.37109375,
      "learning_rate": 6.455999999999999e-07,
      "loss": 0.0004,
      "reward": 0.625,
      "reward_std": 0.13363061845302582,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.625,
      "step": 886
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 118.59375,
      "epoch": 3.548,
      "grad_norm": 59.43404006958008,
      "kl": 0.37158203125,
      "learning_rate": 6.452e-07,
      "loss": 0.0004,
      "reward": 0.875,
      "reward_std": 0.13363061845302582,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.875,
      "step": 887
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 123.0,
      "epoch": 3.552,
      "grad_norm": 0.034085873514413834,
      "kl": 0.42041015625,
      "learning_rate": 6.448000000000001e-07,
      "loss": 0.0004,
      "reward": 0.25,
      "reward_std": 0.0,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.25,
      "step": 888
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 121.15625,
      "epoch": 3.556,
      "grad_norm": 9.283041000366211,
      "kl": 0.3935546875,
      "learning_rate": 6.443999999999999e-07,
      "loss": 0.0004,
      "reward": 0.5625,
      "reward_std": 0.1157275140285492,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.5625,
      "step": 889
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 118.28125,
      "epoch": 3.56,
      "grad_norm": 0.017931606620550156,
      "kl": 0.400390625,
      "learning_rate": 6.44e-07,
      "loss": 0.0004,
      "reward": 0.75,
      "reward_std": 0.0,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.75,
      "step": 890
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 102.59375,
      "epoch": 3.564,
      "grad_norm": 15.351470947265625,
      "kl": 0.4169921875,
      "learning_rate": 6.436e-07,
      "loss": 0.0004,
      "reward": 0.3125,
      "reward_std": 0.1767766922712326,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.3125,
      "step": 891
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 128.875,
      "epoch": 3.568,
      "grad_norm": 6.3688249588012695,
      "kl": 0.7509765625,
      "learning_rate": 6.431999999999999e-07,
      "loss": 0.0007,
      "reward": 0.46875,
      "reward_std": 0.0883883461356163,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.46875,
      "step": 892
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 113.65625,
      "epoch": 3.572,
      "grad_norm": 0.02230403572320938,
      "kl": 0.4072265625,
      "learning_rate": 6.428e-07,
      "loss": 0.0004,
      "reward": 0.25,
      "reward_std": 0.0,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.25,
      "step": 893
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 118.75,
      "epoch": 3.576,
      "grad_norm": 0.03232943266630173,
      "kl": 0.4140625,
      "learning_rate": 6.424e-07,
      "loss": 0.0004,
      "reward": 0.25,
      "reward_std": 0.0,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.25,
      "step": 894
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 120.46875,
      "epoch": 3.58,
      "grad_norm": 0.020676132291555405,
      "kl": 0.40673828125,
      "learning_rate": 6.42e-07,
      "loss": 0.0004,
      "reward": 0.0,
      "reward_std": 0.0,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.0,
      "step": 895
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 109.84375,
      "epoch": 3.584,
      "grad_norm": 0.017553232610225677,
      "kl": 0.400390625,
      "learning_rate": 6.415999999999999e-07,
      "loss": 0.0004,
      "reward": 0.75,
      "reward_std": 0.0,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.75,
      "step": 896
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 113.4375,
      "epoch": 3.588,
      "grad_norm": 3.8409557342529297,
      "kl": 0.43017578125,
      "learning_rate": 6.412e-07,
      "loss": 0.0004,
      "reward": 0.46875,
      "reward_std": 0.0883883461356163,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.46875,
      "step": 897
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 101.15625,
      "epoch": 3.592,
      "grad_norm": 2.7255241870880127,
      "kl": 0.4375,
      "learning_rate": 6.408e-07,
      "loss": 0.0004,
      "reward": 0.21875,
      "reward_std": 0.0883883461356163,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.21875,
      "step": 898
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 103.09375,
      "epoch": 3.596,
      "grad_norm": 0.042015161365270615,
      "kl": 0.45556640625,
      "learning_rate": 6.403999999999999e-07,
      "loss": 0.0005,
      "reward": 0.0,
      "reward_std": 0.0,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.0,
      "step": 899
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 99.125,
      "epoch": 3.6,
      "grad_norm": 9.78475284576416,
      "kl": 0.47802734375,
      "learning_rate": 6.4e-07,
      "loss": 0.0005,
      "reward": 0.21875,
      "reward_std": 0.0883883461356163,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.21875,
      "step": 900
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 115.125,
      "epoch": 3.604,
      "grad_norm": 28.588186264038086,
      "kl": 0.4013671875,
      "learning_rate": 6.395999999999999e-07,
      "loss": 0.0004,
      "reward": 0.28125,
      "reward_std": 0.0883883461356163,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.28125,
      "step": 901
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 99.8125,
      "epoch": 3.608,
      "grad_norm": 9.12941837310791,
      "kl": 0.4296875,
      "learning_rate": 6.392e-07,
      "loss": 0.0004,
      "reward": 0.15625,
      "reward_std": 0.22201896458864212,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.15625,
      "step": 902
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 90.59375,
      "epoch": 3.612,
      "grad_norm": 0.20198337733745575,
      "kl": 0.45458984375,
      "learning_rate": 6.388e-07,
      "loss": 0.0005,
      "reward": 0.75,
      "reward_std": 0.0,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.75,
      "step": 903
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 85.28125,
      "epoch": 3.616,
      "grad_norm": 0.018174080178141594,
      "kl": 0.45654296875,
      "learning_rate": 6.383999999999999e-07,
      "loss": 0.0005,
      "reward": 0.25,
      "reward_std": 0.0,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.25,
      "step": 904
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 101.875,
      "epoch": 3.62,
      "grad_norm": 7.551039218902588,
      "kl": 0.42236328125,
      "learning_rate": 6.38e-07,
      "loss": 0.0004,
      "reward": 0.53125,
      "reward_std": 0.0883883461356163,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.53125,
      "step": 905
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 103.5,
      "epoch": 3.624,
      "grad_norm": 0.15700890123844147,
      "kl": 0.40625,
      "learning_rate": 6.375999999999999e-07,
      "loss": 0.0004,
      "reward": 0.5,
      "reward_std": 0.0,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.5,
      "step": 906
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 111.71875,
      "epoch": 3.628,
      "grad_norm": 3.416736125946045,
      "kl": 0.39453125,
      "learning_rate": 6.371999999999999e-07,
      "loss": 0.0004,
      "reward": 0.53125,
      "reward_std": 0.0883883461356163,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.53125,
      "step": 907
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 90.875,
      "epoch": 3.632,
      "grad_norm": 4.346396446228027,
      "kl": 0.4287109375,
      "learning_rate": 6.368e-07,
      "loss": 0.0004,
      "reward": 0.53125,
      "reward_std": 0.0883883461356163,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.53125,
      "step": 908
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 104.1875,
      "epoch": 3.636,
      "grad_norm": 0.06047122925519943,
      "kl": 0.45361328125,
      "learning_rate": 6.364e-07,
      "loss": 0.0005,
      "reward": 0.5,
      "reward_std": 0.0,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.5,
      "step": 909
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 95.46875,
      "epoch": 3.64,
      "grad_norm": 4.593134880065918,
      "kl": 0.49853515625,
      "learning_rate": 6.36e-07,
      "loss": 0.0005,
      "reward": 0.6875,
      "reward_std": 0.1157275140285492,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.6875,
      "step": 910
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 81.5,
      "epoch": 3.644,
      "grad_norm": 0.031721483916044235,
      "kl": 0.5068359375,
      "learning_rate": 6.356e-07,
      "loss": 0.0005,
      "reward": 0.25,
      "reward_std": 0.0,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.25,
      "step": 911
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 96.125,
      "epoch": 3.648,
      "grad_norm": 7.2418413162231445,
      "kl": 0.46826171875,
      "learning_rate": 6.352e-07,
      "loss": 0.0005,
      "reward": 0.03125,
      "reward_std": 0.0883883461356163,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.03125,
      "step": 912
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 92.0625,
      "epoch": 3.652,
      "grad_norm": 0.04173773527145386,
      "kl": 0.53662109375,
      "learning_rate": 6.348e-07,
      "loss": 0.0005,
      "reward": 0.0,
      "reward_std": 0.0,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.0,
      "step": 913
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 102.46875,
      "epoch": 3.656,
      "grad_norm": 0.024296000599861145,
      "kl": 0.4501953125,
      "learning_rate": 6.343999999999999e-07,
      "loss": 0.0004,
      "reward": 0.25,
      "reward_std": 0.0,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.25,
      "step": 914
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 102.5625,
      "epoch": 3.66,
      "grad_norm": 0.026288993656635284,
      "kl": 0.515625,
      "learning_rate": 6.34e-07,
      "loss": 0.0005,
      "reward": 0.5,
      "reward_std": 0.0,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.5,
      "step": 915
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 92.5,
      "epoch": 3.664,
      "grad_norm": 0.6272538304328918,
      "kl": 0.52490234375,
      "learning_rate": 6.336000000000001e-07,
      "loss": 0.0005,
      "reward": 0.25,
      "reward_std": 0.0,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.25,
      "step": 916
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 95.59375,
      "epoch": 3.668,
      "grad_norm": 3.0255086421966553,
      "kl": 0.50537109375,
      "learning_rate": 6.331999999999999e-07,
      "loss": 0.0005,
      "reward": 0.46875,
      "reward_std": 0.0883883461356163,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.46875,
      "step": 917
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 76.75,
      "epoch": 3.672,
      "grad_norm": 4.8841552734375,
      "kl": 0.55078125,
      "learning_rate": 6.328e-07,
      "loss": 0.0006,
      "reward": 0.6875,
      "reward_std": 0.1767766922712326,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.6875,
      "step": 918
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 86.4375,
      "epoch": 3.676,
      "grad_norm": 5.027050971984863,
      "kl": 0.47802734375,
      "learning_rate": 6.324e-07,
      "loss": 0.0005,
      "reward": 0.28125,
      "reward_std": 0.0883883461356163,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.28125,
      "step": 919
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 103.8125,
      "epoch": 3.68,
      "grad_norm": 13.24487018585205,
      "kl": 0.4423828125,
      "learning_rate": 6.319999999999999e-07,
      "loss": 0.0004,
      "reward": 0.71875,
      "reward_std": 0.0883883461356163,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.71875,
      "step": 920
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 105.53125,
      "epoch": 3.684,
      "grad_norm": 0.01634560152888298,
      "kl": 0.408203125,
      "learning_rate": 6.316e-07,
      "loss": 0.0004,
      "reward": 0.5,
      "reward_std": 0.0,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.5,
      "step": 921
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 77.4375,
      "epoch": 3.6879999999999997,
      "grad_norm": 0.024210568517446518,
      "kl": 0.4921875,
      "learning_rate": 6.312e-07,
      "loss": 0.0005,
      "reward": 0.25,
      "reward_std": 0.0,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.25,
      "step": 922
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 87.65625,
      "epoch": 3.692,
      "grad_norm": 0.024969441816210747,
      "kl": 0.4697265625,
      "learning_rate": 6.308e-07,
      "loss": 0.0005,
      "reward": 0.25,
      "reward_std": 0.0,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.25,
      "step": 923
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 87.34375,
      "epoch": 3.6959999999999997,
      "grad_norm": 14.265741348266602,
      "kl": 0.48095703125,
      "learning_rate": 6.303999999999999e-07,
      "loss": 0.0005,
      "reward": 0.34375,
      "reward_std": 0.1293872892856598,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.34375,
      "step": 924
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 89.875,
      "epoch": 3.7,
      "grad_norm": 2.9257009029388428,
      "kl": 0.45556640625,
      "learning_rate": 6.3e-07,
      "loss": 0.0005,
      "reward": 0.71875,
      "reward_std": 0.0883883461356163,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.71875,
      "step": 925
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 88.78125,
      "epoch": 3.7039999999999997,
      "grad_norm": 0.02930293418467045,
      "kl": 0.4228515625,
      "learning_rate": 6.296e-07,
      "loss": 0.0004,
      "reward": 0.75,
      "reward_std": 0.0,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.75,
      "step": 926
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 91.4375,
      "epoch": 3.708,
      "grad_norm": 0.04982153698801994,
      "kl": 0.47021484375,
      "learning_rate": 6.291999999999999e-07,
      "loss": 0.0005,
      "reward": 0.5,
      "reward_std": 0.0,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.5,
      "step": 927
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 84.65625,
      "epoch": 3.7119999999999997,
      "grad_norm": 17.923877716064453,
      "kl": 0.45654296875,
      "learning_rate": 6.288e-07,
      "loss": 0.0005,
      "reward": 0.84375,
      "reward_std": 0.1293872892856598,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.84375,
      "step": 928
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 94.125,
      "epoch": 3.716,
      "grad_norm": 0.045714374631643295,
      "kl": 0.474609375,
      "learning_rate": 6.283999999999999e-07,
      "loss": 0.0005,
      "reward": 0.0,
      "reward_std": 0.0,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.0,
      "step": 929
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 85.5,
      "epoch": 3.7199999999999998,
      "grad_norm": 0.25551652908325195,
      "kl": 0.48779296875,
      "learning_rate": 6.28e-07,
      "loss": 0.0005,
      "reward": 0.75,
      "reward_std": 0.0,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.75,
      "step": 930
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 85.1875,
      "epoch": 3.724,
      "grad_norm": 0.06522874534130096,
      "kl": 0.5830078125,
      "learning_rate": 6.276e-07,
      "loss": 0.0006,
      "reward": 0.5,
      "reward_std": 0.0,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.5,
      "step": 931
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 101.9375,
      "epoch": 3.7279999999999998,
      "grad_norm": 0.022150270640850067,
      "kl": 0.4189453125,
      "learning_rate": 6.271999999999999e-07,
      "loss": 0.0004,
      "reward": 0.0,
      "reward_std": 0.0,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.0,
      "step": 932
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 95.90625,
      "epoch": 3.732,
      "grad_norm": 3.3656885623931885,
      "kl": 0.4560546875,
      "learning_rate": 6.268e-07,
      "loss": 0.0005,
      "reward": 0.46875,
      "reward_std": 0.0883883461356163,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.46875,
      "step": 933
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 82.625,
      "epoch": 3.7359999999999998,
      "grad_norm": 5.089264392852783,
      "kl": 0.4423828125,
      "learning_rate": 6.263999999999999e-07,
      "loss": 0.0004,
      "reward": 0.34375,
      "reward_std": 0.1293872892856598,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.34375,
      "step": 934
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 93.25,
      "epoch": 3.74,
      "grad_norm": 0.03599882498383522,
      "kl": 0.5029296875,
      "learning_rate": 6.26e-07,
      "loss": 0.0005,
      "reward": 0.5,
      "reward_std": 0.0,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.5,
      "step": 935
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 87.3125,
      "epoch": 3.7439999999999998,
      "grad_norm": 14.653177261352539,
      "kl": 0.47998046875,
      "learning_rate": 6.256e-07,
      "loss": 0.0005,
      "reward": 0.0625,
      "reward_std": 0.1767766922712326,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.0625,
      "step": 936
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 86.875,
      "epoch": 3.748,
      "grad_norm": 2.762113094329834,
      "kl": 0.45068359375,
      "learning_rate": 6.252e-07,
      "loss": 0.0004,
      "reward": 0.21875,
      "reward_std": 0.0883883461356163,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.21875,
      "step": 937
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 86.40625,
      "epoch": 3.752,
      "grad_norm": 157.57774353027344,
      "kl": 0.48583984375,
      "learning_rate": 6.248e-07,
      "loss": 0.0005,
      "reward": 0.15625,
      "reward_std": 0.1293872892856598,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.15625,
      "step": 938
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 74.5625,
      "epoch": 3.7560000000000002,
      "grad_norm": 4.4094929695129395,
      "kl": 0.53271484375,
      "learning_rate": 6.243999999999999e-07,
      "loss": 0.0005,
      "reward": 0.3125,
      "reward_std": 0.1157275140285492,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.3125,
      "step": 939
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 79.875,
      "epoch": 3.76,
      "grad_norm": 0.02680736966431141,
      "kl": 0.5224609375,
      "learning_rate": 6.24e-07,
      "loss": 0.0005,
      "reward": 0.5,
      "reward_std": 0.0,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.5,
      "step": 940
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 88.09375,
      "epoch": 3.7640000000000002,
      "grad_norm": 0.022968243807554245,
      "kl": 0.4638671875,
      "learning_rate": 6.236e-07,
      "loss": 0.0005,
      "reward": 0.25,
      "reward_std": 0.0,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.25,
      "step": 941
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 94.71875,
      "epoch": 3.768,
      "grad_norm": 3.9611449241638184,
      "kl": 0.48486328125,
      "learning_rate": 6.231999999999999e-07,
      "loss": 0.0005,
      "reward": 0.71875,
      "reward_std": 0.0883883461356163,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.71875,
      "step": 942
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 83.53125,
      "epoch": 3.7720000000000002,
      "grad_norm": 4.496031284332275,
      "kl": 0.5693359375,
      "learning_rate": 6.228e-07,
      "loss": 0.0006,
      "reward": 0.40625,
      "reward_std": 0.1293872892856598,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.40625,
      "step": 943
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 102.40625,
      "epoch": 3.776,
      "grad_norm": 6.873812198638916,
      "kl": 0.45068359375,
      "learning_rate": 6.224e-07,
      "loss": 0.0005,
      "reward": 0.375,
      "reward_std": 0.13363061845302582,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.375,
      "step": 944
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 89.03125,
      "epoch": 3.7800000000000002,
      "grad_norm": 0.018681136891245842,
      "kl": 0.5009765625,
      "learning_rate": 6.219999999999999e-07,
      "loss": 0.0005,
      "reward": 0.75,
      "reward_std": 0.0,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.75,
      "step": 945
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 85.71875,
      "epoch": 3.784,
      "grad_norm": 11.728137969970703,
      "kl": 0.49609375,
      "learning_rate": 6.216e-07,
      "loss": 0.0005,
      "reward": 0.375,
      "reward_std": 0.2314550280570984,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.375,
      "step": 946
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 72.15625,
      "epoch": 3.7880000000000003,
      "grad_norm": 0.024085436016321182,
      "kl": 0.5244140625,
      "learning_rate": 6.212e-07,
      "loss": 0.0005,
      "reward": 0.5,
      "reward_std": 0.0,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.5,
      "step": 947
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 76.6875,
      "epoch": 3.792,
      "grad_norm": 15.77620792388916,
      "kl": 0.5263671875,
      "learning_rate": 6.208e-07,
      "loss": 0.0005,
      "reward": 0.5625,
      "reward_std": 0.1767766922712326,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.5625,
      "step": 948
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 93.84375,
      "epoch": 3.7960000000000003,
      "grad_norm": 0.024485567584633827,
      "kl": 0.48974609375,
      "learning_rate": 6.203999999999999e-07,
      "loss": 0.0005,
      "reward": 0.5,
      "reward_std": 0.0,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.5,
      "step": 949
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 80.15625,
      "epoch": 3.8,
      "grad_norm": 9.641373634338379,
      "kl": 0.5712890625,
      "learning_rate": 6.2e-07,
      "loss": 0.0006,
      "reward": 0.34375,
      "reward_std": 0.1293872892856598,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.34375,
      "step": 950
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 91.53125,
      "epoch": 3.8040000000000003,
      "grad_norm": 0.03975791484117508,
      "kl": 0.4365234375,
      "learning_rate": 6.196e-07,
      "loss": 0.0004,
      "reward": 0.5,
      "reward_std": 0.0,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.5,
      "step": 951
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 94.96875,
      "epoch": 3.808,
      "grad_norm": 0.02219507470726967,
      "kl": 0.48095703125,
      "learning_rate": 6.191999999999999e-07,
      "loss": 0.0005,
      "reward": 0.5,
      "reward_std": 0.0,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.5,
      "step": 952
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 82.8125,
      "epoch": 3.8120000000000003,
      "grad_norm": 4.566342830657959,
      "kl": 0.5166015625,
      "learning_rate": 6.188e-07,
      "loss": 0.0005,
      "reward": 0.46875,
      "reward_std": 0.0883883461356163,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.46875,
      "step": 953
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 94.71875,
      "epoch": 3.816,
      "grad_norm": 0.015584059990942478,
      "kl": 0.45556640625,
      "learning_rate": 6.183999999999999e-07,
      "loss": 0.0005,
      "reward": 0.5,
      "reward_std": 0.0,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.5,
      "step": 954
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 93.25,
      "epoch": 3.82,
      "grad_norm": 8.189370155334473,
      "kl": 0.458984375,
      "learning_rate": 6.18e-07,
      "loss": 0.0005,
      "reward": 0.3125,
      "reward_std": 0.1767766922712326,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.3125,
      "step": 955
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 78.65625,
      "epoch": 3.824,
      "grad_norm": 4.044259548187256,
      "kl": 0.4609375,
      "learning_rate": 6.176e-07,
      "loss": 0.0005,
      "reward": 0.71875,
      "reward_std": 0.0883883461356163,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.71875,
      "step": 956
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 101.1875,
      "epoch": 3.828,
      "grad_norm": 6.9039788246154785,
      "kl": 0.404296875,
      "learning_rate": 6.172e-07,
      "loss": 0.0004,
      "reward": 0.40625,
      "reward_std": 0.1293872892856598,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.40625,
      "step": 957
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 84.5625,
      "epoch": 3.832,
      "grad_norm": 0.03601478040218353,
      "kl": 0.45166015625,
      "learning_rate": 6.168e-07,
      "loss": 0.0005,
      "reward": 0.75,
      "reward_std": 0.0,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.75,
      "step": 958
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 78.34375,
      "epoch": 3.836,
      "grad_norm": 0.03652077540755272,
      "kl": 0.53515625,
      "learning_rate": 6.163999999999999e-07,
      "loss": 0.0005,
      "reward": 1.0,
      "reward_std": 0.0,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 1.0,
      "step": 959
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 75.125,
      "epoch": 3.84,
      "grad_norm": 0.04159824922680855,
      "kl": 0.5107421875,
      "learning_rate": 6.16e-07,
      "loss": 0.0005,
      "reward": 0.5,
      "reward_std": 0.0,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.5,
      "step": 960
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 79.3125,
      "epoch": 3.844,
      "grad_norm": 0.05188501998782158,
      "kl": 0.52392578125,
      "learning_rate": 6.156e-07,
      "loss": 0.0005,
      "reward": 0.25,
      "reward_std": 0.0,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.25,
      "step": 961
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 82.28125,
      "epoch": 3.848,
      "grad_norm": 8.360015869140625,
      "kl": 0.44287109375,
      "learning_rate": 6.152e-07,
      "loss": 0.0004,
      "reward": 0.59375,
      "reward_std": 0.1293872892856598,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.59375,
      "step": 962
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 83.28125,
      "epoch": 3.852,
      "grad_norm": 0.032568734139204025,
      "kl": 0.4462890625,
      "learning_rate": 6.148e-07,
      "loss": 0.0004,
      "reward": 0.0,
      "reward_std": 0.0,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.0,
      "step": 963
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 89.84375,
      "epoch": 3.856,
      "grad_norm": 0.02395648881793022,
      "kl": 0.447265625,
      "learning_rate": 6.143999999999999e-07,
      "loss": 0.0004,
      "reward": 0.5,
      "reward_std": 0.0,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.5,
      "step": 964
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 90.84375,
      "epoch": 3.86,
      "grad_norm": 0.01987825147807598,
      "kl": 0.46728515625,
      "learning_rate": 6.14e-07,
      "loss": 0.0005,
      "reward": 0.0,
      "reward_std": 0.0,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.0,
      "step": 965
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 83.15625,
      "epoch": 3.864,
      "grad_norm": 0.10256344825029373,
      "kl": 0.48193359375,
      "learning_rate": 6.136e-07,
      "loss": 0.0005,
      "reward": 0.5,
      "reward_std": 0.0,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.5,
      "step": 966
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 85.0625,
      "epoch": 3.868,
      "grad_norm": 0.05779705569148064,
      "kl": 0.48779296875,
      "learning_rate": 6.131999999999999e-07,
      "loss": 0.0005,
      "reward": 0.25,
      "reward_std": 0.0,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.25,
      "step": 967
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 76.125,
      "epoch": 3.872,
      "grad_norm": 0.01747390255331993,
      "kl": 0.490234375,
      "learning_rate": 6.128e-07,
      "loss": 0.0005,
      "reward": 0.75,
      "reward_std": 0.0,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.75,
      "step": 968
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 81.46875,
      "epoch": 3.876,
      "grad_norm": 0.028449375182390213,
      "kl": 0.51171875,
      "learning_rate": 6.124000000000001e-07,
      "loss": 0.0005,
      "reward": 0.25,
      "reward_std": 0.0,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.25,
      "step": 969
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 86.53125,
      "epoch": 3.88,
      "grad_norm": 0.04495746269822121,
      "kl": 0.52197265625,
      "learning_rate": 6.119999999999999e-07,
      "loss": 0.0005,
      "reward": 0.25,
      "reward_std": 0.0,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.25,
      "step": 970
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 77.625,
      "epoch": 3.884,
      "grad_norm": 3.716093063354492,
      "kl": 0.439453125,
      "learning_rate": 6.116e-07,
      "loss": 0.0004,
      "reward": 0.71875,
      "reward_std": 0.0883883461356163,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.71875,
      "step": 971
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 85.4375,
      "epoch": 3.888,
      "grad_norm": 8.620750427246094,
      "kl": 0.39697265625,
      "learning_rate": 6.112e-07,
      "loss": 0.0004,
      "reward": 0.375,
      "reward_std": 0.13363061845302582,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.375,
      "step": 972
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 65.65625,
      "epoch": 3.892,
      "grad_norm": 0.039972830563783646,
      "kl": 0.55078125,
      "learning_rate": 6.107999999999999e-07,
      "loss": 0.0006,
      "reward": 0.25,
      "reward_std": 0.0,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.25,
      "step": 973
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 80.46875,
      "epoch": 3.896,
      "grad_norm": 0.05005483329296112,
      "kl": 0.4697265625,
      "learning_rate": 6.104e-07,
      "loss": 0.0005,
      "reward": 0.25,
      "reward_std": 0.0,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.25,
      "step": 974
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 83.3125,
      "epoch": 3.9,
      "grad_norm": 0.047045230865478516,
      "kl": 0.50341796875,
      "learning_rate": 6.1e-07,
      "loss": 0.0005,
      "reward": 0.25,
      "reward_std": 0.0,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.25,
      "step": 975
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 88.71875,
      "epoch": 3.904,
      "grad_norm": 0.019349101930856705,
      "kl": 0.453125,
      "learning_rate": 6.096e-07,
      "loss": 0.0005,
      "reward": 0.75,
      "reward_std": 0.0,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.75,
      "step": 976
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 61.75,
      "epoch": 3.908,
      "grad_norm": 0.07856948673725128,
      "kl": 0.52197265625,
      "learning_rate": 6.091999999999999e-07,
      "loss": 0.0005,
      "reward": 0.5,
      "reward_std": 0.0,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.5,
      "step": 977
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 87.46875,
      "epoch": 3.912,
      "grad_norm": 0.02055608481168747,
      "kl": 0.46142578125,
      "learning_rate": 6.088e-07,
      "loss": 0.0005,
      "reward": 0.5,
      "reward_std": 0.0,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.5,
      "step": 978
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 68.0,
      "epoch": 3.916,
      "grad_norm": 7.84152364730835,
      "kl": 0.54345703125,
      "learning_rate": 6.084000000000001e-07,
      "loss": 0.0005,
      "reward": 0.8125,
      "reward_std": 0.1157275140285492,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.8125,
      "step": 979
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 98.53125,
      "epoch": 3.92,
      "grad_norm": 0.041875261813402176,
      "kl": 0.423828125,
      "learning_rate": 6.079999999999999e-07,
      "loss": 0.0004,
      "reward": 0.5,
      "reward_std": 0.0,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.5,
      "step": 980
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 77.96875,
      "epoch": 3.924,
      "grad_norm": 0.028017355129122734,
      "kl": 0.4638671875,
      "learning_rate": 6.076e-07,
      "loss": 0.0005,
      "reward": 0.75,
      "reward_std": 0.0,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.75,
      "step": 981
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 76.96875,
      "epoch": 3.928,
      "grad_norm": 10.669384002685547,
      "kl": 0.498046875,
      "learning_rate": 6.072e-07,
      "loss": 0.0005,
      "reward": 0.71875,
      "reward_std": 0.0883883461356163,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.71875,
      "step": 982
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 85.75,
      "epoch": 3.932,
      "grad_norm": 0.02346782386302948,
      "kl": 0.42626953125,
      "learning_rate": 6.068e-07,
      "loss": 0.0004,
      "reward": 0.5,
      "reward_std": 0.0,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.5,
      "step": 983
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 87.65625,
      "epoch": 3.936,
      "grad_norm": 0.053272150456905365,
      "kl": 0.4482421875,
      "learning_rate": 6.064e-07,
      "loss": 0.0004,
      "reward": 0.5,
      "reward_std": 0.0,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.5,
      "step": 984
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 85.71875,
      "epoch": 3.94,
      "grad_norm": 0.027965044602751732,
      "kl": 0.48486328125,
      "learning_rate": 6.06e-07,
      "loss": 0.0005,
      "reward": 1.0,
      "reward_std": 0.0,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 1.0,
      "step": 985
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 75.4375,
      "epoch": 3.944,
      "grad_norm": 11.373795509338379,
      "kl": 0.4990234375,
      "learning_rate": 6.056e-07,
      "loss": 0.0005,
      "reward": 0.3125,
      "reward_std": 0.1157275140285492,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.3125,
      "step": 986
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 92.3125,
      "epoch": 3.948,
      "grad_norm": 0.03638871759176254,
      "kl": 0.37939453125,
      "learning_rate": 6.051999999999999e-07,
      "loss": 0.0004,
      "reward": 0.75,
      "reward_std": 0.0,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.75,
      "step": 987
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 75.28125,
      "epoch": 3.952,
      "grad_norm": 14.874946594238281,
      "kl": 0.44482421875,
      "learning_rate": 6.048e-07,
      "loss": 0.0004,
      "reward": 0.40625,
      "reward_std": 0.1293872892856598,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.40625,
      "step": 988
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 68.34375,
      "epoch": 3.956,
      "grad_norm": 6.231858730316162,
      "kl": 0.6875,
      "learning_rate": 6.044e-07,
      "loss": 0.0007,
      "reward": 0.3125,
      "reward_std": 0.1157275140285492,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.3125,
      "step": 989
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 87.5625,
      "epoch": 3.96,
      "grad_norm": 0.02317621186375618,
      "kl": 0.40625,
      "learning_rate": 6.04e-07,
      "loss": 0.0004,
      "reward": 0.25,
      "reward_std": 0.0,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.25,
      "step": 990
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 92.40625,
      "epoch": 3.964,
      "grad_norm": 0.032953180372714996,
      "kl": 0.44677734375,
      "learning_rate": 6.036e-07,
      "loss": 0.0004,
      "reward": 0.0,
      "reward_std": 0.0,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.0,
      "step": 991
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 82.0,
      "epoch": 3.968,
      "grad_norm": 0.04599643126130104,
      "kl": 0.482421875,
      "learning_rate": 6.031999999999999e-07,
      "loss": 0.0005,
      "reward": 0.5,
      "reward_std": 0.0,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.5,
      "step": 992
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 88.96875,
      "epoch": 3.972,
      "grad_norm": 37.198795318603516,
      "kl": 0.45166015625,
      "learning_rate": 6.028e-07,
      "loss": 0.0005,
      "reward": 0.71875,
      "reward_std": 0.0883883461356163,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.71875,
      "step": 993
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 87.28125,
      "epoch": 3.976,
      "grad_norm": 7.460580348968506,
      "kl": 0.4873046875,
      "learning_rate": 6.024e-07,
      "loss": 0.0005,
      "reward": 0.1875,
      "reward_std": 0.1157275140285492,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.1875,
      "step": 994
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 74.59375,
      "epoch": 3.98,
      "grad_norm": 0.06895284354686737,
      "kl": 0.45751953125,
      "learning_rate": 6.019999999999999e-07,
      "loss": 0.0005,
      "reward": 0.0,
      "reward_std": 0.0,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.0,
      "step": 995
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 83.3125,
      "epoch": 3.984,
      "grad_norm": 6.996845245361328,
      "kl": 0.453125,
      "learning_rate": 6.016e-07,
      "loss": 0.0005,
      "reward": 0.6875,
      "reward_std": 0.1157275140285492,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.6875,
      "step": 996
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 101.6875,
      "epoch": 3.988,
      "grad_norm": 0.015215347521007061,
      "kl": 0.39599609375,
      "learning_rate": 6.012e-07,
      "loss": 0.0004,
      "reward": 0.25,
      "reward_std": 0.0,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.25,
      "step": 997
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 93.46875,
      "epoch": 3.992,
      "grad_norm": 3.8644797801971436,
      "kl": 0.7294921875,
      "learning_rate": 6.007999999999999e-07,
      "loss": 0.0007,
      "reward": 0.6875,
      "reward_std": 0.1157275140285492,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.6875,
      "step": 998
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 99.875,
      "epoch": 3.996,
      "grad_norm": 0.023438185453414917,
      "kl": 0.373046875,
      "learning_rate": 6.004e-07,
      "loss": 0.0004,
      "reward": 0.5,
      "reward_std": 0.0,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.5,
      "step": 999
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 88.71875,
      "epoch": 4.0,
      "grad_norm": 0.025715671479701996,
      "kl": 0.4033203125,
      "learning_rate": 6e-07,
      "loss": 0.0004,
      "reward": 0.25,
      "reward_std": 0.0,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.25,
      "step": 1000
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 106.90625,
      "epoch": 4.004,
      "grad_norm": 0.028165020048618317,
      "kl": 0.3896484375,
      "learning_rate": 5.995999999999999e-07,
      "loss": 0.0004,
      "reward": 0.25,
      "reward_std": 0.0,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.25,
      "step": 1001
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 99.53125,
      "epoch": 4.008,
      "grad_norm": 0.02348499931395054,
      "kl": 0.41650390625,
      "learning_rate": 5.991999999999999e-07,
      "loss": 0.0004,
      "reward": 0.75,
      "reward_std": 0.0,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.75,
      "step": 1002
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 97.5,
      "epoch": 4.012,
      "grad_norm": 0.024073129519820213,
      "kl": 0.396484375,
      "learning_rate": 5.988e-07,
      "loss": 0.0004,
      "reward": 0.25,
      "reward_std": 0.0,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.25,
      "step": 1003
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 78.0,
      "epoch": 4.016,
      "grad_norm": 10.424686431884766,
      "kl": 0.41455078125,
      "learning_rate": 5.984000000000001e-07,
      "loss": 0.0004,
      "reward": 0.46875,
      "reward_std": 0.2651650384068489,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.46875,
      "step": 1004
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 96.28125,
      "epoch": 4.02,
      "grad_norm": 0.07880790531635284,
      "kl": 0.4091796875,
      "learning_rate": 5.979999999999999e-07,
      "loss": 0.0004,
      "reward": 0.5,
      "reward_std": 0.0,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.5,
      "step": 1005
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 92.875,
      "epoch": 4.024,
      "grad_norm": 0.020534174516797066,
      "kl": 0.455078125,
      "learning_rate": 5.976e-07,
      "loss": 0.0005,
      "reward": 0.75,
      "reward_std": 0.0,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.75,
      "step": 1006
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 95.34375,
      "epoch": 4.028,
      "grad_norm": 6.856741428375244,
      "kl": 0.40673828125,
      "learning_rate": 5.972e-07,
      "loss": 0.0004,
      "reward": 0.3125,
      "reward_std": 0.1157275140285492,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.3125,
      "step": 1007
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 97.09375,
      "epoch": 4.032,
      "grad_norm": 5.802669048309326,
      "kl": 0.36181640625,
      "learning_rate": 5.967999999999999e-07,
      "loss": 0.0004,
      "reward": 0.28125,
      "reward_std": 0.0883883461356163,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.28125,
      "step": 1008
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 79.5625,
      "epoch": 4.036,
      "grad_norm": 17.41197395324707,
      "kl": 0.4091796875,
      "learning_rate": 5.964e-07,
      "loss": 0.0004,
      "reward": 0.71875,
      "reward_std": 0.0883883461356163,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.71875,
      "step": 1009
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 113.1875,
      "epoch": 4.04,
      "grad_norm": 0.03334508463740349,
      "kl": 0.35791015625,
      "learning_rate": 5.96e-07,
      "loss": 0.0004,
      "reward": 0.25,
      "reward_std": 0.0,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.25,
      "step": 1010
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 103.71875,
      "epoch": 4.044,
      "grad_norm": 10.246638298034668,
      "kl": 0.41943359375,
      "learning_rate": 5.956e-07,
      "loss": 0.0004,
      "reward": 0.59375,
      "reward_std": 0.2041158601641655,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.59375,
      "step": 1011
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 97.21875,
      "epoch": 4.048,
      "grad_norm": 0.021599557250738144,
      "kl": 0.4111328125,
      "learning_rate": 5.951999999999999e-07,
      "loss": 0.0004,
      "reward": 0.75,
      "reward_std": 0.0,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.75,
      "step": 1012
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 93.90625,
      "epoch": 4.052,
      "grad_norm": 0.01371154747903347,
      "kl": 0.38037109375,
      "learning_rate": 5.948e-07,
      "loss": 0.0004,
      "reward": 0.75,
      "reward_std": 0.0,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.75,
      "step": 1013
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 94.9375,
      "epoch": 4.056,
      "grad_norm": 0.028121687471866608,
      "kl": 0.3671875,
      "learning_rate": 5.944e-07,
      "loss": 0.0004,
      "reward": 0.25,
      "reward_std": 0.0,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.25,
      "step": 1014
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 96.21875,
      "epoch": 4.06,
      "grad_norm": 4.1240715980529785,
      "kl": 0.35546875,
      "learning_rate": 5.939999999999999e-07,
      "loss": 0.0004,
      "reward": 0.09375,
      "reward_std": 0.1293872892856598,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.09375,
      "step": 1015
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 109.71875,
      "epoch": 4.064,
      "grad_norm": 0.11714218556880951,
      "kl": 0.39013671875,
      "learning_rate": 5.936e-07,
      "loss": 0.0004,
      "reward": 0.5,
      "reward_std": 0.0,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.5,
      "step": 1016
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 100.4375,
      "epoch": 4.068,
      "grad_norm": 0.041518840938806534,
      "kl": 0.392578125,
      "learning_rate": 5.931999999999999e-07,
      "loss": 0.0004,
      "reward": 0.75,
      "reward_std": 0.0,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.75,
      "step": 1017
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 79.5,
      "epoch": 4.072,
      "grad_norm": 0.04562797397375107,
      "kl": 0.47021484375,
      "learning_rate": 5.928e-07,
      "loss": 0.0005,
      "reward": 0.5,
      "reward_std": 0.0,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.5,
      "step": 1018
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 90.53125,
      "epoch": 4.076,
      "grad_norm": 0.0457560196518898,
      "kl": 0.40234375,
      "learning_rate": 5.924e-07,
      "loss": 0.0004,
      "reward": 0.5,
      "reward_std": 0.0,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.5,
      "step": 1019
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 101.59375,
      "epoch": 4.08,
      "grad_norm": 15.262734413146973,
      "kl": 0.40087890625,
      "learning_rate": 5.919999999999999e-07,
      "loss": 0.0004,
      "reward": 0.40625,
      "reward_std": 0.1293872892856598,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.40625,
      "step": 1020
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 103.0625,
      "epoch": 4.084,
      "grad_norm": 11.54024887084961,
      "kl": 0.40185546875,
      "learning_rate": 5.916e-07,
      "loss": 0.0004,
      "reward": 0.1875,
      "reward_std": 0.2177756354212761,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.1875,
      "step": 1021
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 93.125,
      "epoch": 4.088,
      "grad_norm": 10.655007362365723,
      "kl": 0.369140625,
      "learning_rate": 5.911999999999999e-07,
      "loss": 0.0004,
      "reward": 0.6875,
      "reward_std": 0.1157275140285492,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.6875,
      "step": 1022
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 80.28125,
      "epoch": 4.092,
      "grad_norm": 6.9213032722473145,
      "kl": 0.41015625,
      "learning_rate": 5.907999999999999e-07,
      "loss": 0.0004,
      "reward": 0.71875,
      "reward_std": 0.0883883461356163,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.71875,
      "step": 1023
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 96.875,
      "epoch": 4.096,
      "grad_norm": 17.385894775390625,
      "kl": 0.46337890625,
      "learning_rate": 5.904e-07,
      "loss": 0.0005,
      "reward": 0.6875,
      "reward_std": 0.1157275140285492,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.6875,
      "step": 1024
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 91.09375,
      "epoch": 4.1,
      "grad_norm": 17.795957565307617,
      "kl": 0.4208984375,
      "learning_rate": 5.9e-07,
      "loss": 0.0004,
      "reward": 0.6875,
      "reward_std": 0.1157275140285492,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.6875,
      "step": 1025
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 103.46875,
      "epoch": 4.104,
      "grad_norm": 0.06932353973388672,
      "kl": 0.42724609375,
      "learning_rate": 5.896e-07,
      "loss": 0.0004,
      "reward": 0.5,
      "reward_std": 0.0,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.5,
      "step": 1026
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 87.46875,
      "epoch": 4.108,
      "grad_norm": 6.8877692222595215,
      "kl": 0.4482421875,
      "learning_rate": 5.891999999999999e-07,
      "loss": 0.0004,
      "reward": 0.34375,
      "reward_std": 0.1293872892856598,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.34375,
      "step": 1027
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 107.375,
      "epoch": 4.112,
      "grad_norm": 0.03306640312075615,
      "kl": 0.34130859375,
      "learning_rate": 5.888e-07,
      "loss": 0.0003,
      "reward": 0.75,
      "reward_std": 0.0,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.75,
      "step": 1028
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 93.8125,
      "epoch": 4.116,
      "grad_norm": 3.8260140419006348,
      "kl": 0.40234375,
      "learning_rate": 5.884000000000001e-07,
      "loss": 0.0004,
      "reward": 0.09375,
      "reward_std": 0.1293872892856598,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.09375,
      "step": 1029
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 107.21875,
      "epoch": 4.12,
      "grad_norm": 0.03710152581334114,
      "kl": 0.4384765625,
      "learning_rate": 5.879999999999999e-07,
      "loss": 0.0004,
      "reward": 0.0,
      "reward_std": 0.0,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.0,
      "step": 1030
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 90.40625,
      "epoch": 4.124,
      "grad_norm": 0.03203753009438515,
      "kl": 0.43359375,
      "learning_rate": 5.876e-07,
      "loss": 0.0004,
      "reward": 0.75,
      "reward_std": 0.0,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.75,
      "step": 1031
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 95.28125,
      "epoch": 4.128,
      "grad_norm": 0.018461469560861588,
      "kl": 0.3740234375,
      "learning_rate": 5.872000000000001e-07,
      "loss": 0.0004,
      "reward": 0.25,
      "reward_std": 0.0,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.25,
      "step": 1032
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 82.375,
      "epoch": 4.132,
      "grad_norm": 6.8474225997924805,
      "kl": 0.45654296875,
      "learning_rate": 5.867999999999999e-07,
      "loss": 0.0005,
      "reward": 0.875,
      "reward_std": 0.2177756354212761,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.875,
      "step": 1033
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 84.75,
      "epoch": 4.136,
      "grad_norm": 5.176055908203125,
      "kl": 0.48095703125,
      "learning_rate": 5.864e-07,
      "loss": 0.0005,
      "reward": 0.28125,
      "reward_std": 0.0883883461356163,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.28125,
      "step": 1034
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 90.09375,
      "epoch": 4.14,
      "grad_norm": 8.81025505065918,
      "kl": 0.42138671875,
      "learning_rate": 5.86e-07,
      "loss": 0.0004,
      "reward": 0.1875,
      "reward_std": 0.1157275140285492,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.1875,
      "step": 1035
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 94.8125,
      "epoch": 4.144,
      "grad_norm": 3.071901798248291,
      "kl": 0.52587890625,
      "learning_rate": 5.856e-07,
      "loss": 0.0005,
      "reward": 0.34375,
      "reward_std": 0.1293872892856598,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.34375,
      "step": 1036
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 101.53125,
      "epoch": 4.148,
      "grad_norm": 5.3059539794921875,
      "kl": 0.44091796875,
      "learning_rate": 5.852e-07,
      "loss": 0.0004,
      "reward": 0.71875,
      "reward_std": 0.0883883461356163,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.71875,
      "step": 1037
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 82.15625,
      "epoch": 4.152,
      "grad_norm": 8.074084281921387,
      "kl": 0.404296875,
      "learning_rate": 5.848e-07,
      "loss": 0.0004,
      "reward": 0.46875,
      "reward_std": 0.24511480331420898,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.46875,
      "step": 1038
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 105.53125,
      "epoch": 4.156,
      "grad_norm": 9.092409133911133,
      "kl": 0.361328125,
      "learning_rate": 5.844e-07,
      "loss": 0.0004,
      "reward": 0.375,
      "reward_std": 0.13363061845302582,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.375,
      "step": 1039
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 100.625,
      "epoch": 4.16,
      "grad_norm": 0.0583561547100544,
      "kl": 0.55029296875,
      "learning_rate": 5.839999999999999e-07,
      "loss": 0.0006,
      "reward": 0.5,
      "reward_std": 0.0,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.5,
      "step": 1040
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 85.84375,
      "epoch": 4.164,
      "grad_norm": 7.650039196014404,
      "kl": 0.486328125,
      "learning_rate": 5.836e-07,
      "loss": 0.0005,
      "reward": 0.71875,
      "reward_std": 0.0883883461356163,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.71875,
      "step": 1041
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 98.78125,
      "epoch": 4.168,
      "grad_norm": 5.11639928817749,
      "kl": 0.37646484375,
      "learning_rate": 5.832e-07,
      "loss": 0.0004,
      "reward": 0.625,
      "reward_std": 0.13363061845302582,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.625,
      "step": 1042
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 101.21875,
      "epoch": 4.172,
      "grad_norm": 13.872001647949219,
      "kl": 0.38427734375,
      "learning_rate": 5.828e-07,
      "loss": 0.0004,
      "reward": 0.6875,
      "reward_std": 0.1157275140285492,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.6875,
      "step": 1043
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 100.75,
      "epoch": 4.176,
      "grad_norm": 0.059585072100162506,
      "kl": 0.42431640625,
      "learning_rate": 5.824e-07,
      "loss": 0.0004,
      "reward": 0.25,
      "reward_std": 0.0,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.25,
      "step": 1044
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 87.65625,
      "epoch": 4.18,
      "grad_norm": 4.06519079208374,
      "kl": 0.4208984375,
      "learning_rate": 5.819999999999999e-07,
      "loss": 0.0004,
      "reward": 0.09375,
      "reward_std": 0.1293872892856598,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.09375,
      "step": 1045
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 96.84375,
      "epoch": 4.184,
      "grad_norm": 7.095196723937988,
      "kl": 0.38427734375,
      "learning_rate": 5.816e-07,
      "loss": 0.0004,
      "reward": 0.71875,
      "reward_std": 0.0883883461356163,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.71875,
      "step": 1046
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 79.28125,
      "epoch": 4.188,
      "grad_norm": 3.555328369140625,
      "kl": 0.4697265625,
      "learning_rate": 5.812e-07,
      "loss": 0.0005,
      "reward": 0.96875,
      "reward_std": 0.0883883461356163,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.96875,
      "step": 1047
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 98.65625,
      "epoch": 4.192,
      "grad_norm": 0.01599174737930298,
      "kl": 0.3681640625,
      "learning_rate": 5.807999999999999e-07,
      "loss": 0.0004,
      "reward": 0.5,
      "reward_std": 0.0,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.5,
      "step": 1048
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 97.5625,
      "epoch": 4.196,
      "grad_norm": 129.0567626953125,
      "kl": 0.36083984375,
      "learning_rate": 5.804e-07,
      "loss": 0.0004,
      "reward": 0.65625,
      "reward_std": 0.1293872892856598,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.65625,
      "step": 1049
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 90.59375,
      "epoch": 4.2,
      "grad_norm": 0.21417531371116638,
      "kl": 0.85791015625,
      "learning_rate": 5.8e-07,
      "loss": 0.0009,
      "reward": 1.0,
      "reward_std": 0.0,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 1.0,
      "step": 1050
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 110.59375,
      "epoch": 4.204,
      "grad_norm": 0.05474923178553581,
      "kl": 0.3818359375,
      "learning_rate": 5.796e-07,
      "loss": 0.0004,
      "reward": 0.75,
      "reward_std": 0.0,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.75,
      "step": 1051
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 86.5625,
      "epoch": 4.208,
      "grad_norm": 13.118707656860352,
      "kl": 0.42333984375,
      "learning_rate": 5.792e-07,
      "loss": 0.0004,
      "reward": 0.59375,
      "reward_std": 0.1293872892856598,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.59375,
      "step": 1052
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 92.625,
      "epoch": 4.212,
      "grad_norm": 0.019875887781381607,
      "kl": 0.4140625,
      "learning_rate": 5.788e-07,
      "loss": 0.0004,
      "reward": 0.75,
      "reward_std": 0.0,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.75,
      "step": 1053
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 100.65625,
      "epoch": 4.216,
      "grad_norm": 0.021807491779327393,
      "kl": 0.35009765625,
      "learning_rate": 5.784e-07,
      "loss": 0.0003,
      "reward": 0.5,
      "reward_std": 0.0,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.5,
      "step": 1054
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 89.34375,
      "epoch": 4.22,
      "grad_norm": 0.017294036224484444,
      "kl": 0.4287109375,
      "learning_rate": 5.779999999999999e-07,
      "loss": 0.0004,
      "reward": 0.5,
      "reward_std": 0.0,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.5,
      "step": 1055
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 110.34375,
      "epoch": 4.224,
      "grad_norm": 0.02823016233742237,
      "kl": 0.38671875,
      "learning_rate": 5.776e-07,
      "loss": 0.0004,
      "reward": 0.75,
      "reward_std": 0.0,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.75,
      "step": 1056
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 98.3125,
      "epoch": 4.228,
      "grad_norm": 0.24307022988796234,
      "kl": 0.48291015625,
      "learning_rate": 5.772000000000001e-07,
      "loss": 0.0005,
      "reward": 0.5,
      "reward_std": 0.0,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.5,
      "step": 1057
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 108.71875,
      "epoch": 4.232,
      "grad_norm": 4.628800868988037,
      "kl": 0.38330078125,
      "learning_rate": 5.767999999999999e-07,
      "loss": 0.0004,
      "reward": 0.03125,
      "reward_std": 0.0883883461356163,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.03125,
      "step": 1058
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 104.9375,
      "epoch": 4.236,
      "grad_norm": 5.48722505569458,
      "kl": 0.357421875,
      "learning_rate": 5.764e-07,
      "loss": 0.0004,
      "reward": 0.5625,
      "reward_std": 0.1157275140285492,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.5625,
      "step": 1059
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 99.625,
      "epoch": 4.24,
      "grad_norm": 5.535810470581055,
      "kl": 0.4375,
      "learning_rate": 5.76e-07,
      "loss": 0.0004,
      "reward": 0.65625,
      "reward_std": 0.1293872892856598,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.65625,
      "step": 1060
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 106.4375,
      "epoch": 4.244,
      "grad_norm": 0.09390097856521606,
      "kl": 0.36962890625,
      "learning_rate": 5.755999999999999e-07,
      "loss": 0.0004,
      "reward": 0.25,
      "reward_std": 0.0,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.25,
      "step": 1061
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 98.625,
      "epoch": 4.248,
      "grad_norm": 4.879156112670898,
      "kl": 0.41455078125,
      "learning_rate": 5.752e-07,
      "loss": 0.0004,
      "reward": 0.71875,
      "reward_std": 0.0883883461356163,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.71875,
      "step": 1062
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 119.1875,
      "epoch": 4.252,
      "grad_norm": 0.03136778250336647,
      "kl": 0.3642578125,
      "learning_rate": 5.748e-07,
      "loss": 0.0004,
      "reward": 0.75,
      "reward_std": 0.0,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.75,
      "step": 1063
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 87.0,
      "epoch": 4.256,
      "grad_norm": 7.648971080780029,
      "kl": 0.4091796875,
      "learning_rate": 5.744e-07,
      "loss": 0.0004,
      "reward": 0.46875,
      "reward_std": 0.0883883461356163,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.46875,
      "step": 1064
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 107.71875,
      "epoch": 4.26,
      "grad_norm": 0.014682465232908726,
      "kl": 0.35791015625,
      "learning_rate": 5.739999999999999e-07,
      "loss": 0.0004,
      "reward": 0.25,
      "reward_std": 0.0,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.25,
      "step": 1065
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 114.25,
      "epoch": 4.264,
      "grad_norm": 0.014329612255096436,
      "kl": 0.3232421875,
      "learning_rate": 5.736e-07,
      "loss": 0.0003,
      "reward": 0.25,
      "reward_std": 0.0,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.25,
      "step": 1066
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 118.59375,
      "epoch": 4.268,
      "grad_norm": 6.315824508666992,
      "kl": 0.3076171875,
      "learning_rate": 5.732e-07,
      "loss": 0.0003,
      "reward": 0.46875,
      "reward_std": 0.0883883461356163,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.46875,
      "step": 1067
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 132.46875,
      "epoch": 4.272,
      "grad_norm": 0.019951844587922096,
      "kl": 0.28759765625,
      "learning_rate": 5.727999999999999e-07,
      "loss": 0.0003,
      "reward": 0.25,
      "reward_std": 0.0,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.25,
      "step": 1068
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 119.84375,
      "epoch": 4.276,
      "grad_norm": 10.391695976257324,
      "kl": 0.337890625,
      "learning_rate": 5.724e-07,
      "loss": 0.0003,
      "reward": 0.1875,
      "reward_std": 0.2177756354212761,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.1875,
      "step": 1069
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 104.78125,
      "epoch": 4.28,
      "grad_norm": 0.017950357869267464,
      "kl": 0.34716796875,
      "learning_rate": 5.719999999999999e-07,
      "loss": 0.0003,
      "reward": 0.75,
      "reward_std": 0.0,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.75,
      "step": 1070
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 121.0625,
      "epoch": 4.284,
      "grad_norm": 2.3389358520507812,
      "kl": 0.34619140625,
      "learning_rate": 5.716e-07,
      "loss": 0.0003,
      "reward": 0.53125,
      "reward_std": 0.0883883461356163,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.53125,
      "step": 1071
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 142.90625,
      "epoch": 4.288,
      "grad_norm": 0.018405856564641,
      "kl": 0.29541015625,
      "learning_rate": 5.712e-07,
      "loss": 0.0003,
      "reward": 0.25,
      "reward_std": 0.0,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.25,
      "step": 1072
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 117.0,
      "epoch": 4.292,
      "grad_norm": 0.015511310659348965,
      "kl": 0.33447265625,
      "learning_rate": 5.707999999999999e-07,
      "loss": 0.0003,
      "reward": 0.25,
      "reward_std": 0.0,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.25,
      "step": 1073
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 113.0625,
      "epoch": 4.296,
      "grad_norm": 0.021742764860391617,
      "kl": 0.367431640625,
      "learning_rate": 5.704e-07,
      "loss": 0.0004,
      "reward": 0.75,
      "reward_std": 0.0,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.75,
      "step": 1074
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 115.875,
      "epoch": 4.3,
      "grad_norm": 0.020581087097525597,
      "kl": 0.35107421875,
      "learning_rate": 5.699999999999999e-07,
      "loss": 0.0004,
      "reward": 0.5,
      "reward_std": 0.0,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.5,
      "step": 1075
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 111.1875,
      "epoch": 4.304,
      "grad_norm": 41.18159103393555,
      "kl": 0.4345703125,
      "learning_rate": 5.696e-07,
      "loss": 0.0004,
      "reward": 0.59375,
      "reward_std": 0.2041158601641655,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.59375,
      "step": 1076
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 112.46875,
      "epoch": 4.308,
      "grad_norm": 3.167879104614258,
      "kl": 0.34228515625,
      "learning_rate": 5.692e-07,
      "loss": 0.0003,
      "reward": 0.59375,
      "reward_std": 0.1293872892856598,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.59375,
      "step": 1077
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 141.375,
      "epoch": 4.312,
      "grad_norm": 4.036911964416504,
      "kl": 0.358154296875,
      "learning_rate": 5.688e-07,
      "loss": 0.0004,
      "reward": 0.4375,
      "reward_std": 0.1767766922712326,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.4375,
      "step": 1078
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 113.90625,
      "epoch": 4.316,
      "grad_norm": 42.841827392578125,
      "kl": 0.33056640625,
      "learning_rate": 5.684e-07,
      "loss": 0.0003,
      "reward": 0.3125,
      "reward_std": 0.1767766922712326,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.3125,
      "step": 1079
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 122.625,
      "epoch": 4.32,
      "grad_norm": 3.454944133758545,
      "kl": 0.28662109375,
      "learning_rate": 5.679999999999999e-07,
      "loss": 0.0003,
      "reward": 0.78125,
      "reward_std": 0.0883883461356163,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.78125,
      "step": 1080
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 131.875,
      "epoch": 4.324,
      "grad_norm": 0.23900356888771057,
      "kl": 0.31494140625,
      "learning_rate": 5.676e-07,
      "loss": 0.0003,
      "reward": 0.5,
      "reward_std": 0.0,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.5,
      "step": 1081
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 119.6875,
      "epoch": 4.328,
      "grad_norm": 3.832806348800659,
      "kl": 0.3564453125,
      "learning_rate": 5.672e-07,
      "loss": 0.0004,
      "reward": 0.53125,
      "reward_std": 0.0883883461356163,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.53125,
      "step": 1082
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 92.5,
      "epoch": 4.332,
      "grad_norm": 4.399573802947998,
      "kl": 0.4384765625,
      "learning_rate": 5.667999999999999e-07,
      "loss": 0.0004,
      "reward": 0.6875,
      "reward_std": 0.1157275140285492,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.6875,
      "step": 1083
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 121.90625,
      "epoch": 4.336,
      "grad_norm": 4.044859886169434,
      "kl": 0.37060546875,
      "learning_rate": 5.664e-07,
      "loss": 0.0004,
      "reward": 0.15625,
      "reward_std": 0.1293872892856598,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.15625,
      "step": 1084
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 100.875,
      "epoch": 4.34,
      "grad_norm": 6.003587245941162,
      "kl": 0.37890625,
      "learning_rate": 5.66e-07,
      "loss": 0.0004,
      "reward": 0.25,
      "reward_std": 0.1767766922712326,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.25,
      "step": 1085
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 126.96875,
      "epoch": 4.344,
      "grad_norm": 5.561252593994141,
      "kl": 0.298828125,
      "learning_rate": 5.655999999999999e-07,
      "loss": 0.0003,
      "reward": 0.53125,
      "reward_std": 0.0883883461356163,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.53125,
      "step": 1086
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 116.625,
      "epoch": 4.348,
      "grad_norm": 2.8689191341400146,
      "kl": 0.3154296875,
      "learning_rate": 5.652e-07,
      "loss": 0.0003,
      "reward": 0.46875,
      "reward_std": 0.0883883461356163,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.46875,
      "step": 1087
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 125.96875,
      "epoch": 4.352,
      "grad_norm": 8.750130653381348,
      "kl": 0.42529296875,
      "learning_rate": 5.648e-07,
      "loss": 0.0004,
      "reward": 0.875,
      "reward_std": 0.2177756354212761,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.875,
      "step": 1088
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 121.46875,
      "epoch": 4.356,
      "grad_norm": 6.625911712646484,
      "kl": 0.43017578125,
      "learning_rate": 5.643999999999999e-07,
      "loss": 0.0004,
      "reward": 0.28125,
      "reward_std": 0.0883883461356163,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.28125,
      "step": 1089
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 133.78125,
      "epoch": 4.36,
      "grad_norm": 0.02224709466099739,
      "kl": 0.340576171875,
      "learning_rate": 5.639999999999999e-07,
      "loss": 0.0003,
      "reward": 0.25,
      "reward_std": 0.0,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.25,
      "step": 1090
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 121.78125,
      "epoch": 4.364,
      "grad_norm": 0.014431355521082878,
      "kl": 0.324462890625,
      "learning_rate": 5.636e-07,
      "loss": 0.0003,
      "reward": 0.5,
      "reward_std": 0.0,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.5,
      "step": 1091
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 121.90625,
      "epoch": 4.368,
      "grad_norm": 5.893221855163574,
      "kl": 0.2978515625,
      "learning_rate": 5.632e-07,
      "loss": 0.0003,
      "reward": 0.40625,
      "reward_std": 0.1293872892856598,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.40625,
      "step": 1092
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 105.4375,
      "epoch": 4.372,
      "grad_norm": 0.03661732003092766,
      "kl": 0.3515625,
      "learning_rate": 5.627999999999999e-07,
      "loss": 0.0004,
      "reward": 0.25,
      "reward_std": 0.0,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.25,
      "step": 1093
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 117.28125,
      "epoch": 4.376,
      "grad_norm": 0.015532216057181358,
      "kl": 0.318115234375,
      "learning_rate": 5.624e-07,
      "loss": 0.0003,
      "reward": 0.0,
      "reward_std": 0.0,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.0,
      "step": 1094
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 124.0,
      "epoch": 4.38,
      "grad_norm": 3.2747490406036377,
      "kl": 0.328857421875,
      "learning_rate": 5.620000000000001e-07,
      "loss": 0.0003,
      "reward": 0.375,
      "reward_std": 0.13363061845302582,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.375,
      "step": 1095
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 145.875,
      "epoch": 4.384,
      "grad_norm": 0.02865215390920639,
      "kl": 0.347900390625,
      "learning_rate": 5.615999999999999e-07,
      "loss": 0.0003,
      "reward": 0.0,
      "reward_std": 0.0,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.0,
      "step": 1096
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 118.90625,
      "epoch": 4.388,
      "grad_norm": 0.02989412285387516,
      "kl": 0.32861328125,
      "learning_rate": 5.612e-07,
      "loss": 0.0003,
      "reward": 0.25,
      "reward_std": 0.0,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.25,
      "step": 1097
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 120.53125,
      "epoch": 4.392,
      "grad_norm": 4.302076816558838,
      "kl": 0.31640625,
      "learning_rate": 5.608e-07,
      "loss": 0.0003,
      "reward": 0.0625,
      "reward_std": 0.1157275140285492,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.0625,
      "step": 1098
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 134.96875,
      "epoch": 4.396,
      "grad_norm": 0.018360061571002007,
      "kl": 0.243896484375,
      "learning_rate": 5.604e-07,
      "loss": 0.0002,
      "reward": 0.5,
      "reward_std": 0.0,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.5,
      "step": 1099
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 124.40625,
      "epoch": 4.4,
      "grad_norm": 0.0953860804438591,
      "kl": 0.322021484375,
      "learning_rate": 5.6e-07,
      "loss": 0.0003,
      "reward": 0.5,
      "reward_std": 0.0,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.5,
      "step": 1100
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 113.3125,
      "epoch": 4.404,
      "grad_norm": 6.554188251495361,
      "kl": 0.302734375,
      "learning_rate": 5.596e-07,
      "loss": 0.0003,
      "reward": 0.75,
      "reward_std": 0.1767766922712326,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.75,
      "step": 1101
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 129.03125,
      "epoch": 4.408,
      "grad_norm": 4.409978866577148,
      "kl": 0.30712890625,
      "learning_rate": 5.592e-07,
      "loss": 0.0003,
      "reward": 0.34375,
      "reward_std": 0.1293872892856598,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.34375,
      "step": 1102
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 151.84375,
      "epoch": 4.412,
      "grad_norm": 3.4362077713012695,
      "kl": 0.260498046875,
      "learning_rate": 5.588e-07,
      "loss": 0.0003,
      "reward": 0.71875,
      "reward_std": 0.0883883461356163,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.71875,
      "step": 1103
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 128.4375,
      "epoch": 4.416,
      "grad_norm": 0.017261173576116562,
      "kl": 0.3330078125,
      "learning_rate": 5.584e-07,
      "loss": 0.0003,
      "reward": 0.0,
      "reward_std": 0.0,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.0,
      "step": 1104
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 151.875,
      "epoch": 4.42,
      "grad_norm": 12.046651840209961,
      "kl": 0.299072265625,
      "learning_rate": 5.58e-07,
      "loss": 0.0003,
      "reward": 0.0625,
      "reward_std": 0.1157275140285492,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.0625,
      "step": 1105
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 127.5,
      "epoch": 4.424,
      "grad_norm": 0.04971817508339882,
      "kl": 0.3076171875,
      "learning_rate": 5.576e-07,
      "loss": 0.0003,
      "reward": 0.5,
      "reward_std": 0.0,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.5,
      "step": 1106
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 119.5,
      "epoch": 4.428,
      "grad_norm": 0.014673988334834576,
      "kl": 0.30126953125,
      "learning_rate": 5.572e-07,
      "loss": 0.0003,
      "reward": 0.25,
      "reward_std": 0.0,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.25,
      "step": 1107
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 154.90625,
      "epoch": 4.432,
      "grad_norm": 2.1451337337493896,
      "kl": 0.266845703125,
      "learning_rate": 5.567999999999999e-07,
      "loss": 0.0003,
      "reward": 0.46875,
      "reward_std": 0.0883883461356163,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.46875,
      "step": 1108
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 150.1875,
      "epoch": 4.436,
      "grad_norm": 0.012567956931889057,
      "kl": 0.2841796875,
      "learning_rate": 5.564e-07,
      "loss": 0.0003,
      "reward": 0.5,
      "reward_std": 0.0,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.5,
      "step": 1109
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 133.03125,
      "epoch": 4.44,
      "grad_norm": 10.358048439025879,
      "kl": 0.263916015625,
      "learning_rate": 5.560000000000001e-07,
      "loss": 0.0003,
      "reward": 0.6875,
      "reward_std": 0.1767766922712326,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.6875,
      "step": 1110
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 140.0,
      "epoch": 4.444,
      "grad_norm": 0.01186215877532959,
      "kl": 0.281494140625,
      "learning_rate": 5.555999999999999e-07,
      "loss": 0.0003,
      "reward": 0.25,
      "reward_std": 0.0,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.25,
      "step": 1111
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 149.25,
      "epoch": 4.448,
      "grad_norm": 0.01055728830397129,
      "kl": 0.265380859375,
      "learning_rate": 5.552e-07,
      "loss": 0.0003,
      "reward": 0.5,
      "reward_std": 0.0,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.5,
      "step": 1112
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 131.40625,
      "epoch": 4.452,
      "grad_norm": 0.019849078729748726,
      "kl": 0.291259765625,
      "learning_rate": 5.548e-07,
      "loss": 0.0003,
      "reward": 0.0,
      "reward_std": 0.0,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.0,
      "step": 1113
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 150.125,
      "epoch": 4.456,
      "grad_norm": 16.895469665527344,
      "kl": 0.24072265625,
      "learning_rate": 5.543999999999999e-07,
      "loss": 0.0002,
      "reward": 0.5,
      "reward_std": 0.3514062538743019,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.5,
      "step": 1114
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 127.03125,
      "epoch": 4.46,
      "grad_norm": 0.01886427029967308,
      "kl": 0.33251953125,
      "learning_rate": 5.54e-07,
      "loss": 0.0003,
      "reward": 0.5,
      "reward_std": 0.0,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.5,
      "step": 1115
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 127.4375,
      "epoch": 4.464,
      "grad_norm": 2.4929914474487305,
      "kl": 0.32958984375,
      "learning_rate": 5.536e-07,
      "loss": 0.0003,
      "reward": 0.46875,
      "reward_std": 0.0883883461356163,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.46875,
      "step": 1116
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 126.5,
      "epoch": 4.468,
      "grad_norm": 0.025320416316390038,
      "kl": 0.289794921875,
      "learning_rate": 5.532e-07,
      "loss": 0.0003,
      "reward": 0.5,
      "reward_std": 0.0,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.5,
      "step": 1117
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 139.4375,
      "epoch": 4.4719999999999995,
      "grad_norm": 0.04268677905201912,
      "kl": 0.32177734375,
      "learning_rate": 5.527999999999999e-07,
      "loss": 0.0003,
      "reward": 0.5,
      "reward_std": 0.0,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.5,
      "step": 1118
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 118.5,
      "epoch": 4.476,
      "grad_norm": 0.054494209587574005,
      "kl": 0.3642578125,
      "learning_rate": 5.524e-07,
      "loss": 0.0004,
      "reward": 0.0,
      "reward_std": 0.0,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.0,
      "step": 1119
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 157.5,
      "epoch": 4.48,
      "grad_norm": 5.896949291229248,
      "kl": 0.24755859375,
      "learning_rate": 5.520000000000001e-07,
      "loss": 0.0002,
      "reward": 0.46875,
      "reward_std": 0.0883883461356163,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.46875,
      "step": 1120
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 139.21875,
      "epoch": 4.484,
      "grad_norm": 4.089102268218994,
      "kl": 0.32763671875,
      "learning_rate": 5.515999999999999e-07,
      "loss": 0.0003,
      "reward": 0.3125,
      "reward_std": 0.1157275140285492,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.3125,
      "step": 1121
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 144.59375,
      "epoch": 4.4879999999999995,
      "grad_norm": 3.3508760929107666,
      "kl": 0.30712890625,
      "learning_rate": 5.512e-07,
      "loss": 0.0003,
      "reward": 0.21875,
      "reward_std": 0.0883883461356163,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.21875,
      "step": 1122
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 146.8125,
      "epoch": 4.492,
      "grad_norm": 0.016674378886818886,
      "kl": 0.27978515625,
      "learning_rate": 5.508e-07,
      "loss": 0.0003,
      "reward": 0.75,
      "reward_std": 0.0,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.75,
      "step": 1123
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 172.8125,
      "epoch": 4.496,
      "grad_norm": 0.015236934646964073,
      "kl": 0.229248046875,
      "learning_rate": 5.504e-07,
      "loss": 0.0002,
      "reward": 0.25,
      "reward_std": 0.0,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.25,
      "step": 1124
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 143.75,
      "epoch": 4.5,
      "grad_norm": 3.5608510971069336,
      "kl": 0.29541015625,
      "learning_rate": 5.5e-07,
      "loss": 0.0003,
      "reward": 0.65625,
      "reward_std": 0.1293872892856598,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.65625,
      "step": 1125
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 155.46875,
      "epoch": 4.504,
      "grad_norm": 0.04151947423815727,
      "kl": 0.305908203125,
      "learning_rate": 5.496e-07,
      "loss": 0.0003,
      "reward": 0.75,
      "reward_std": 0.0,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.75,
      "step": 1126
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 135.21875,
      "epoch": 4.508,
      "grad_norm": 0.025374554097652435,
      "kl": 0.33544921875,
      "learning_rate": 5.492e-07,
      "loss": 0.0003,
      "reward": 0.5,
      "reward_std": 0.0,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.5,
      "step": 1127
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 161.71875,
      "epoch": 4.5120000000000005,
      "grad_norm": 4.039253234863281,
      "kl": 0.2421875,
      "learning_rate": 5.487999999999999e-07,
      "loss": 0.0002,
      "reward": 0.4375,
      "reward_std": 0.1157275140285492,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.4375,
      "step": 1128
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 150.46875,
      "epoch": 4.516,
      "grad_norm": 0.0404631569981575,
      "kl": 0.28857421875,
      "learning_rate": 5.484e-07,
      "loss": 0.0003,
      "reward": 1.0,
      "reward_std": 0.0,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 1.0,
      "step": 1129
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 168.125,
      "epoch": 4.52,
      "grad_norm": 0.014968291856348515,
      "kl": 0.2412109375,
      "learning_rate": 5.48e-07,
      "loss": 0.0002,
      "reward": 0.75,
      "reward_std": 0.0,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.75,
      "step": 1130
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 146.0625,
      "epoch": 4.524,
      "grad_norm": 4.139593124389648,
      "kl": 0.29248046875,
      "learning_rate": 5.476e-07,
      "loss": 0.0003,
      "reward": 0.40625,
      "reward_std": 0.1293872892856598,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.40625,
      "step": 1131
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 155.53125,
      "epoch": 4.5280000000000005,
      "grad_norm": 0.015036333352327347,
      "kl": 0.265869140625,
      "learning_rate": 5.472e-07,
      "loss": 0.0003,
      "reward": 0.75,
      "reward_std": 0.0,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.75,
      "step": 1132
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 160.21875,
      "epoch": 4.532,
      "grad_norm": 0.016429008916020393,
      "kl": 0.270751953125,
      "learning_rate": 5.467999999999999e-07,
      "loss": 0.0003,
      "reward": 0.25,
      "reward_std": 0.0,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.25,
      "step": 1133
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 165.25,
      "epoch": 4.536,
      "grad_norm": 11.153393745422363,
      "kl": 0.27685546875,
      "learning_rate": 5.464e-07,
      "loss": 0.0003,
      "reward": 0.0625,
      "reward_std": 0.1157275140285492,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.0625,
      "step": 1134
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 159.0,
      "epoch": 4.54,
      "grad_norm": 0.06211073324084282,
      "kl": 0.2939453125,
      "learning_rate": 5.46e-07,
      "loss": 0.0003,
      "reward": 0.5,
      "reward_std": 0.0,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.5,
      "step": 1135
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 156.125,
      "epoch": 4.5440000000000005,
      "grad_norm": 0.030909527093172073,
      "kl": 0.277587890625,
      "learning_rate": 5.455999999999999e-07,
      "loss": 0.0003,
      "reward": 0.25,
      "reward_std": 0.0,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.25,
      "step": 1136
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 162.75,
      "epoch": 4.548,
      "grad_norm": 23.891538619995117,
      "kl": 0.262451171875,
      "learning_rate": 5.452e-07,
      "loss": 0.0003,
      "reward": 0.875,
      "reward_std": 0.13363061845302582,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.875,
      "step": 1137
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 144.84375,
      "epoch": 4.552,
      "grad_norm": 0.05398315191268921,
      "kl": 0.299560546875,
      "learning_rate": 5.448e-07,
      "loss": 0.0003,
      "reward": 0.75,
      "reward_std": 0.0,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.75,
      "step": 1138
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 157.125,
      "epoch": 4.556,
      "grad_norm": 1.7841356992721558,
      "kl": 0.2744140625,
      "learning_rate": 5.443999999999999e-07,
      "loss": 0.0003,
      "reward": 0.71875,
      "reward_std": 0.0883883461356163,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.71875,
      "step": 1139
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 178.3125,
      "epoch": 4.5600000000000005,
      "grad_norm": 0.10616292804479599,
      "kl": 0.23095703125,
      "learning_rate": 5.44e-07,
      "loss": 0.0002,
      "reward": 0.25,
      "reward_std": 0.0,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.25,
      "step": 1140
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 156.84375,
      "epoch": 4.564,
      "grad_norm": 2.414126396179199,
      "kl": 0.28271484375,
      "learning_rate": 5.436e-07,
      "loss": 0.0003,
      "reward": 0.53125,
      "reward_std": 0.0883883461356163,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.53125,
      "step": 1141
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 132.125,
      "epoch": 4.568,
      "grad_norm": 0.022762814536690712,
      "kl": 0.333984375,
      "learning_rate": 5.431999999999999e-07,
      "loss": 0.0003,
      "reward": 1.0,
      "reward_std": 0.0,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 1.0,
      "step": 1142
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 155.8125,
      "epoch": 4.572,
      "grad_norm": 0.023758282884955406,
      "kl": 0.28857421875,
      "learning_rate": 5.427999999999999e-07,
      "loss": 0.0003,
      "reward": 0.25,
      "reward_std": 0.0,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.25,
      "step": 1143
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 165.65625,
      "epoch": 4.576,
      "grad_norm": 0.01728799194097519,
      "kl": 0.242431640625,
      "learning_rate": 5.424e-07,
      "loss": 0.0002,
      "reward": 0.75,
      "reward_std": 0.0,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.75,
      "step": 1144
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 167.0,
      "epoch": 4.58,
      "grad_norm": 6.379558086395264,
      "kl": 0.29541015625,
      "learning_rate": 5.420000000000001e-07,
      "loss": 0.0003,
      "reward": 0.71875,
      "reward_std": 0.0883883461356163,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.71875,
      "step": 1145
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 146.0,
      "epoch": 4.584,
      "grad_norm": 0.11052519083023071,
      "kl": 0.29541015625,
      "learning_rate": 5.415999999999999e-07,
      "loss": 0.0003,
      "reward": 0.5,
      "reward_std": 0.0,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.5,
      "step": 1146
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 148.78125,
      "epoch": 4.588,
      "grad_norm": 4.697267055511475,
      "kl": 0.26806640625,
      "learning_rate": 5.412e-07,
      "loss": 0.0003,
      "reward": 0.5625,
      "reward_std": 0.1157275140285492,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.5625,
      "step": 1147
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 179.96875,
      "epoch": 4.592,
      "grad_norm": 0.043529871851205826,
      "kl": 0.22119140625,
      "learning_rate": 5.408e-07,
      "loss": 0.0002,
      "reward": 0.25,
      "reward_std": 0.0,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.25,
      "step": 1148
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 158.34375,
      "epoch": 4.596,
      "grad_norm": 5.729726314544678,
      "kl": 0.26953125,
      "learning_rate": 5.403999999999999e-07,
      "loss": 0.0003,
      "reward": 0.3125,
      "reward_std": 0.1157275140285492,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.3125,
      "step": 1149
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 170.53125,
      "epoch": 4.6,
      "grad_norm": 2.999812126159668,
      "kl": 0.244873046875,
      "learning_rate": 5.4e-07,
      "loss": 0.0002,
      "reward": 0.46875,
      "reward_std": 0.0883883461356163,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.46875,
      "step": 1150
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 145.0625,
      "epoch": 4.604,
      "grad_norm": 3.113231897354126,
      "kl": 0.268310546875,
      "learning_rate": 5.396e-07,
      "loss": 0.0003,
      "reward": 0.5625,
      "reward_std": 0.1157275140285492,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.5625,
      "step": 1151
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 156.03125,
      "epoch": 4.608,
      "grad_norm": 0.012986867688596249,
      "kl": 0.2568359375,
      "learning_rate": 5.392e-07,
      "loss": 0.0003,
      "reward": 0.0,
      "reward_std": 0.0,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.0,
      "step": 1152
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 162.15625,
      "epoch": 4.612,
      "grad_norm": 25.76164436340332,
      "kl": 0.262451171875,
      "learning_rate": 5.387999999999999e-07,
      "loss": 0.0003,
      "reward": 0.6875,
      "reward_std": 0.2925042062997818,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.6875,
      "step": 1153
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 147.59375,
      "epoch": 4.616,
      "grad_norm": 0.011444277130067348,
      "kl": 0.25830078125,
      "learning_rate": 5.384e-07,
      "loss": 0.0003,
      "reward": 0.25,
      "reward_std": 0.0,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.25,
      "step": 1154
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 152.625,
      "epoch": 4.62,
      "grad_norm": 0.013514154590666294,
      "kl": 0.277587890625,
      "learning_rate": 5.38e-07,
      "loss": 0.0003,
      "reward": 1.0,
      "reward_std": 0.0,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 1.0,
      "step": 1155
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 143.9375,
      "epoch": 4.624,
      "grad_norm": 6.5501508712768555,
      "kl": 0.2734375,
      "learning_rate": 5.375999999999999e-07,
      "loss": 0.0003,
      "reward": 0.5625,
      "reward_std": 0.1157275140285492,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.5625,
      "step": 1156
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 160.25,
      "epoch": 4.628,
      "grad_norm": 4.9734063148498535,
      "kl": 0.24853515625,
      "learning_rate": 5.372e-07,
      "loss": 0.0002,
      "reward": 0.5625,
      "reward_std": 0.1767766922712326,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.5625,
      "step": 1157
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 136.84375,
      "epoch": 4.632,
      "grad_norm": 11.077103614807129,
      "kl": 0.286865234375,
      "learning_rate": 5.368e-07,
      "loss": 0.0003,
      "reward": 0.71875,
      "reward_std": 0.0883883461356163,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.71875,
      "step": 1158
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 148.0,
      "epoch": 4.636,
      "grad_norm": 0.02520548552274704,
      "kl": 0.3271484375,
      "learning_rate": 5.364e-07,
      "loss": 0.0003,
      "reward": 0.5,
      "reward_std": 0.0,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.5,
      "step": 1159
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 153.21875,
      "epoch": 4.64,
      "grad_norm": 0.027134539559483528,
      "kl": 0.30126953125,
      "learning_rate": 5.36e-07,
      "loss": 0.0003,
      "reward": 0.5,
      "reward_std": 0.0,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.5,
      "step": 1160
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 149.0625,
      "epoch": 4.644,
      "grad_norm": 0.019688010215759277,
      "kl": 0.271484375,
      "learning_rate": 5.355999999999999e-07,
      "loss": 0.0003,
      "reward": 0.25,
      "reward_std": 0.0,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.25,
      "step": 1161
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 153.0,
      "epoch": 4.648,
      "grad_norm": 0.024623429402709007,
      "kl": 0.239501953125,
      "learning_rate": 5.352e-07,
      "loss": 0.0002,
      "reward": 0.5,
      "reward_std": 0.0,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.5,
      "step": 1162
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 147.25,
      "epoch": 4.652,
      "grad_norm": 5.193655490875244,
      "kl": 0.270751953125,
      "learning_rate": 5.348e-07,
      "loss": 0.0003,
      "reward": 0.6875,
      "reward_std": 0.1767766922712326,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.6875,
      "step": 1163
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 146.4375,
      "epoch": 4.656,
      "grad_norm": 0.01629042439162731,
      "kl": 0.3095703125,
      "learning_rate": 5.343999999999999e-07,
      "loss": 0.0003,
      "reward": 0.75,
      "reward_std": 0.0,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.75,
      "step": 1164
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 149.75,
      "epoch": 4.66,
      "grad_norm": 0.01183979120105505,
      "kl": 0.27294921875,
      "learning_rate": 5.34e-07,
      "loss": 0.0003,
      "reward": 0.75,
      "reward_std": 0.0,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.75,
      "step": 1165
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 160.875,
      "epoch": 4.664,
      "grad_norm": 6.2848944664001465,
      "kl": 0.51513671875,
      "learning_rate": 5.336e-07,
      "loss": 0.0005,
      "reward": 0.6875,
      "reward_std": 0.1157275140285492,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.6875,
      "step": 1166
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 159.3125,
      "epoch": 4.668,
      "grad_norm": 0.17453256249427795,
      "kl": 0.29248046875,
      "learning_rate": 5.331999999999999e-07,
      "loss": 0.0003,
      "reward": 0.5,
      "reward_std": 0.0,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.5,
      "step": 1167
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 132.4375,
      "epoch": 4.672,
      "grad_norm": 0.04194831848144531,
      "kl": 0.314697265625,
      "learning_rate": 5.328e-07,
      "loss": 0.0003,
      "reward": 0.5,
      "reward_std": 0.0,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.5,
      "step": 1168
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 145.75,
      "epoch": 4.676,
      "grad_norm": 0.029039116576313972,
      "kl": 0.2998046875,
      "learning_rate": 5.324e-07,
      "loss": 0.0003,
      "reward": 0.75,
      "reward_std": 0.0,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.75,
      "step": 1169
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 160.15625,
      "epoch": 4.68,
      "grad_norm": 3.237999439239502,
      "kl": 0.280517578125,
      "learning_rate": 5.32e-07,
      "loss": 0.0003,
      "reward": 0.21875,
      "reward_std": 0.0883883461356163,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.21875,
      "step": 1170
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 170.5,
      "epoch": 4.684,
      "grad_norm": 0.017721598967909813,
      "kl": 0.250244140625,
      "learning_rate": 5.315999999999999e-07,
      "loss": 0.0003,
      "reward": 0.0,
      "reward_std": 0.0,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.0,
      "step": 1171
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 143.34375,
      "epoch": 4.688,
      "grad_norm": 8.563695907592773,
      "kl": 0.328125,
      "learning_rate": 5.312e-07,
      "loss": 0.0003,
      "reward": 0.40625,
      "reward_std": 0.22201896458864212,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.40625,
      "step": 1172
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 157.71875,
      "epoch": 4.692,
      "grad_norm": 0.014651010744273663,
      "kl": 0.25,
      "learning_rate": 5.308000000000001e-07,
      "loss": 0.0002,
      "reward": 0.25,
      "reward_std": 0.0,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.25,
      "step": 1173
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 159.78125,
      "epoch": 4.696,
      "grad_norm": 0.014831765554845333,
      "kl": 0.2890625,
      "learning_rate": 5.303999999999999e-07,
      "loss": 0.0003,
      "reward": 0.5,
      "reward_std": 0.0,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.5,
      "step": 1174
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 148.03125,
      "epoch": 4.7,
      "grad_norm": 5.07773494720459,
      "kl": 0.30517578125,
      "learning_rate": 5.3e-07,
      "loss": 0.0003,
      "reward": 0.71875,
      "reward_std": 0.0883883461356163,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.71875,
      "step": 1175
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 143.71875,
      "epoch": 4.704,
      "grad_norm": 0.0156971737742424,
      "kl": 0.243408203125,
      "learning_rate": 5.296e-07,
      "loss": 0.0002,
      "reward": 0.0,
      "reward_std": 0.0,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.0,
      "step": 1176
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 160.53125,
      "epoch": 4.708,
      "grad_norm": 0.03695749491453171,
      "kl": 0.271240234375,
      "learning_rate": 5.292e-07,
      "loss": 0.0003,
      "reward": 0.5,
      "reward_std": 0.0,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.5,
      "step": 1177
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 152.96875,
      "epoch": 4.712,
      "grad_norm": 0.020503485575318336,
      "kl": 0.241455078125,
      "learning_rate": 5.288e-07,
      "loss": 0.0002,
      "reward": 1.0,
      "reward_std": 0.0,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 1.0,
      "step": 1178
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 143.96875,
      "epoch": 4.716,
      "grad_norm": 4.993830680847168,
      "kl": 0.263916015625,
      "learning_rate": 5.284e-07,
      "loss": 0.0003,
      "reward": 0.65625,
      "reward_std": 0.1293872892856598,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.65625,
      "step": 1179
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 152.71875,
      "epoch": 4.72,
      "grad_norm": 2.388988971710205,
      "kl": 0.254638671875,
      "learning_rate": 5.28e-07,
      "loss": 0.0003,
      "reward": 0.71875,
      "reward_std": 0.0883883461356163,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.71875,
      "step": 1180
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 140.15625,
      "epoch": 4.724,
      "grad_norm": 4.1817216873168945,
      "kl": 0.300048828125,
      "learning_rate": 5.275999999999999e-07,
      "loss": 0.0003,
      "reward": 0.4375,
      "reward_std": 0.1157275140285492,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.4375,
      "step": 1181
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 139.5,
      "epoch": 4.728,
      "grad_norm": 16.10909652709961,
      "kl": 0.2822265625,
      "learning_rate": 5.272e-07,
      "loss": 0.0003,
      "reward": 0.75,
      "reward_std": 0.2587745785713196,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.75,
      "step": 1182
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 129.40625,
      "epoch": 4.732,
      "grad_norm": 0.042968470603227615,
      "kl": 0.296630859375,
      "learning_rate": 5.268e-07,
      "loss": 0.0003,
      "reward": 0.75,
      "reward_std": 0.0,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.75,
      "step": 1183
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 144.09375,
      "epoch": 4.736,
      "grad_norm": 7.72847318649292,
      "kl": 0.2802734375,
      "learning_rate": 5.264e-07,
      "loss": 0.0003,
      "reward": 0.4375,
      "reward_std": 0.1157275140285492,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.4375,
      "step": 1184
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 125.71875,
      "epoch": 4.74,
      "grad_norm": 5.490642070770264,
      "kl": 0.2861328125,
      "learning_rate": 5.26e-07,
      "loss": 0.0003,
      "reward": 0.53125,
      "reward_std": 0.0883883461356163,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.53125,
      "step": 1185
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 134.65625,
      "epoch": 4.744,
      "grad_norm": 0.05201601982116699,
      "kl": 0.387451171875,
      "learning_rate": 5.255999999999999e-07,
      "loss": 0.0004,
      "reward": 0.5,
      "reward_std": 0.0,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.5,
      "step": 1186
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 144.46875,
      "epoch": 4.748,
      "grad_norm": 0.011890291236341,
      "kl": 0.252685546875,
      "learning_rate": 5.252e-07,
      "loss": 0.0003,
      "reward": 0.25,
      "reward_std": 0.0,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.25,
      "step": 1187
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 121.78125,
      "epoch": 4.752,
      "grad_norm": 0.021535459905862808,
      "kl": 0.2802734375,
      "learning_rate": 5.248e-07,
      "loss": 0.0003,
      "reward": 0.5,
      "reward_std": 0.0,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.5,
      "step": 1188
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 141.40625,
      "epoch": 4.756,
      "grad_norm": 0.010472171008586884,
      "kl": 0.26123046875,
      "learning_rate": 5.243999999999999e-07,
      "loss": 0.0003,
      "reward": 0.5,
      "reward_std": 0.0,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.5,
      "step": 1189
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 142.71875,
      "epoch": 4.76,
      "grad_norm": 0.022479629144072533,
      "kl": 0.24267578125,
      "learning_rate": 5.24e-07,
      "loss": 0.0002,
      "reward": 0.25,
      "reward_std": 0.0,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.25,
      "step": 1190
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 135.53125,
      "epoch": 4.764,
      "grad_norm": 0.02011249028146267,
      "kl": 0.26318359375,
      "learning_rate": 5.236e-07,
      "loss": 0.0003,
      "reward": 0.5,
      "reward_std": 0.0,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.5,
      "step": 1191
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 132.40625,
      "epoch": 4.768,
      "grad_norm": 0.021441668272018433,
      "kl": 0.3212890625,
      "learning_rate": 5.232e-07,
      "loss": 0.0003,
      "reward": 0.75,
      "reward_std": 0.0,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.75,
      "step": 1192
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 116.3125,
      "epoch": 4.772,
      "grad_norm": 0.01715943031013012,
      "kl": 0.31298828125,
      "learning_rate": 5.228e-07,
      "loss": 0.0003,
      "reward": 0.5,
      "reward_std": 0.0,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.5,
      "step": 1193
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 140.15625,
      "epoch": 4.776,
      "grad_norm": 4.664139270782471,
      "kl": 0.2490234375,
      "learning_rate": 5.224e-07,
      "loss": 0.0002,
      "reward": 0.375,
      "reward_std": 0.13363061845302582,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.375,
      "step": 1194
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 117.21875,
      "epoch": 4.78,
      "grad_norm": 3.334951877593994,
      "kl": 0.32275390625,
      "learning_rate": 5.22e-07,
      "loss": 0.0003,
      "reward": 0.375,
      "reward_std": 0.13363061845302582,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.375,
      "step": 1195
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 142.65625,
      "epoch": 4.784,
      "grad_norm": 0.010868766345083714,
      "kl": 0.264404296875,
      "learning_rate": 5.215999999999999e-07,
      "loss": 0.0003,
      "reward": 1.0,
      "reward_std": 0.0,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 1.0,
      "step": 1196
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 130.78125,
      "epoch": 4.788,
      "grad_norm": 0.033785078674554825,
      "kl": 0.298828125,
      "learning_rate": 5.212e-07,
      "loss": 0.0003,
      "reward": 0.25,
      "reward_std": 0.0,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.25,
      "step": 1197
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 113.25,
      "epoch": 4.792,
      "grad_norm": 3.678492784500122,
      "kl": 0.35888671875,
      "learning_rate": 5.208000000000001e-07,
      "loss": 0.0004,
      "reward": 0.625,
      "reward_std": 0.13363061845302582,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.625,
      "step": 1198
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 129.75,
      "epoch": 4.796,
      "grad_norm": 0.01994801126420498,
      "kl": 0.296875,
      "learning_rate": 5.203999999999999e-07,
      "loss": 0.0003,
      "reward": 0.5,
      "reward_std": 0.0,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.5,
      "step": 1199
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 138.0,
      "epoch": 4.8,
      "grad_norm": 0.013591584749519825,
      "kl": 0.26708984375,
      "learning_rate": 5.2e-07,
      "loss": 0.0003,
      "reward": 0.5,
      "reward_std": 0.0,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.5,
      "step": 1200
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 127.8125,
      "epoch": 4.804,
      "grad_norm": 0.035540465265512466,
      "kl": 0.28955078125,
      "learning_rate": 5.196e-07,
      "loss": 0.0003,
      "reward": 0.25,
      "reward_std": 0.0,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.25,
      "step": 1201
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 140.71875,
      "epoch": 4.808,
      "grad_norm": 2.8226571083068848,
      "kl": 0.281982421875,
      "learning_rate": 5.191999999999999e-07,
      "loss": 0.0003,
      "reward": 0.4375,
      "reward_std": 0.1157275140285492,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.4375,
      "step": 1202
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 136.4375,
      "epoch": 4.812,
      "grad_norm": 0.15687893331050873,
      "kl": 0.344482421875,
      "learning_rate": 5.188e-07,
      "loss": 0.0003,
      "reward": 0.75,
      "reward_std": 0.0,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.75,
      "step": 1203
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 124.25,
      "epoch": 4.816,
      "grad_norm": 0.09211844205856323,
      "kl": 0.2890625,
      "learning_rate": 5.184e-07,
      "loss": 0.0003,
      "reward": 0.5,
      "reward_std": 0.0,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.5,
      "step": 1204
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 113.3125,
      "epoch": 4.82,
      "grad_norm": 0.01810278184711933,
      "kl": 0.32421875,
      "learning_rate": 5.18e-07,
      "loss": 0.0003,
      "reward": 0.25,
      "reward_std": 0.0,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.25,
      "step": 1205
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 132.96875,
      "epoch": 4.824,
      "grad_norm": 0.016102444380521774,
      "kl": 0.30322265625,
      "learning_rate": 5.175999999999999e-07,
      "loss": 0.0003,
      "reward": 0.75,
      "reward_std": 0.0,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.75,
      "step": 1206
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 109.15625,
      "epoch": 4.828,
      "grad_norm": 0.026123980060219765,
      "kl": 0.29541015625,
      "learning_rate": 5.172e-07,
      "loss": 0.0003,
      "reward": 0.5,
      "reward_std": 0.0,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.5,
      "step": 1207
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 130.75,
      "epoch": 4.832,
      "grad_norm": 46.64488220214844,
      "kl": 0.3076171875,
      "learning_rate": 5.168e-07,
      "loss": 0.0003,
      "reward": 0.84375,
      "reward_std": 0.1293872892856598,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.84375,
      "step": 1208
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 144.3125,
      "epoch": 4.836,
      "grad_norm": 0.06699591130018234,
      "kl": 0.26806640625,
      "learning_rate": 5.163999999999999e-07,
      "loss": 0.0003,
      "reward": 0.75,
      "reward_std": 0.0,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.75,
      "step": 1209
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 130.4375,
      "epoch": 4.84,
      "grad_norm": 3.1910650730133057,
      "kl": 0.31396484375,
      "learning_rate": 5.16e-07,
      "loss": 0.0003,
      "reward": 0.28125,
      "reward_std": 0.0883883461356163,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.28125,
      "step": 1210
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 142.5,
      "epoch": 4.844,
      "grad_norm": 0.011820975691080093,
      "kl": 0.261474609375,
      "learning_rate": 5.155999999999999e-07,
      "loss": 0.0003,
      "reward": 0.25,
      "reward_std": 0.0,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.25,
      "step": 1211
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 145.21875,
      "epoch": 4.848,
      "grad_norm": 0.015639204531908035,
      "kl": 0.27197265625,
      "learning_rate": 5.152e-07,
      "loss": 0.0003,
      "reward": 0.25,
      "reward_std": 0.0,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.25,
      "step": 1212
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 128.40625,
      "epoch": 4.852,
      "grad_norm": 0.01515189278870821,
      "kl": 0.2724609375,
      "learning_rate": 5.148e-07,
      "loss": 0.0003,
      "reward": 0.0,
      "reward_std": 0.0,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.0,
      "step": 1213
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 113.75,
      "epoch": 4.856,
      "grad_norm": 0.05278758332133293,
      "kl": 0.31298828125,
      "learning_rate": 5.143999999999999e-07,
      "loss": 0.0003,
      "reward": 0.5,
      "reward_std": 0.0,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.5,
      "step": 1214
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 144.625,
      "epoch": 4.86,
      "grad_norm": 4.175487518310547,
      "kl": 0.279296875,
      "learning_rate": 5.14e-07,
      "loss": 0.0003,
      "reward": 0.1875,
      "reward_std": 0.1157275140285492,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.1875,
      "step": 1215
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 150.28125,
      "epoch": 4.864,
      "grad_norm": 0.012961026281118393,
      "kl": 0.25830078125,
      "learning_rate": 5.135999999999999e-07,
      "loss": 0.0003,
      "reward": 1.0,
      "reward_std": 0.0,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 1.0,
      "step": 1216
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 158.625,
      "epoch": 4.868,
      "grad_norm": 3.8442583084106445,
      "kl": 0.27099609375,
      "learning_rate": 5.132e-07,
      "loss": 0.0003,
      "reward": 0.0625,
      "reward_std": 0.1157275140285492,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.0625,
      "step": 1217
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 139.625,
      "epoch": 4.872,
      "grad_norm": 27.46537971496582,
      "kl": 0.337158203125,
      "learning_rate": 5.128e-07,
      "loss": 0.0003,
      "reward": 0.4375,
      "reward_std": 0.1157275140285492,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.4375,
      "step": 1218
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 129.8125,
      "epoch": 4.876,
      "grad_norm": 0.04781090095639229,
      "kl": 0.30908203125,
      "learning_rate": 5.124e-07,
      "loss": 0.0003,
      "reward": 0.75,
      "reward_std": 0.0,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.75,
      "step": 1219
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 159.46875,
      "epoch": 4.88,
      "grad_norm": 0.01112871989607811,
      "kl": 0.229736328125,
      "learning_rate": 5.12e-07,
      "loss": 0.0002,
      "reward": 0.75,
      "reward_std": 0.0,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.75,
      "step": 1220
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 142.15625,
      "epoch": 4.884,
      "grad_norm": 0.08341070264577866,
      "kl": 0.276611328125,
      "learning_rate": 5.116e-07,
      "loss": 0.0003,
      "reward": 0.75,
      "reward_std": 0.0,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.75,
      "step": 1221
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 138.875,
      "epoch": 4.888,
      "grad_norm": 3.5840585231781006,
      "kl": 0.268310546875,
      "learning_rate": 5.112e-07,
      "loss": 0.0003,
      "reward": 0.21875,
      "reward_std": 0.0883883461356163,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.21875,
      "step": 1222
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 145.9375,
      "epoch": 4.892,
      "grad_norm": 0.015838317573070526,
      "kl": 0.240234375,
      "learning_rate": 5.108e-07,
      "loss": 0.0002,
      "reward": 0.75,
      "reward_std": 0.0,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.75,
      "step": 1223
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 136.1875,
      "epoch": 4.896,
      "grad_norm": 0.19625888764858246,
      "kl": 0.31298828125,
      "learning_rate": 5.103999999999999e-07,
      "loss": 0.0003,
      "reward": 0.75,
      "reward_std": 0.0,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.75,
      "step": 1224
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 137.15625,
      "epoch": 4.9,
      "grad_norm": 0.0170870553702116,
      "kl": 0.268798828125,
      "learning_rate": 5.1e-07,
      "loss": 0.0003,
      "reward": 0.5,
      "reward_std": 0.0,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.5,
      "step": 1225
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 153.21875,
      "epoch": 4.904,
      "grad_norm": 0.01228098664432764,
      "kl": 0.231689453125,
      "learning_rate": 5.096000000000001e-07,
      "loss": 0.0002,
      "reward": 0.5,
      "reward_std": 0.0,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.5,
      "step": 1226
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 126.40625,
      "epoch": 4.908,
      "grad_norm": 0.017006240785121918,
      "kl": 0.2939453125,
      "learning_rate": 5.091999999999999e-07,
      "loss": 0.0003,
      "reward": 0.75,
      "reward_std": 0.0,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.75,
      "step": 1227
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 120.03125,
      "epoch": 4.912,
      "grad_norm": 6.2305450439453125,
      "kl": 0.367919921875,
      "learning_rate": 5.088e-07,
      "loss": 0.0004,
      "reward": 0.28125,
      "reward_std": 0.0883883461356163,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.28125,
      "step": 1228
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 135.84375,
      "epoch": 4.916,
      "grad_norm": 0.02266404964029789,
      "kl": 0.26611328125,
      "learning_rate": 5.084e-07,
      "loss": 0.0003,
      "reward": 0.5,
      "reward_std": 0.0,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.5,
      "step": 1229
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 138.90625,
      "epoch": 4.92,
      "grad_norm": 3.1232662200927734,
      "kl": 0.299560546875,
      "learning_rate": 5.079999999999999e-07,
      "loss": 0.0003,
      "reward": 0.1875,
      "reward_std": 0.1157275140285492,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.1875,
      "step": 1230
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 142.8125,
      "epoch": 4.924,
      "grad_norm": 0.09912146627902985,
      "kl": 0.256103515625,
      "learning_rate": 5.076e-07,
      "loss": 0.0003,
      "reward": 0.75,
      "reward_std": 0.0,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.75,
      "step": 1231
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 161.46875,
      "epoch": 4.928,
      "grad_norm": 4.011734962463379,
      "kl": 0.26806640625,
      "learning_rate": 5.072e-07,
      "loss": 0.0003,
      "reward": 0.4375,
      "reward_std": 0.1157275140285492,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.4375,
      "step": 1232
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 146.34375,
      "epoch": 4.932,
      "grad_norm": 0.07319404929876328,
      "kl": 0.3330078125,
      "learning_rate": 5.068e-07,
      "loss": 0.0003,
      "reward": 0.25,
      "reward_std": 0.0,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.25,
      "step": 1233
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 136.125,
      "epoch": 4.936,
      "grad_norm": 3.3835110664367676,
      "kl": 0.294189453125,
      "learning_rate": 5.063999999999999e-07,
      "loss": 0.0003,
      "reward": 0.65625,
      "reward_std": 0.1293872892856598,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.65625,
      "step": 1234
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 145.0625,
      "epoch": 4.9399999999999995,
      "grad_norm": 0.014413079246878624,
      "kl": 0.262939453125,
      "learning_rate": 5.06e-07,
      "loss": 0.0003,
      "reward": 0.5,
      "reward_std": 0.0,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.5,
      "step": 1235
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 144.09375,
      "epoch": 4.944,
      "grad_norm": 0.20096711814403534,
      "kl": 0.267333984375,
      "learning_rate": 5.056e-07,
      "loss": 0.0003,
      "reward": 0.75,
      "reward_std": 0.0,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.75,
      "step": 1236
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 124.625,
      "epoch": 4.948,
      "grad_norm": 0.040523186326026917,
      "kl": 0.31640625,
      "learning_rate": 5.051999999999999e-07,
      "loss": 0.0003,
      "reward": 0.25,
      "reward_std": 0.0,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.25,
      "step": 1237
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 131.875,
      "epoch": 4.952,
      "grad_norm": 0.0352230928838253,
      "kl": 0.27587890625,
      "learning_rate": 5.048e-07,
      "loss": 0.0003,
      "reward": 0.25,
      "reward_std": 0.0,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.25,
      "step": 1238
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 124.28125,
      "epoch": 4.9559999999999995,
      "grad_norm": 0.01514040119946003,
      "kl": 0.274658203125,
      "learning_rate": 5.043999999999999e-07,
      "loss": 0.0003,
      "reward": 0.25,
      "reward_std": 0.0,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.25,
      "step": 1239
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 138.34375,
      "epoch": 4.96,
      "grad_norm": 2.1852774620056152,
      "kl": 0.29833984375,
      "learning_rate": 5.04e-07,
      "loss": 0.0003,
      "reward": 0.71875,
      "reward_std": 0.0883883461356163,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.71875,
      "step": 1240
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 141.34375,
      "epoch": 4.964,
      "grad_norm": 0.022987544536590576,
      "kl": 0.260009765625,
      "learning_rate": 5.036e-07,
      "loss": 0.0003,
      "reward": 0.5,
      "reward_std": 0.0,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.5,
      "step": 1241
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 149.53125,
      "epoch": 4.968,
      "grad_norm": 0.015245127491652966,
      "kl": 0.2578125,
      "learning_rate": 5.032e-07,
      "loss": 0.0003,
      "reward": 0.75,
      "reward_std": 0.0,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.75,
      "step": 1242
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 149.625,
      "epoch": 4.9719999999999995,
      "grad_norm": 0.022085487842559814,
      "kl": 0.2705078125,
      "learning_rate": 5.028e-07,
      "loss": 0.0003,
      "reward": 0.75,
      "reward_std": 0.0,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.75,
      "step": 1243
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 131.71875,
      "epoch": 4.976,
      "grad_norm": 0.015508446842432022,
      "kl": 0.29541015625,
      "learning_rate": 5.023999999999999e-07,
      "loss": 0.0003,
      "reward": 0.5,
      "reward_std": 0.0,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.5,
      "step": 1244
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 137.96875,
      "epoch": 4.98,
      "grad_norm": 3.951266050338745,
      "kl": 0.264892578125,
      "learning_rate": 5.02e-07,
      "loss": 0.0003,
      "reward": 0.125,
      "reward_std": 0.13363061845302582,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.125,
      "step": 1245
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 153.625,
      "epoch": 4.984,
      "grad_norm": 0.016056444495916367,
      "kl": 0.2578125,
      "learning_rate": 5.016e-07,
      "loss": 0.0003,
      "reward": 0.25,
      "reward_std": 0.0,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.25,
      "step": 1246
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 139.1875,
      "epoch": 4.9879999999999995,
      "grad_norm": 0.012911013327538967,
      "kl": 0.27685546875,
      "learning_rate": 5.012e-07,
      "loss": 0.0003,
      "reward": 0.5,
      "reward_std": 0.0,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.5,
      "step": 1247
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 148.90625,
      "epoch": 4.992,
      "grad_norm": 0.01337132416665554,
      "kl": 0.237548828125,
      "learning_rate": 5.008e-07,
      "loss": 0.0002,
      "reward": 0.25,
      "reward_std": 0.0,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.25,
      "step": 1248
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 135.90625,
      "epoch": 4.996,
      "grad_norm": 20.10576057434082,
      "kl": 0.244384765625,
      "learning_rate": 5.003999999999999e-07,
      "loss": 0.0002,
      "reward": 0.4375,
      "reward_std": 0.1157275140285492,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.4375,
      "step": 1249
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 146.5,
      "epoch": 5.0,
      "grad_norm": 0.021885152906179428,
      "kl": 0.2666015625,
      "learning_rate": 5e-07,
      "loss": 0.0003,
      "reward": 0.75,
      "reward_std": 0.0,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.75,
      "step": 1250
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 138.09375,
      "epoch": 5.004,
      "grad_norm": 0.04491958022117615,
      "kl": 0.33447265625,
      "learning_rate": 4.996e-07,
      "loss": 0.0003,
      "reward": 0.75,
      "reward_std": 0.0,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.75,
      "step": 1251
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 154.96875,
      "epoch": 5.008,
      "grad_norm": 0.02253258042037487,
      "kl": 0.243896484375,
      "learning_rate": 4.991999999999999e-07,
      "loss": 0.0002,
      "reward": 0.5,
      "reward_std": 0.0,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.5,
      "step": 1252
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 144.375,
      "epoch": 5.012,
      "grad_norm": 0.03096994198858738,
      "kl": 0.250732421875,
      "learning_rate": 4.988e-07,
      "loss": 0.0003,
      "reward": 0.25,
      "reward_std": 0.0,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.25,
      "step": 1253
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 135.03125,
      "epoch": 5.016,
      "grad_norm": 0.022731071338057518,
      "kl": 0.26513671875,
      "learning_rate": 4.984e-07,
      "loss": 0.0003,
      "reward": 0.5,
      "reward_std": 0.0,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.5,
      "step": 1254
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 149.96875,
      "epoch": 5.02,
      "grad_norm": 0.009650071151554585,
      "kl": 0.224853515625,
      "learning_rate": 4.979999999999999e-07,
      "loss": 0.0002,
      "reward": 0.5,
      "reward_std": 0.0,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.5,
      "step": 1255
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 160.40625,
      "epoch": 5.024,
      "grad_norm": 2.596329927444458,
      "kl": 0.23095703125,
      "learning_rate": 4.976e-07,
      "loss": 0.0002,
      "reward": 0.71875,
      "reward_std": 0.0883883461356163,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.71875,
      "step": 1256
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 137.84375,
      "epoch": 5.028,
      "grad_norm": 0.2113117128610611,
      "kl": 0.28857421875,
      "learning_rate": 4.972e-07,
      "loss": 0.0003,
      "reward": 1.0,
      "reward_std": 0.0,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 1.0,
      "step": 1257
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 136.6875,
      "epoch": 5.032,
      "grad_norm": 0.011437843553721905,
      "kl": 0.240234375,
      "learning_rate": 4.968e-07,
      "loss": 0.0002,
      "reward": 0.5,
      "reward_std": 0.0,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.5,
      "step": 1258
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 150.4375,
      "epoch": 5.036,
      "grad_norm": 0.032690178602933884,
      "kl": 0.2548828125,
      "learning_rate": 4.964e-07,
      "loss": 0.0003,
      "reward": 0.5,
      "reward_std": 0.0,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.5,
      "step": 1259
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 131.90625,
      "epoch": 5.04,
      "grad_norm": 0.017420809715986252,
      "kl": 0.3017578125,
      "learning_rate": 4.96e-07,
      "loss": 0.0003,
      "reward": 0.5,
      "reward_std": 0.0,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.5,
      "step": 1260
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 140.15625,
      "epoch": 5.044,
      "grad_norm": 0.022388502955436707,
      "kl": 0.263427734375,
      "learning_rate": 4.956e-07,
      "loss": 0.0003,
      "reward": 0.5,
      "reward_std": 0.0,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.5,
      "step": 1261
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 149.53125,
      "epoch": 5.048,
      "grad_norm": 0.013329915702342987,
      "kl": 0.216064453125,
      "learning_rate": 4.951999999999999e-07,
      "loss": 0.0002,
      "reward": 0.5,
      "reward_std": 0.0,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.5,
      "step": 1262
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 158.0,
      "epoch": 5.052,
      "grad_norm": 4.0549139976501465,
      "kl": 0.23193359375,
      "learning_rate": 4.948e-07,
      "loss": 0.0002,
      "reward": 0.28125,
      "reward_std": 0.0883883461356163,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.28125,
      "step": 1263
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 132.8125,
      "epoch": 5.056,
      "grad_norm": 0.01824198290705681,
      "kl": 0.25537109375,
      "learning_rate": 4.944e-07,
      "loss": 0.0003,
      "reward": 0.5,
      "reward_std": 0.0,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.5,
      "step": 1264
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 139.0625,
      "epoch": 5.06,
      "grad_norm": 0.01272667944431305,
      "kl": 0.2470703125,
      "learning_rate": 4.94e-07,
      "loss": 0.0002,
      "reward": 0.0,
      "reward_std": 0.0,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.0,
      "step": 1265
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 102.375,
      "epoch": 5.064,
      "grad_norm": 10.764348030090332,
      "kl": 0.3193359375,
      "learning_rate": 4.935999999999999e-07,
      "loss": 0.0003,
      "reward": 0.75,
      "reward_std": 0.1767766922712326,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.75,
      "step": 1266
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 131.875,
      "epoch": 5.068,
      "grad_norm": 0.019407546147704124,
      "kl": 0.301025390625,
      "learning_rate": 4.932e-07,
      "loss": 0.0003,
      "reward": 0.75,
      "reward_std": 0.0,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.75,
      "step": 1267
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 133.96875,
      "epoch": 5.072,
      "grad_norm": 0.013575092889368534,
      "kl": 0.245361328125,
      "learning_rate": 4.928e-07,
      "loss": 0.0002,
      "reward": 0.5,
      "reward_std": 0.0,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.5,
      "step": 1268
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 125.21875,
      "epoch": 5.076,
      "grad_norm": 0.03519083559513092,
      "kl": 0.293701171875,
      "learning_rate": 4.923999999999999e-07,
      "loss": 0.0003,
      "reward": 0.5,
      "reward_std": 0.0,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.5,
      "step": 1269
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 152.5,
      "epoch": 5.08,
      "grad_norm": 0.17260581254959106,
      "kl": 0.23583984375,
      "learning_rate": 4.92e-07,
      "loss": 0.0002,
      "reward": 0.75,
      "reward_std": 0.0,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.75,
      "step": 1270
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 120.875,
      "epoch": 5.084,
      "grad_norm": 0.05579555779695511,
      "kl": 0.39599609375,
      "learning_rate": 4.916e-07,
      "loss": 0.0004,
      "reward": 0.75,
      "reward_std": 0.0,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.75,
      "step": 1271
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 122.375,
      "epoch": 5.088,
      "grad_norm": 30.387306213378906,
      "kl": 0.284423828125,
      "learning_rate": 4.912e-07,
      "loss": 0.0003,
      "reward": 0.6875,
      "reward_std": 0.1767766922712326,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.6875,
      "step": 1272
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 126.46875,
      "epoch": 5.092,
      "grad_norm": 10.023923873901367,
      "kl": 0.267578125,
      "learning_rate": 4.908e-07,
      "loss": 0.0003,
      "reward": 0.21875,
      "reward_std": 0.2041158601641655,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.21875,
      "step": 1273
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 121.0625,
      "epoch": 5.096,
      "grad_norm": 0.01640361174941063,
      "kl": 0.30224609375,
      "learning_rate": 4.904e-07,
      "loss": 0.0003,
      "reward": 0.5,
      "reward_std": 0.0,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.5,
      "step": 1274
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 131.25,
      "epoch": 5.1,
      "grad_norm": 8.803305625915527,
      "kl": 0.357666015625,
      "learning_rate": 4.9e-07,
      "loss": 0.0004,
      "reward": 0.78125,
      "reward_std": 0.0883883461356163,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.78125,
      "step": 1275
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 131.25,
      "epoch": 5.104,
      "grad_norm": 0.10217951238155365,
      "kl": 0.27734375,
      "learning_rate": 4.895999999999999e-07,
      "loss": 0.0003,
      "reward": 0.75,
      "reward_std": 0.0,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.75,
      "step": 1276
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 138.78125,
      "epoch": 5.108,
      "grad_norm": 15.798462867736816,
      "kl": 0.223388671875,
      "learning_rate": 4.892e-07,
      "loss": 0.0002,
      "reward": 0.40625,
      "reward_std": 0.22201896458864212,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.40625,
      "step": 1277
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 140.75,
      "epoch": 5.112,
      "grad_norm": 0.05122094973921776,
      "kl": 0.2958984375,
      "learning_rate": 4.888e-07,
      "loss": 0.0003,
      "reward": 0.25,
      "reward_std": 0.0,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.25,
      "step": 1278
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 136.71875,
      "epoch": 5.116,
      "grad_norm": 7.983183860778809,
      "kl": 0.285888671875,
      "learning_rate": 4.884e-07,
      "loss": 0.0003,
      "reward": 0.65625,
      "reward_std": 0.1293872892856598,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.65625,
      "step": 1279
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 148.65625,
      "epoch": 5.12,
      "grad_norm": 0.07215754687786102,
      "kl": 0.24462890625,
      "learning_rate": 4.879999999999999e-07,
      "loss": 0.0002,
      "reward": 0.5,
      "reward_std": 0.0,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.5,
      "step": 1280
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 109.21875,
      "epoch": 5.124,
      "grad_norm": 5.287899494171143,
      "kl": 0.29443359375,
      "learning_rate": 4.876e-07,
      "loss": 0.0003,
      "reward": 0.40625,
      "reward_std": 0.1293872892856598,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.40625,
      "step": 1281
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 124.75,
      "epoch": 5.128,
      "grad_norm": 0.020164377987384796,
      "kl": 0.3046875,
      "learning_rate": 4.872e-07,
      "loss": 0.0003,
      "reward": 0.75,
      "reward_std": 0.0,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.75,
      "step": 1282
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 137.375,
      "epoch": 5.132,
      "grad_norm": 0.013421082869172096,
      "kl": 0.26123046875,
      "learning_rate": 4.867999999999999e-07,
      "loss": 0.0003,
      "reward": 0.75,
      "reward_std": 0.0,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.75,
      "step": 1283
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 122.03125,
      "epoch": 5.136,
      "grad_norm": 0.013570448383688927,
      "kl": 0.27783203125,
      "learning_rate": 4.864e-07,
      "loss": 0.0003,
      "reward": 1.0,
      "reward_std": 0.0,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 1.0,
      "step": 1284
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 154.03125,
      "epoch": 5.14,
      "grad_norm": 0.012936098501086235,
      "kl": 0.230224609375,
      "learning_rate": 4.86e-07,
      "loss": 0.0002,
      "reward": 1.0,
      "reward_std": 0.0,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 1.0,
      "step": 1285
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 110.625,
      "epoch": 5.144,
      "grad_norm": 0.015391627326607704,
      "kl": 0.2607421875,
      "learning_rate": 4.856e-07,
      "loss": 0.0003,
      "reward": 0.5,
      "reward_std": 0.0,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.5,
      "step": 1286
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 137.4375,
      "epoch": 5.148,
      "grad_norm": 3.80214786529541,
      "kl": 0.270263671875,
      "learning_rate": 4.852e-07,
      "loss": 0.0003,
      "reward": 0.96875,
      "reward_std": 0.0883883461356163,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.96875,
      "step": 1287
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 127.90625,
      "epoch": 5.152,
      "grad_norm": 0.014751072973012924,
      "kl": 0.241943359375,
      "learning_rate": 4.848e-07,
      "loss": 0.0002,
      "reward": 0.75,
      "reward_std": 0.0,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.75,
      "step": 1288
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 123.6875,
      "epoch": 5.156,
      "grad_norm": 3.2137176990509033,
      "kl": 0.3037109375,
      "learning_rate": 4.844e-07,
      "loss": 0.0003,
      "reward": 0.28125,
      "reward_std": 0.0883883461356163,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.28125,
      "step": 1289
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 126.875,
      "epoch": 5.16,
      "grad_norm": 0.019958974793553352,
      "kl": 0.259521484375,
      "learning_rate": 4.839999999999999e-07,
      "loss": 0.0003,
      "reward": 0.75,
      "reward_std": 0.0,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.75,
      "step": 1290
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 139.78125,
      "epoch": 5.164,
      "grad_norm": 0.019287146627902985,
      "kl": 0.236572265625,
      "learning_rate": 4.835999999999999e-07,
      "loss": 0.0002,
      "reward": 0.5,
      "reward_std": 0.0,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.5,
      "step": 1291
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 120.4375,
      "epoch": 5.168,
      "grad_norm": 0.02103956788778305,
      "kl": 0.23095703125,
      "learning_rate": 4.832e-07,
      "loss": 0.0002,
      "reward": 0.0,
      "reward_std": 0.0,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.0,
      "step": 1292
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 144.34375,
      "epoch": 5.172,
      "grad_norm": 0.010608724318444729,
      "kl": 0.22216796875,
      "learning_rate": 4.828e-07,
      "loss": 0.0002,
      "reward": 0.75,
      "reward_std": 0.0,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.75,
      "step": 1293
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 131.09375,
      "epoch": 5.176,
      "grad_norm": 0.019097983837127686,
      "kl": 0.279541015625,
      "learning_rate": 4.823999999999999e-07,
      "loss": 0.0003,
      "reward": 1.0,
      "reward_std": 0.0,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 1.0,
      "step": 1294
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 112.03125,
      "epoch": 5.18,
      "grad_norm": 2.9960885047912598,
      "kl": 0.267822265625,
      "learning_rate": 4.82e-07,
      "loss": 0.0003,
      "reward": 0.4375,
      "reward_std": 0.1157275140285492,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.4375,
      "step": 1295
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 148.875,
      "epoch": 5.184,
      "grad_norm": 3.545438289642334,
      "kl": 0.22509765625,
      "learning_rate": 4.816e-07,
      "loss": 0.0002,
      "reward": 0.6875,
      "reward_std": 0.1157275140285492,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.6875,
      "step": 1296
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 105.28125,
      "epoch": 5.188,
      "grad_norm": 0.31230002641677856,
      "kl": 0.3515625,
      "learning_rate": 4.812e-07,
      "loss": 0.0004,
      "reward": 0.75,
      "reward_std": 0.0,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.75,
      "step": 1297
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 138.1875,
      "epoch": 5.192,
      "grad_norm": 0.07331553101539612,
      "kl": 0.267333984375,
      "learning_rate": 4.808e-07,
      "loss": 0.0003,
      "reward": 0.0,
      "reward_std": 0.0,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.0,
      "step": 1298
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 126.53125,
      "epoch": 5.196,
      "grad_norm": 0.012874067761003971,
      "kl": 0.251708984375,
      "learning_rate": 4.804e-07,
      "loss": 0.0003,
      "reward": 0.5,
      "reward_std": 0.0,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.5,
      "step": 1299
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 136.78125,
      "epoch": 5.2,
      "grad_norm": 0.02414083294570446,
      "kl": 0.302001953125,
      "learning_rate": 4.8e-07,
      "loss": 0.0003,
      "reward": 0.25,
      "reward_std": 0.0,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.25,
      "step": 1300
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 122.1875,
      "epoch": 5.204,
      "grad_norm": 0.01956980675458908,
      "kl": 0.2978515625,
      "learning_rate": 4.796e-07,
      "loss": 0.0003,
      "reward": 0.0,
      "reward_std": 0.0,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.0,
      "step": 1301
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 124.375,
      "epoch": 5.208,
      "grad_norm": 0.0632576122879982,
      "kl": 0.2861328125,
      "learning_rate": 4.792e-07,
      "loss": 0.0003,
      "reward": 0.75,
      "reward_std": 0.0,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.75,
      "step": 1302
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 117.125,
      "epoch": 5.212,
      "grad_norm": 13.517343521118164,
      "kl": 0.2841796875,
      "learning_rate": 4.788e-07,
      "loss": 0.0003,
      "reward": 0.53125,
      "reward_std": 0.0883883461356163,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.53125,
      "step": 1303
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 116.59375,
      "epoch": 5.216,
      "grad_norm": 74.23445129394531,
      "kl": 0.267333984375,
      "learning_rate": 4.783999999999999e-07,
      "loss": 0.0003,
      "reward": 0.75,
      "reward_std": 0.1767766922712326,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.75,
      "step": 1304
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 120.75,
      "epoch": 5.22,
      "grad_norm": 12.984841346740723,
      "kl": 0.315185546875,
      "learning_rate": 4.779999999999999e-07,
      "loss": 0.0003,
      "reward": 0.375,
      "reward_std": 0.2314550280570984,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.375,
      "step": 1305
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 118.28125,
      "epoch": 5.224,
      "grad_norm": 0.02908492088317871,
      "kl": 0.286376953125,
      "learning_rate": 4.776e-07,
      "loss": 0.0003,
      "reward": 0.75,
      "reward_std": 0.0,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.75,
      "step": 1306
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 124.71875,
      "epoch": 5.228,
      "grad_norm": 0.09594712406396866,
      "kl": 0.2783203125,
      "learning_rate": 4.772e-07,
      "loss": 0.0003,
      "reward": 0.25,
      "reward_std": 0.0,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.25,
      "step": 1307
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 110.25,
      "epoch": 5.232,
      "grad_norm": 3.8894386291503906,
      "kl": 0.308349609375,
      "learning_rate": 4.768e-07,
      "loss": 0.0003,
      "reward": 0.1875,
      "reward_std": 0.1157275140285492,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.1875,
      "step": 1308
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 115.4375,
      "epoch": 5.236,
      "grad_norm": 0.021537786349654198,
      "kl": 0.292236328125,
      "learning_rate": 4.7639999999999995e-07,
      "loss": 0.0003,
      "reward": 0.75,
      "reward_std": 0.0,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.75,
      "step": 1309
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 116.625,
      "epoch": 5.24,
      "grad_norm": 11.736676216125488,
      "kl": 0.3056640625,
      "learning_rate": 4.76e-07,
      "loss": 0.0003,
      "reward": 0.03125,
      "reward_std": 0.0883883461356163,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.03125,
      "step": 1310
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 106.5625,
      "epoch": 5.244,
      "grad_norm": 0.032925479114055634,
      "kl": 0.34814453125,
      "learning_rate": 4.756e-07,
      "loss": 0.0003,
      "reward": 0.75,
      "reward_std": 0.0,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.75,
      "step": 1311
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 111.96875,
      "epoch": 5.248,
      "grad_norm": 0.01813489943742752,
      "kl": 0.28466796875,
      "learning_rate": 4.7519999999999997e-07,
      "loss": 0.0003,
      "reward": 0.25,
      "reward_std": 0.0,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.25,
      "step": 1312
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 113.6875,
      "epoch": 5.252,
      "grad_norm": 0.02953912504017353,
      "kl": 0.296875,
      "learning_rate": 4.748e-07,
      "loss": 0.0003,
      "reward": 0.0,
      "reward_std": 0.0,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.0,
      "step": 1313
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 101.96875,
      "epoch": 5.256,
      "grad_norm": 5.140154838562012,
      "kl": 0.32470703125,
      "learning_rate": 4.7439999999999996e-07,
      "loss": 0.0003,
      "reward": 0.71875,
      "reward_std": 0.0883883461356163,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.71875,
      "step": 1314
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 106.75,
      "epoch": 5.26,
      "grad_norm": 0.013369467109441757,
      "kl": 0.291748046875,
      "learning_rate": 4.7399999999999993e-07,
      "loss": 0.0003,
      "reward": 0.0,
      "reward_std": 0.0,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.0,
      "step": 1315
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 117.875,
      "epoch": 5.264,
      "grad_norm": 0.07846423983573914,
      "kl": 0.317138671875,
      "learning_rate": 4.736e-07,
      "loss": 0.0003,
      "reward": 0.5,
      "reward_std": 0.0,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.5,
      "step": 1316
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 104.84375,
      "epoch": 5.268,
      "grad_norm": 17.26542091369629,
      "kl": 0.308349609375,
      "learning_rate": 4.732e-07,
      "loss": 0.0003,
      "reward": 0.53125,
      "reward_std": 0.0883883461356163,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.53125,
      "step": 1317
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 117.46875,
      "epoch": 5.272,
      "grad_norm": 0.038358062505722046,
      "kl": 0.272705078125,
      "learning_rate": 4.728e-07,
      "loss": 0.0003,
      "reward": 0.0,
      "reward_std": 0.0,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.0,
      "step": 1318
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 104.5,
      "epoch": 5.276,
      "grad_norm": 0.01997431553900242,
      "kl": 0.28955078125,
      "learning_rate": 4.7239999999999997e-07,
      "loss": 0.0003,
      "reward": 0.75,
      "reward_std": 0.0,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.75,
      "step": 1319
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 121.15625,
      "epoch": 5.28,
      "grad_norm": 0.024043850600719452,
      "kl": 0.281494140625,
      "learning_rate": 4.7199999999999994e-07,
      "loss": 0.0003,
      "reward": 1.0,
      "reward_std": 0.0,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 1.0,
      "step": 1320
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 91.84375,
      "epoch": 5.284,
      "grad_norm": 0.07115339487791061,
      "kl": 0.33935546875,
      "learning_rate": 4.716e-07,
      "loss": 0.0003,
      "reward": 0.75,
      "reward_std": 0.0,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.75,
      "step": 1321
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 96.3125,
      "epoch": 5.288,
      "grad_norm": 0.018591202795505524,
      "kl": 0.3193359375,
      "learning_rate": 4.712e-07,
      "loss": 0.0003,
      "reward": 1.0,
      "reward_std": 0.0,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 1.0,
      "step": 1322
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 115.96875,
      "epoch": 5.292,
      "grad_norm": 0.034662384539842606,
      "kl": 0.3310546875,
      "learning_rate": 4.7079999999999995e-07,
      "loss": 0.0003,
      "reward": 0.5,
      "reward_std": 0.0,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.5,
      "step": 1323
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 100.0,
      "epoch": 5.296,
      "grad_norm": 0.03057856857776642,
      "kl": 0.3447265625,
      "learning_rate": 4.704e-07,
      "loss": 0.0003,
      "reward": 0.75,
      "reward_std": 0.0,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.75,
      "step": 1324
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 98.75,
      "epoch": 5.3,
      "grad_norm": 0.01605205051600933,
      "kl": 0.35107421875,
      "learning_rate": 4.6999999999999995e-07,
      "loss": 0.0004,
      "reward": 0.25,
      "reward_std": 0.0,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.25,
      "step": 1325
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 90.28125,
      "epoch": 5.304,
      "grad_norm": 15.290507316589355,
      "kl": 0.32373046875,
      "learning_rate": 4.6959999999999997e-07,
      "loss": 0.0003,
      "reward": 0.8125,
      "reward_std": 0.249358132481575,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.8125,
      "step": 1326
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 122.0625,
      "epoch": 5.308,
      "grad_norm": 0.016233794391155243,
      "kl": 0.264892578125,
      "learning_rate": 4.692e-07,
      "loss": 0.0003,
      "reward": 0.5,
      "reward_std": 0.0,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.5,
      "step": 1327
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 116.9375,
      "epoch": 5.312,
      "grad_norm": 0.019229408353567123,
      "kl": 0.287109375,
      "learning_rate": 4.6879999999999996e-07,
      "loss": 0.0003,
      "reward": 0.5,
      "reward_std": 0.0,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.5,
      "step": 1328
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 110.75,
      "epoch": 5.316,
      "grad_norm": 0.02307780273258686,
      "kl": 0.3203125,
      "learning_rate": 4.684e-07,
      "loss": 0.0003,
      "reward": 0.5,
      "reward_std": 0.0,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.5,
      "step": 1329
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 122.15625,
      "epoch": 5.32,
      "grad_norm": 0.02725479006767273,
      "kl": 0.2734375,
      "learning_rate": 4.68e-07,
      "loss": 0.0003,
      "reward": 0.25,
      "reward_std": 0.0,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.25,
      "step": 1330
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 117.25,
      "epoch": 5.324,
      "grad_norm": 0.017006712034344673,
      "kl": 0.297119140625,
      "learning_rate": 4.676e-07,
      "loss": 0.0003,
      "reward": 0.75,
      "reward_std": 0.0,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.75,
      "step": 1331
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 119.09375,
      "epoch": 5.328,
      "grad_norm": 6.460864067077637,
      "kl": 0.32373046875,
      "learning_rate": 4.672e-07,
      "loss": 0.0003,
      "reward": 0.625,
      "reward_std": 0.13363061845302582,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.625,
      "step": 1332
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 118.9375,
      "epoch": 5.332,
      "grad_norm": 0.036992259323596954,
      "kl": 0.25830078125,
      "learning_rate": 4.6679999999999997e-07,
      "loss": 0.0003,
      "reward": 0.25,
      "reward_std": 0.0,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.25,
      "step": 1333
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 112.84375,
      "epoch": 5.336,
      "grad_norm": 16.82443618774414,
      "kl": 0.307861328125,
      "learning_rate": 4.6639999999999994e-07,
      "loss": 0.0003,
      "reward": 0.53125,
      "reward_std": 0.0883883461356163,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.53125,
      "step": 1334
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 115.21875,
      "epoch": 5.34,
      "grad_norm": 5.29921817779541,
      "kl": 0.25830078125,
      "learning_rate": 4.66e-07,
      "loss": 0.0003,
      "reward": 0.5625,
      "reward_std": 0.1157275140285492,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.5625,
      "step": 1335
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 122.84375,
      "epoch": 5.344,
      "grad_norm": 0.01829020492732525,
      "kl": 0.274169921875,
      "learning_rate": 4.656e-07,
      "loss": 0.0003,
      "reward": 1.0,
      "reward_std": 0.0,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 1.0,
      "step": 1336
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 111.875,
      "epoch": 5.348,
      "grad_norm": 0.08917669951915741,
      "kl": 0.46435546875,
      "learning_rate": 4.6519999999999996e-07,
      "loss": 0.0005,
      "reward": 0.25,
      "reward_std": 0.0,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.25,
      "step": 1337
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 123.78125,
      "epoch": 5.352,
      "grad_norm": 0.012455848976969719,
      "kl": 0.241455078125,
      "learning_rate": 4.648e-07,
      "loss": 0.0002,
      "reward": 0.75,
      "reward_std": 0.0,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.75,
      "step": 1338
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 117.25,
      "epoch": 5.356,
      "grad_norm": 0.014067823998630047,
      "kl": 0.276611328125,
      "learning_rate": 4.6439999999999995e-07,
      "loss": 0.0003,
      "reward": 1.0,
      "reward_std": 0.0,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 1.0,
      "step": 1339
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 124.8125,
      "epoch": 5.36,
      "grad_norm": 0.2574978768825531,
      "kl": 0.328125,
      "learning_rate": 4.64e-07,
      "loss": 0.0003,
      "reward": 0.5,
      "reward_std": 0.0,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.5,
      "step": 1340
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 117.15625,
      "epoch": 5.364,
      "grad_norm": 0.025305943563580513,
      "kl": 0.308349609375,
      "learning_rate": 4.636e-07,
      "loss": 0.0003,
      "reward": 1.0,
      "reward_std": 0.0,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 1.0,
      "step": 1341
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 109.34375,
      "epoch": 5.368,
      "grad_norm": 0.01303054392337799,
      "kl": 0.265625,
      "learning_rate": 4.6319999999999997e-07,
      "loss": 0.0003,
      "reward": 0.5,
      "reward_std": 0.0,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.5,
      "step": 1342
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 125.71875,
      "epoch": 5.372,
      "grad_norm": 0.029574358835816383,
      "kl": 0.253662109375,
      "learning_rate": 4.628e-07,
      "loss": 0.0003,
      "reward": 0.25,
      "reward_std": 0.0,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.25,
      "step": 1343
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 130.09375,
      "epoch": 5.376,
      "grad_norm": 0.01669168844819069,
      "kl": 0.275634765625,
      "learning_rate": 4.6239999999999996e-07,
      "loss": 0.0003,
      "reward": 0.75,
      "reward_std": 0.0,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.75,
      "step": 1344
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 103.59375,
      "epoch": 5.38,
      "grad_norm": 0.020334647968411446,
      "kl": 0.30029296875,
      "learning_rate": 4.62e-07,
      "loss": 0.0003,
      "reward": 0.5,
      "reward_std": 0.0,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.5,
      "step": 1345
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 124.8125,
      "epoch": 5.384,
      "grad_norm": 2.8787789344787598,
      "kl": 0.267822265625,
      "learning_rate": 4.616e-07,
      "loss": 0.0003,
      "reward": 0.46875,
      "reward_std": 0.0883883461356163,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.46875,
      "step": 1346
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 99.46875,
      "epoch": 5.388,
      "grad_norm": 0.017947504296898842,
      "kl": 0.2763671875,
      "learning_rate": 4.612e-07,
      "loss": 0.0003,
      "reward": 0.75,
      "reward_std": 0.0,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.75,
      "step": 1347
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 130.625,
      "epoch": 5.392,
      "grad_norm": 7.526956558227539,
      "kl": 0.2431640625,
      "learning_rate": 4.6079999999999994e-07,
      "loss": 0.0002,
      "reward": 0.71875,
      "reward_std": 0.0883883461356163,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.71875,
      "step": 1348
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 127.65625,
      "epoch": 5.396,
      "grad_norm": 0.016020270064473152,
      "kl": 0.270263671875,
      "learning_rate": 4.6039999999999997e-07,
      "loss": 0.0003,
      "reward": 0.75,
      "reward_std": 0.0,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.75,
      "step": 1349
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 123.5625,
      "epoch": 5.4,
      "grad_norm": 0.0346699096262455,
      "kl": 0.281982421875,
      "learning_rate": 4.6e-07,
      "loss": 0.0003,
      "reward": 0.25,
      "reward_std": 0.0,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.25,
      "step": 1350
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 115.46875,
      "epoch": 5.404,
      "grad_norm": 0.027214795351028442,
      "kl": 0.28076171875,
      "learning_rate": 4.596e-07,
      "loss": 0.0003,
      "reward": 0.25,
      "reward_std": 0.0,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.25,
      "step": 1351
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 124.71875,
      "epoch": 5.408,
      "grad_norm": 0.014406121335923672,
      "kl": 0.26513671875,
      "learning_rate": 4.592e-07,
      "loss": 0.0003,
      "reward": 0.25,
      "reward_std": 0.0,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.25,
      "step": 1352
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 123.59375,
      "epoch": 5.412,
      "grad_norm": 0.013721935451030731,
      "kl": 0.2900390625,
      "learning_rate": 4.5879999999999995e-07,
      "loss": 0.0003,
      "reward": 0.25,
      "reward_std": 0.0,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.25,
      "step": 1353
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 121.25,
      "epoch": 5.416,
      "grad_norm": 0.014371649362146854,
      "kl": 0.25244140625,
      "learning_rate": 4.584e-07,
      "loss": 0.0003,
      "reward": 1.0,
      "reward_std": 0.0,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 1.0,
      "step": 1354
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 128.90625,
      "epoch": 5.42,
      "grad_norm": 5.574648380279541,
      "kl": 0.2587890625,
      "learning_rate": 4.58e-07,
      "loss": 0.0003,
      "reward": 0.40625,
      "reward_std": 0.1293872892856598,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.40625,
      "step": 1355
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 123.59375,
      "epoch": 5.424,
      "grad_norm": 7.619860649108887,
      "kl": 0.3642578125,
      "learning_rate": 4.5759999999999997e-07,
      "loss": 0.0004,
      "reward": 0.71875,
      "reward_std": 0.0883883461356163,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.71875,
      "step": 1356
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 132.28125,
      "epoch": 5.428,
      "grad_norm": 0.015486681833863258,
      "kl": 0.2646484375,
      "learning_rate": 4.572e-07,
      "loss": 0.0003,
      "reward": 0.75,
      "reward_std": 0.0,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.75,
      "step": 1357
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 126.1875,
      "epoch": 5.432,
      "grad_norm": 0.021418945863842964,
      "kl": 0.2763671875,
      "learning_rate": 4.5679999999999996e-07,
      "loss": 0.0003,
      "reward": 0.5,
      "reward_std": 0.0,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.5,
      "step": 1358
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 117.03125,
      "epoch": 5.436,
      "grad_norm": 0.012791316956281662,
      "kl": 0.259033203125,
      "learning_rate": 4.5639999999999993e-07,
      "loss": 0.0003,
      "reward": 0.75,
      "reward_std": 0.0,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.75,
      "step": 1359
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 108.875,
      "epoch": 5.44,
      "grad_norm": 3.0429584980010986,
      "kl": 0.275634765625,
      "learning_rate": 4.56e-07,
      "loss": 0.0003,
      "reward": 0.5625,
      "reward_std": 0.1157275140285492,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.5625,
      "step": 1360
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 125.0625,
      "epoch": 5.444,
      "grad_norm": 0.01512636337429285,
      "kl": 0.253173828125,
      "learning_rate": 4.556e-07,
      "loss": 0.0003,
      "reward": 0.5,
      "reward_std": 0.0,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.5,
      "step": 1361
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 128.3125,
      "epoch": 5.448,
      "grad_norm": 0.014309265650808811,
      "kl": 0.26416015625,
      "learning_rate": 4.5519999999999995e-07,
      "loss": 0.0003,
      "reward": 0.75,
      "reward_std": 0.0,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.75,
      "step": 1362
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 131.90625,
      "epoch": 5.452,
      "grad_norm": 22.00885772705078,
      "kl": 0.248291015625,
      "learning_rate": 4.5479999999999997e-07,
      "loss": 0.0002,
      "reward": 0.65625,
      "reward_std": 0.2041158601641655,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.65625,
      "step": 1363
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 125.65625,
      "epoch": 5.456,
      "grad_norm": 0.015618322417140007,
      "kl": 0.2705078125,
      "learning_rate": 4.544e-07,
      "loss": 0.0003,
      "reward": 0.75,
      "reward_std": 0.0,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.75,
      "step": 1364
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 106.0625,
      "epoch": 5.46,
      "grad_norm": 7.478302478790283,
      "kl": 0.291015625,
      "learning_rate": 4.54e-07,
      "loss": 0.0003,
      "reward": 0.375,
      "reward_std": 0.13363061845302582,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.375,
      "step": 1365
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 130.78125,
      "epoch": 5.464,
      "grad_norm": 0.01786927506327629,
      "kl": 0.26220703125,
      "learning_rate": 4.536e-07,
      "loss": 0.0003,
      "reward": 0.0,
      "reward_std": 0.0,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.0,
      "step": 1366
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 116.875,
      "epoch": 5.468,
      "grad_norm": 3.8085968494415283,
      "kl": 0.254150390625,
      "learning_rate": 4.5319999999999996e-07,
      "loss": 0.0003,
      "reward": 0.28125,
      "reward_std": 0.0883883461356163,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.28125,
      "step": 1367
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 119.59375,
      "epoch": 5.4719999999999995,
      "grad_norm": 4.416646957397461,
      "kl": 0.29296875,
      "learning_rate": 4.528e-07,
      "loss": 0.0003,
      "reward": 0.46875,
      "reward_std": 0.0883883461356163,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.46875,
      "step": 1368
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 107.1875,
      "epoch": 5.476,
      "grad_norm": 0.017520559951663017,
      "kl": 0.288818359375,
      "learning_rate": 4.524e-07,
      "loss": 0.0003,
      "reward": 1.0,
      "reward_std": 0.0,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 1.0,
      "step": 1369
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 116.21875,
      "epoch": 5.48,
      "grad_norm": 0.02807965874671936,
      "kl": 0.30078125,
      "learning_rate": 4.5199999999999997e-07,
      "loss": 0.0003,
      "reward": 1.0,
      "reward_std": 0.0,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 1.0,
      "step": 1370
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 132.0625,
      "epoch": 5.484,
      "grad_norm": 0.019850078970193863,
      "kl": 0.265625,
      "learning_rate": 4.516e-07,
      "loss": 0.0003,
      "reward": 0.25,
      "reward_std": 0.0,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.25,
      "step": 1371
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 126.0625,
      "epoch": 5.4879999999999995,
      "grad_norm": 0.01606827788054943,
      "kl": 0.24169921875,
      "learning_rate": 4.5119999999999996e-07,
      "loss": 0.0002,
      "reward": 0.25,
      "reward_std": 0.0,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.25,
      "step": 1372
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 121.1875,
      "epoch": 5.492,
      "grad_norm": 0.03711554408073425,
      "kl": 0.29736328125,
      "learning_rate": 4.5079999999999993e-07,
      "loss": 0.0003,
      "reward": 1.0,
      "reward_std": 0.0,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 1.0,
      "step": 1373
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 119.34375,
      "epoch": 5.496,
      "grad_norm": 0.019352754577994347,
      "kl": 0.3203125,
      "learning_rate": 4.504e-07,
      "loss": 0.0003,
      "reward": 0.25,
      "reward_std": 0.0,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.25,
      "step": 1374
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 124.375,
      "epoch": 5.5,
      "grad_norm": 0.01793813332915306,
      "kl": 0.296875,
      "learning_rate": 4.5e-07,
      "loss": 0.0003,
      "reward": 0.25,
      "reward_std": 0.0,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.25,
      "step": 1375
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 119.53125,
      "epoch": 5.504,
      "grad_norm": 6.705965995788574,
      "kl": 0.29736328125,
      "learning_rate": 4.496e-07,
      "loss": 0.0003,
      "reward": 0.5625,
      "reward_std": 0.2177756354212761,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.5625,
      "step": 1376
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 123.0,
      "epoch": 5.508,
      "grad_norm": 0.04789667949080467,
      "kl": 0.234375,
      "learning_rate": 4.4919999999999997e-07,
      "loss": 0.0002,
      "reward": 0.5,
      "reward_std": 0.0,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.5,
      "step": 1377
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 106.8125,
      "epoch": 5.5120000000000005,
      "grad_norm": 9.72686767578125,
      "kl": 0.32861328125,
      "learning_rate": 4.4879999999999994e-07,
      "loss": 0.0003,
      "reward": 0.71875,
      "reward_std": 0.0883883461356163,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.71875,
      "step": 1378
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 130.6875,
      "epoch": 5.516,
      "grad_norm": 0.5167446732521057,
      "kl": 0.252197265625,
      "learning_rate": 4.484e-07,
      "loss": 0.0003,
      "reward": 0.5,
      "reward_std": 0.0,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.5,
      "step": 1379
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 137.5,
      "epoch": 5.52,
      "grad_norm": 9.791504859924316,
      "kl": 0.2724609375,
      "learning_rate": 4.48e-07,
      "loss": 0.0003,
      "reward": 0.71875,
      "reward_std": 0.0883883461356163,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.71875,
      "step": 1380
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 129.46875,
      "epoch": 5.524,
      "grad_norm": 0.020339520648121834,
      "kl": 0.30908203125,
      "learning_rate": 4.4759999999999996e-07,
      "loss": 0.0003,
      "reward": 0.75,
      "reward_std": 0.0,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.75,
      "step": 1381
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 125.0,
      "epoch": 5.5280000000000005,
      "grad_norm": 31.193042755126953,
      "kl": 0.259033203125,
      "learning_rate": 4.472e-07,
      "loss": 0.0003,
      "reward": 0.71875,
      "reward_std": 0.0883883461356163,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.71875,
      "step": 1382
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 121.03125,
      "epoch": 5.532,
      "grad_norm": 0.022690288722515106,
      "kl": 0.331298828125,
      "learning_rate": 4.4679999999999995e-07,
      "loss": 0.0003,
      "reward": 0.25,
      "reward_std": 0.0,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.25,
      "step": 1383
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 139.1875,
      "epoch": 5.536,
      "grad_norm": 0.020180946215987206,
      "kl": 0.23974609375,
      "learning_rate": 4.464e-07,
      "loss": 0.0002,
      "reward": 0.75,
      "reward_std": 0.0,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.75,
      "step": 1384
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 115.03125,
      "epoch": 5.54,
      "grad_norm": 0.02811640128493309,
      "kl": 0.2900390625,
      "learning_rate": 4.46e-07,
      "loss": 0.0003,
      "reward": 0.5,
      "reward_std": 0.0,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.5,
      "step": 1385
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 147.53125,
      "epoch": 5.5440000000000005,
      "grad_norm": 5.117380142211914,
      "kl": 0.24853515625,
      "learning_rate": 4.4559999999999997e-07,
      "loss": 0.0002,
      "reward": 0.46875,
      "reward_std": 0.0883883461356163,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.46875,
      "step": 1386
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 110.59375,
      "epoch": 5.548,
      "grad_norm": 15.822004318237305,
      "kl": 0.29541015625,
      "learning_rate": 4.452e-07,
      "loss": 0.0003,
      "reward": 0.3125,
      "reward_std": 0.1157275140285492,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.3125,
      "step": 1387
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 121.9375,
      "epoch": 5.552,
      "grad_norm": 0.02592463605105877,
      "kl": 0.28955078125,
      "learning_rate": 4.4479999999999996e-07,
      "loss": 0.0003,
      "reward": 0.25,
      "reward_std": 0.0,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.25,
      "step": 1388
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 126.78125,
      "epoch": 5.556,
      "grad_norm": 0.019850684329867363,
      "kl": 0.2919921875,
      "learning_rate": 4.444e-07,
      "loss": 0.0003,
      "reward": 0.75,
      "reward_std": 0.0,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.75,
      "step": 1389
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 122.4375,
      "epoch": 5.5600000000000005,
      "grad_norm": 0.016988864168524742,
      "kl": 0.276123046875,
      "learning_rate": 4.44e-07,
      "loss": 0.0003,
      "reward": 0.75,
      "reward_std": 0.0,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.75,
      "step": 1390
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 147.875,
      "epoch": 5.564,
      "grad_norm": 14.205126762390137,
      "kl": 0.21337890625,
      "learning_rate": 4.436e-07,
      "loss": 0.0002,
      "reward": 0.4375,
      "reward_std": 0.1157275140285492,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.4375,
      "step": 1391
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 142.71875,
      "epoch": 5.568,
      "grad_norm": 0.014212789945304394,
      "kl": 0.23193359375,
      "learning_rate": 4.4319999999999995e-07,
      "loss": 0.0002,
      "reward": 0.5,
      "reward_std": 0.0,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.5,
      "step": 1392
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 122.90625,
      "epoch": 5.572,
      "grad_norm": 0.06460288912057877,
      "kl": 0.289306640625,
      "learning_rate": 4.428e-07,
      "loss": 0.0003,
      "reward": 0.75,
      "reward_std": 0.0,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.75,
      "step": 1393
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 141.6875,
      "epoch": 5.576,
      "grad_norm": 0.015372205525636673,
      "kl": 0.231689453125,
      "learning_rate": 4.424e-07,
      "loss": 0.0002,
      "reward": 0.25,
      "reward_std": 0.0,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.25,
      "step": 1394
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 126.71875,
      "epoch": 5.58,
      "grad_norm": 0.03167112544178963,
      "kl": 0.2587890625,
      "learning_rate": 4.4199999999999996e-07,
      "loss": 0.0003,
      "reward": 0.5,
      "reward_std": 0.0,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.5,
      "step": 1395
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 137.28125,
      "epoch": 5.584,
      "grad_norm": 2.416749954223633,
      "kl": 0.25390625,
      "learning_rate": 4.416e-07,
      "loss": 0.0003,
      "reward": 0.21875,
      "reward_std": 0.0883883461356163,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.21875,
      "step": 1396
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 137.40625,
      "epoch": 5.588,
      "grad_norm": 0.0339285172522068,
      "kl": 0.25048828125,
      "learning_rate": 4.4119999999999995e-07,
      "loss": 0.0003,
      "reward": 0.5,
      "reward_std": 0.0,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.5,
      "step": 1397
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 148.03125,
      "epoch": 5.592,
      "grad_norm": 19.303815841674805,
      "kl": 0.2470703125,
      "learning_rate": 4.4080000000000003e-07,
      "loss": 0.0002,
      "reward": 0.96875,
      "reward_std": 0.0883883461356163,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.96875,
      "step": 1398
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 129.65625,
      "epoch": 5.596,
      "grad_norm": 24.951007843017578,
      "kl": 0.24755859375,
      "learning_rate": 4.404e-07,
      "loss": 0.0002,
      "reward": 0.6875,
      "reward_std": 0.1157275140285492,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.6875,
      "step": 1399
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 125.8125,
      "epoch": 5.6,
      "grad_norm": 3.073333978652954,
      "kl": 0.299072265625,
      "learning_rate": 4.3999999999999997e-07,
      "loss": 0.0003,
      "reward": 0.78125,
      "reward_std": 0.0883883461356163,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.78125,
      "step": 1400
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 135.21875,
      "epoch": 5.604,
      "grad_norm": 0.012348666787147522,
      "kl": 0.23291015625,
      "learning_rate": 4.396e-07,
      "loss": 0.0002,
      "reward": 0.5,
      "reward_std": 0.0,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.5,
      "step": 1401
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 139.25,
      "epoch": 5.608,
      "grad_norm": 0.012078620493412018,
      "kl": 0.244384765625,
      "learning_rate": 4.3919999999999996e-07,
      "loss": 0.0002,
      "reward": 0.5,
      "reward_std": 0.0,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.5,
      "step": 1402
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 132.09375,
      "epoch": 5.612,
      "grad_norm": 8.18929386138916,
      "kl": 0.257080078125,
      "learning_rate": 4.388e-07,
      "loss": 0.0003,
      "reward": 0.46875,
      "reward_std": 0.0883883461356163,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.46875,
      "step": 1403
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 123.65625,
      "epoch": 5.616,
      "grad_norm": 0.01962558925151825,
      "kl": 0.254150390625,
      "learning_rate": 4.384e-07,
      "loss": 0.0003,
      "reward": 0.75,
      "reward_std": 0.0,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.75,
      "step": 1404
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 121.1875,
      "epoch": 5.62,
      "grad_norm": 14.552216529846191,
      "kl": 0.278076171875,
      "learning_rate": 4.38e-07,
      "loss": 0.0003,
      "reward": 0.375,
      "reward_std": 0.2177756354212761,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.375,
      "step": 1405
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 129.25,
      "epoch": 5.624,
      "grad_norm": 6.222041130065918,
      "kl": 0.251708984375,
      "learning_rate": 4.3759999999999995e-07,
      "loss": 0.0003,
      "reward": 0.71875,
      "reward_std": 0.0883883461356163,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.71875,
      "step": 1406
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 128.875,
      "epoch": 5.628,
      "grad_norm": 0.016271695494651794,
      "kl": 0.265380859375,
      "learning_rate": 4.3719999999999997e-07,
      "loss": 0.0003,
      "reward": 1.0,
      "reward_std": 0.0,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 1.0,
      "step": 1407
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 117.28125,
      "epoch": 5.632,
      "grad_norm": 0.028270702809095383,
      "kl": 0.32177734375,
      "learning_rate": 4.368e-07,
      "loss": 0.0003,
      "reward": 0.75,
      "reward_std": 0.0,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.75,
      "step": 1408
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 143.71875,
      "epoch": 5.636,
      "grad_norm": 0.0161314494907856,
      "kl": 0.212158203125,
      "learning_rate": 4.364e-07,
      "loss": 0.0002,
      "reward": 0.5,
      "reward_std": 0.0,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.5,
      "step": 1409
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 120.71875,
      "epoch": 5.64,
      "grad_norm": 0.015298144891858101,
      "kl": 0.24072265625,
      "learning_rate": 4.36e-07,
      "loss": 0.0002,
      "reward": 0.75,
      "reward_std": 0.0,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.75,
      "step": 1410
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 129.34375,
      "epoch": 5.644,
      "grad_norm": 0.014223064295947552,
      "kl": 0.255859375,
      "learning_rate": 4.3559999999999996e-07,
      "loss": 0.0003,
      "reward": 0.5,
      "reward_std": 0.0,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.5,
      "step": 1411
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 116.4375,
      "epoch": 5.648,
      "grad_norm": 0.10899148136377335,
      "kl": 0.27197265625,
      "learning_rate": 4.352e-07,
      "loss": 0.0003,
      "reward": 0.75,
      "reward_std": 0.0,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.75,
      "step": 1412
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 124.59375,
      "epoch": 5.652,
      "grad_norm": 0.06161722168326378,
      "kl": 0.276123046875,
      "learning_rate": 4.348e-07,
      "loss": 0.0003,
      "reward": 0.0,
      "reward_std": 0.0,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.0,
      "step": 1413
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 127.125,
      "epoch": 5.656,
      "grad_norm": 5.6334147453308105,
      "kl": 0.23486328125,
      "learning_rate": 4.3439999999999997e-07,
      "loss": 0.0002,
      "reward": 0.71875,
      "reward_std": 0.2041158601641655,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.71875,
      "step": 1414
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 106.84375,
      "epoch": 5.66,
      "grad_norm": 10.188370704650879,
      "kl": 0.308837890625,
      "learning_rate": 4.34e-07,
      "loss": 0.0003,
      "reward": 0.5625,
      "reward_std": 0.1767766922712326,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.5625,
      "step": 1415
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 124.71875,
      "epoch": 5.664,
      "grad_norm": 3.2052812576293945,
      "kl": 0.26708984375,
      "learning_rate": 4.3359999999999997e-07,
      "loss": 0.0003,
      "reward": 0.3125,
      "reward_std": 0.1157275140285492,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.3125,
      "step": 1416
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 115.125,
      "epoch": 5.668,
      "grad_norm": 3.812560796737671,
      "kl": 0.259033203125,
      "learning_rate": 4.3319999999999994e-07,
      "loss": 0.0003,
      "reward": 0.65625,
      "reward_std": 0.1293872892856598,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.65625,
      "step": 1417
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 115.09375,
      "epoch": 5.672,
      "grad_norm": 0.023214712738990784,
      "kl": 0.283447265625,
      "learning_rate": 4.328e-07,
      "loss": 0.0003,
      "reward": 0.25,
      "reward_std": 0.0,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.25,
      "step": 1418
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 114.125,
      "epoch": 5.676,
      "grad_norm": 4.322704315185547,
      "kl": 0.30322265625,
      "learning_rate": 4.324e-07,
      "loss": 0.0003,
      "reward": 0.03125,
      "reward_std": 0.0883883461356163,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.03125,
      "step": 1419
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 116.375,
      "epoch": 5.68,
      "grad_norm": 0.017529631033539772,
      "kl": 0.283447265625,
      "learning_rate": 4.3199999999999995e-07,
      "loss": 0.0003,
      "reward": 0.75,
      "reward_std": 0.0,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.75,
      "step": 1420
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 89.1875,
      "epoch": 5.684,
      "grad_norm": 6.940032005310059,
      "kl": 0.3125,
      "learning_rate": 4.316e-07,
      "loss": 0.0003,
      "reward": 0.3125,
      "reward_std": 0.1157275140285492,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.3125,
      "step": 1421
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 100.21875,
      "epoch": 5.688,
      "grad_norm": 0.03592343628406525,
      "kl": 0.31103515625,
      "learning_rate": 4.312e-07,
      "loss": 0.0003,
      "reward": 0.5,
      "reward_std": 0.0,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.5,
      "step": 1422
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 101.65625,
      "epoch": 5.692,
      "grad_norm": 0.023512819781899452,
      "kl": 0.30419921875,
      "learning_rate": 4.308e-07,
      "loss": 0.0003,
      "reward": 1.0,
      "reward_std": 0.0,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 1.0,
      "step": 1423
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 96.8125,
      "epoch": 5.696,
      "grad_norm": 0.03242960944771767,
      "kl": 0.303955078125,
      "learning_rate": 4.304e-07,
      "loss": 0.0003,
      "reward": 0.5,
      "reward_std": 0.0,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.5,
      "step": 1424
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 118.15625,
      "epoch": 5.7,
      "grad_norm": 7.978532791137695,
      "kl": 0.258056640625,
      "learning_rate": 4.2999999999999996e-07,
      "loss": 0.0003,
      "reward": 0.375,
      "reward_std": 0.2314550280570984,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.375,
      "step": 1425
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 119.65625,
      "epoch": 5.704,
      "grad_norm": 0.038134608417749405,
      "kl": 0.274658203125,
      "learning_rate": 4.296e-07,
      "loss": 0.0003,
      "reward": 0.5,
      "reward_std": 0.0,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.5,
      "step": 1426
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 114.71875,
      "epoch": 5.708,
      "grad_norm": 0.02023491822183132,
      "kl": 0.266357421875,
      "learning_rate": 4.292e-07,
      "loss": 0.0003,
      "reward": 0.75,
      "reward_std": 0.0,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.75,
      "step": 1427
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 112.5,
      "epoch": 5.712,
      "grad_norm": 0.011425556614995003,
      "kl": 0.289306640625,
      "learning_rate": 4.288e-07,
      "loss": 0.0003,
      "reward": 0.5,
      "reward_std": 0.0,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.5,
      "step": 1428
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 113.03125,
      "epoch": 5.716,
      "grad_norm": 4.867710590362549,
      "kl": 0.31005859375,
      "learning_rate": 4.284e-07,
      "loss": 0.0003,
      "reward": 0.25,
      "reward_std": 0.1767766922712326,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.25,
      "step": 1429
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 118.78125,
      "epoch": 5.72,
      "grad_norm": 0.0579223595559597,
      "kl": 0.348876953125,
      "learning_rate": 4.2799999999999997e-07,
      "loss": 0.0003,
      "reward": 0.25,
      "reward_std": 0.0,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.25,
      "step": 1430
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 119.90625,
      "epoch": 5.724,
      "grad_norm": 26.91503143310547,
      "kl": 0.2587890625,
      "learning_rate": 4.2759999999999994e-07,
      "loss": 0.0003,
      "reward": 0.25,
      "reward_std": 0.1767766922712326,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.25,
      "step": 1431
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 132.78125,
      "epoch": 5.728,
      "grad_norm": 12.275778770446777,
      "kl": 0.24560546875,
      "learning_rate": 4.272e-07,
      "loss": 0.0002,
      "reward": 0.6875,
      "reward_std": 0.2925042062997818,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.6875,
      "step": 1432
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 118.375,
      "epoch": 5.732,
      "grad_norm": 4.2581048011779785,
      "kl": 0.258056640625,
      "learning_rate": 4.268e-07,
      "loss": 0.0003,
      "reward": 0.71875,
      "reward_std": 0.0883883461356163,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.71875,
      "step": 1433
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 126.0625,
      "epoch": 5.736,
      "grad_norm": 9.373467445373535,
      "kl": 0.235107421875,
      "learning_rate": 4.264e-07,
      "loss": 0.0002,
      "reward": 0.15625,
      "reward_std": 0.1293872892856598,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.15625,
      "step": 1434
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 108.40625,
      "epoch": 5.74,
      "grad_norm": 0.018158825114369392,
      "kl": 0.279541015625,
      "learning_rate": 4.26e-07,
      "loss": 0.0003,
      "reward": 0.5,
      "reward_std": 0.0,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.5,
      "step": 1435
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 118.5625,
      "epoch": 5.744,
      "grad_norm": 0.03807520866394043,
      "kl": 0.2763671875,
      "learning_rate": 4.2559999999999995e-07,
      "loss": 0.0003,
      "reward": 0.5,
      "reward_std": 0.0,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.5,
      "step": 1436
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 122.53125,
      "epoch": 5.748,
      "grad_norm": 0.017027022317051888,
      "kl": 0.282470703125,
      "learning_rate": 4.252e-07,
      "loss": 0.0003,
      "reward": 0.5,
      "reward_std": 0.0,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.5,
      "step": 1437
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 119.71875,
      "epoch": 5.752,
      "grad_norm": 0.02188960462808609,
      "kl": 0.261962890625,
      "learning_rate": 4.248e-07,
      "loss": 0.0003,
      "reward": 0.5,
      "reward_std": 0.0,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.5,
      "step": 1438
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 136.96875,
      "epoch": 5.756,
      "grad_norm": 4.000221252441406,
      "kl": 0.231201171875,
      "learning_rate": 4.2439999999999996e-07,
      "loss": 0.0002,
      "reward": 0.46875,
      "reward_std": 0.0883883461356163,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.46875,
      "step": 1439
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 129.4375,
      "epoch": 5.76,
      "grad_norm": 3.5156147480010986,
      "kl": 0.4384765625,
      "learning_rate": 4.24e-07,
      "loss": 0.0004,
      "reward": 0.71875,
      "reward_std": 0.0883883461356163,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.71875,
      "step": 1440
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 125.875,
      "epoch": 5.764,
      "grad_norm": 0.015104636549949646,
      "kl": 0.275390625,
      "learning_rate": 4.2359999999999995e-07,
      "loss": 0.0003,
      "reward": 0.75,
      "reward_std": 0.0,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.75,
      "step": 1441
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 112.4375,
      "epoch": 5.768,
      "grad_norm": 4.537987232208252,
      "kl": 0.281005859375,
      "learning_rate": 4.232e-07,
      "loss": 0.0003,
      "reward": 0.1875,
      "reward_std": 0.1157275140285492,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.1875,
      "step": 1442
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 134.09375,
      "epoch": 5.772,
      "grad_norm": 0.1709776669740677,
      "kl": 0.41357421875,
      "learning_rate": 4.228e-07,
      "loss": 0.0004,
      "reward": 1.0,
      "reward_std": 0.0,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 1.0,
      "step": 1443
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 136.15625,
      "epoch": 5.776,
      "grad_norm": 15.879798889160156,
      "kl": 0.349609375,
      "learning_rate": 4.2239999999999997e-07,
      "loss": 0.0003,
      "reward": 0.40625,
      "reward_std": 0.22201896458864212,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.40625,
      "step": 1444
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 140.65625,
      "epoch": 5.78,
      "grad_norm": 3.6717405319213867,
      "kl": 0.259521484375,
      "learning_rate": 4.2199999999999994e-07,
      "loss": 0.0003,
      "reward": 0.0625,
      "reward_std": 0.1157275140285492,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.0625,
      "step": 1445
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 149.125,
      "epoch": 5.784,
      "grad_norm": 0.009855804964900017,
      "kl": 0.22314453125,
      "learning_rate": 4.2159999999999996e-07,
      "loss": 0.0002,
      "reward": 0.5,
      "reward_std": 0.0,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.5,
      "step": 1446
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 135.9375,
      "epoch": 5.788,
      "grad_norm": 2.943847179412842,
      "kl": 0.23193359375,
      "learning_rate": 4.212e-07,
      "loss": 0.0002,
      "reward": 0.71875,
      "reward_std": 0.0883883461356163,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.71875,
      "step": 1447
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 136.03125,
      "epoch": 5.792,
      "grad_norm": 4.524595737457275,
      "kl": 0.283935546875,
      "learning_rate": 4.208e-07,
      "loss": 0.0003,
      "reward": 0.71875,
      "reward_std": 0.0883883461356163,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.71875,
      "step": 1448
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 122.375,
      "epoch": 5.796,
      "grad_norm": 5.786934852600098,
      "kl": 0.26953125,
      "learning_rate": 4.204e-07,
      "loss": 0.0003,
      "reward": 0.71875,
      "reward_std": 0.0883883461356163,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.71875,
      "step": 1449
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 132.6875,
      "epoch": 5.8,
      "grad_norm": 2.3982131481170654,
      "kl": 0.22802734375,
      "learning_rate": 4.1999999999999995e-07,
      "loss": 0.0002,
      "reward": 0.96875,
      "reward_std": 0.0883883461356163,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.96875,
      "step": 1450
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 141.875,
      "epoch": 5.804,
      "grad_norm": 0.020155563950538635,
      "kl": 0.215087890625,
      "learning_rate": 4.1959999999999997e-07,
      "loss": 0.0002,
      "reward": 0.5,
      "reward_std": 0.0,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.5,
      "step": 1451
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 140.4375,
      "epoch": 5.808,
      "grad_norm": 15.559988021850586,
      "kl": 0.216796875,
      "learning_rate": 4.192e-07,
      "loss": 0.0002,
      "reward": 0.4375,
      "reward_std": 0.1157275140285492,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.4375,
      "step": 1452
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 119.1875,
      "epoch": 5.812,
      "grad_norm": 0.024397611618041992,
      "kl": 0.22998046875,
      "learning_rate": 4.1879999999999996e-07,
      "loss": 0.0002,
      "reward": 0.5,
      "reward_std": 0.0,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.5,
      "step": 1453
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 110.875,
      "epoch": 5.816,
      "grad_norm": 0.03885308653116226,
      "kl": 0.295166015625,
      "learning_rate": 4.184e-07,
      "loss": 0.0003,
      "reward": 0.75,
      "reward_std": 0.0,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.75,
      "step": 1454
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 107.0625,
      "epoch": 5.82,
      "grad_norm": 0.06271590292453766,
      "kl": 0.3564453125,
      "learning_rate": 4.1799999999999996e-07,
      "loss": 0.0004,
      "reward": 0.25,
      "reward_std": 0.0,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.25,
      "step": 1455
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 136.8125,
      "epoch": 5.824,
      "grad_norm": 4.7433905601501465,
      "kl": 0.232177734375,
      "learning_rate": 4.1760000000000003e-07,
      "loss": 0.0002,
      "reward": 0.03125,
      "reward_std": 0.0883883461356163,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.03125,
      "step": 1456
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 152.375,
      "epoch": 5.828,
      "grad_norm": 3.7032761573791504,
      "kl": 0.27734375,
      "learning_rate": 4.172e-07,
      "loss": 0.0003,
      "reward": 0.5625,
      "reward_std": 0.1767766922712326,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.5625,
      "step": 1457
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 131.90625,
      "epoch": 5.832,
      "grad_norm": 8.936473846435547,
      "kl": 0.24462890625,
      "learning_rate": 4.1679999999999997e-07,
      "loss": 0.0002,
      "reward": 0.71875,
      "reward_std": 0.0883883461356163,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.71875,
      "step": 1458
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 139.53125,
      "epoch": 5.836,
      "grad_norm": 0.011547540314495564,
      "kl": 0.239013671875,
      "learning_rate": 4.164e-07,
      "loss": 0.0002,
      "reward": 0.75,
      "reward_std": 0.0,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.75,
      "step": 1459
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 138.53125,
      "epoch": 5.84,
      "grad_norm": 0.012623059563338757,
      "kl": 0.2470703125,
      "learning_rate": 4.1599999999999997e-07,
      "loss": 0.0002,
      "reward": 0.25,
      "reward_std": 0.0,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.25,
      "step": 1460
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 153.0,
      "epoch": 5.844,
      "grad_norm": 0.01233961246907711,
      "kl": 0.22509765625,
      "learning_rate": 4.156e-07,
      "loss": 0.0002,
      "reward": 0.75,
      "reward_std": 0.0,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.75,
      "step": 1461
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 146.1875,
      "epoch": 5.848,
      "grad_norm": 0.041940413415431976,
      "kl": 0.2353515625,
      "learning_rate": 4.152e-07,
      "loss": 0.0002,
      "reward": 0.25,
      "reward_std": 0.0,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.25,
      "step": 1462
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 146.8125,
      "epoch": 5.852,
      "grad_norm": 0.015926353633403778,
      "kl": 0.26025390625,
      "learning_rate": 4.148e-07,
      "loss": 0.0003,
      "reward": 0.25,
      "reward_std": 0.0,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.25,
      "step": 1463
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 153.21875,
      "epoch": 5.856,
      "grad_norm": 0.09476323425769806,
      "kl": 0.246826171875,
      "learning_rate": 4.1439999999999995e-07,
      "loss": 0.0002,
      "reward": 0.5,
      "reward_std": 0.0,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.5,
      "step": 1464
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 116.9375,
      "epoch": 5.86,
      "grad_norm": 0.014341183006763458,
      "kl": 0.29736328125,
      "learning_rate": 4.14e-07,
      "loss": 0.0003,
      "reward": 0.5,
      "reward_std": 0.0,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.5,
      "step": 1465
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 136.8125,
      "epoch": 5.864,
      "grad_norm": 0.02640533819794655,
      "kl": 0.24365234375,
      "learning_rate": 4.136e-07,
      "loss": 0.0002,
      "reward": 0.25,
      "reward_std": 0.0,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.25,
      "step": 1466
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 141.25,
      "epoch": 5.868,
      "grad_norm": 0.15583530068397522,
      "kl": 0.286376953125,
      "learning_rate": 4.1319999999999997e-07,
      "loss": 0.0003,
      "reward": 0.25,
      "reward_std": 0.0,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.25,
      "step": 1467
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 146.03125,
      "epoch": 5.872,
      "grad_norm": 0.023009290918707848,
      "kl": 0.207275390625,
      "learning_rate": 4.128e-07,
      "loss": 0.0002,
      "reward": 0.5,
      "reward_std": 0.0,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.5,
      "step": 1468
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 146.46875,
      "epoch": 5.876,
      "grad_norm": 0.04910910502076149,
      "kl": 0.270751953125,
      "learning_rate": 4.1239999999999996e-07,
      "loss": 0.0003,
      "reward": 0.75,
      "reward_std": 0.0,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.75,
      "step": 1469
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 151.9375,
      "epoch": 5.88,
      "grad_norm": 0.01728554256260395,
      "kl": 0.26025390625,
      "learning_rate": 4.12e-07,
      "loss": 0.0003,
      "reward": 0.5,
      "reward_std": 0.0,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.5,
      "step": 1470
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 103.59375,
      "epoch": 5.884,
      "grad_norm": 5.218578815460205,
      "kl": 0.330810546875,
      "learning_rate": 4.116e-07,
      "loss": 0.0003,
      "reward": 0.8125,
      "reward_std": 0.1157275140285492,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.8125,
      "step": 1471
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 136.0,
      "epoch": 5.888,
      "grad_norm": 0.02677454985678196,
      "kl": 0.25146484375,
      "learning_rate": 4.112e-07,
      "loss": 0.0003,
      "reward": 0.0,
      "reward_std": 0.0,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.0,
      "step": 1472
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 139.46875,
      "epoch": 5.892,
      "grad_norm": 2.7134621143341064,
      "kl": 0.23828125,
      "learning_rate": 4.108e-07,
      "loss": 0.0002,
      "reward": 0.71875,
      "reward_std": 0.0883883461356163,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.71875,
      "step": 1473
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 124.4375,
      "epoch": 5.896,
      "grad_norm": 3.046168327331543,
      "kl": 0.26953125,
      "learning_rate": 4.1039999999999997e-07,
      "loss": 0.0003,
      "reward": 0.53125,
      "reward_std": 0.0883883461356163,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.53125,
      "step": 1474
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 145.53125,
      "epoch": 5.9,
      "grad_norm": 0.024896908551454544,
      "kl": 0.2568359375,
      "learning_rate": 4.0999999999999994e-07,
      "loss": 0.0003,
      "reward": 0.75,
      "reward_std": 0.0,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.75,
      "step": 1475
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 150.1875,
      "epoch": 5.904,
      "grad_norm": 10.011385917663574,
      "kl": 0.255615234375,
      "learning_rate": 4.096e-07,
      "loss": 0.0003,
      "reward": 0.96875,
      "reward_std": 0.0883883461356163,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.96875,
      "step": 1476
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 149.15625,
      "epoch": 5.908,
      "grad_norm": 2.341022491455078,
      "kl": 0.220703125,
      "learning_rate": 4.092e-07,
      "loss": 0.0002,
      "reward": 0.71875,
      "reward_std": 0.0883883461356163,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.71875,
      "step": 1477
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 140.28125,
      "epoch": 5.912,
      "grad_norm": 0.045798491686582565,
      "kl": 0.322021484375,
      "learning_rate": 4.0879999999999995e-07,
      "loss": 0.0003,
      "reward": 0.25,
      "reward_std": 0.0,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.25,
      "step": 1478
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 142.90625,
      "epoch": 5.916,
      "grad_norm": 14.667131423950195,
      "kl": 0.27978515625,
      "learning_rate": 4.084e-07,
      "loss": 0.0003,
      "reward": 0.28125,
      "reward_std": 0.0883883461356163,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.28125,
      "step": 1479
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 149.09375,
      "epoch": 5.92,
      "grad_norm": 0.01748691126704216,
      "kl": 0.226318359375,
      "learning_rate": 4.0799999999999995e-07,
      "loss": 0.0002,
      "reward": 0.75,
      "reward_std": 0.0,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.75,
      "step": 1480
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 131.59375,
      "epoch": 5.924,
      "grad_norm": 7.990422248840332,
      "kl": 0.258544921875,
      "learning_rate": 4.076e-07,
      "loss": 0.0003,
      "reward": 0.21875,
      "reward_std": 0.0883883461356163,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.21875,
      "step": 1481
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 134.5,
      "epoch": 5.928,
      "grad_norm": 5.347377777099609,
      "kl": 0.278076171875,
      "learning_rate": 4.072e-07,
      "loss": 0.0003,
      "reward": 0.625,
      "reward_std": 0.13363061845302582,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.625,
      "step": 1482
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 141.0625,
      "epoch": 5.932,
      "grad_norm": 0.017506249248981476,
      "kl": 0.28515625,
      "learning_rate": 4.0679999999999996e-07,
      "loss": 0.0003,
      "reward": 0.5,
      "reward_std": 0.0,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.5,
      "step": 1483
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 145.65625,
      "epoch": 5.936,
      "grad_norm": 0.012642715126276016,
      "kl": 0.25537109375,
      "learning_rate": 4.064e-07,
      "loss": 0.0003,
      "reward": 0.5,
      "reward_std": 0.0,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.5,
      "step": 1484
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 149.0625,
      "epoch": 5.9399999999999995,
      "grad_norm": 6.140988349914551,
      "kl": 0.316162109375,
      "learning_rate": 4.06e-07,
      "loss": 0.0003,
      "reward": 0.09375,
      "reward_std": 0.1293872892856598,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.09375,
      "step": 1485
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 158.6875,
      "epoch": 5.944,
      "grad_norm": 4.951878547668457,
      "kl": 0.229736328125,
      "learning_rate": 4.056e-07,
      "loss": 0.0002,
      "reward": 0.40625,
      "reward_std": 0.2041158601641655,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.40625,
      "step": 1486
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 132.625,
      "epoch": 5.948,
      "grad_norm": 51.99997329711914,
      "kl": 0.2744140625,
      "learning_rate": 4.052e-07,
      "loss": 0.0003,
      "reward": 0.625,
      "reward_std": 0.2177756354212761,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.625,
      "step": 1487
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 138.90625,
      "epoch": 5.952,
      "grad_norm": 0.027056338265538216,
      "kl": 0.2275390625,
      "learning_rate": 4.0479999999999997e-07,
      "loss": 0.0002,
      "reward": 0.25,
      "reward_std": 0.0,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.25,
      "step": 1488
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 119.65625,
      "epoch": 5.9559999999999995,
      "grad_norm": 4.917178153991699,
      "kl": 0.2939453125,
      "learning_rate": 4.0439999999999994e-07,
      "loss": 0.0003,
      "reward": 0.40625,
      "reward_std": 0.2041158601641655,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.40625,
      "step": 1489
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 155.46875,
      "epoch": 5.96,
      "grad_norm": 0.08113057166337967,
      "kl": 0.222412109375,
      "learning_rate": 4.04e-07,
      "loss": 0.0002,
      "reward": 0.5,
      "reward_std": 0.0,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.5,
      "step": 1490
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 155.03125,
      "epoch": 5.964,
      "grad_norm": 2.7024013996124268,
      "kl": 0.235595703125,
      "learning_rate": 4.036e-07,
      "loss": 0.0002,
      "reward": 0.78125,
      "reward_std": 0.0883883461356163,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.78125,
      "step": 1491
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 150.71875,
      "epoch": 5.968,
      "grad_norm": 0.015600664541125298,
      "kl": 0.216796875,
      "learning_rate": 4.032e-07,
      "loss": 0.0002,
      "reward": 0.75,
      "reward_std": 0.0,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.75,
      "step": 1492
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 147.0625,
      "epoch": 5.9719999999999995,
      "grad_norm": 0.010895362123847008,
      "kl": 0.226806640625,
      "learning_rate": 4.028e-07,
      "loss": 0.0002,
      "reward": 0.75,
      "reward_std": 0.0,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.75,
      "step": 1493
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 160.15625,
      "epoch": 5.976,
      "grad_norm": 0.02662215754389763,
      "kl": 0.23681640625,
      "learning_rate": 4.0239999999999995e-07,
      "loss": 0.0002,
      "reward": 0.75,
      "reward_std": 0.0,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.75,
      "step": 1494
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 123.34375,
      "epoch": 5.98,
      "grad_norm": 0.023051710799336433,
      "kl": 0.255615234375,
      "learning_rate": 4.02e-07,
      "loss": 0.0003,
      "reward": 0.75,
      "reward_std": 0.0,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.75,
      "step": 1495
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 169.4375,
      "epoch": 5.984,
      "grad_norm": 4.173442840576172,
      "kl": 0.269287109375,
      "learning_rate": 4.016e-07,
      "loss": 0.0003,
      "reward": 0.4375,
      "reward_std": 0.1157275140285492,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.4375,
      "step": 1496
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 146.8125,
      "epoch": 5.9879999999999995,
      "grad_norm": 0.021261505782604218,
      "kl": 0.238525390625,
      "learning_rate": 4.0119999999999997e-07,
      "loss": 0.0002,
      "reward": 0.5,
      "reward_std": 0.0,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.5,
      "step": 1497
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 140.0,
      "epoch": 5.992,
      "grad_norm": 0.018028948456048965,
      "kl": 0.2529296875,
      "learning_rate": 4.008e-07,
      "loss": 0.0003,
      "reward": 0.5,
      "reward_std": 0.0,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.5,
      "step": 1498
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 138.3125,
      "epoch": 5.996,
      "grad_norm": 0.010460511781275272,
      "kl": 0.2529296875,
      "learning_rate": 4.0039999999999996e-07,
      "loss": 0.0003,
      "reward": 0.75,
      "reward_std": 0.0,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.75,
      "step": 1499
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 139.21875,
      "epoch": 6.0,
      "grad_norm": 0.012725908309221268,
      "kl": 0.22314453125,
      "learning_rate": 4e-07,
      "loss": 0.0002,
      "reward": 0.5,
      "reward_std": 0.0,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.5,
      "step": 1500
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 161.34375,
      "epoch": 6.004,
      "grad_norm": 0.09431967884302139,
      "kl": 0.2158203125,
      "learning_rate": 3.996e-07,
      "loss": 0.0002,
      "reward": 0.75,
      "reward_std": 0.0,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.75,
      "step": 1501
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 143.28125,
      "epoch": 6.008,
      "grad_norm": 0.09922438114881516,
      "kl": 0.245849609375,
      "learning_rate": 3.992e-07,
      "loss": 0.0002,
      "reward": 0.5,
      "reward_std": 0.0,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.5,
      "step": 1502
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 148.21875,
      "epoch": 6.012,
      "grad_norm": 0.025248998776078224,
      "kl": 0.227294921875,
      "learning_rate": 3.9879999999999994e-07,
      "loss": 0.0002,
      "reward": 0.5,
      "reward_std": 0.0,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.5,
      "step": 1503
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 142.65625,
      "epoch": 6.016,
      "grad_norm": 0.012519005686044693,
      "kl": 0.24560546875,
      "learning_rate": 3.9839999999999997e-07,
      "loss": 0.0002,
      "reward": 0.75,
      "reward_std": 0.0,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.75,
      "step": 1504
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 144.5,
      "epoch": 6.02,
      "grad_norm": 0.010999331250786781,
      "kl": 0.2431640625,
      "learning_rate": 3.98e-07,
      "loss": 0.0002,
      "reward": 0.75,
      "reward_std": 0.0,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.75,
      "step": 1505
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 163.125,
      "epoch": 6.024,
      "grad_norm": 3.0499050617218018,
      "kl": 0.22412109375,
      "learning_rate": 3.976e-07,
      "loss": 0.0002,
      "reward": 0.40625,
      "reward_std": 0.1293872892856598,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.40625,
      "step": 1506
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 129.84375,
      "epoch": 6.028,
      "grad_norm": 0.03026980720460415,
      "kl": 0.294921875,
      "learning_rate": 3.972e-07,
      "loss": 0.0003,
      "reward": 0.25,
      "reward_std": 0.0,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.25,
      "step": 1507
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 156.8125,
      "epoch": 6.032,
      "grad_norm": 0.018876824527978897,
      "kl": 0.200439453125,
      "learning_rate": 3.9679999999999995e-07,
      "loss": 0.0002,
      "reward": 0.75,
      "reward_std": 0.0,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.75,
      "step": 1508
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 138.34375,
      "epoch": 6.036,
      "grad_norm": 0.026277733966708183,
      "kl": 0.263671875,
      "learning_rate": 3.964e-07,
      "loss": 0.0003,
      "reward": 0.75,
      "reward_std": 0.0,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.75,
      "step": 1509
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 130.5,
      "epoch": 6.04,
      "grad_norm": 0.01999865472316742,
      "kl": 0.25830078125,
      "learning_rate": 3.96e-07,
      "loss": 0.0003,
      "reward": 0.75,
      "reward_std": 0.0,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.75,
      "step": 1510
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 144.65625,
      "epoch": 6.044,
      "grad_norm": 4.469531536102295,
      "kl": 0.221923828125,
      "learning_rate": 3.9559999999999997e-07,
      "loss": 0.0002,
      "reward": 0.34375,
      "reward_std": 0.1293872892856598,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.34375,
      "step": 1511
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 117.84375,
      "epoch": 6.048,
      "grad_norm": 0.01932692714035511,
      "kl": 0.25537109375,
      "learning_rate": 3.952e-07,
      "loss": 0.0003,
      "reward": 0.5,
      "reward_std": 0.0,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.5,
      "step": 1512
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 133.375,
      "epoch": 6.052,
      "grad_norm": 0.010463076643645763,
      "kl": 0.2373046875,
      "learning_rate": 3.9479999999999996e-07,
      "loss": 0.0002,
      "reward": 0.5,
      "reward_std": 0.0,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.5,
      "step": 1513
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 153.40625,
      "epoch": 6.056,
      "grad_norm": 3.0791492462158203,
      "kl": 0.2177734375,
      "learning_rate": 3.9439999999999993e-07,
      "loss": 0.0002,
      "reward": 0.6875,
      "reward_std": 0.1157275140285492,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.6875,
      "step": 1514
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 153.90625,
      "epoch": 6.06,
      "grad_norm": 5.025982856750488,
      "kl": 0.210205078125,
      "learning_rate": 3.94e-07,
      "loss": 0.0002,
      "reward": 0.375,
      "reward_std": 0.13363061845302582,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.375,
      "step": 1515
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 141.53125,
      "epoch": 6.064,
      "grad_norm": 2.0357513427734375,
      "kl": 0.240478515625,
      "learning_rate": 3.936e-07,
      "loss": 0.0002,
      "reward": 0.46875,
      "reward_std": 0.0883883461356163,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.46875,
      "step": 1516
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 148.5625,
      "epoch": 6.068,
      "grad_norm": 0.013776174746453762,
      "kl": 0.25146484375,
      "learning_rate": 3.932e-07,
      "loss": 0.0003,
      "reward": 0.5,
      "reward_std": 0.0,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.5,
      "step": 1517
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 138.15625,
      "epoch": 6.072,
      "grad_norm": 0.017277352511882782,
      "kl": 0.214111328125,
      "learning_rate": 3.9279999999999997e-07,
      "loss": 0.0002,
      "reward": 0.5,
      "reward_std": 0.0,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.5,
      "step": 1518
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 141.71875,
      "epoch": 6.076,
      "grad_norm": 6.314394474029541,
      "kl": 0.234130859375,
      "learning_rate": 3.924e-07,
      "loss": 0.0002,
      "reward": 0.25,
      "reward_std": 0.1767766922712326,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.25,
      "step": 1519
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 148.25,
      "epoch": 6.08,
      "grad_norm": 9.867873191833496,
      "kl": 0.25390625,
      "learning_rate": 3.92e-07,
      "loss": 0.0003,
      "reward": 0.71875,
      "reward_std": 0.0883883461356163,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.71875,
      "step": 1520
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 155.15625,
      "epoch": 6.084,
      "grad_norm": 6.222536563873291,
      "kl": 0.275634765625,
      "learning_rate": 3.916e-07,
      "loss": 0.0003,
      "reward": 0.28125,
      "reward_std": 0.0883883461356163,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.28125,
      "step": 1521
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 134.25,
      "epoch": 6.088,
      "grad_norm": 0.10638325661420822,
      "kl": 0.260986328125,
      "learning_rate": 3.9119999999999996e-07,
      "loss": 0.0003,
      "reward": 0.5,
      "reward_std": 0.0,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.5,
      "step": 1522
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 140.1875,
      "epoch": 6.092,
      "grad_norm": 0.02681051194667816,
      "kl": 0.26611328125,
      "learning_rate": 3.908e-07,
      "loss": 0.0003,
      "reward": 0.75,
      "reward_std": 0.0,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.75,
      "step": 1523
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 135.8125,
      "epoch": 6.096,
      "grad_norm": 2.5863921642303467,
      "kl": 0.2333984375,
      "learning_rate": 3.904e-07,
      "loss": 0.0002,
      "reward": 0.9375,
      "reward_std": 0.1157275140285492,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.9375,
      "step": 1524
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 152.03125,
      "epoch": 6.1,
      "grad_norm": 0.01757073402404785,
      "kl": 0.22607421875,
      "learning_rate": 3.8999999999999997e-07,
      "loss": 0.0002,
      "reward": 0.25,
      "reward_std": 0.0,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.25,
      "step": 1525
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 143.65625,
      "epoch": 6.104,
      "grad_norm": 0.016751131042838097,
      "kl": 0.238037109375,
      "learning_rate": 3.896e-07,
      "loss": 0.0002,
      "reward": 0.75,
      "reward_std": 0.0,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.75,
      "step": 1526
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 131.53125,
      "epoch": 6.108,
      "grad_norm": 0.03731915354728699,
      "kl": 0.291259765625,
      "learning_rate": 3.8919999999999996e-07,
      "loss": 0.0003,
      "reward": 0.25,
      "reward_std": 0.0,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.25,
      "step": 1527
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 141.375,
      "epoch": 6.112,
      "grad_norm": 0.02277294546365738,
      "kl": 0.22998046875,
      "learning_rate": 3.888e-07,
      "loss": 0.0002,
      "reward": 1.0,
      "reward_std": 0.0,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 1.0,
      "step": 1528
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 123.3125,
      "epoch": 6.116,
      "grad_norm": 10.101119995117188,
      "kl": 0.274658203125,
      "learning_rate": 3.884e-07,
      "loss": 0.0003,
      "reward": 0.40625,
      "reward_std": 0.1293872892856598,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.40625,
      "step": 1529
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 136.21875,
      "epoch": 6.12,
      "grad_norm": 5.7409868240356445,
      "kl": 0.24609375,
      "learning_rate": 3.88e-07,
      "loss": 0.0002,
      "reward": 0.40625,
      "reward_std": 0.1293872892856598,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.40625,
      "step": 1530
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 137.84375,
      "epoch": 6.124,
      "grad_norm": 4.028714179992676,
      "kl": 0.264404296875,
      "learning_rate": 3.876e-07,
      "loss": 0.0003,
      "reward": 0.59375,
      "reward_std": 0.1293872892856598,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.59375,
      "step": 1531
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 136.25,
      "epoch": 6.128,
      "grad_norm": 3.5504932403564453,
      "kl": 0.246826171875,
      "learning_rate": 3.8719999999999997e-07,
      "loss": 0.0002,
      "reward": 0.4375,
      "reward_std": 0.1157275140285492,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.4375,
      "step": 1532
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 138.59375,
      "epoch": 6.132,
      "grad_norm": 0.019457193091511726,
      "kl": 0.2373046875,
      "learning_rate": 3.8679999999999994e-07,
      "loss": 0.0002,
      "reward": 0.25,
      "reward_std": 0.0,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.25,
      "step": 1533
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 135.34375,
      "epoch": 6.136,
      "grad_norm": 0.03133631870150566,
      "kl": 0.244140625,
      "learning_rate": 3.864e-07,
      "loss": 0.0002,
      "reward": 0.25,
      "reward_std": 0.0,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.25,
      "step": 1534
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 110.90625,
      "epoch": 6.14,
      "grad_norm": 6.78147029876709,
      "kl": 0.267822265625,
      "learning_rate": 3.86e-07,
      "loss": 0.0003,
      "reward": 0.6875,
      "reward_std": 0.249358132481575,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.6875,
      "step": 1535
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 136.9375,
      "epoch": 6.144,
      "grad_norm": 0.016054855659604073,
      "kl": 0.21728515625,
      "learning_rate": 3.8559999999999996e-07,
      "loss": 0.0002,
      "reward": 1.0,
      "reward_std": 0.0,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 1.0,
      "step": 1536
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 112.90625,
      "epoch": 6.148,
      "grad_norm": 0.02238682471215725,
      "kl": 0.294921875,
      "learning_rate": 3.852e-07,
      "loss": 0.0003,
      "reward": 0.5,
      "reward_std": 0.0,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.5,
      "step": 1537
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 103.0,
      "epoch": 6.152,
      "grad_norm": 3.2338173389434814,
      "kl": 0.308349609375,
      "learning_rate": 3.8479999999999995e-07,
      "loss": 0.0003,
      "reward": 0.53125,
      "reward_std": 0.0883883461356163,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.53125,
      "step": 1538
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 132.9375,
      "epoch": 6.156,
      "grad_norm": 0.014229514636099339,
      "kl": 0.23388671875,
      "learning_rate": 3.8440000000000003e-07,
      "loss": 0.0002,
      "reward": 0.5,
      "reward_std": 0.0,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.5,
      "step": 1539
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 106.25,
      "epoch": 6.16,
      "grad_norm": 0.16671757400035858,
      "kl": 0.313720703125,
      "learning_rate": 3.84e-07,
      "loss": 0.0003,
      "reward": 0.5,
      "reward_std": 0.0,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.5,
      "step": 1540
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 120.84375,
      "epoch": 6.164,
      "grad_norm": 0.015546483919024467,
      "kl": 0.256591796875,
      "learning_rate": 3.8359999999999997e-07,
      "loss": 0.0003,
      "reward": 0.25,
      "reward_std": 0.0,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.25,
      "step": 1541
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 110.90625,
      "epoch": 6.168,
      "grad_norm": 11.513829231262207,
      "kl": 0.26171875,
      "learning_rate": 3.832e-07,
      "loss": 0.0003,
      "reward": 0.5625,
      "reward_std": 0.1767766922712326,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.5625,
      "step": 1542
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 134.40625,
      "epoch": 6.172,
      "grad_norm": 0.021982690319418907,
      "kl": 0.2275390625,
      "learning_rate": 3.8279999999999996e-07,
      "loss": 0.0002,
      "reward": 0.25,
      "reward_std": 0.0,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.25,
      "step": 1543
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 121.84375,
      "epoch": 6.176,
      "grad_norm": 0.030291661620140076,
      "kl": 0.324951171875,
      "learning_rate": 3.824e-07,
      "loss": 0.0003,
      "reward": 0.5,
      "reward_std": 0.0,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.5,
      "step": 1544
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 126.46875,
      "epoch": 6.18,
      "grad_norm": 4.345338344573975,
      "kl": 0.244140625,
      "learning_rate": 3.82e-07,
      "loss": 0.0002,
      "reward": 0.28125,
      "reward_std": 0.0883883461356163,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.28125,
      "step": 1545
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 115.96875,
      "epoch": 6.184,
      "grad_norm": 8.077239990234375,
      "kl": 0.251708984375,
      "learning_rate": 3.816e-07,
      "loss": 0.0003,
      "reward": 0.65625,
      "reward_std": 0.22201896458864212,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.65625,
      "step": 1546
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 121.90625,
      "epoch": 6.188,
      "grad_norm": 0.025769701227545738,
      "kl": 0.257080078125,
      "learning_rate": 3.8119999999999995e-07,
      "loss": 0.0003,
      "reward": 0.5,
      "reward_std": 0.0,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.5,
      "step": 1547
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 116.03125,
      "epoch": 6.192,
      "grad_norm": 0.011613396927714348,
      "kl": 0.28662109375,
      "learning_rate": 3.808e-07,
      "loss": 0.0003,
      "reward": 0.5,
      "reward_std": 0.0,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.5,
      "step": 1548
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 115.0625,
      "epoch": 6.196,
      "grad_norm": 0.18783621490001678,
      "kl": 0.29833984375,
      "learning_rate": 3.804e-07,
      "loss": 0.0003,
      "reward": 0.25,
      "reward_std": 0.0,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.25,
      "step": 1549
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 110.46875,
      "epoch": 6.2,
      "grad_norm": 0.012086768634617329,
      "kl": 0.288818359375,
      "learning_rate": 3.7999999999999996e-07,
      "loss": 0.0003,
      "reward": 0.75,
      "reward_std": 0.0,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.75,
      "step": 1550
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 118.21875,
      "epoch": 6.204,
      "grad_norm": 0.012049547396600246,
      "kl": 0.280029296875,
      "learning_rate": 3.796e-07,
      "loss": 0.0003,
      "reward": 0.5,
      "reward_std": 0.0,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.5,
      "step": 1551
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 115.59375,
      "epoch": 6.208,
      "grad_norm": 6.670135974884033,
      "kl": 0.30078125,
      "learning_rate": 3.7919999999999995e-07,
      "loss": 0.0003,
      "reward": 0.1875,
      "reward_std": 0.1157275140285492,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.1875,
      "step": 1552
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 111.625,
      "epoch": 6.212,
      "grad_norm": 5.236369609832764,
      "kl": 0.262451171875,
      "learning_rate": 3.7880000000000003e-07,
      "loss": 0.0003,
      "reward": 0.625,
      "reward_std": 0.13363061845302582,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.625,
      "step": 1553
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 131.75,
      "epoch": 6.216,
      "grad_norm": 0.034675970673561096,
      "kl": 0.2919921875,
      "learning_rate": 3.784e-07,
      "loss": 0.0003,
      "reward": 0.75,
      "reward_std": 0.0,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.75,
      "step": 1554
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 131.90625,
      "epoch": 6.22,
      "grad_norm": 0.01678628660738468,
      "kl": 0.23974609375,
      "learning_rate": 3.7799999999999997e-07,
      "loss": 0.0002,
      "reward": 0.75,
      "reward_std": 0.0,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.75,
      "step": 1555
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 111.46875,
      "epoch": 6.224,
      "grad_norm": 0.021899383515119553,
      "kl": 0.26904296875,
      "learning_rate": 3.776e-07,
      "loss": 0.0003,
      "reward": 0.25,
      "reward_std": 0.0,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.25,
      "step": 1556
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 121.25,
      "epoch": 6.228,
      "grad_norm": 0.01460999995470047,
      "kl": 0.270751953125,
      "learning_rate": 3.7719999999999996e-07,
      "loss": 0.0003,
      "reward": 0.5,
      "reward_std": 0.0,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.5,
      "step": 1557
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 120.15625,
      "epoch": 6.232,
      "grad_norm": 0.021711671724915504,
      "kl": 0.23974609375,
      "learning_rate": 3.768e-07,
      "loss": 0.0002,
      "reward": 1.0,
      "reward_std": 0.0,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 1.0,
      "step": 1558
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 101.75,
      "epoch": 6.236,
      "grad_norm": 0.01255778782069683,
      "kl": 0.31396484375,
      "learning_rate": 3.764e-07,
      "loss": 0.0003,
      "reward": 0.75,
      "reward_std": 0.0,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.75,
      "step": 1559
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 118.53125,
      "epoch": 6.24,
      "grad_norm": 0.20877426862716675,
      "kl": 0.248046875,
      "learning_rate": 3.76e-07,
      "loss": 0.0002,
      "reward": 0.25,
      "reward_std": 0.0,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.25,
      "step": 1560
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 130.875,
      "epoch": 6.244,
      "grad_norm": 0.012842838652431965,
      "kl": 0.272216796875,
      "learning_rate": 3.7559999999999995e-07,
      "loss": 0.0003,
      "reward": 0.25,
      "reward_std": 0.0,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.25,
      "step": 1561
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 128.71875,
      "epoch": 6.248,
      "grad_norm": 4.114800930023193,
      "kl": 0.238525390625,
      "learning_rate": 3.7519999999999997e-07,
      "loss": 0.0002,
      "reward": 0.71875,
      "reward_std": 0.0883883461356163,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.71875,
      "step": 1562
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 114.3125,
      "epoch": 6.252,
      "grad_norm": 0.0288203377276659,
      "kl": 0.25732421875,
      "learning_rate": 3.748e-07,
      "loss": 0.0003,
      "reward": 0.75,
      "reward_std": 0.0,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.75,
      "step": 1563
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 113.65625,
      "epoch": 6.256,
      "grad_norm": 0.014990667812526226,
      "kl": 0.278564453125,
      "learning_rate": 3.744e-07,
      "loss": 0.0003,
      "reward": 0.5,
      "reward_std": 0.0,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.5,
      "step": 1564
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 126.625,
      "epoch": 6.26,
      "grad_norm": 0.0191346425563097,
      "kl": 0.267578125,
      "learning_rate": 3.74e-07,
      "loss": 0.0003,
      "reward": 0.5,
      "reward_std": 0.0,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.5,
      "step": 1565
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 125.625,
      "epoch": 6.264,
      "grad_norm": 0.021278565749526024,
      "kl": 0.30029296875,
      "learning_rate": 3.7359999999999996e-07,
      "loss": 0.0003,
      "reward": 0.75,
      "reward_std": 0.0,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.75,
      "step": 1566
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 133.25,
      "epoch": 6.268,
      "grad_norm": 0.009844094514846802,
      "kl": 0.2421875,
      "learning_rate": 3.732e-07,
      "loss": 0.0002,
      "reward": 0.75,
      "reward_std": 0.0,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.75,
      "step": 1567
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 117.9375,
      "epoch": 6.272,
      "grad_norm": 0.010759477503597736,
      "kl": 0.26123046875,
      "learning_rate": 3.728e-07,
      "loss": 0.0003,
      "reward": 0.25,
      "reward_std": 0.0,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.25,
      "step": 1568
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 117.96875,
      "epoch": 6.276,
      "grad_norm": 0.014643261209130287,
      "kl": 0.25634765625,
      "learning_rate": 3.7239999999999997e-07,
      "loss": 0.0003,
      "reward": 1.0,
      "reward_std": 0.0,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 1.0,
      "step": 1569
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 93.34375,
      "epoch": 6.28,
      "grad_norm": 0.016045663505792618,
      "kl": 0.31591796875,
      "learning_rate": 3.72e-07,
      "loss": 0.0003,
      "reward": 0.75,
      "reward_std": 0.0,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.75,
      "step": 1570
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 137.9375,
      "epoch": 6.284,
      "grad_norm": 0.015440396964550018,
      "kl": 0.2529296875,
      "learning_rate": 3.7159999999999997e-07,
      "loss": 0.0003,
      "reward": 0.75,
      "reward_std": 0.0,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.75,
      "step": 1571
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 114.5625,
      "epoch": 6.288,
      "grad_norm": 0.010787459090352058,
      "kl": 0.243896484375,
      "learning_rate": 3.7119999999999994e-07,
      "loss": 0.0002,
      "reward": 0.25,
      "reward_std": 0.0,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.25,
      "step": 1572
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 128.59375,
      "epoch": 6.292,
      "grad_norm": 0.029563790187239647,
      "kl": 0.235107421875,
      "learning_rate": 3.708e-07,
      "loss": 0.0002,
      "reward": 0.25,
      "reward_std": 0.0,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.25,
      "step": 1573
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 115.5,
      "epoch": 6.296,
      "grad_norm": 0.01302887313067913,
      "kl": 0.241455078125,
      "learning_rate": 3.704e-07,
      "loss": 0.0002,
      "reward": 0.0,
      "reward_std": 0.0,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.0,
      "step": 1574
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 102.6875,
      "epoch": 6.3,
      "grad_norm": 10.8460693359375,
      "kl": 0.2958984375,
      "learning_rate": 3.7e-07,
      "loss": 0.0003,
      "reward": 0.5625,
      "reward_std": 0.1157275140285492,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.5625,
      "step": 1575
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 113.34375,
      "epoch": 6.304,
      "grad_norm": 0.01626221463084221,
      "kl": 0.279296875,
      "learning_rate": 3.696e-07,
      "loss": 0.0003,
      "reward": 0.25,
      "reward_std": 0.0,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.25,
      "step": 1576
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 119.5,
      "epoch": 6.308,
      "grad_norm": 0.014014032669365406,
      "kl": 0.23828125,
      "learning_rate": 3.6919999999999994e-07,
      "loss": 0.0002,
      "reward": 0.5,
      "reward_std": 0.0,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.5,
      "step": 1577
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 86.21875,
      "epoch": 6.312,
      "grad_norm": 0.045053623616695404,
      "kl": 0.31689453125,
      "learning_rate": 3.688e-07,
      "loss": 0.0003,
      "reward": 0.75,
      "reward_std": 0.0,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.75,
      "step": 1578
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 119.90625,
      "epoch": 6.316,
      "grad_norm": 12.00381088256836,
      "kl": 0.290283203125,
      "learning_rate": 3.684e-07,
      "loss": 0.0003,
      "reward": 0.59375,
      "reward_std": 0.1293872892856598,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.59375,
      "step": 1579
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 108.875,
      "epoch": 6.32,
      "grad_norm": 0.03300373628735542,
      "kl": 0.302001953125,
      "learning_rate": 3.6799999999999996e-07,
      "loss": 0.0003,
      "reward": 0.25,
      "reward_std": 0.0,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.25,
      "step": 1580
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 128.28125,
      "epoch": 6.324,
      "grad_norm": 12.245332717895508,
      "kl": 0.2373046875,
      "learning_rate": 3.676e-07,
      "loss": 0.0002,
      "reward": 0.53125,
      "reward_std": 0.0883883461356163,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.53125,
      "step": 1581
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 121.5625,
      "epoch": 6.328,
      "grad_norm": 8.192108154296875,
      "kl": 0.264892578125,
      "learning_rate": 3.672e-07,
      "loss": 0.0003,
      "reward": 0.5625,
      "reward_std": 0.1157275140285492,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.5625,
      "step": 1582
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 118.15625,
      "epoch": 6.332,
      "grad_norm": 0.011171386577188969,
      "kl": 0.255126953125,
      "learning_rate": 3.668e-07,
      "loss": 0.0003,
      "reward": 0.25,
      "reward_std": 0.0,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.25,
      "step": 1583
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 125.1875,
      "epoch": 6.336,
      "grad_norm": 5.237426280975342,
      "kl": 0.248779296875,
      "learning_rate": 3.664e-07,
      "loss": 0.0002,
      "reward": 0.3125,
      "reward_std": 0.1157275140285492,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.3125,
      "step": 1584
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 115.46875,
      "epoch": 6.34,
      "grad_norm": 3.1701931953430176,
      "kl": 0.283203125,
      "learning_rate": 3.6599999999999997e-07,
      "loss": 0.0003,
      "reward": 0.28125,
      "reward_std": 0.0883883461356163,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.28125,
      "step": 1585
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 121.4375,
      "epoch": 6.344,
      "grad_norm": 0.03177843987941742,
      "kl": 0.256103515625,
      "learning_rate": 3.6559999999999994e-07,
      "loss": 0.0003,
      "reward": 1.0,
      "reward_std": 0.0,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 1.0,
      "step": 1586
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 130.875,
      "epoch": 6.348,
      "grad_norm": 0.030346285551786423,
      "kl": 0.24853515625,
      "learning_rate": 3.652e-07,
      "loss": 0.0002,
      "reward": 0.75,
      "reward_std": 0.0,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.75,
      "step": 1587
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 98.8125,
      "epoch": 6.352,
      "grad_norm": 0.023983564227819443,
      "kl": 0.283203125,
      "learning_rate": 3.648e-07,
      "loss": 0.0003,
      "reward": 0.5,
      "reward_std": 0.0,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.5,
      "step": 1588
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 119.46875,
      "epoch": 6.356,
      "grad_norm": 0.01813400909304619,
      "kl": 0.253662109375,
      "learning_rate": 3.644e-07,
      "loss": 0.0003,
      "reward": 0.5,
      "reward_std": 0.0,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.5,
      "step": 1589
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 112.03125,
      "epoch": 6.36,
      "grad_norm": 0.041276965290308,
      "kl": 0.247802734375,
      "learning_rate": 3.64e-07,
      "loss": 0.0002,
      "reward": 0.5,
      "reward_std": 0.0,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.5,
      "step": 1590
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 108.75,
      "epoch": 6.364,
      "grad_norm": 3.3416824340820312,
      "kl": 0.260986328125,
      "learning_rate": 3.6359999999999995e-07,
      "loss": 0.0003,
      "reward": 0.3125,
      "reward_std": 0.1157275140285492,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.3125,
      "step": 1591
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 119.625,
      "epoch": 6.368,
      "grad_norm": 0.014476221986114979,
      "kl": 0.245849609375,
      "learning_rate": 3.632e-07,
      "loss": 0.0002,
      "reward": 1.0,
      "reward_std": 0.0,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 1.0,
      "step": 1592
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 115.0625,
      "epoch": 6.372,
      "grad_norm": 6.685598373413086,
      "kl": 0.26220703125,
      "learning_rate": 3.628e-07,
      "loss": 0.0003,
      "reward": 0.46875,
      "reward_std": 0.0883883461356163,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.46875,
      "step": 1593
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 100.0,
      "epoch": 6.376,
      "grad_norm": 7.291464328765869,
      "kl": 0.271240234375,
      "learning_rate": 3.6239999999999996e-07,
      "loss": 0.0003,
      "reward": 0.46875,
      "reward_std": 0.0883883461356163,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.46875,
      "step": 1594
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 121.15625,
      "epoch": 6.38,
      "grad_norm": 0.04222020506858826,
      "kl": 0.2705078125,
      "learning_rate": 3.62e-07,
      "loss": 0.0003,
      "reward": 0.5,
      "reward_std": 0.0,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.5,
      "step": 1595
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 120.78125,
      "epoch": 6.384,
      "grad_norm": 2.879133701324463,
      "kl": 0.232421875,
      "learning_rate": 3.6159999999999996e-07,
      "loss": 0.0002,
      "reward": 0.53125,
      "reward_std": 0.0883883461356163,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.53125,
      "step": 1596
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 111.53125,
      "epoch": 6.388,
      "grad_norm": 0.1174178197979927,
      "kl": 0.27001953125,
      "learning_rate": 3.612e-07,
      "loss": 0.0003,
      "reward": 0.75,
      "reward_std": 0.0,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.75,
      "step": 1597
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 125.84375,
      "epoch": 6.392,
      "grad_norm": 3.6118648052215576,
      "kl": 0.264892578125,
      "learning_rate": 3.608e-07,
      "loss": 0.0003,
      "reward": 0.34375,
      "reward_std": 0.1293872892856598,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.34375,
      "step": 1598
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 108.375,
      "epoch": 6.396,
      "grad_norm": 22.451313018798828,
      "kl": 0.250732421875,
      "learning_rate": 3.6039999999999997e-07,
      "loss": 0.0003,
      "reward": 0.375,
      "reward_std": 0.13363061845302582,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.375,
      "step": 1599
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 137.40625,
      "epoch": 6.4,
      "grad_norm": 0.021385841071605682,
      "kl": 0.23974609375,
      "learning_rate": 3.6e-07,
      "loss": 0.0002,
      "reward": 0.25,
      "reward_std": 0.0,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.25,
      "step": 1600
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 99.96875,
      "epoch": 6.404,
      "grad_norm": 0.16467660665512085,
      "kl": 0.2890625,
      "learning_rate": 3.5959999999999996e-07,
      "loss": 0.0003,
      "reward": 0.75,
      "reward_std": 0.0,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.75,
      "step": 1601
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 133.5,
      "epoch": 6.408,
      "grad_norm": 0.011968264356255531,
      "kl": 0.231689453125,
      "learning_rate": 3.592e-07,
      "loss": 0.0002,
      "reward": 0.5,
      "reward_std": 0.0,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.5,
      "step": 1602
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 109.125,
      "epoch": 6.412,
      "grad_norm": 21.922290802001953,
      "kl": 0.276123046875,
      "learning_rate": 3.588e-07,
      "loss": 0.0003,
      "reward": 0.71875,
      "reward_std": 0.0883883461356163,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.71875,
      "step": 1603
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 128.0,
      "epoch": 6.416,
      "grad_norm": 4.997833251953125,
      "kl": 0.211181640625,
      "learning_rate": 3.584e-07,
      "loss": 0.0002,
      "reward": 0.59375,
      "reward_std": 0.1293872892856598,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.59375,
      "step": 1604
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 118.25,
      "epoch": 6.42,
      "grad_norm": 3.5824813842773438,
      "kl": 0.26025390625,
      "learning_rate": 3.5799999999999995e-07,
      "loss": 0.0003,
      "reward": 0.28125,
      "reward_std": 0.0883883461356163,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.28125,
      "step": 1605
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 139.03125,
      "epoch": 6.424,
      "grad_norm": 7.136850833892822,
      "kl": 0.24560546875,
      "learning_rate": 3.5759999999999997e-07,
      "loss": 0.0002,
      "reward": 0.625,
      "reward_std": 0.13363061845302582,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.625,
      "step": 1606
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 119.4375,
      "epoch": 6.428,
      "grad_norm": 0.04120960459113121,
      "kl": 0.21923828125,
      "learning_rate": 3.572e-07,
      "loss": 0.0002,
      "reward": 0.5,
      "reward_std": 0.0,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.5,
      "step": 1607
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 113.5,
      "epoch": 6.432,
      "grad_norm": 8.48258113861084,
      "kl": 0.250732421875,
      "learning_rate": 3.5679999999999997e-07,
      "loss": 0.0003,
      "reward": 0.53125,
      "reward_std": 0.24511480331420898,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.53125,
      "step": 1608
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 120.15625,
      "epoch": 6.436,
      "grad_norm": 3.9160797595977783,
      "kl": 0.2861328125,
      "learning_rate": 3.564e-07,
      "loss": 0.0003,
      "reward": 0.125,
      "reward_std": 0.13363061845302582,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.125,
      "step": 1609
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 127.09375,
      "epoch": 6.44,
      "grad_norm": 3.2609615325927734,
      "kl": 0.235595703125,
      "learning_rate": 3.5599999999999996e-07,
      "loss": 0.0002,
      "reward": 0.65625,
      "reward_std": 0.1293872892856598,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.65625,
      "step": 1610
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 126.6875,
      "epoch": 6.444,
      "grad_norm": 0.013134521432220936,
      "kl": 0.24755859375,
      "learning_rate": 3.5560000000000003e-07,
      "loss": 0.0002,
      "reward": 0.75,
      "reward_std": 0.0,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.75,
      "step": 1611
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 117.75,
      "epoch": 6.448,
      "grad_norm": 2.9193601608276367,
      "kl": 0.2529296875,
      "learning_rate": 3.552e-07,
      "loss": 0.0003,
      "reward": 0.71875,
      "reward_std": 0.0883883461356163,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.71875,
      "step": 1612
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 113.40625,
      "epoch": 6.452,
      "grad_norm": 0.02556130290031433,
      "kl": 0.259521484375,
      "learning_rate": 3.548e-07,
      "loss": 0.0003,
      "reward": 0.5,
      "reward_std": 0.0,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.5,
      "step": 1613
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 121.0,
      "epoch": 6.456,
      "grad_norm": 0.015901606529951096,
      "kl": 0.239013671875,
      "learning_rate": 3.544e-07,
      "loss": 0.0002,
      "reward": 0.5,
      "reward_std": 0.0,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.5,
      "step": 1614
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 135.46875,
      "epoch": 6.46,
      "grad_norm": 5.322353839874268,
      "kl": 0.2333984375,
      "learning_rate": 3.5399999999999997e-07,
      "loss": 0.0002,
      "reward": 0.09375,
      "reward_std": 0.2041158601641655,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.09375,
      "step": 1615
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 117.5625,
      "epoch": 6.464,
      "grad_norm": 6.865261554718018,
      "kl": 0.30224609375,
      "learning_rate": 3.536e-07,
      "loss": 0.0003,
      "reward": 0.4375,
      "reward_std": 0.1767766922712326,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.4375,
      "step": 1616
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 123.0,
      "epoch": 6.468,
      "grad_norm": 0.015447596088051796,
      "kl": 0.2353515625,
      "learning_rate": 3.532e-07,
      "loss": 0.0002,
      "reward": 0.25,
      "reward_std": 0.0,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.25,
      "step": 1617
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 120.1875,
      "epoch": 6.4719999999999995,
      "grad_norm": 0.010904483497142792,
      "kl": 0.255615234375,
      "learning_rate": 3.528e-07,
      "loss": 0.0003,
      "reward": 1.0,
      "reward_std": 0.0,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 1.0,
      "step": 1618
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 112.90625,
      "epoch": 6.476,
      "grad_norm": 4.7979960441589355,
      "kl": 0.2978515625,
      "learning_rate": 3.5239999999999995e-07,
      "loss": 0.0003,
      "reward": 0.625,
      "reward_std": 0.13363061845302582,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.625,
      "step": 1619
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 121.15625,
      "epoch": 6.48,
      "grad_norm": 0.331534206867218,
      "kl": 0.31591796875,
      "learning_rate": 3.52e-07,
      "loss": 0.0003,
      "reward": 0.75,
      "reward_std": 0.0,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.75,
      "step": 1620
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 143.46875,
      "epoch": 6.484,
      "grad_norm": 0.0417492613196373,
      "kl": 0.22216796875,
      "learning_rate": 3.516e-07,
      "loss": 0.0002,
      "reward": 0.5,
      "reward_std": 0.0,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.5,
      "step": 1621
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 129.90625,
      "epoch": 6.4879999999999995,
      "grad_norm": 0.013283985666930676,
      "kl": 0.25,
      "learning_rate": 3.512e-07,
      "loss": 0.0002,
      "reward": 0.25,
      "reward_std": 0.0,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.25,
      "step": 1622
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 114.875,
      "epoch": 6.492,
      "grad_norm": 6.892928123474121,
      "kl": 0.2724609375,
      "learning_rate": 3.508e-07,
      "loss": 0.0003,
      "reward": 0.3125,
      "reward_std": 0.1767766922712326,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.3125,
      "step": 1623
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 147.46875,
      "epoch": 6.496,
      "grad_norm": 0.011722794733941555,
      "kl": 0.222412109375,
      "learning_rate": 3.5039999999999996e-07,
      "loss": 0.0002,
      "reward": 0.0,
      "reward_std": 0.0,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.0,
      "step": 1624
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 84.03125,
      "epoch": 6.5,
      "grad_norm": 0.16105572879314423,
      "kl": 0.33935546875,
      "learning_rate": 3.5e-07,
      "loss": 0.0003,
      "reward": 0.5,
      "reward_std": 0.0,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.5,
      "step": 1625
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 116.5625,
      "epoch": 6.504,
      "grad_norm": 0.02802308276295662,
      "kl": 0.267333984375,
      "learning_rate": 3.496e-07,
      "loss": 0.0003,
      "reward": 0.25,
      "reward_std": 0.0,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.25,
      "step": 1626
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 120.53125,
      "epoch": 6.508,
      "grad_norm": 0.016088781878352165,
      "kl": 0.2802734375,
      "learning_rate": 3.492e-07,
      "loss": 0.0003,
      "reward": 0.75,
      "reward_std": 0.0,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.75,
      "step": 1627
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 132.3125,
      "epoch": 6.5120000000000005,
      "grad_norm": 0.014105676673352718,
      "kl": 0.261474609375,
      "learning_rate": 3.488e-07,
      "loss": 0.0003,
      "reward": 0.75,
      "reward_std": 0.0,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.75,
      "step": 1628
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 136.40625,
      "epoch": 6.516,
      "grad_norm": 8.322484970092773,
      "kl": 0.26123046875,
      "learning_rate": 3.4839999999999997e-07,
      "loss": 0.0003,
      "reward": 0.96875,
      "reward_std": 0.0883883461356163,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.96875,
      "step": 1629
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 113.03125,
      "epoch": 6.52,
      "grad_norm": 0.028809091076254845,
      "kl": 0.2998046875,
      "learning_rate": 3.4799999999999994e-07,
      "loss": 0.0003,
      "reward": 0.75,
      "reward_std": 0.0,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.75,
      "step": 1630
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 120.875,
      "epoch": 6.524,
      "grad_norm": 5.110529899597168,
      "kl": 0.263916015625,
      "learning_rate": 3.476e-07,
      "loss": 0.0003,
      "reward": 0.46875,
      "reward_std": 0.0883883461356163,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.46875,
      "step": 1631
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 126.125,
      "epoch": 6.5280000000000005,
      "grad_norm": 4.825544834136963,
      "kl": 0.279541015625,
      "learning_rate": 3.472e-07,
      "loss": 0.0003,
      "reward": 0.46875,
      "reward_std": 0.0883883461356163,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.46875,
      "step": 1632
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 131.40625,
      "epoch": 6.532,
      "grad_norm": 0.02405603416264057,
      "kl": 0.252685546875,
      "learning_rate": 3.4679999999999996e-07,
      "loss": 0.0003,
      "reward": 0.5,
      "reward_std": 0.0,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.5,
      "step": 1633
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 121.875,
      "epoch": 6.536,
      "grad_norm": 0.037923168390989304,
      "kl": 0.31005859375,
      "learning_rate": 3.464e-07,
      "loss": 0.0003,
      "reward": 0.5,
      "reward_std": 0.0,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.5,
      "step": 1634
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 123.59375,
      "epoch": 6.54,
      "grad_norm": 0.014340847730636597,
      "kl": 0.255126953125,
      "learning_rate": 3.4599999999999995e-07,
      "loss": 0.0003,
      "reward": 0.75,
      "reward_std": 0.0,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.75,
      "step": 1635
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 114.21875,
      "epoch": 6.5440000000000005,
      "grad_norm": 3.843825578689575,
      "kl": 0.27734375,
      "learning_rate": 3.456e-07,
      "loss": 0.0003,
      "reward": 0.71875,
      "reward_std": 0.0883883461356163,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.71875,
      "step": 1636
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 136.25,
      "epoch": 6.548,
      "grad_norm": 3.9917640686035156,
      "kl": 0.232421875,
      "learning_rate": 3.452e-07,
      "loss": 0.0002,
      "reward": 0.3125,
      "reward_std": 0.1157275140285492,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.3125,
      "step": 1637
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 136.9375,
      "epoch": 6.552,
      "grad_norm": 0.023825135082006454,
      "kl": 0.228759765625,
      "learning_rate": 3.4479999999999996e-07,
      "loss": 0.0002,
      "reward": 1.0,
      "reward_std": 0.0,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 1.0,
      "step": 1638
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 120.84375,
      "epoch": 6.556,
      "grad_norm": 0.01872715726494789,
      "kl": 0.256591796875,
      "learning_rate": 3.444e-07,
      "loss": 0.0003,
      "reward": 0.25,
      "reward_std": 0.0,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.25,
      "step": 1639
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 130.15625,
      "epoch": 6.5600000000000005,
      "grad_norm": 0.021696532145142555,
      "kl": 0.2705078125,
      "learning_rate": 3.4399999999999996e-07,
      "loss": 0.0003,
      "reward": 0.75,
      "reward_std": 0.0,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.75,
      "step": 1640
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 97.59375,
      "epoch": 6.564,
      "grad_norm": 0.015612926334142685,
      "kl": 0.2958984375,
      "learning_rate": 3.436e-07,
      "loss": 0.0003,
      "reward": 0.5,
      "reward_std": 0.0,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.5,
      "step": 1641
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 136.4375,
      "epoch": 6.568,
      "grad_norm": 0.034007567912340164,
      "kl": 0.2470703125,
      "learning_rate": 3.432e-07,
      "loss": 0.0002,
      "reward": 0.75,
      "reward_std": 0.0,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.75,
      "step": 1642
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 117.3125,
      "epoch": 6.572,
      "grad_norm": 0.01821989193558693,
      "kl": 0.249755859375,
      "learning_rate": 3.4279999999999997e-07,
      "loss": 0.0002,
      "reward": 0.25,
      "reward_std": 0.0,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.25,
      "step": 1643
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 117.21875,
      "epoch": 6.576,
      "grad_norm": 4.127259731292725,
      "kl": 0.271728515625,
      "learning_rate": 3.4239999999999994e-07,
      "loss": 0.0003,
      "reward": 0.28125,
      "reward_std": 0.0883883461356163,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.28125,
      "step": 1644
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 131.40625,
      "epoch": 6.58,
      "grad_norm": 3.971961498260498,
      "kl": 0.235107421875,
      "learning_rate": 3.42e-07,
      "loss": 0.0002,
      "reward": 0.4375,
      "reward_std": 0.1157275140285492,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.4375,
      "step": 1645
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 136.71875,
      "epoch": 6.584,
      "grad_norm": 0.012673835270106792,
      "kl": 0.2255859375,
      "learning_rate": 3.416e-07,
      "loss": 0.0002,
      "reward": 0.5,
      "reward_std": 0.0,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.5,
      "step": 1646
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 128.5625,
      "epoch": 6.588,
      "grad_norm": 4.5228400230407715,
      "kl": 0.259521484375,
      "learning_rate": 3.412e-07,
      "loss": 0.0003,
      "reward": 0.375,
      "reward_std": 0.13363061845302582,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.375,
      "step": 1647
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 122.8125,
      "epoch": 6.592,
      "grad_norm": 4.939081192016602,
      "kl": 0.260009765625,
      "learning_rate": 3.408e-07,
      "loss": 0.0003,
      "reward": 0.5,
      "reward_std": 0.1767766922712326,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.5,
      "step": 1648
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 126.0625,
      "epoch": 6.596,
      "grad_norm": 0.026103483512997627,
      "kl": 0.266845703125,
      "learning_rate": 3.4039999999999995e-07,
      "loss": 0.0003,
      "reward": 0.75,
      "reward_std": 0.0,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.75,
      "step": 1649
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 130.375,
      "epoch": 6.6,
      "grad_norm": 0.040323834866285324,
      "kl": 0.262451171875,
      "learning_rate": 3.4000000000000003e-07,
      "loss": 0.0003,
      "reward": 0.25,
      "reward_std": 0.0,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.25,
      "step": 1650
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 133.28125,
      "epoch": 6.604,
      "grad_norm": 4.574068546295166,
      "kl": 0.2646484375,
      "learning_rate": 3.396e-07,
      "loss": 0.0003,
      "reward": 0.90625,
      "reward_std": 0.1293872892856598,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.90625,
      "step": 1651
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 138.84375,
      "epoch": 6.608,
      "grad_norm": 0.01749238930642605,
      "kl": 0.249267578125,
      "learning_rate": 3.3919999999999997e-07,
      "loss": 0.0002,
      "reward": 1.0,
      "reward_std": 0.0,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 1.0,
      "step": 1652
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 121.9375,
      "epoch": 6.612,
      "grad_norm": 7.682275295257568,
      "kl": 0.3369140625,
      "learning_rate": 3.388e-07,
      "loss": 0.0003,
      "reward": 0.96875,
      "reward_std": 0.0883883461356163,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.96875,
      "step": 1653
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 134.90625,
      "epoch": 6.616,
      "grad_norm": 4.5393218994140625,
      "kl": 0.237548828125,
      "learning_rate": 3.3839999999999996e-07,
      "loss": 0.0002,
      "reward": 0.28125,
      "reward_std": 0.0883883461356163,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.28125,
      "step": 1654
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 112.15625,
      "epoch": 6.62,
      "grad_norm": 7.715600490570068,
      "kl": 0.32958984375,
      "learning_rate": 3.38e-07,
      "loss": 0.0003,
      "reward": 0.6875,
      "reward_std": 0.1767766922712326,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.6875,
      "step": 1655
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 138.84375,
      "epoch": 6.624,
      "grad_norm": 0.02325483411550522,
      "kl": 0.258544921875,
      "learning_rate": 3.376e-07,
      "loss": 0.0003,
      "reward": 0.5,
      "reward_std": 0.0,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.5,
      "step": 1656
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 130.15625,
      "epoch": 6.628,
      "grad_norm": 3.471364974975586,
      "kl": 0.265380859375,
      "learning_rate": 3.372e-07,
      "loss": 0.0003,
      "reward": 0.90625,
      "reward_std": 0.1293872892856598,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.90625,
      "step": 1657
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 110.53125,
      "epoch": 6.632,
      "grad_norm": 26.9646053314209,
      "kl": 0.26025390625,
      "learning_rate": 3.368e-07,
      "loss": 0.0003,
      "reward": 0.75,
      "reward_std": 0.2314550280570984,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.75,
      "step": 1658
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 131.0625,
      "epoch": 6.636,
      "grad_norm": 3.3357322216033936,
      "kl": 0.244384765625,
      "learning_rate": 3.3639999999999997e-07,
      "loss": 0.0002,
      "reward": 0.59375,
      "reward_std": 0.1293872892856598,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.59375,
      "step": 1659
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 139.21875,
      "epoch": 6.64,
      "grad_norm": 0.04232488200068474,
      "kl": 0.2978515625,
      "learning_rate": 3.36e-07,
      "loss": 0.0003,
      "reward": 0.5,
      "reward_std": 0.0,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.5,
      "step": 1660
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 121.15625,
      "epoch": 6.644,
      "grad_norm": 9.698933601379395,
      "kl": 0.263427734375,
      "learning_rate": 3.356e-07,
      "loss": 0.0003,
      "reward": 0.53125,
      "reward_std": 0.0883883461356163,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.53125,
      "step": 1661
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 121.15625,
      "epoch": 6.648,
      "grad_norm": 8.163617134094238,
      "kl": 0.249267578125,
      "learning_rate": 3.352e-07,
      "loss": 0.0002,
      "reward": 0.34375,
      "reward_std": 0.1293872892856598,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.34375,
      "step": 1662
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 116.8125,
      "epoch": 6.652,
      "grad_norm": 6.082509994506836,
      "kl": 0.23974609375,
      "learning_rate": 3.3479999999999995e-07,
      "loss": 0.0002,
      "reward": 0.34375,
      "reward_std": 0.1293872892856598,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.34375,
      "step": 1663
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 125.25,
      "epoch": 6.656,
      "grad_norm": 5.705850601196289,
      "kl": 0.25341796875,
      "learning_rate": 3.344e-07,
      "loss": 0.0003,
      "reward": 0.71875,
      "reward_std": 0.0883883461356163,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.71875,
      "step": 1664
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 139.25,
      "epoch": 6.66,
      "grad_norm": 9.513586044311523,
      "kl": 0.22900390625,
      "learning_rate": 3.34e-07,
      "loss": 0.0002,
      "reward": 0.375,
      "reward_std": 0.13363061845302582,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.375,
      "step": 1665
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 140.875,
      "epoch": 6.664,
      "grad_norm": 4.371448516845703,
      "kl": 0.253662109375,
      "learning_rate": 3.3359999999999997e-07,
      "loss": 0.0003,
      "reward": 0.28125,
      "reward_std": 0.0883883461356163,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.28125,
      "step": 1666
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 114.3125,
      "epoch": 6.668,
      "grad_norm": 5.456398010253906,
      "kl": 0.27392578125,
      "learning_rate": 3.332e-07,
      "loss": 0.0003,
      "reward": 0.6875,
      "reward_std": 0.1157275140285492,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.6875,
      "step": 1667
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 131.3125,
      "epoch": 6.672,
      "grad_norm": 0.018667636439204216,
      "kl": 0.23046875,
      "learning_rate": 3.3279999999999996e-07,
      "loss": 0.0002,
      "reward": 1.0,
      "reward_std": 0.0,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 1.0,
      "step": 1668
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 129.6875,
      "epoch": 6.676,
      "grad_norm": 8.642349243164062,
      "kl": 0.264892578125,
      "learning_rate": 3.3239999999999993e-07,
      "loss": 0.0003,
      "reward": 0.53125,
      "reward_std": 0.0883883461356163,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.53125,
      "step": 1669
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 108.4375,
      "epoch": 6.68,
      "grad_norm": 3.1424806118011475,
      "kl": 0.2666015625,
      "learning_rate": 3.32e-07,
      "loss": 0.0003,
      "reward": 0.71875,
      "reward_std": 0.0883883461356163,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.71875,
      "step": 1670
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 120.5,
      "epoch": 6.684,
      "grad_norm": 0.010927255265414715,
      "kl": 0.232666015625,
      "learning_rate": 3.316e-07,
      "loss": 0.0002,
      "reward": 1.0,
      "reward_std": 0.0,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 1.0,
      "step": 1671
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 117.59375,
      "epoch": 6.688,
      "grad_norm": 0.015416597947478294,
      "kl": 0.25927734375,
      "learning_rate": 3.312e-07,
      "loss": 0.0003,
      "reward": 0.5,
      "reward_std": 0.0,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.5,
      "step": 1672
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 120.59375,
      "epoch": 6.692,
      "grad_norm": 0.046456098556518555,
      "kl": 0.2802734375,
      "learning_rate": 3.3079999999999997e-07,
      "loss": 0.0003,
      "reward": 0.25,
      "reward_std": 0.0,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.25,
      "step": 1673
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 130.375,
      "epoch": 6.696,
      "grad_norm": 0.043725769966840744,
      "kl": 0.2783203125,
      "learning_rate": 3.304e-07,
      "loss": 0.0003,
      "reward": 0.5,
      "reward_std": 0.0,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.5,
      "step": 1674
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 132.8125,
      "epoch": 6.7,
      "grad_norm": 4.785642623901367,
      "kl": 0.230224609375,
      "learning_rate": 3.3e-07,
      "loss": 0.0002,
      "reward": 0.34375,
      "reward_std": 0.1293872892856598,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.34375,
      "step": 1675
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 135.125,
      "epoch": 6.704,
      "grad_norm": 3.8570468425750732,
      "kl": 0.211181640625,
      "learning_rate": 3.296e-07,
      "loss": 0.0002,
      "reward": 0.59375,
      "reward_std": 0.1293872892856598,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.59375,
      "step": 1676
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 141.90625,
      "epoch": 6.708,
      "grad_norm": 13.3823881149292,
      "kl": 0.226806640625,
      "learning_rate": 3.2919999999999996e-07,
      "loss": 0.0002,
      "reward": 0.4375,
      "reward_std": 0.1157275140285492,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.4375,
      "step": 1677
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 145.59375,
      "epoch": 6.712,
      "grad_norm": 0.02153557725250721,
      "kl": 0.2646484375,
      "learning_rate": 3.288e-07,
      "loss": 0.0003,
      "reward": 0.75,
      "reward_std": 0.0,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.75,
      "step": 1678
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 140.15625,
      "epoch": 6.716,
      "grad_norm": 0.011307768523693085,
      "kl": 0.23681640625,
      "learning_rate": 3.284e-07,
      "loss": 0.0002,
      "reward": 0.25,
      "reward_std": 0.0,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.25,
      "step": 1679
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 138.875,
      "epoch": 6.72,
      "grad_norm": 0.10024478286504745,
      "kl": 0.252685546875,
      "learning_rate": 3.28e-07,
      "loss": 0.0003,
      "reward": 0.5,
      "reward_std": 0.0,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.5,
      "step": 1680
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 146.5625,
      "epoch": 6.724,
      "grad_norm": 3.667468786239624,
      "kl": 0.23974609375,
      "learning_rate": 3.276e-07,
      "loss": 0.0002,
      "reward": 0.375,
      "reward_std": 0.13363061845302582,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.375,
      "step": 1681
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 123.4375,
      "epoch": 6.728,
      "grad_norm": 3.920100450515747,
      "kl": 0.239013671875,
      "learning_rate": 3.2719999999999997e-07,
      "loss": 0.0002,
      "reward": 0.71875,
      "reward_std": 0.0883883461356163,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.71875,
      "step": 1682
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 144.4375,
      "epoch": 6.732,
      "grad_norm": 0.03040585294365883,
      "kl": 0.228515625,
      "learning_rate": 3.268e-07,
      "loss": 0.0002,
      "reward": 0.5,
      "reward_std": 0.0,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.5,
      "step": 1683
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 150.6875,
      "epoch": 6.736,
      "grad_norm": 0.012111141346395016,
      "kl": 0.230224609375,
      "learning_rate": 3.264e-07,
      "loss": 0.0002,
      "reward": 0.0,
      "reward_std": 0.0,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.0,
      "step": 1684
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 134.84375,
      "epoch": 6.74,
      "grad_norm": 0.009474484249949455,
      "kl": 0.248779296875,
      "learning_rate": 3.26e-07,
      "loss": 0.0002,
      "reward": 0.5,
      "reward_std": 0.0,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.5,
      "step": 1685
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 143.6875,
      "epoch": 6.744,
      "grad_norm": 0.021335192024707794,
      "kl": 0.22021484375,
      "learning_rate": 3.256e-07,
      "loss": 0.0002,
      "reward": 0.75,
      "reward_std": 0.0,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.75,
      "step": 1686
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 152.75,
      "epoch": 6.748,
      "grad_norm": 0.010034766048192978,
      "kl": 0.19580078125,
      "learning_rate": 3.252e-07,
      "loss": 0.0002,
      "reward": 0.0,
      "reward_std": 0.0,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.0,
      "step": 1687
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 160.15625,
      "epoch": 6.752,
      "grad_norm": 0.013335257768630981,
      "kl": 0.203857421875,
      "learning_rate": 3.2479999999999994e-07,
      "loss": 0.0002,
      "reward": 0.5,
      "reward_std": 0.0,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.5,
      "step": 1688
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 142.40625,
      "epoch": 6.756,
      "grad_norm": 7.195217132568359,
      "kl": 0.22705078125,
      "learning_rate": 3.244e-07,
      "loss": 0.0002,
      "reward": 0.5625,
      "reward_std": 0.1157275140285492,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.5625,
      "step": 1689
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 158.25,
      "epoch": 6.76,
      "grad_norm": 7.4089035987854,
      "kl": 0.203125,
      "learning_rate": 3.24e-07,
      "loss": 0.0002,
      "reward": 0.3125,
      "reward_std": 0.1157275140285492,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.3125,
      "step": 1690
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 144.65625,
      "epoch": 6.764,
      "grad_norm": 0.010112687945365906,
      "kl": 0.226318359375,
      "learning_rate": 3.2359999999999996e-07,
      "loss": 0.0002,
      "reward": 0.25,
      "reward_std": 0.0,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.25,
      "step": 1691
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 134.09375,
      "epoch": 6.768,
      "grad_norm": 5.197105407714844,
      "kl": 0.253662109375,
      "learning_rate": 3.232e-07,
      "loss": 0.0003,
      "reward": 0.28125,
      "reward_std": 0.0883883461356163,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.28125,
      "step": 1692
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 130.53125,
      "epoch": 6.772,
      "grad_norm": 7.619472503662109,
      "kl": 0.245849609375,
      "learning_rate": 3.2279999999999995e-07,
      "loss": 0.0002,
      "reward": 0.4375,
      "reward_std": 0.2177756354212761,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.4375,
      "step": 1693
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 144.34375,
      "epoch": 6.776,
      "grad_norm": 0.03138521686196327,
      "kl": 0.234619140625,
      "learning_rate": 3.2240000000000003e-07,
      "loss": 0.0002,
      "reward": 1.0,
      "reward_std": 0.0,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 1.0,
      "step": 1694
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 141.78125,
      "epoch": 6.78,
      "grad_norm": 3.3987510204315186,
      "kl": 0.251953125,
      "learning_rate": 3.22e-07,
      "loss": 0.0003,
      "reward": 0.53125,
      "reward_std": 0.0883883461356163,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.53125,
      "step": 1695
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 167.28125,
      "epoch": 6.784,
      "grad_norm": 0.057530857622623444,
      "kl": 0.22607421875,
      "learning_rate": 3.2159999999999997e-07,
      "loss": 0.0002,
      "reward": 0.75,
      "reward_std": 0.0,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.75,
      "step": 1696
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 144.28125,
      "epoch": 6.788,
      "grad_norm": 0.019927602261304855,
      "kl": 0.2509765625,
      "learning_rate": 3.212e-07,
      "loss": 0.0003,
      "reward": 0.75,
      "reward_std": 0.0,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.75,
      "step": 1697
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 147.84375,
      "epoch": 6.792,
      "grad_norm": 13.615362167358398,
      "kl": 0.2431640625,
      "learning_rate": 3.2079999999999996e-07,
      "loss": 0.0002,
      "reward": 0.3125,
      "reward_std": 0.1157275140285492,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.3125,
      "step": 1698
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 154.375,
      "epoch": 6.796,
      "grad_norm": 2.8694865703582764,
      "kl": 0.214599609375,
      "learning_rate": 3.204e-07,
      "loss": 0.0002,
      "reward": 0.46875,
      "reward_std": 0.0883883461356163,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.46875,
      "step": 1699
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 138.5625,
      "epoch": 6.8,
      "grad_norm": 0.02882728911936283,
      "kl": 0.24072265625,
      "learning_rate": 3.2e-07,
      "loss": 0.0002,
      "reward": 0.5,
      "reward_std": 0.0,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.5,
      "step": 1700
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 134.75,
      "epoch": 6.804,
      "grad_norm": 4.48323392868042,
      "kl": 0.257080078125,
      "learning_rate": 3.196e-07,
      "loss": 0.0003,
      "reward": 0.5625,
      "reward_std": 0.1157275140285492,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.5625,
      "step": 1701
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 135.46875,
      "epoch": 6.808,
      "grad_norm": 7.7014875411987305,
      "kl": 0.254638671875,
      "learning_rate": 3.1919999999999995e-07,
      "loss": 0.0003,
      "reward": 0.59375,
      "reward_std": 0.1293872892856598,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.59375,
      "step": 1702
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 151.90625,
      "epoch": 6.812,
      "grad_norm": 0.015006392262876034,
      "kl": 0.232177734375,
      "learning_rate": 3.1879999999999997e-07,
      "loss": 0.0002,
      "reward": 0.75,
      "reward_std": 0.0,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.75,
      "step": 1703
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 117.71875,
      "epoch": 6.816,
      "grad_norm": 4.717138767242432,
      "kl": 0.2587890625,
      "learning_rate": 3.184e-07,
      "loss": 0.0003,
      "reward": 0.96875,
      "reward_std": 0.0883883461356163,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.96875,
      "step": 1704
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 129.1875,
      "epoch": 6.82,
      "grad_norm": 0.010588133707642555,
      "kl": 0.249755859375,
      "learning_rate": 3.18e-07,
      "loss": 0.0002,
      "reward": 0.75,
      "reward_std": 0.0,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.75,
      "step": 1705
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 127.40625,
      "epoch": 6.824,
      "grad_norm": 0.010078756138682365,
      "kl": 0.267578125,
      "learning_rate": 3.176e-07,
      "loss": 0.0003,
      "reward": 0.25,
      "reward_std": 0.0,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.25,
      "step": 1706
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 144.46875,
      "epoch": 6.828,
      "grad_norm": 0.021512340754270554,
      "kl": 0.241943359375,
      "learning_rate": 3.1719999999999996e-07,
      "loss": 0.0002,
      "reward": 1.0,
      "reward_std": 0.0,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 1.0,
      "step": 1707
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 145.5625,
      "epoch": 6.832,
      "grad_norm": 0.022671356797218323,
      "kl": 0.21923828125,
      "learning_rate": 3.1680000000000003e-07,
      "loss": 0.0002,
      "reward": 1.0,
      "reward_std": 0.0,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 1.0,
      "step": 1708
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 137.78125,
      "epoch": 6.836,
      "grad_norm": 0.012268491089344025,
      "kl": 0.223876953125,
      "learning_rate": 3.164e-07,
      "loss": 0.0002,
      "reward": 0.25,
      "reward_std": 0.0,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.25,
      "step": 1709
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 147.03125,
      "epoch": 6.84,
      "grad_norm": 0.011735421605408192,
      "kl": 0.21533203125,
      "learning_rate": 3.1599999999999997e-07,
      "loss": 0.0002,
      "reward": 0.25,
      "reward_std": 0.0,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.25,
      "step": 1710
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 130.71875,
      "epoch": 6.844,
      "grad_norm": 4.961607456207275,
      "kl": 0.24560546875,
      "learning_rate": 3.156e-07,
      "loss": 0.0002,
      "reward": 0.65625,
      "reward_std": 0.1293872892856598,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.65625,
      "step": 1711
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 121.09375,
      "epoch": 6.848,
      "grad_norm": 0.02071729302406311,
      "kl": 0.261474609375,
      "learning_rate": 3.1519999999999996e-07,
      "loss": 0.0003,
      "reward": 0.75,
      "reward_std": 0.0,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.75,
      "step": 1712
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 112.625,
      "epoch": 6.852,
      "grad_norm": 0.010273251682519913,
      "kl": 0.245849609375,
      "learning_rate": 3.148e-07,
      "loss": 0.0002,
      "reward": 0.75,
      "reward_std": 0.0,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.75,
      "step": 1713
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 140.59375,
      "epoch": 6.856,
      "grad_norm": 33.23796081542969,
      "kl": 0.251708984375,
      "learning_rate": 3.144e-07,
      "loss": 0.0003,
      "reward": 0.3125,
      "reward_std": 0.1767766922712326,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.3125,
      "step": 1714
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 117.46875,
      "epoch": 6.86,
      "grad_norm": 0.033448725938797,
      "kl": 0.311279296875,
      "learning_rate": 3.14e-07,
      "loss": 0.0003,
      "reward": 1.0,
      "reward_std": 0.0,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 1.0,
      "step": 1715
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 136.6875,
      "epoch": 6.864,
      "grad_norm": 34.9610481262207,
      "kl": 0.2412109375,
      "learning_rate": 3.1359999999999995e-07,
      "loss": 0.0002,
      "reward": 0.53125,
      "reward_std": 0.2041158601641655,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.53125,
      "step": 1716
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 131.28125,
      "epoch": 6.868,
      "grad_norm": 0.02433558739721775,
      "kl": 0.225341796875,
      "learning_rate": 3.1319999999999997e-07,
      "loss": 0.0002,
      "reward": 0.25,
      "reward_std": 0.0,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.25,
      "step": 1717
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 128.46875,
      "epoch": 6.872,
      "grad_norm": 6.025835037231445,
      "kl": 0.252197265625,
      "learning_rate": 3.128e-07,
      "loss": 0.0003,
      "reward": 0.8125,
      "reward_std": 0.1157275140285492,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.8125,
      "step": 1718
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 118.875,
      "epoch": 6.876,
      "grad_norm": 0.011774418875575066,
      "kl": 0.26220703125,
      "learning_rate": 3.124e-07,
      "loss": 0.0003,
      "reward": 0.75,
      "reward_std": 0.0,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.75,
      "step": 1719
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 129.75,
      "epoch": 6.88,
      "grad_norm": 0.013408509083092213,
      "kl": 0.2275390625,
      "learning_rate": 3.12e-07,
      "loss": 0.0002,
      "reward": 1.0,
      "reward_std": 0.0,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 1.0,
      "step": 1720
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 139.625,
      "epoch": 6.884,
      "grad_norm": 0.08474773168563843,
      "kl": 0.249267578125,
      "learning_rate": 3.1159999999999996e-07,
      "loss": 0.0002,
      "reward": 0.5,
      "reward_std": 0.0,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.5,
      "step": 1721
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 137.3125,
      "epoch": 6.888,
      "grad_norm": 0.020842982456088066,
      "kl": 0.2177734375,
      "learning_rate": 3.112e-07,
      "loss": 0.0002,
      "reward": 0.25,
      "reward_std": 0.0,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.25,
      "step": 1722
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 133.6875,
      "epoch": 6.892,
      "grad_norm": 0.025626204907894135,
      "kl": 0.2177734375,
      "learning_rate": 3.108e-07,
      "loss": 0.0002,
      "reward": 0.5,
      "reward_std": 0.0,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.5,
      "step": 1723
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 127.78125,
      "epoch": 6.896,
      "grad_norm": 0.03614748269319534,
      "kl": 0.262451171875,
      "learning_rate": 3.104e-07,
      "loss": 0.0003,
      "reward": 0.5,
      "reward_std": 0.0,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.5,
      "step": 1724
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 142.21875,
      "epoch": 6.9,
      "grad_norm": 0.024518979713320732,
      "kl": 0.2353515625,
      "learning_rate": 3.1e-07,
      "loss": 0.0002,
      "reward": 0.25,
      "reward_std": 0.0,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.25,
      "step": 1725
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 135.1875,
      "epoch": 6.904,
      "grad_norm": 5.823258399963379,
      "kl": 0.252197265625,
      "learning_rate": 3.0959999999999997e-07,
      "loss": 0.0003,
      "reward": 0.75,
      "reward_std": 0.2314550280570984,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.75,
      "step": 1726
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 122.84375,
      "epoch": 6.908,
      "grad_norm": 83.99097442626953,
      "kl": 0.2958984375,
      "learning_rate": 3.0919999999999994e-07,
      "loss": 0.0003,
      "reward": 0.96875,
      "reward_std": 0.0883883461356163,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.96875,
      "step": 1727
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 126.84375,
      "epoch": 6.912,
      "grad_norm": 17.042682647705078,
      "kl": 0.255126953125,
      "learning_rate": 3.088e-07,
      "loss": 0.0003,
      "reward": 0.46875,
      "reward_std": 0.0883883461356163,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.46875,
      "step": 1728
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 121.0,
      "epoch": 6.916,
      "grad_norm": 2.770087242126465,
      "kl": 0.270263671875,
      "learning_rate": 3.084e-07,
      "loss": 0.0003,
      "reward": 0.90625,
      "reward_std": 0.1293872892856598,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.90625,
      "step": 1729
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 112.15625,
      "epoch": 6.92,
      "grad_norm": 5.1743950843811035,
      "kl": 0.249755859375,
      "learning_rate": 3.08e-07,
      "loss": 0.0002,
      "reward": 0.53125,
      "reward_std": 0.0883883461356163,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.53125,
      "step": 1730
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 114.15625,
      "epoch": 6.924,
      "grad_norm": 9.435473442077637,
      "kl": 0.28076171875,
      "learning_rate": 3.076e-07,
      "loss": 0.0003,
      "reward": 0.34375,
      "reward_std": 0.2041158601641655,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.34375,
      "step": 1731
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 132.4375,
      "epoch": 6.928,
      "grad_norm": 0.019298089668154716,
      "kl": 0.20654296875,
      "learning_rate": 3.0719999999999995e-07,
      "loss": 0.0002,
      "reward": 0.25,
      "reward_std": 0.0,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.25,
      "step": 1732
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 123.09375,
      "epoch": 6.932,
      "grad_norm": 2.2248311042785645,
      "kl": 0.258544921875,
      "learning_rate": 3.068e-07,
      "loss": 0.0003,
      "reward": 0.78125,
      "reward_std": 0.0883883461356163,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.78125,
      "step": 1733
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 120.28125,
      "epoch": 6.936,
      "grad_norm": 0.03255412355065346,
      "kl": 0.26171875,
      "learning_rate": 3.064e-07,
      "loss": 0.0003,
      "reward": 0.25,
      "reward_std": 0.0,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.25,
      "step": 1734
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 119.53125,
      "epoch": 6.9399999999999995,
      "grad_norm": 0.020736834034323692,
      "kl": 0.279541015625,
      "learning_rate": 3.0599999999999996e-07,
      "loss": 0.0003,
      "reward": 1.0,
      "reward_std": 0.0,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 1.0,
      "step": 1735
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 129.21875,
      "epoch": 6.944,
      "grad_norm": 0.03552312031388283,
      "kl": 0.2998046875,
      "learning_rate": 3.056e-07,
      "loss": 0.0003,
      "reward": 0.25,
      "reward_std": 0.0,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.25,
      "step": 1736
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 113.375,
      "epoch": 6.948,
      "grad_norm": 0.024283356964588165,
      "kl": 0.268798828125,
      "learning_rate": 3.052e-07,
      "loss": 0.0003,
      "reward": 1.0,
      "reward_std": 0.0,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 1.0,
      "step": 1737
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 127.65625,
      "epoch": 6.952,
      "grad_norm": 3.449219226837158,
      "kl": 0.2373046875,
      "learning_rate": 3.048e-07,
      "loss": 0.0002,
      "reward": 0.03125,
      "reward_std": 0.0883883461356163,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.03125,
      "step": 1738
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 144.28125,
      "epoch": 6.9559999999999995,
      "grad_norm": 39.45863342285156,
      "kl": 0.23828125,
      "learning_rate": 3.044e-07,
      "loss": 0.0002,
      "reward": 0.28125,
      "reward_std": 0.0883883461356163,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.28125,
      "step": 1739
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 143.75,
      "epoch": 6.96,
      "grad_norm": 4.161734104156494,
      "kl": 0.20361328125,
      "learning_rate": 3.0399999999999997e-07,
      "loss": 0.0002,
      "reward": 0.4375,
      "reward_std": 0.1157275140285492,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.4375,
      "step": 1740
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 129.78125,
      "epoch": 6.964,
      "grad_norm": 9.146801948547363,
      "kl": 0.247314453125,
      "learning_rate": 3.036e-07,
      "loss": 0.0002,
      "reward": 0.1875,
      "reward_std": 0.2177756354212761,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.1875,
      "step": 1741
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 110.1875,
      "epoch": 6.968,
      "grad_norm": 5.464874267578125,
      "kl": 0.263671875,
      "learning_rate": 3.032e-07,
      "loss": 0.0003,
      "reward": 0.46875,
      "reward_std": 0.0883883461356163,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.46875,
      "step": 1742
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 108.53125,
      "epoch": 6.9719999999999995,
      "grad_norm": 0.018157022073864937,
      "kl": 0.248291015625,
      "learning_rate": 3.028e-07,
      "loss": 0.0002,
      "reward": 0.5,
      "reward_std": 0.0,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.5,
      "step": 1743
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 144.90625,
      "epoch": 6.976,
      "grad_norm": 0.015170086175203323,
      "kl": 0.213623046875,
      "learning_rate": 3.024e-07,
      "loss": 0.0002,
      "reward": 0.5,
      "reward_std": 0.0,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.5,
      "step": 1744
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 140.84375,
      "epoch": 6.98,
      "grad_norm": 8.443598747253418,
      "kl": 0.22998046875,
      "learning_rate": 3.02e-07,
      "loss": 0.0002,
      "reward": 0.6875,
      "reward_std": 0.249358132481575,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.6875,
      "step": 1745
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 129.1875,
      "epoch": 6.984,
      "grad_norm": 0.02584746666252613,
      "kl": 0.33984375,
      "learning_rate": 3.0159999999999995e-07,
      "loss": 0.0003,
      "reward": 1.0,
      "reward_std": 0.0,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 1.0,
      "step": 1746
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 135.09375,
      "epoch": 6.9879999999999995,
      "grad_norm": 2.5752437114715576,
      "kl": 0.248291015625,
      "learning_rate": 3.012e-07,
      "loss": 0.0002,
      "reward": 0.78125,
      "reward_std": 0.0883883461356163,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.78125,
      "step": 1747
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 139.375,
      "epoch": 6.992,
      "grad_norm": 3.975189208984375,
      "kl": 0.2275390625,
      "learning_rate": 3.008e-07,
      "loss": 0.0002,
      "reward": 0.53125,
      "reward_std": 0.0883883461356163,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.53125,
      "step": 1748
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 105.875,
      "epoch": 6.996,
      "grad_norm": 8.94530200958252,
      "kl": 0.30078125,
      "learning_rate": 3.0039999999999996e-07,
      "loss": 0.0003,
      "reward": 0.8125,
      "reward_std": 0.249358132481575,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.8125,
      "step": 1749
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 138.78125,
      "epoch": 7.0,
      "grad_norm": 29.34798812866211,
      "kl": 0.240478515625,
      "learning_rate": 3e-07,
      "loss": 0.0002,
      "reward": 0.65625,
      "reward_std": 0.2041158601641655,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.65625,
      "step": 1750
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 138.65625,
      "epoch": 7.004,
      "grad_norm": 0.015926005318760872,
      "kl": 0.22509765625,
      "learning_rate": 2.9959999999999996e-07,
      "loss": 0.0002,
      "reward": 0.25,
      "reward_std": 0.0,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.25,
      "step": 1751
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 140.0,
      "epoch": 7.008,
      "grad_norm": 0.016799021512269974,
      "kl": 0.234375,
      "learning_rate": 2.9920000000000003e-07,
      "loss": 0.0002,
      "reward": 0.25,
      "reward_std": 0.0,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.25,
      "step": 1752
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 124.4375,
      "epoch": 7.012,
      "grad_norm": 3.331233263015747,
      "kl": 0.265625,
      "learning_rate": 2.988e-07,
      "loss": 0.0003,
      "reward": 0.625,
      "reward_std": 0.13363061845302582,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.625,
      "step": 1753
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 140.03125,
      "epoch": 7.016,
      "grad_norm": 2.5669376850128174,
      "kl": 0.246826171875,
      "learning_rate": 2.9839999999999997e-07,
      "loss": 0.0002,
      "reward": 0.5625,
      "reward_std": 0.1157275140285492,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.5625,
      "step": 1754
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 124.8125,
      "epoch": 7.02,
      "grad_norm": 4.277744293212891,
      "kl": 0.218505859375,
      "learning_rate": 2.98e-07,
      "loss": 0.0002,
      "reward": 0.59375,
      "reward_std": 0.1293872892856598,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.59375,
      "step": 1755
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 140.59375,
      "epoch": 7.024,
      "grad_norm": 11.192852020263672,
      "kl": 0.275390625,
      "learning_rate": 2.9759999999999996e-07,
      "loss": 0.0003,
      "reward": 0.46875,
      "reward_std": 0.2041158601641655,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.46875,
      "step": 1756
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 135.28125,
      "epoch": 7.028,
      "grad_norm": 0.02714182808995247,
      "kl": 0.268310546875,
      "learning_rate": 2.972e-07,
      "loss": 0.0003,
      "reward": 0.75,
      "reward_std": 0.0,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.75,
      "step": 1757
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 119.75,
      "epoch": 7.032,
      "grad_norm": 0.01661909557878971,
      "kl": 0.2578125,
      "learning_rate": 2.968e-07,
      "loss": 0.0003,
      "reward": 1.0,
      "reward_std": 0.0,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 1.0,
      "step": 1758
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 135.96875,
      "epoch": 7.036,
      "grad_norm": 2.4527981281280518,
      "kl": 0.2216796875,
      "learning_rate": 2.964e-07,
      "loss": 0.0002,
      "reward": 0.5625,
      "reward_std": 0.1157275140285492,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.5625,
      "step": 1759
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 146.40625,
      "epoch": 7.04,
      "grad_norm": 0.01741536520421505,
      "kl": 0.23046875,
      "learning_rate": 2.9599999999999995e-07,
      "loss": 0.0002,
      "reward": 0.5,
      "reward_std": 0.0,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.5,
      "step": 1760
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 146.875,
      "epoch": 7.044,
      "grad_norm": 0.011578344739973545,
      "kl": 0.2294921875,
      "learning_rate": 2.9559999999999997e-07,
      "loss": 0.0002,
      "reward": 0.75,
      "reward_std": 0.0,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.75,
      "step": 1761
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 139.3125,
      "epoch": 7.048,
      "grad_norm": 6.758629322052002,
      "kl": 0.239013671875,
      "learning_rate": 2.952e-07,
      "loss": 0.0002,
      "reward": 0.96875,
      "reward_std": 0.0883883461356163,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.96875,
      "step": 1762
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 127.96875,
      "epoch": 7.052,
      "grad_norm": 0.03515255078673363,
      "kl": 0.279052734375,
      "learning_rate": 2.948e-07,
      "loss": 0.0003,
      "reward": 1.0,
      "reward_std": 0.0,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 1.0,
      "step": 1763
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 145.0625,
      "epoch": 7.056,
      "grad_norm": 27.444110870361328,
      "kl": 0.23046875,
      "learning_rate": 2.944e-07,
      "loss": 0.0002,
      "reward": 0.875,
      "reward_std": 0.13363061845302582,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.875,
      "step": 1764
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 166.90625,
      "epoch": 7.06,
      "grad_norm": 3.919861316680908,
      "kl": 0.236083984375,
      "learning_rate": 2.9399999999999996e-07,
      "loss": 0.0002,
      "reward": 0.65625,
      "reward_std": 0.1293872892856598,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.65625,
      "step": 1765
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 133.09375,
      "epoch": 7.064,
      "grad_norm": 3.479031801223755,
      "kl": 0.251708984375,
      "learning_rate": 2.9360000000000003e-07,
      "loss": 0.0003,
      "reward": 0.53125,
      "reward_std": 0.0883883461356163,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.53125,
      "step": 1766
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 122.6875,
      "epoch": 7.068,
      "grad_norm": 5.610877990722656,
      "kl": 0.29345703125,
      "learning_rate": 2.932e-07,
      "loss": 0.0003,
      "reward": 0.53125,
      "reward_std": 0.2041158601641655,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.53125,
      "step": 1767
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 133.8125,
      "epoch": 7.072,
      "grad_norm": 0.011953371576964855,
      "kl": 0.241943359375,
      "learning_rate": 2.928e-07,
      "loss": 0.0002,
      "reward": 0.5,
      "reward_std": 0.0,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.5,
      "step": 1768
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 133.71875,
      "epoch": 7.076,
      "grad_norm": 3.6290597915649414,
      "kl": 0.269775390625,
      "learning_rate": 2.924e-07,
      "loss": 0.0003,
      "reward": 0.53125,
      "reward_std": 0.0883883461356163,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.53125,
      "step": 1769
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 118.9375,
      "epoch": 7.08,
      "grad_norm": 0.012909610755741596,
      "kl": 0.26171875,
      "learning_rate": 2.9199999999999997e-07,
      "loss": 0.0003,
      "reward": 1.0,
      "reward_std": 0.0,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 1.0,
      "step": 1770
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 130.53125,
      "epoch": 7.084,
      "grad_norm": 0.01084019523113966,
      "kl": 0.24951171875,
      "learning_rate": 2.916e-07,
      "loss": 0.0002,
      "reward": 0.25,
      "reward_std": 0.0,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.25,
      "step": 1771
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 127.34375,
      "epoch": 7.088,
      "grad_norm": 0.01240465510636568,
      "kl": 0.241943359375,
      "learning_rate": 2.912e-07,
      "loss": 0.0002,
      "reward": 0.5,
      "reward_std": 0.0,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.5,
      "step": 1772
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 127.71875,
      "epoch": 7.092,
      "grad_norm": 2.420382022857666,
      "kl": 0.303466796875,
      "learning_rate": 2.908e-07,
      "loss": 0.0003,
      "reward": 0.71875,
      "reward_std": 0.0883883461356163,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.71875,
      "step": 1773
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 122.65625,
      "epoch": 7.096,
      "grad_norm": 0.06588277965784073,
      "kl": 0.380126953125,
      "learning_rate": 2.9039999999999995e-07,
      "loss": 0.0004,
      "reward": 0.25,
      "reward_std": 0.0,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.25,
      "step": 1774
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 132.625,
      "epoch": 7.1,
      "grad_norm": 0.012310614809393883,
      "kl": 0.240966796875,
      "learning_rate": 2.9e-07,
      "loss": 0.0002,
      "reward": 0.5,
      "reward_std": 0.0,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.5,
      "step": 1775
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 141.6875,
      "epoch": 7.104,
      "grad_norm": 0.0857868492603302,
      "kl": 0.248046875,
      "learning_rate": 2.896e-07,
      "loss": 0.0002,
      "reward": 0.75,
      "reward_std": 0.0,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.75,
      "step": 1776
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 139.28125,
      "epoch": 7.108,
      "grad_norm": 0.035870615392923355,
      "kl": 0.227783203125,
      "learning_rate": 2.892e-07,
      "loss": 0.0002,
      "reward": 1.0,
      "reward_std": 0.0,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 1.0,
      "step": 1777
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 130.25,
      "epoch": 7.112,
      "grad_norm": 0.022740250453352928,
      "kl": 0.22802734375,
      "learning_rate": 2.888e-07,
      "loss": 0.0002,
      "reward": 0.5,
      "reward_std": 0.0,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.5,
      "step": 1778
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 145.53125,
      "epoch": 7.116,
      "grad_norm": 0.010278968140482903,
      "kl": 0.20947265625,
      "learning_rate": 2.8839999999999996e-07,
      "loss": 0.0002,
      "reward": 0.75,
      "reward_std": 0.0,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.75,
      "step": 1779
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 146.03125,
      "epoch": 7.12,
      "grad_norm": 4.470780372619629,
      "kl": 0.24951171875,
      "learning_rate": 2.88e-07,
      "loss": 0.0002,
      "reward": 0.96875,
      "reward_std": 0.0883883461356163,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.96875,
      "step": 1780
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 135.1875,
      "epoch": 7.124,
      "grad_norm": 14.300722122192383,
      "kl": 0.25146484375,
      "learning_rate": 2.876e-07,
      "loss": 0.0003,
      "reward": 0.71875,
      "reward_std": 0.0883883461356163,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.71875,
      "step": 1781
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 111.15625,
      "epoch": 7.128,
      "grad_norm": 9.489889144897461,
      "kl": 0.27099609375,
      "learning_rate": 2.872e-07,
      "loss": 0.0003,
      "reward": 0.6875,
      "reward_std": 0.1157275140285492,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.6875,
      "step": 1782
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 124.5,
      "epoch": 7.132,
      "grad_norm": 0.027017172425985336,
      "kl": 0.29931640625,
      "learning_rate": 2.868e-07,
      "loss": 0.0003,
      "reward": 0.25,
      "reward_std": 0.0,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.25,
      "step": 1783
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 131.4375,
      "epoch": 7.136,
      "grad_norm": 3.6589062213897705,
      "kl": 0.254638671875,
      "learning_rate": 2.8639999999999997e-07,
      "loss": 0.0003,
      "reward": 0.03125,
      "reward_std": 0.0883883461356163,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.03125,
      "step": 1784
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 132.53125,
      "epoch": 7.14,
      "grad_norm": 0.011488151736557484,
      "kl": 0.27685546875,
      "learning_rate": 2.8599999999999994e-07,
      "loss": 0.0003,
      "reward": 0.5,
      "reward_std": 0.0,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.5,
      "step": 1785
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 133.78125,
      "epoch": 7.144,
      "grad_norm": 0.02643558382987976,
      "kl": 0.2373046875,
      "learning_rate": 2.856e-07,
      "loss": 0.0002,
      "reward": 1.0,
      "reward_std": 0.0,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 1.0,
      "step": 1786
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 125.40625,
      "epoch": 7.148,
      "grad_norm": 0.016571881249547005,
      "kl": 0.260009765625,
      "learning_rate": 2.852e-07,
      "loss": 0.0003,
      "reward": 0.75,
      "reward_std": 0.0,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.75,
      "step": 1787
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 149.71875,
      "epoch": 7.152,
      "grad_norm": 5.302474498748779,
      "kl": 0.23095703125,
      "learning_rate": 2.848e-07,
      "loss": 0.0002,
      "reward": 0.34375,
      "reward_std": 0.22201896458864212,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.34375,
      "step": 1788
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 145.84375,
      "epoch": 7.156,
      "grad_norm": 0.012218784540891647,
      "kl": 0.2578125,
      "learning_rate": 2.844e-07,
      "loss": 0.0003,
      "reward": 0.75,
      "reward_std": 0.0,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.75,
      "step": 1789
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 136.84375,
      "epoch": 7.16,
      "grad_norm": 3.5980210304260254,
      "kl": 0.22607421875,
      "learning_rate": 2.8399999999999995e-07,
      "loss": 0.0002,
      "reward": 0.34375,
      "reward_std": 0.1293872892856598,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.34375,
      "step": 1790
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 123.78125,
      "epoch": 7.164,
      "grad_norm": 0.018439585343003273,
      "kl": 0.2421875,
      "learning_rate": 2.836e-07,
      "loss": 0.0002,
      "reward": 0.75,
      "reward_std": 0.0,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.75,
      "step": 1791
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 121.875,
      "epoch": 7.168,
      "grad_norm": 0.02534143626689911,
      "kl": 0.253662109375,
      "learning_rate": 2.832e-07,
      "loss": 0.0003,
      "reward": 0.25,
      "reward_std": 0.0,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.25,
      "step": 1792
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 134.1875,
      "epoch": 7.172,
      "grad_norm": 0.010159192606806755,
      "kl": 0.27392578125,
      "learning_rate": 2.8279999999999996e-07,
      "loss": 0.0003,
      "reward": 0.75,
      "reward_std": 0.0,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.75,
      "step": 1793
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 151.03125,
      "epoch": 7.176,
      "grad_norm": 0.011022322811186314,
      "kl": 0.226806640625,
      "learning_rate": 2.824e-07,
      "loss": 0.0002,
      "reward": 0.25,
      "reward_std": 0.0,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.25,
      "step": 1794
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 144.0625,
      "epoch": 7.18,
      "grad_norm": 0.017136024311184883,
      "kl": 0.276611328125,
      "learning_rate": 2.8199999999999996e-07,
      "loss": 0.0003,
      "reward": 0.25,
      "reward_std": 0.0,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.25,
      "step": 1795
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 127.90625,
      "epoch": 7.184,
      "grad_norm": 2.502056121826172,
      "kl": 0.271240234375,
      "learning_rate": 2.816e-07,
      "loss": 0.0003,
      "reward": 0.71875,
      "reward_std": 0.0883883461356163,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.71875,
      "step": 1796
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 140.53125,
      "epoch": 7.188,
      "grad_norm": 0.025581011548638344,
      "kl": 0.243896484375,
      "learning_rate": 2.812e-07,
      "loss": 0.0002,
      "reward": 0.5,
      "reward_std": 0.0,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.5,
      "step": 1797
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 137.9375,
      "epoch": 7.192,
      "grad_norm": 0.025997832417488098,
      "kl": 0.27001953125,
      "learning_rate": 2.8079999999999997e-07,
      "loss": 0.0003,
      "reward": 0.25,
      "reward_std": 0.0,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.25,
      "step": 1798
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 140.59375,
      "epoch": 7.196,
      "grad_norm": 3.0310513973236084,
      "kl": 0.2529296875,
      "learning_rate": 2.804e-07,
      "loss": 0.0003,
      "reward": 0.09375,
      "reward_std": 0.1293872892856598,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.09375,
      "step": 1799
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 131.34375,
      "epoch": 7.2,
      "grad_norm": 0.013194085098803043,
      "kl": 0.24365234375,
      "learning_rate": 2.8e-07,
      "loss": 0.0002,
      "reward": 0.75,
      "reward_std": 0.0,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.75,
      "step": 1800
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 139.3125,
      "epoch": 7.204,
      "grad_norm": 3.7839457988739014,
      "kl": 0.21875,
      "learning_rate": 2.796e-07,
      "loss": 0.0002,
      "reward": 0.5625,
      "reward_std": 0.1157275140285492,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.5625,
      "step": 1801
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 150.53125,
      "epoch": 7.208,
      "grad_norm": 0.008167166262865067,
      "kl": 0.224365234375,
      "learning_rate": 2.792e-07,
      "loss": 0.0002,
      "reward": 0.75,
      "reward_std": 0.0,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.75,
      "step": 1802
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 141.84375,
      "epoch": 7.212,
      "grad_norm": 7.576340675354004,
      "kl": 0.2685546875,
      "learning_rate": 2.788e-07,
      "loss": 0.0003,
      "reward": 0.625,
      "reward_std": 0.2314550280570984,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.625,
      "step": 1803
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 130.5625,
      "epoch": 7.216,
      "grad_norm": 5.645166397094727,
      "kl": 0.26025390625,
      "learning_rate": 2.7839999999999995e-07,
      "loss": 0.0003,
      "reward": 0.5,
      "reward_std": 0.2587745785713196,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.5,
      "step": 1804
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 159.125,
      "epoch": 7.22,
      "grad_norm": 0.018675118684768677,
      "kl": 0.252685546875,
      "learning_rate": 2.7800000000000003e-07,
      "loss": 0.0003,
      "reward": 0.75,
      "reward_std": 0.0,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.75,
      "step": 1805
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 149.09375,
      "epoch": 7.224,
      "grad_norm": 0.018949370831251144,
      "kl": 0.25830078125,
      "learning_rate": 2.776e-07,
      "loss": 0.0003,
      "reward": 0.25,
      "reward_std": 0.0,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.25,
      "step": 1806
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 132.25,
      "epoch": 7.228,
      "grad_norm": 0.021013256162405014,
      "kl": 0.264404296875,
      "learning_rate": 2.7719999999999997e-07,
      "loss": 0.0003,
      "reward": 0.5,
      "reward_std": 0.0,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.5,
      "step": 1807
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 145.90625,
      "epoch": 7.232,
      "grad_norm": 21.497236251831055,
      "kl": 0.22998046875,
      "learning_rate": 2.768e-07,
      "loss": 0.0002,
      "reward": 0.65625,
      "reward_std": 0.2041158601641655,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.65625,
      "step": 1808
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 139.5625,
      "epoch": 7.236,
      "grad_norm": 0.010374129749834538,
      "kl": 0.232666015625,
      "learning_rate": 2.7639999999999996e-07,
      "loss": 0.0002,
      "reward": 0.5,
      "reward_std": 0.0,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.5,
      "step": 1809
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 141.4375,
      "epoch": 7.24,
      "grad_norm": 3.038699150085449,
      "kl": 0.2373046875,
      "learning_rate": 2.7600000000000004e-07,
      "loss": 0.0002,
      "reward": 0.71875,
      "reward_std": 0.0883883461356163,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.71875,
      "step": 1810
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 144.71875,
      "epoch": 7.244,
      "grad_norm": 0.028039082884788513,
      "kl": 0.24072265625,
      "learning_rate": 2.756e-07,
      "loss": 0.0002,
      "reward": 0.75,
      "reward_std": 0.0,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.75,
      "step": 1811
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 132.3125,
      "epoch": 7.248,
      "grad_norm": 4.714695930480957,
      "kl": 0.2392578125,
      "learning_rate": 2.752e-07,
      "loss": 0.0002,
      "reward": 0.5625,
      "reward_std": 0.1157275140285492,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.5625,
      "step": 1812
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 134.34375,
      "epoch": 7.252,
      "grad_norm": 10.04321002960205,
      "kl": 0.26611328125,
      "learning_rate": 2.748e-07,
      "loss": 0.0003,
      "reward": 0.53125,
      "reward_std": 0.0883883461356163,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.53125,
      "step": 1813
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 154.875,
      "epoch": 7.256,
      "grad_norm": 0.19845528900623322,
      "kl": 0.25146484375,
      "learning_rate": 2.7439999999999997e-07,
      "loss": 0.0003,
      "reward": 0.75,
      "reward_std": 0.0,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.75,
      "step": 1814
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 133.84375,
      "epoch": 7.26,
      "grad_norm": 0.01072254590690136,
      "kl": 0.228271484375,
      "learning_rate": 2.74e-07,
      "loss": 0.0002,
      "reward": 0.25,
      "reward_std": 0.0,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.25,
      "step": 1815
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 139.625,
      "epoch": 7.264,
      "grad_norm": 9.315225601196289,
      "kl": 0.23876953125,
      "learning_rate": 2.736e-07,
      "loss": 0.0002,
      "reward": 0.375,
      "reward_std": 0.2177756354212761,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.375,
      "step": 1816
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 159.78125,
      "epoch": 7.268,
      "grad_norm": 0.05398958921432495,
      "kl": 0.216552734375,
      "learning_rate": 2.732e-07,
      "loss": 0.0002,
      "reward": 0.5,
      "reward_std": 0.0,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.5,
      "step": 1817
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 135.0625,
      "epoch": 7.272,
      "grad_norm": 0.009782551787793636,
      "kl": 0.235107421875,
      "learning_rate": 2.7279999999999995e-07,
      "loss": 0.0002,
      "reward": 0.5,
      "reward_std": 0.0,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.5,
      "step": 1818
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 135.9375,
      "epoch": 7.276,
      "grad_norm": 0.02469274215400219,
      "kl": 0.2890625,
      "learning_rate": 2.724e-07,
      "loss": 0.0003,
      "reward": 0.5,
      "reward_std": 0.0,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.5,
      "step": 1819
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 134.40625,
      "epoch": 7.28,
      "grad_norm": 2.8159947395324707,
      "kl": 0.231689453125,
      "learning_rate": 2.72e-07,
      "loss": 0.0002,
      "reward": 0.53125,
      "reward_std": 0.0883883461356163,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.53125,
      "step": 1820
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 142.9375,
      "epoch": 7.284,
      "grad_norm": 5.9620137214660645,
      "kl": 0.223388671875,
      "learning_rate": 2.7159999999999997e-07,
      "loss": 0.0002,
      "reward": 0.9375,
      "reward_std": 0.1157275140285492,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.9375,
      "step": 1821
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 165.625,
      "epoch": 7.288,
      "grad_norm": 5.891270160675049,
      "kl": 0.23779296875,
      "learning_rate": 2.712e-07,
      "loss": 0.0002,
      "reward": 0.3125,
      "reward_std": 0.1157275140285492,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.3125,
      "step": 1822
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 147.9375,
      "epoch": 7.292,
      "grad_norm": 0.4249248802661896,
      "kl": 0.251953125,
      "learning_rate": 2.7079999999999996e-07,
      "loss": 0.0003,
      "reward": 1.0,
      "reward_std": 0.0,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 1.0,
      "step": 1823
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 147.84375,
      "epoch": 7.296,
      "grad_norm": 0.012278322130441666,
      "kl": 0.2490234375,
      "learning_rate": 2.704e-07,
      "loss": 0.0002,
      "reward": 0.5,
      "reward_std": 0.0,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.5,
      "step": 1824
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 158.75,
      "epoch": 7.3,
      "grad_norm": 0.012475477531552315,
      "kl": 0.235107421875,
      "learning_rate": 2.7e-07,
      "loss": 0.0002,
      "reward": 0.25,
      "reward_std": 0.0,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.25,
      "step": 1825
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 149.25,
      "epoch": 7.304,
      "grad_norm": 4.129249572753906,
      "kl": 0.1767578125,
      "learning_rate": 2.696e-07,
      "loss": 0.0002,
      "reward": 0.28125,
      "reward_std": 0.2041158601641655,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.28125,
      "step": 1826
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 146.1875,
      "epoch": 7.308,
      "grad_norm": 0.11612062156200409,
      "kl": 0.23779296875,
      "learning_rate": 2.692e-07,
      "loss": 0.0002,
      "reward": 0.25,
      "reward_std": 0.0,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.25,
      "step": 1827
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 150.34375,
      "epoch": 7.312,
      "grad_norm": 21.594154357910156,
      "kl": 0.25537109375,
      "learning_rate": 2.6879999999999997e-07,
      "loss": 0.0003,
      "reward": 0.375,
      "reward_std": 0.13363061845302582,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.375,
      "step": 1828
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 149.375,
      "epoch": 7.316,
      "grad_norm": 1.7644016742706299,
      "kl": 0.2548828125,
      "learning_rate": 2.684e-07,
      "loss": 0.0003,
      "reward": 0.71875,
      "reward_std": 0.0883883461356163,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.71875,
      "step": 1829
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 128.375,
      "epoch": 7.32,
      "grad_norm": 0.010569787584245205,
      "kl": 0.25927734375,
      "learning_rate": 2.68e-07,
      "loss": 0.0003,
      "reward": 1.0,
      "reward_std": 0.0,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 1.0,
      "step": 1830
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 144.78125,
      "epoch": 7.324,
      "grad_norm": 0.009858982637524605,
      "kl": 0.221435546875,
      "learning_rate": 2.676e-07,
      "loss": 0.0002,
      "reward": 0.75,
      "reward_std": 0.0,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.75,
      "step": 1831
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 132.84375,
      "epoch": 7.328,
      "grad_norm": 19.585786819458008,
      "kl": 0.231689453125,
      "learning_rate": 2.6719999999999996e-07,
      "loss": 0.0002,
      "reward": 0.96875,
      "reward_std": 0.0883883461356163,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.96875,
      "step": 1832
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 143.0,
      "epoch": 7.332,
      "grad_norm": 0.016479607671499252,
      "kl": 0.2490234375,
      "learning_rate": 2.668e-07,
      "loss": 0.0002,
      "reward": 0.75,
      "reward_std": 0.0,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.75,
      "step": 1833
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 167.625,
      "epoch": 7.336,
      "grad_norm": 0.01886342652142048,
      "kl": 0.214599609375,
      "learning_rate": 2.664e-07,
      "loss": 0.0002,
      "reward": 0.75,
      "reward_std": 0.0,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.75,
      "step": 1834
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 139.0625,
      "epoch": 7.34,
      "grad_norm": 4.9278669357299805,
      "kl": 0.232421875,
      "learning_rate": 2.66e-07,
      "loss": 0.0002,
      "reward": 0.96875,
      "reward_std": 0.0883883461356163,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.96875,
      "step": 1835
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 130.53125,
      "epoch": 7.344,
      "grad_norm": 7.495325088500977,
      "kl": 0.2451171875,
      "learning_rate": 2.656e-07,
      "loss": 0.0002,
      "reward": 0.53125,
      "reward_std": 0.0883883461356163,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.53125,
      "step": 1836
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 134.3125,
      "epoch": 7.348,
      "grad_norm": 0.016268165782094002,
      "kl": 0.24462890625,
      "learning_rate": 2.6519999999999997e-07,
      "loss": 0.0002,
      "reward": 0.5,
      "reward_std": 0.0,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.5,
      "step": 1837
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 144.59375,
      "epoch": 7.352,
      "grad_norm": 0.025646265596151352,
      "kl": 0.240234375,
      "learning_rate": 2.648e-07,
      "loss": 0.0002,
      "reward": 0.5,
      "reward_std": 0.0,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.5,
      "step": 1838
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 147.25,
      "epoch": 7.356,
      "grad_norm": 3.833749532699585,
      "kl": 0.266845703125,
      "learning_rate": 2.644e-07,
      "loss": 0.0003,
      "reward": 0.3125,
      "reward_std": 0.1157275140285492,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.3125,
      "step": 1839
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 130.0,
      "epoch": 7.36,
      "grad_norm": 6.705461502075195,
      "kl": 0.244873046875,
      "learning_rate": 2.64e-07,
      "loss": 0.0002,
      "reward": 0.125,
      "reward_std": 0.2177756354212761,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.125,
      "step": 1840
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 147.125,
      "epoch": 7.364,
      "grad_norm": 0.0143667533993721,
      "kl": 0.2314453125,
      "learning_rate": 2.636e-07,
      "loss": 0.0002,
      "reward": 0.5,
      "reward_std": 0.0,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.5,
      "step": 1841
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 137.375,
      "epoch": 7.368,
      "grad_norm": 4.071930408477783,
      "kl": 0.260498046875,
      "learning_rate": 2.632e-07,
      "loss": 0.0003,
      "reward": 0.53125,
      "reward_std": 0.0883883461356163,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.53125,
      "step": 1842
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 144.75,
      "epoch": 7.372,
      "grad_norm": 0.014131944626569748,
      "kl": 0.238525390625,
      "learning_rate": 2.6279999999999994e-07,
      "loss": 0.0002,
      "reward": 0.75,
      "reward_std": 0.0,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.75,
      "step": 1843
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 145.40625,
      "epoch": 7.376,
      "grad_norm": 0.011142823845148087,
      "kl": 0.228759765625,
      "learning_rate": 2.624e-07,
      "loss": 0.0002,
      "reward": 1.0,
      "reward_std": 0.0,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 1.0,
      "step": 1844
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 145.9375,
      "epoch": 7.38,
      "grad_norm": 0.015368065796792507,
      "kl": 0.239501953125,
      "learning_rate": 2.62e-07,
      "loss": 0.0002,
      "reward": 1.0,
      "reward_std": 0.0,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 1.0,
      "step": 1845
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 138.28125,
      "epoch": 7.384,
      "grad_norm": 6.953902721405029,
      "kl": 0.2548828125,
      "learning_rate": 2.616e-07,
      "loss": 0.0003,
      "reward": 0.375,
      "reward_std": 0.13363061845302582,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.375,
      "step": 1846
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 157.59375,
      "epoch": 7.388,
      "grad_norm": 3.6508946418762207,
      "kl": 0.23046875,
      "learning_rate": 2.612e-07,
      "loss": 0.0002,
      "reward": 0.28125,
      "reward_std": 0.0883883461356163,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.28125,
      "step": 1847
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 148.125,
      "epoch": 7.392,
      "grad_norm": 0.018701639026403427,
      "kl": 0.21923828125,
      "learning_rate": 2.6079999999999995e-07,
      "loss": 0.0002,
      "reward": 0.5,
      "reward_std": 0.0,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.5,
      "step": 1848
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 136.59375,
      "epoch": 7.396,
      "grad_norm": 6.276679992675781,
      "kl": 0.2705078125,
      "learning_rate": 2.6040000000000003e-07,
      "loss": 0.0003,
      "reward": 0.59375,
      "reward_std": 0.24511480331420898,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.59375,
      "step": 1849
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 152.15625,
      "epoch": 7.4,
      "grad_norm": 0.023353656753897667,
      "kl": 0.237060546875,
      "learning_rate": 2.6e-07,
      "loss": 0.0002,
      "reward": 0.5,
      "reward_std": 0.0,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.5,
      "step": 1850
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 131.0625,
      "epoch": 7.404,
      "grad_norm": 8.78076457977295,
      "kl": 0.242431640625,
      "learning_rate": 2.5959999999999997e-07,
      "loss": 0.0002,
      "reward": 0.53125,
      "reward_std": 0.0883883461356163,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.53125,
      "step": 1851
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 136.3125,
      "epoch": 7.408,
      "grad_norm": 0.008420605212450027,
      "kl": 0.230224609375,
      "learning_rate": 2.592e-07,
      "loss": 0.0002,
      "reward": 0.75,
      "reward_std": 0.0,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.75,
      "step": 1852
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 118.15625,
      "epoch": 7.412,
      "grad_norm": 0.026488205417990685,
      "kl": 0.25146484375,
      "learning_rate": 2.5879999999999996e-07,
      "loss": 0.0003,
      "reward": 0.75,
      "reward_std": 0.0,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.75,
      "step": 1853
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 154.75,
      "epoch": 7.416,
      "grad_norm": 9.024503707885742,
      "kl": 0.247802734375,
      "learning_rate": 2.584e-07,
      "loss": 0.0002,
      "reward": 0.125,
      "reward_std": 0.2177756354212761,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.125,
      "step": 1854
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 142.6875,
      "epoch": 7.42,
      "grad_norm": 4.075369358062744,
      "kl": 0.227294921875,
      "learning_rate": 2.58e-07,
      "loss": 0.0002,
      "reward": 0.6875,
      "reward_std": 0.1157275140285492,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.6875,
      "step": 1855
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 124.34375,
      "epoch": 7.424,
      "grad_norm": 0.03375690430402756,
      "kl": 0.28076171875,
      "learning_rate": 2.576e-07,
      "loss": 0.0003,
      "reward": 0.75,
      "reward_std": 0.0,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.75,
      "step": 1856
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 153.375,
      "epoch": 7.428,
      "grad_norm": 0.013101683929562569,
      "kl": 0.235107421875,
      "learning_rate": 2.5719999999999995e-07,
      "loss": 0.0002,
      "reward": 0.5,
      "reward_std": 0.0,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.5,
      "step": 1857
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 153.9375,
      "epoch": 7.432,
      "grad_norm": 0.021614953875541687,
      "kl": 0.2392578125,
      "learning_rate": 2.5679999999999997e-07,
      "loss": 0.0002,
      "reward": 0.25,
      "reward_std": 0.0,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.25,
      "step": 1858
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 156.09375,
      "epoch": 7.436,
      "grad_norm": 0.016671855002641678,
      "kl": 0.254638671875,
      "learning_rate": 2.564e-07,
      "loss": 0.0003,
      "reward": 0.5,
      "reward_std": 0.0,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.5,
      "step": 1859
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 136.65625,
      "epoch": 7.44,
      "grad_norm": 0.011686879210174084,
      "kl": 0.2314453125,
      "learning_rate": 2.56e-07,
      "loss": 0.0002,
      "reward": 0.75,
      "reward_std": 0.0,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.75,
      "step": 1860
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 145.34375,
      "epoch": 7.444,
      "grad_norm": 0.06134292110800743,
      "kl": 0.23583984375,
      "learning_rate": 2.556e-07,
      "loss": 0.0002,
      "reward": 0.5,
      "reward_std": 0.0,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.5,
      "step": 1861
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 139.125,
      "epoch": 7.448,
      "grad_norm": 0.01656407117843628,
      "kl": 0.249755859375,
      "learning_rate": 2.5519999999999996e-07,
      "loss": 0.0002,
      "reward": 0.5,
      "reward_std": 0.0,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.5,
      "step": 1862
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 146.09375,
      "epoch": 7.452,
      "grad_norm": 10.695383071899414,
      "kl": 0.2548828125,
      "learning_rate": 2.5480000000000003e-07,
      "loss": 0.0003,
      "reward": 0.34375,
      "reward_std": 0.1293872892856598,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.34375,
      "step": 1863
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 150.59375,
      "epoch": 7.456,
      "grad_norm": 0.01177735161036253,
      "kl": 0.231689453125,
      "learning_rate": 2.544e-07,
      "loss": 0.0002,
      "reward": 0.75,
      "reward_std": 0.0,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.75,
      "step": 1864
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 148.3125,
      "epoch": 7.46,
      "grad_norm": 5.133981227874756,
      "kl": 0.2275390625,
      "learning_rate": 2.5399999999999997e-07,
      "loss": 0.0002,
      "reward": 0.71875,
      "reward_std": 0.0883883461356163,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.71875,
      "step": 1865
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 150.875,
      "epoch": 7.464,
      "grad_norm": 11.371137619018555,
      "kl": 0.2470703125,
      "learning_rate": 2.536e-07,
      "loss": 0.0002,
      "reward": 0.65625,
      "reward_std": 0.2041158601641655,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.65625,
      "step": 1866
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 143.78125,
      "epoch": 7.468,
      "grad_norm": 1.879071831703186,
      "kl": 0.261962890625,
      "learning_rate": 2.5319999999999996e-07,
      "loss": 0.0003,
      "reward": 0.71875,
      "reward_std": 0.0883883461356163,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.71875,
      "step": 1867
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 145.78125,
      "epoch": 7.4719999999999995,
      "grad_norm": 3.2217588424682617,
      "kl": 0.286865234375,
      "learning_rate": 2.528e-07,
      "loss": 0.0003,
      "reward": 0.96875,
      "reward_std": 0.0883883461356163,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.96875,
      "step": 1868
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 144.4375,
      "epoch": 7.476,
      "grad_norm": 3.6826796531677246,
      "kl": 0.24560546875,
      "learning_rate": 2.524e-07,
      "loss": 0.0002,
      "reward": 0.65625,
      "reward_std": 0.1293872892856598,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.65625,
      "step": 1869
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 114.25,
      "epoch": 7.48,
      "grad_norm": 2.8793160915374756,
      "kl": 0.270263671875,
      "learning_rate": 2.52e-07,
      "loss": 0.0003,
      "reward": 0.5625,
      "reward_std": 0.1157275140285492,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.5625,
      "step": 1870
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 129.53125,
      "epoch": 7.484,
      "grad_norm": 4.199337959289551,
      "kl": 0.32421875,
      "learning_rate": 2.516e-07,
      "loss": 0.0003,
      "reward": 0.6875,
      "reward_std": 0.1157275140285492,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.6875,
      "step": 1871
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 147.8125,
      "epoch": 7.4879999999999995,
      "grad_norm": 0.009505903348326683,
      "kl": 0.22998046875,
      "learning_rate": 2.5119999999999997e-07,
      "loss": 0.0002,
      "reward": 0.5,
      "reward_std": 0.0,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.5,
      "step": 1872
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 122.25,
      "epoch": 7.492,
      "grad_norm": 0.02206607535481453,
      "kl": 0.263671875,
      "learning_rate": 2.508e-07,
      "loss": 0.0003,
      "reward": 0.75,
      "reward_std": 0.0,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.75,
      "step": 1873
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 142.1875,
      "epoch": 7.496,
      "grad_norm": 0.03191354125738144,
      "kl": 0.232177734375,
      "learning_rate": 2.504e-07,
      "loss": 0.0002,
      "reward": 0.75,
      "reward_std": 0.0,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.75,
      "step": 1874
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 156.3125,
      "epoch": 7.5,
      "grad_norm": 29.48009490966797,
      "kl": 0.2216796875,
      "learning_rate": 2.5e-07,
      "loss": 0.0002,
      "reward": 0.25,
      "reward_std": 0.1767766922712326,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.25,
      "step": 1875
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 140.21875,
      "epoch": 7.504,
      "grad_norm": 2.530576467514038,
      "kl": 0.262939453125,
      "learning_rate": 2.4959999999999996e-07,
      "loss": 0.0003,
      "reward": 0.71875,
      "reward_std": 0.0883883461356163,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.71875,
      "step": 1876
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 148.5625,
      "epoch": 7.508,
      "grad_norm": 0.025454092770814896,
      "kl": 0.2666015625,
      "learning_rate": 2.492e-07,
      "loss": 0.0003,
      "reward": 0.5,
      "reward_std": 0.0,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.5,
      "step": 1877
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 148.28125,
      "epoch": 7.5120000000000005,
      "grad_norm": 0.021782230585813522,
      "kl": 0.242431640625,
      "learning_rate": 2.488e-07,
      "loss": 0.0002,
      "reward": 0.25,
      "reward_std": 0.0,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.25,
      "step": 1878
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 113.6875,
      "epoch": 7.516,
      "grad_norm": 18.299837112426758,
      "kl": 0.297607421875,
      "learning_rate": 2.484e-07,
      "loss": 0.0003,
      "reward": 0.5,
      "reward_std": 0.1767766922712326,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.5,
      "step": 1879
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 146.53125,
      "epoch": 7.52,
      "grad_norm": 0.02207281067967415,
      "kl": 0.209716796875,
      "learning_rate": 2.48e-07,
      "loss": 0.0002,
      "reward": 0.75,
      "reward_std": 0.0,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.75,
      "step": 1880
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 139.5625,
      "epoch": 7.524,
      "grad_norm": 0.01175529696047306,
      "kl": 0.234619140625,
      "learning_rate": 2.4759999999999997e-07,
      "loss": 0.0002,
      "reward": 0.5,
      "reward_std": 0.0,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.5,
      "step": 1881
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 141.53125,
      "epoch": 7.5280000000000005,
      "grad_norm": 0.01930665411055088,
      "kl": 0.248046875,
      "learning_rate": 2.472e-07,
      "loss": 0.0002,
      "reward": 0.5,
      "reward_std": 0.0,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.5,
      "step": 1882
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 131.53125,
      "epoch": 7.532,
      "grad_norm": 0.03325871750712395,
      "kl": 0.2373046875,
      "learning_rate": 2.4679999999999996e-07,
      "loss": 0.0002,
      "reward": 0.5,
      "reward_std": 0.0,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.5,
      "step": 1883
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 123.34375,
      "epoch": 7.536,
      "grad_norm": 3.0264883041381836,
      "kl": 0.2314453125,
      "learning_rate": 2.464e-07,
      "loss": 0.0002,
      "reward": 0.4375,
      "reward_std": 0.1157275140285492,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.4375,
      "step": 1884
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 124.875,
      "epoch": 7.54,
      "grad_norm": 24.308670043945312,
      "kl": 0.273193359375,
      "learning_rate": 2.46e-07,
      "loss": 0.0003,
      "reward": 0.96875,
      "reward_std": 0.0883883461356163,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.96875,
      "step": 1885
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 135.75,
      "epoch": 7.5440000000000005,
      "grad_norm": 0.027888264507055283,
      "kl": 0.28515625,
      "learning_rate": 2.456e-07,
      "loss": 0.0003,
      "reward": 0.5,
      "reward_std": 0.0,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.5,
      "step": 1886
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 127.625,
      "epoch": 7.548,
      "grad_norm": 0.01135960966348648,
      "kl": 0.248046875,
      "learning_rate": 2.452e-07,
      "loss": 0.0002,
      "reward": 1.0,
      "reward_std": 0.0,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 1.0,
      "step": 1887
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 131.15625,
      "epoch": 7.552,
      "grad_norm": 6.592331886291504,
      "kl": 0.2197265625,
      "learning_rate": 2.4479999999999997e-07,
      "loss": 0.0002,
      "reward": 0.46875,
      "reward_std": 0.0883883461356163,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.46875,
      "step": 1888
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 123.0,
      "epoch": 7.556,
      "grad_norm": 38.33015060424805,
      "kl": 0.2431640625,
      "learning_rate": 2.444e-07,
      "loss": 0.0002,
      "reward": 0.34375,
      "reward_std": 0.3966485261917114,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.34375,
      "step": 1889
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 144.96875,
      "epoch": 7.5600000000000005,
      "grad_norm": 0.008862972259521484,
      "kl": 0.20703125,
      "learning_rate": 2.4399999999999996e-07,
      "loss": 0.0002,
      "reward": 0.5,
      "reward_std": 0.0,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.5,
      "step": 1890
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 138.0,
      "epoch": 7.564,
      "grad_norm": 9.667668342590332,
      "kl": 0.2529296875,
      "learning_rate": 2.436e-07,
      "loss": 0.0003,
      "reward": 0.5,
      "reward_std": 0.1767766922712326,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.5,
      "step": 1891
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 145.8125,
      "epoch": 7.568,
      "grad_norm": 4.47056770324707,
      "kl": 0.24609375,
      "learning_rate": 2.432e-07,
      "loss": 0.0002,
      "reward": 0.28125,
      "reward_std": 0.0883883461356163,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.28125,
      "step": 1892
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 129.625,
      "epoch": 7.572,
      "grad_norm": 0.05386551842093468,
      "kl": 0.341796875,
      "learning_rate": 2.428e-07,
      "loss": 0.0003,
      "reward": 0.75,
      "reward_std": 0.0,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.75,
      "step": 1893
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 128.71875,
      "epoch": 7.576,
      "grad_norm": 0.010915116406977177,
      "kl": 0.24853515625,
      "learning_rate": 2.424e-07,
      "loss": 0.0002,
      "reward": 0.75,
      "reward_std": 0.0,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.75,
      "step": 1894
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 122.9375,
      "epoch": 7.58,
      "grad_norm": 0.011938932351768017,
      "kl": 0.27978515625,
      "learning_rate": 2.4199999999999997e-07,
      "loss": 0.0003,
      "reward": 1.0,
      "reward_std": 0.0,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 1.0,
      "step": 1895
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 139.625,
      "epoch": 7.584,
      "grad_norm": 0.03858279809355736,
      "kl": 0.27587890625,
      "learning_rate": 2.416e-07,
      "loss": 0.0003,
      "reward": 0.0,
      "reward_std": 0.0,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.0,
      "step": 1896
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 145.75,
      "epoch": 7.588,
      "grad_norm": 0.01846129633486271,
      "kl": 0.226318359375,
      "learning_rate": 2.4119999999999996e-07,
      "loss": 0.0002,
      "reward": 1.0,
      "reward_std": 0.0,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 1.0,
      "step": 1897
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 149.6875,
      "epoch": 7.592,
      "grad_norm": 0.018052654340863228,
      "kl": 0.226318359375,
      "learning_rate": 2.408e-07,
      "loss": 0.0002,
      "reward": 0.25,
      "reward_std": 0.0,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.25,
      "step": 1898
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 141.5625,
      "epoch": 7.596,
      "grad_norm": 0.013747518882155418,
      "kl": 0.23193359375,
      "learning_rate": 2.404e-07,
      "loss": 0.0002,
      "reward": 0.75,
      "reward_std": 0.0,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.75,
      "step": 1899
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 137.59375,
      "epoch": 7.6,
      "grad_norm": 5.019984722137451,
      "kl": 0.254150390625,
      "learning_rate": 2.4e-07,
      "loss": 0.0003,
      "reward": 0.96875,
      "reward_std": 0.0883883461356163,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.96875,
      "step": 1900
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 129.21875,
      "epoch": 7.604,
      "grad_norm": 3.4700429439544678,
      "kl": 0.281494140625,
      "learning_rate": 2.396e-07,
      "loss": 0.0003,
      "reward": 0.96875,
      "reward_std": 0.0883883461356163,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.96875,
      "step": 1901
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 139.15625,
      "epoch": 7.608,
      "grad_norm": 12.041154861450195,
      "kl": 0.27197265625,
      "learning_rate": 2.3919999999999997e-07,
      "loss": 0.0003,
      "reward": 0.59375,
      "reward_std": 0.1293872892856598,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.59375,
      "step": 1902
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 131.0625,
      "epoch": 7.612,
      "grad_norm": 3.186065912246704,
      "kl": 0.26708984375,
      "learning_rate": 2.388e-07,
      "loss": 0.0003,
      "reward": 0.5625,
      "reward_std": 0.1157275140285492,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.5625,
      "step": 1903
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 152.25,
      "epoch": 7.616,
      "grad_norm": 14.463269233703613,
      "kl": 0.222412109375,
      "learning_rate": 2.384e-07,
      "loss": 0.0002,
      "reward": 0.6875,
      "reward_std": 0.1157275140285492,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.6875,
      "step": 1904
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 138.75,
      "epoch": 7.62,
      "grad_norm": 0.01525117177516222,
      "kl": 0.27490234375,
      "learning_rate": 2.38e-07,
      "loss": 0.0003,
      "reward": 0.75,
      "reward_std": 0.0,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.75,
      "step": 1905
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 125.875,
      "epoch": 7.624,
      "grad_norm": 4.901587009429932,
      "kl": 0.21875,
      "learning_rate": 2.3759999999999998e-07,
      "loss": 0.0002,
      "reward": 0.65625,
      "reward_std": 0.1293872892856598,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.65625,
      "step": 1906
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 117.5625,
      "epoch": 7.628,
      "grad_norm": 0.06975147128105164,
      "kl": 0.30029296875,
      "learning_rate": 2.3719999999999998e-07,
      "loss": 0.0003,
      "reward": 0.75,
      "reward_std": 0.0,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.75,
      "step": 1907
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 129.15625,
      "epoch": 7.632,
      "grad_norm": 0.026855183765292168,
      "kl": 0.262939453125,
      "learning_rate": 2.368e-07,
      "loss": 0.0003,
      "reward": 0.5,
      "reward_std": 0.0,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.5,
      "step": 1908
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 157.25,
      "epoch": 7.636,
      "grad_norm": 0.0099476994946599,
      "kl": 0.221923828125,
      "learning_rate": 2.364e-07,
      "loss": 0.0002,
      "reward": 1.0,
      "reward_std": 0.0,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 1.0,
      "step": 1909
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 144.625,
      "epoch": 7.64,
      "grad_norm": 3.703361749649048,
      "kl": 0.28857421875,
      "learning_rate": 2.3599999999999997e-07,
      "loss": 0.0003,
      "reward": 0.375,
      "reward_std": 0.13363061845302582,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.375,
      "step": 1910
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 114.78125,
      "epoch": 7.644,
      "grad_norm": 0.016570964828133583,
      "kl": 0.2587890625,
      "learning_rate": 2.356e-07,
      "loss": 0.0003,
      "reward": 0.5,
      "reward_std": 0.0,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.5,
      "step": 1911
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 139.125,
      "epoch": 7.648,
      "grad_norm": 0.02191745862364769,
      "kl": 0.240478515625,
      "learning_rate": 2.352e-07,
      "loss": 0.0002,
      "reward": 0.25,
      "reward_std": 0.0,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.25,
      "step": 1912
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 140.25,
      "epoch": 7.652,
      "grad_norm": 14.985285758972168,
      "kl": 0.24365234375,
      "learning_rate": 2.3479999999999998e-07,
      "loss": 0.0002,
      "reward": 0.40625,
      "reward_std": 0.24511480331420898,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.40625,
      "step": 1913
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 143.8125,
      "epoch": 7.656,
      "grad_norm": 7.7383246421813965,
      "kl": 0.23486328125,
      "learning_rate": 2.3439999999999998e-07,
      "loss": 0.0002,
      "reward": 0.28125,
      "reward_std": 0.0883883461356163,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.28125,
      "step": 1914
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 134.21875,
      "epoch": 7.66,
      "grad_norm": 0.0742357075214386,
      "kl": 0.253173828125,
      "learning_rate": 2.34e-07,
      "loss": 0.0003,
      "reward": 1.0,
      "reward_std": 0.0,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 1.0,
      "step": 1915
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 155.125,
      "epoch": 7.664,
      "grad_norm": 7.616150379180908,
      "kl": 0.213134765625,
      "learning_rate": 2.336e-07,
      "loss": 0.0002,
      "reward": 0.375,
      "reward_std": 0.13363061845302582,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.375,
      "step": 1916
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 145.78125,
      "epoch": 7.668,
      "grad_norm": 0.01790645904839039,
      "kl": 0.258544921875,
      "learning_rate": 2.3319999999999997e-07,
      "loss": 0.0003,
      "reward": 0.5,
      "reward_std": 0.0,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.5,
      "step": 1917
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 121.75,
      "epoch": 7.672,
      "grad_norm": 0.029581710696220398,
      "kl": 0.29736328125,
      "learning_rate": 2.328e-07,
      "loss": 0.0003,
      "reward": 0.5,
      "reward_std": 0.0,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.5,
      "step": 1918
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 149.75,
      "epoch": 7.676,
      "grad_norm": 0.00842868909239769,
      "kl": 0.236328125,
      "learning_rate": 2.324e-07,
      "loss": 0.0002,
      "reward": 0.75,
      "reward_std": 0.0,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.75,
      "step": 1919
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 123.40625,
      "epoch": 7.68,
      "grad_norm": 0.020408790558576584,
      "kl": 0.2958984375,
      "learning_rate": 2.32e-07,
      "loss": 0.0003,
      "reward": 0.5,
      "reward_std": 0.0,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.5,
      "step": 1920
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 138.1875,
      "epoch": 7.684,
      "grad_norm": 9.475417137145996,
      "kl": 0.253662109375,
      "learning_rate": 2.3159999999999998e-07,
      "loss": 0.0003,
      "reward": 0.40625,
      "reward_std": 0.2041158601641655,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.40625,
      "step": 1921
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 142.0,
      "epoch": 7.688,
      "grad_norm": 0.008864258415997028,
      "kl": 0.22021484375,
      "learning_rate": 2.3119999999999998e-07,
      "loss": 0.0002,
      "reward": 0.75,
      "reward_std": 0.0,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.75,
      "step": 1922
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 147.15625,
      "epoch": 7.692,
      "grad_norm": 0.008828671649098396,
      "kl": 0.2353515625,
      "learning_rate": 2.308e-07,
      "loss": 0.0002,
      "reward": 0.75,
      "reward_std": 0.0,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.75,
      "step": 1923
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 142.78125,
      "epoch": 7.696,
      "grad_norm": 0.010452870279550552,
      "kl": 0.26953125,
      "learning_rate": 2.3039999999999997e-07,
      "loss": 0.0003,
      "reward": 0.5,
      "reward_std": 0.0,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.5,
      "step": 1924
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 162.0,
      "epoch": 7.7,
      "grad_norm": 0.019362686201930046,
      "kl": 0.213134765625,
      "learning_rate": 2.3e-07,
      "loss": 0.0002,
      "reward": 0.5,
      "reward_std": 0.0,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.5,
      "step": 1925
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 147.875,
      "epoch": 7.704,
      "grad_norm": 0.014581533148884773,
      "kl": 0.236328125,
      "learning_rate": 2.296e-07,
      "loss": 0.0002,
      "reward": 0.75,
      "reward_std": 0.0,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.75,
      "step": 1926
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 158.28125,
      "epoch": 7.708,
      "grad_norm": 4.2602763175964355,
      "kl": 0.20947265625,
      "learning_rate": 2.292e-07,
      "loss": 0.0002,
      "reward": 0.6875,
      "reward_std": 0.1157275140285492,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.6875,
      "step": 1927
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 139.375,
      "epoch": 7.712,
      "grad_norm": 5.5056867599487305,
      "kl": 0.232666015625,
      "learning_rate": 2.2879999999999998e-07,
      "loss": 0.0002,
      "reward": 0.5625,
      "reward_std": 0.1157275140285492,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.5625,
      "step": 1928
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 157.4375,
      "epoch": 7.716,
      "grad_norm": 0.037104617804288864,
      "kl": 0.24609375,
      "learning_rate": 2.2839999999999998e-07,
      "loss": 0.0002,
      "reward": 0.25,
      "reward_std": 0.0,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.25,
      "step": 1929
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 149.9375,
      "epoch": 7.72,
      "grad_norm": 5.277840614318848,
      "kl": 0.225830078125,
      "learning_rate": 2.28e-07,
      "loss": 0.0002,
      "reward": 0.28125,
      "reward_std": 0.2041158601641655,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.28125,
      "step": 1930
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 148.5,
      "epoch": 7.724,
      "grad_norm": 2.965268611907959,
      "kl": 0.2080078125,
      "learning_rate": 2.2759999999999997e-07,
      "loss": 0.0002,
      "reward": 0.46875,
      "reward_std": 0.0883883461356163,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.46875,
      "step": 1931
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 144.34375,
      "epoch": 7.728,
      "grad_norm": 8.214702606201172,
      "kl": 0.199462890625,
      "learning_rate": 2.272e-07,
      "loss": 0.0002,
      "reward": 0.40625,
      "reward_std": 0.3061639815568924,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.40625,
      "step": 1932
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 128.03125,
      "epoch": 7.732,
      "grad_norm": 5.006229877471924,
      "kl": 0.256103515625,
      "learning_rate": 2.268e-07,
      "loss": 0.0003,
      "reward": 0.5625,
      "reward_std": 0.2177756354212761,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.5625,
      "step": 1933
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 151.65625,
      "epoch": 7.736,
      "grad_norm": 0.036480680108070374,
      "kl": 0.2841796875,
      "learning_rate": 2.264e-07,
      "loss": 0.0003,
      "reward": 0.75,
      "reward_std": 0.0,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.75,
      "step": 1934
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 141.71875,
      "epoch": 7.74,
      "grad_norm": 0.04855004698038101,
      "kl": 0.244140625,
      "learning_rate": 2.2599999999999999e-07,
      "loss": 0.0002,
      "reward": 0.5,
      "reward_std": 0.0,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.5,
      "step": 1935
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 126.9375,
      "epoch": 7.744,
      "grad_norm": 60.03561019897461,
      "kl": 0.23974609375,
      "learning_rate": 2.2559999999999998e-07,
      "loss": 0.0002,
      "reward": 0.53125,
      "reward_std": 0.0883883461356163,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.53125,
      "step": 1936
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 147.75,
      "epoch": 7.748,
      "grad_norm": 2.2336316108703613,
      "kl": 0.2255859375,
      "learning_rate": 2.252e-07,
      "loss": 0.0002,
      "reward": 0.53125,
      "reward_std": 0.0883883461356163,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.53125,
      "step": 1937
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 150.5625,
      "epoch": 7.752,
      "grad_norm": 0.016191931441426277,
      "kl": 0.235595703125,
      "learning_rate": 2.248e-07,
      "loss": 0.0002,
      "reward": 0.5,
      "reward_std": 0.0,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.5,
      "step": 1938
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 157.9375,
      "epoch": 7.756,
      "grad_norm": 10.064260482788086,
      "kl": 0.201171875,
      "learning_rate": 2.2439999999999997e-07,
      "loss": 0.0002,
      "reward": 0.6875,
      "reward_std": 0.1157275140285492,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.6875,
      "step": 1939
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 138.53125,
      "epoch": 7.76,
      "grad_norm": 4.292687892913818,
      "kl": 0.274169921875,
      "learning_rate": 2.24e-07,
      "loss": 0.0003,
      "reward": 0.53125,
      "reward_std": 0.0883883461356163,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.53125,
      "step": 1940
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 157.1875,
      "epoch": 7.764,
      "grad_norm": 3.657590866088867,
      "kl": 0.21484375,
      "learning_rate": 2.236e-07,
      "loss": 0.0002,
      "reward": 0.8125,
      "reward_std": 0.1157275140285492,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.8125,
      "step": 1941
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 157.21875,
      "epoch": 7.768,
      "grad_norm": 0.012756803072988987,
      "kl": 0.218994140625,
      "learning_rate": 2.232e-07,
      "loss": 0.0002,
      "reward": 0.5,
      "reward_std": 0.0,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.5,
      "step": 1942
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 125.5,
      "epoch": 7.772,
      "grad_norm": 4.151922702789307,
      "kl": 0.2666015625,
      "learning_rate": 2.2279999999999998e-07,
      "loss": 0.0003,
      "reward": 0.8125,
      "reward_std": 0.1157275140285492,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.8125,
      "step": 1943
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 161.125,
      "epoch": 7.776,
      "grad_norm": 1.983794093132019,
      "kl": 0.270751953125,
      "learning_rate": 2.2239999999999998e-07,
      "loss": 0.0003,
      "reward": 0.28125,
      "reward_std": 0.0883883461356163,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.28125,
      "step": 1944
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 140.34375,
      "epoch": 7.78,
      "grad_norm": 24.126968383789062,
      "kl": 0.2373046875,
      "learning_rate": 2.22e-07,
      "loss": 0.0002,
      "reward": 0.875,
      "reward_std": 0.13363061845302582,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.875,
      "step": 1945
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 153.5625,
      "epoch": 7.784,
      "grad_norm": 0.018322832882404327,
      "kl": 0.21923828125,
      "learning_rate": 2.2159999999999997e-07,
      "loss": 0.0002,
      "reward": 0.75,
      "reward_std": 0.0,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.75,
      "step": 1946
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 165.25,
      "epoch": 7.788,
      "grad_norm": 3.5054235458374023,
      "kl": 0.21826171875,
      "learning_rate": 2.212e-07,
      "loss": 0.0002,
      "reward": 0.09375,
      "reward_std": 0.1293872892856598,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.09375,
      "step": 1947
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 152.53125,
      "epoch": 7.792,
      "grad_norm": 4.478523254394531,
      "kl": 0.22314453125,
      "learning_rate": 2.208e-07,
      "loss": 0.0002,
      "reward": 0.78125,
      "reward_std": 0.0883883461356163,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.78125,
      "step": 1948
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 144.21875,
      "epoch": 7.796,
      "grad_norm": 0.09022127836942673,
      "kl": 0.266845703125,
      "learning_rate": 2.2040000000000001e-07,
      "loss": 0.0003,
      "reward": 0.25,
      "reward_std": 0.0,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.25,
      "step": 1949
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 148.90625,
      "epoch": 7.8,
      "grad_norm": 2.8682868480682373,
      "kl": 0.22265625,
      "learning_rate": 2.1999999999999998e-07,
      "loss": 0.0002,
      "reward": 0.3125,
      "reward_std": 0.1157275140285492,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.3125,
      "step": 1950
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 164.53125,
      "epoch": 7.804,
      "grad_norm": 0.01168039720505476,
      "kl": 0.203369140625,
      "learning_rate": 2.1959999999999998e-07,
      "loss": 0.0002,
      "reward": 0.25,
      "reward_std": 0.0,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.25,
      "step": 1951
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 144.21875,
      "epoch": 7.808,
      "grad_norm": 0.03067288175225258,
      "kl": 0.2578125,
      "learning_rate": 2.192e-07,
      "loss": 0.0003,
      "reward": 0.75,
      "reward_std": 0.0,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.75,
      "step": 1952
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 160.1875,
      "epoch": 7.812,
      "grad_norm": 0.01002495177090168,
      "kl": 0.228515625,
      "learning_rate": 2.1879999999999997e-07,
      "loss": 0.0002,
      "reward": 0.25,
      "reward_std": 0.0,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.25,
      "step": 1953
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 149.96875,
      "epoch": 7.816,
      "grad_norm": 4.154101848602295,
      "kl": 0.228515625,
      "learning_rate": 2.184e-07,
      "loss": 0.0002,
      "reward": 0.71875,
      "reward_std": 0.0883883461356163,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.71875,
      "step": 1954
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 144.84375,
      "epoch": 7.82,
      "grad_norm": 5.7108378410339355,
      "kl": 0.2138671875,
      "learning_rate": 2.18e-07,
      "loss": 0.0002,
      "reward": 0.4375,
      "reward_std": 0.2177756354212761,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.4375,
      "step": 1955
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 152.125,
      "epoch": 7.824,
      "grad_norm": 7.467862129211426,
      "kl": 0.219970703125,
      "learning_rate": 2.176e-07,
      "loss": 0.0002,
      "reward": 0.78125,
      "reward_std": 0.0883883461356163,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.78125,
      "step": 1956
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 132.40625,
      "epoch": 7.828,
      "grad_norm": 12.400018692016602,
      "kl": 0.2880859375,
      "learning_rate": 2.1719999999999999e-07,
      "loss": 0.0003,
      "reward": 0.46875,
      "reward_std": 0.0883883461356163,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.46875,
      "step": 1957
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 158.625,
      "epoch": 7.832,
      "grad_norm": 0.012539033778011799,
      "kl": 0.231689453125,
      "learning_rate": 2.1679999999999998e-07,
      "loss": 0.0002,
      "reward": 1.0,
      "reward_std": 0.0,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 1.0,
      "step": 1958
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 130.9375,
      "epoch": 7.836,
      "grad_norm": 0.029422098770737648,
      "kl": 0.23388671875,
      "learning_rate": 2.164e-07,
      "loss": 0.0002,
      "reward": 0.75,
      "reward_std": 0.0,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.75,
      "step": 1959
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 170.375,
      "epoch": 7.84,
      "grad_norm": 0.009244058281183243,
      "kl": 0.1982421875,
      "learning_rate": 2.1599999999999998e-07,
      "loss": 0.0002,
      "reward": 0.25,
      "reward_std": 0.0,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.25,
      "step": 1960
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 182.90625,
      "epoch": 7.844,
      "grad_norm": 3.6143999099731445,
      "kl": 0.1953125,
      "learning_rate": 2.156e-07,
      "loss": 0.0002,
      "reward": 0.71875,
      "reward_std": 0.0883883461356163,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.71875,
      "step": 1961
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 152.9375,
      "epoch": 7.848,
      "grad_norm": 0.10127752274274826,
      "kl": 0.344970703125,
      "learning_rate": 2.152e-07,
      "loss": 0.0003,
      "reward": 0.25,
      "reward_std": 0.0,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.25,
      "step": 1962
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 158.96875,
      "epoch": 7.852,
      "grad_norm": 4.056572914123535,
      "kl": 0.212890625,
      "learning_rate": 2.148e-07,
      "loss": 0.0002,
      "reward": 0.0625,
      "reward_std": 0.1767766922712326,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.0625,
      "step": 1963
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 139.84375,
      "epoch": 7.856,
      "grad_norm": 3.686461925506592,
      "kl": 0.257568359375,
      "learning_rate": 2.144e-07,
      "loss": 0.0003,
      "reward": 0.96875,
      "reward_std": 0.0883883461356163,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.96875,
      "step": 1964
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 146.46875,
      "epoch": 7.86,
      "grad_norm": 0.01108139380812645,
      "kl": 0.235595703125,
      "learning_rate": 2.1399999999999998e-07,
      "loss": 0.0002,
      "reward": 0.5,
      "reward_std": 0.0,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.5,
      "step": 1965
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 148.125,
      "epoch": 7.864,
      "grad_norm": 5.642823219299316,
      "kl": 0.216796875,
      "learning_rate": 2.136e-07,
      "loss": 0.0002,
      "reward": 0.0625,
      "reward_std": 0.1767766922712326,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.0625,
      "step": 1966
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 163.1875,
      "epoch": 7.868,
      "grad_norm": 0.01909429021179676,
      "kl": 0.22705078125,
      "learning_rate": 2.132e-07,
      "loss": 0.0002,
      "reward": 0.75,
      "reward_std": 0.0,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.75,
      "step": 1967
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 163.9375,
      "epoch": 7.872,
      "grad_norm": 3.205712080001831,
      "kl": 0.23095703125,
      "learning_rate": 2.1279999999999997e-07,
      "loss": 0.0002,
      "reward": 0.46875,
      "reward_std": 0.0883883461356163,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.46875,
      "step": 1968
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 158.3125,
      "epoch": 7.876,
      "grad_norm": 4.996670246124268,
      "kl": 0.2431640625,
      "learning_rate": 2.124e-07,
      "loss": 0.0002,
      "reward": 0.625,
      "reward_std": 0.2177756354212761,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.625,
      "step": 1969
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 126.3125,
      "epoch": 7.88,
      "grad_norm": 9.479227066040039,
      "kl": 0.229248046875,
      "learning_rate": 2.12e-07,
      "loss": 0.0002,
      "reward": 0.59375,
      "reward_std": 0.2041158601641655,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.59375,
      "step": 1970
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 140.125,
      "epoch": 7.884,
      "grad_norm": 4.485763072967529,
      "kl": 0.2607421875,
      "learning_rate": 2.116e-07,
      "loss": 0.0003,
      "reward": 0.09375,
      "reward_std": 0.2041158601641655,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.09375,
      "step": 1971
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 160.34375,
      "epoch": 7.888,
      "grad_norm": 0.025699200108647346,
      "kl": 0.224609375,
      "learning_rate": 2.1119999999999999e-07,
      "loss": 0.0002,
      "reward": 0.75,
      "reward_std": 0.0,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.75,
      "step": 1972
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 147.96875,
      "epoch": 7.892,
      "grad_norm": 5.059063911437988,
      "kl": 0.21630859375,
      "learning_rate": 2.1079999999999998e-07,
      "loss": 0.0002,
      "reward": 0.4375,
      "reward_std": 0.1157275140285492,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.4375,
      "step": 1973
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 157.84375,
      "epoch": 7.896,
      "grad_norm": 0.009619446471333504,
      "kl": 0.2255859375,
      "learning_rate": 2.104e-07,
      "loss": 0.0002,
      "reward": 0.5,
      "reward_std": 0.0,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.5,
      "step": 1974
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 135.53125,
      "epoch": 7.9,
      "grad_norm": 16.93857192993164,
      "kl": 0.25927734375,
      "learning_rate": 2.0999999999999997e-07,
      "loss": 0.0003,
      "reward": 0.75,
      "reward_std": 0.2587745785713196,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.75,
      "step": 1975
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 139.21875,
      "epoch": 7.904,
      "grad_norm": 2.993839979171753,
      "kl": 0.238525390625,
      "learning_rate": 2.096e-07,
      "loss": 0.0002,
      "reward": 0.78125,
      "reward_std": 0.0883883461356163,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.78125,
      "step": 1976
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 153.25,
      "epoch": 7.908,
      "grad_norm": 3.5786585807800293,
      "kl": 0.24609375,
      "learning_rate": 2.092e-07,
      "loss": 0.0002,
      "reward": 0.28125,
      "reward_std": 0.0883883461356163,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.28125,
      "step": 1977
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 162.5625,
      "epoch": 7.912,
      "grad_norm": 0.00976133719086647,
      "kl": 0.192626953125,
      "learning_rate": 2.0880000000000002e-07,
      "loss": 0.0002,
      "reward": 0.5,
      "reward_std": 0.0,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.5,
      "step": 1978
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 148.53125,
      "epoch": 7.916,
      "grad_norm": 10.226258277893066,
      "kl": 0.242431640625,
      "learning_rate": 2.0839999999999999e-07,
      "loss": 0.0002,
      "reward": 0.625,
      "reward_std": 0.2177756354212761,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.625,
      "step": 1979
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 136.9375,
      "epoch": 7.92,
      "grad_norm": 2.807443618774414,
      "kl": 0.27587890625,
      "learning_rate": 2.0799999999999998e-07,
      "loss": 0.0003,
      "reward": 0.78125,
      "reward_std": 0.0883883461356163,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.78125,
      "step": 1980
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 162.5,
      "epoch": 7.924,
      "grad_norm": 0.011735876090824604,
      "kl": 0.22314453125,
      "learning_rate": 2.076e-07,
      "loss": 0.0002,
      "reward": 0.25,
      "reward_std": 0.0,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.25,
      "step": 1981
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 159.78125,
      "epoch": 7.928,
      "grad_norm": 0.01989402063190937,
      "kl": 0.222900390625,
      "learning_rate": 2.0719999999999998e-07,
      "loss": 0.0002,
      "reward": 0.75,
      "reward_std": 0.0,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.75,
      "step": 1982
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 153.78125,
      "epoch": 7.932,
      "grad_norm": 6.905916213989258,
      "kl": 0.214111328125,
      "learning_rate": 2.068e-07,
      "loss": 0.0002,
      "reward": 0.34375,
      "reward_std": 0.1293872892856598,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.34375,
      "step": 1983
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 134.15625,
      "epoch": 7.936,
      "grad_norm": 0.013006472028791904,
      "kl": 0.27783203125,
      "learning_rate": 2.064e-07,
      "loss": 0.0003,
      "reward": 0.25,
      "reward_std": 0.0,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.25,
      "step": 1984
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 137.0,
      "epoch": 7.9399999999999995,
      "grad_norm": 4.519454479217529,
      "kl": 0.24853515625,
      "learning_rate": 2.06e-07,
      "loss": 0.0002,
      "reward": 0.53125,
      "reward_std": 0.0883883461356163,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.53125,
      "step": 1985
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 155.09375,
      "epoch": 7.944,
      "grad_norm": 0.0074937655590474606,
      "kl": 0.22021484375,
      "learning_rate": 2.056e-07,
      "loss": 0.0002,
      "reward": 0.5,
      "reward_std": 0.0,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.5,
      "step": 1986
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 151.4375,
      "epoch": 7.948,
      "grad_norm": 0.06854574382305145,
      "kl": 0.260986328125,
      "learning_rate": 2.0519999999999998e-07,
      "loss": 0.0003,
      "reward": 0.75,
      "reward_std": 0.0,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.75,
      "step": 1987
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 149.5,
      "epoch": 7.952,
      "grad_norm": 0.011434744112193584,
      "kl": 0.2255859375,
      "learning_rate": 2.048e-07,
      "loss": 0.0002,
      "reward": 0.5,
      "reward_std": 0.0,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.5,
      "step": 1988
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 140.1875,
      "epoch": 7.9559999999999995,
      "grad_norm": 6.404064178466797,
      "kl": 0.27734375,
      "learning_rate": 2.0439999999999998e-07,
      "loss": 0.0003,
      "reward": 0.90625,
      "reward_std": 0.1293872892856598,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.90625,
      "step": 1989
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 143.8125,
      "epoch": 7.96,
      "grad_norm": 0.01260036788880825,
      "kl": 0.23681640625,
      "learning_rate": 2.0399999999999997e-07,
      "loss": 0.0002,
      "reward": 0.5,
      "reward_std": 0.0,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.5,
      "step": 1990
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 147.65625,
      "epoch": 7.964,
      "grad_norm": 3.680248975753784,
      "kl": 0.243896484375,
      "learning_rate": 2.036e-07,
      "loss": 0.0002,
      "reward": 0.46875,
      "reward_std": 0.0883883461356163,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.46875,
      "step": 1991
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 157.09375,
      "epoch": 7.968,
      "grad_norm": 4.196675777435303,
      "kl": 0.219482421875,
      "learning_rate": 2.032e-07,
      "loss": 0.0002,
      "reward": 0.09375,
      "reward_std": 0.1293872892856598,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.09375,
      "step": 1992
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 151.71875,
      "epoch": 7.9719999999999995,
      "grad_norm": 5.907729148864746,
      "kl": 0.249267578125,
      "learning_rate": 2.028e-07,
      "loss": 0.0002,
      "reward": 0.34375,
      "reward_std": 0.3061639815568924,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.34375,
      "step": 1993
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 138.4375,
      "epoch": 7.976,
      "grad_norm": 3.481532335281372,
      "kl": 0.2724609375,
      "learning_rate": 2.0239999999999999e-07,
      "loss": 0.0003,
      "reward": 0.78125,
      "reward_std": 0.0883883461356163,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.78125,
      "step": 1994
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 155.25,
      "epoch": 7.98,
      "grad_norm": 5.639888763427734,
      "kl": 0.246337890625,
      "learning_rate": 2.02e-07,
      "loss": 0.0002,
      "reward": 0.53125,
      "reward_std": 0.2630179077386856,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.53125,
      "step": 1995
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 147.5,
      "epoch": 7.984,
      "grad_norm": 5.939465522766113,
      "kl": 0.23876953125,
      "learning_rate": 2.016e-07,
      "loss": 0.0002,
      "reward": 0.40625,
      "reward_std": 0.22201896458864212,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.40625,
      "step": 1996
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 168.53125,
      "epoch": 7.9879999999999995,
      "grad_norm": 6.994452476501465,
      "kl": 0.2109375,
      "learning_rate": 2.0119999999999998e-07,
      "loss": 0.0002,
      "reward": 0.6875,
      "reward_std": 0.2925042062997818,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.6875,
      "step": 1997
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 149.5,
      "epoch": 7.992,
      "grad_norm": 6.502929210662842,
      "kl": 0.23974609375,
      "learning_rate": 2.008e-07,
      "loss": 0.0002,
      "reward": 0.5625,
      "reward_std": 0.1157275140285492,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.5625,
      "step": 1998
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 161.875,
      "epoch": 7.996,
      "grad_norm": 0.008142419159412384,
      "kl": 0.1962890625,
      "learning_rate": 2.004e-07,
      "loss": 0.0002,
      "reward": 0.5,
      "reward_std": 0.0,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.5,
      "step": 1999
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 177.125,
      "epoch": 8.0,
      "grad_norm": 10.10364818572998,
      "kl": 0.1796875,
      "learning_rate": 2e-07,
      "loss": 0.0002,
      "reward": 0.40625,
      "reward_std": 0.1293872892856598,
      "rewards/format_reward": 0.0,
      "rewards/score_reward": 0.40625,
      "step": 2000
    }
  ],
  "logging_steps": 1.0,
  "max_steps": 2500,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 10,
  "save_steps": 500,
  "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
}