{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 0.9893390191897654,
  "eval_steps": 100,
  "global_step": 58,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "completion_length": 605.8158782958984,
      "epoch": 0.08528784648187633,
      "grad_norm": 0.792238175868988,
      "kl": 0.00026590824127197265,
      "learning_rate": 2.5e-06,
      "loss": 0.0376,
      "reward": 0.6513393133878708,
      "reward_std": 0.3232782337814569,
      "rewards/accuracy_reward": 0.6511160984635354,
      "rewards/format_reward": 0.00022321429569274187,
      "step": 5
    },
    {
      "completion_length": 608.7732391357422,
      "epoch": 0.17057569296375266,
      "grad_norm": 4.090766906738281,
      "kl": 1868.8025625228881,
      "learning_rate": 2.956412726139078e-06,
      "loss": 92.3117,
      "reward": 0.7147321745753288,
      "reward_std": 0.27289656847715377,
      "rewards/accuracy_reward": 0.714285746216774,
      "rewards/format_reward": 0.00044642859138548373,
      "step": 10
    },
    {
      "completion_length": 596.5480163574218,
      "epoch": 0.255863539445629,
      "grad_norm": 0.19806408882141113,
      "kl": 0.0032785415649414064,
      "learning_rate": 2.7836719084521715e-06,
      "loss": 0.0645,
      "reward": 0.7618303894996643,
      "reward_std": 0.20630390886217356,
      "rewards/accuracy_reward": 0.7618303894996643,
      "rewards/format_reward": 0.0,
      "step": 15
    },
    {
      "completion_length": 579.8969009399414,
      "epoch": 0.3411513859275053,
      "grad_norm": 0.12550625205039978,
      "kl": 0.0029325485229492188,
      "learning_rate": 2.4946839873611927e-06,
      "loss": 0.0465,
      "reward": 0.7792411088943482,
      "reward_std": 0.18276286944746972,
      "rewards/accuracy_reward": 0.7792411088943482,
      "rewards/format_reward": 0.0,
      "step": 20
    },
    {
      "completion_length": 587.1710090637207,
      "epoch": 0.42643923240938164,
      "grad_norm": 0.18487855792045593,
      "kl": 0.004139328002929687,
      "learning_rate": 2.1156192081791355e-06,
      "loss": 0.0452,
      "reward": 0.7696428924798966,
      "reward_std": 0.17610765052959323,
      "rewards/accuracy_reward": 0.7696428924798966,
      "rewards/format_reward": 0.0,
      "step": 25
    },
    {
      "completion_length": 593.3131958007813,
      "epoch": 0.511727078891258,
      "grad_norm": 0.1600646674633026,
      "kl": 0.004147720336914062,
      "learning_rate": 1.6808050203829845e-06,
      "loss": 0.0428,
      "reward": 0.7513393238186836,
      "reward_std": 0.17458505025133492,
      "rewards/accuracy_reward": 0.7513393238186836,
      "rewards/format_reward": 0.0,
      "step": 30
    },
    {
      "completion_length": 582.7629692077637,
      "epoch": 0.5970149253731343,
      "grad_norm": 0.1672583967447281,
      "kl": 0.004720306396484375,
      "learning_rate": 1.2296174432791415e-06,
      "loss": 0.0355,
      "reward": 0.7491071790456771,
      "reward_std": 0.17336862673982978,
      "rewards/accuracy_reward": 0.7491071790456771,
      "rewards/format_reward": 0.0,
      "step": 35
    },
    {
      "completion_length": 575.660513305664,
      "epoch": 0.6823027718550106,
      "grad_norm": 0.1312212347984314,
      "kl": 0.004215621948242187,
      "learning_rate": 8.029152419343472e-07,
      "loss": 0.0387,
      "reward": 0.7794643193483353,
      "reward_std": 0.16920664254575968,
      "rewards/accuracy_reward": 0.7794643193483353,
      "rewards/format_reward": 0.0,
      "step": 40
    },
    {
      "completion_length": 588.4040451049805,
      "epoch": 0.767590618336887,
      "grad_norm": 0.18345855176448822,
      "kl": 0.004205322265625,
      "learning_rate": 4.3933982822017883e-07,
      "loss": 0.036,
      "reward": 0.7571428939700127,
      "reward_std": 0.18678655978292227,
      "rewards/accuracy_reward": 0.7571428939700127,
      "rewards/format_reward": 0.0,
      "step": 45
    },
    {
      "completion_length": 590.6636367797852,
      "epoch": 0.8528784648187633,
      "grad_norm": 0.17137926816940308,
      "kl": 0.004085159301757813,
      "learning_rate": 1.718159615201853e-07,
      "loss": 0.0329,
      "reward": 0.759151816368103,
      "reward_std": 0.17406445033848286,
      "rewards/accuracy_reward": 0.759151816368103,
      "rewards/format_reward": 0.0,
      "step": 50
    },
    {
      "completion_length": 580.3337303161621,
      "epoch": 0.9381663113006397,
      "grad_norm": 0.3547234833240509,
      "kl": 0.0044841766357421875,
      "learning_rate": 2.4570139579284723e-08,
      "loss": 0.0341,
      "reward": 0.7834821850061416,
      "reward_std": 0.17811276931315662,
      "rewards/accuracy_reward": 0.7834821850061416,
      "rewards/format_reward": 0.0,
      "step": 55
    },
    {
      "completion_length": 572.3296394348145,
      "epoch": 0.9893390191897654,
      "kl": 0.0043417612711588545,
      "reward": 0.771205392976602,
      "reward_std": 0.1717855976894498,
      "rewards/accuracy_reward": 0.771205392976602,
      "rewards/format_reward": 0.0,
      "step": 58,
      "total_flos": 0.0,
      "train_loss": 7.9953470033561365,
      "train_runtime": 16497.7515,
      "train_samples_per_second": 0.455,
      "train_steps_per_second": 0.004
    }
  ],
  "logging_steps": 5,
  "max_steps": 58,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 1,
  "save_steps": 500,
  "stateful_callbacks": {
    "TrainerControl": {
      "args": {
        "should_epoch_stop": false,
        "should_evaluate": false,
        "should_log": false,
        "should_save": false,
        "should_training_stop": false
      },
      "attributes": {}
    }
  },
  "total_flos": 0.0,
  "train_batch_size": 16,
  "trial_name": null,
  "trial_params": null
}