{
  "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": 614.1698936462402,
      "epoch": 0.08528784648187633,
      "grad_norm": 1.6420103311538696,
      "kl": 0.00021245479583740235,
      "learning_rate": 2.5e-06,
      "loss": 0.0,
      "reward": 0.6379464596509934,
      "reward_std": 0.3210102315992117,
      "rewards/accuracy_reward": 0.6375000312924385,
      "rewards/format_reward": 0.00044642859138548373,
      "step": 5
    },
    {
      "completion_length": 624.264535522461,
      "epoch": 0.17057569296375266,
      "grad_norm": 1526.457275390625,
      "kl": 0.14271965026855468,
      "learning_rate": 2.956412726139078e-06,
      "loss": 0.0057,
      "reward": 0.6959821730852127,
      "reward_std": 0.28622329905629157,
      "rewards/accuracy_reward": 0.6955357447266579,
      "rewards/format_reward": 0.00044642859138548373,
      "step": 10
    },
    {
      "completion_length": 616.0730186462403,
      "epoch": 0.255863539445629,
      "grad_norm": 0.11626364290714264,
      "kl": 0.0034526824951171876,
      "learning_rate": 2.7836719084521715e-06,
      "loss": 0.0001,
      "reward": 0.7558036059141159,
      "reward_std": 0.22891067173331975,
      "rewards/accuracy_reward": 0.7555803924798965,
      "rewards/format_reward": 0.00022321429569274187,
      "step": 15
    },
    {
      "completion_length": 602.7257011413574,
      "epoch": 0.3411513859275053,
      "grad_norm": 0.3678954541683197,
      "kl": 0.0035373687744140623,
      "learning_rate": 2.4946839873611927e-06,
      "loss": 0.0001,
      "reward": 0.7736607491970062,
      "reward_std": 0.1879386292770505,
      "rewards/accuracy_reward": 0.7736607491970062,
      "rewards/format_reward": 0.0,
      "step": 20
    },
    {
      "completion_length": 606.3076171875,
      "epoch": 0.42643923240938164,
      "grad_norm": 0.08185073733329773,
      "kl": 0.008475685119628906,
      "learning_rate": 2.1156192081791355e-06,
      "loss": 0.0003,
      "reward": 0.7618303924798966,
      "reward_std": 0.18782664239406585,
      "rewards/accuracy_reward": 0.7618303924798966,
      "rewards/format_reward": 0.0,
      "step": 25
    },
    {
      "completion_length": 615.8890892028809,
      "epoch": 0.511727078891258,
      "grad_norm": 0.1566159576177597,
      "kl": 0.0039794921875,
      "learning_rate": 1.6808050203829845e-06,
      "loss": 0.0002,
      "reward": 0.7493303924798965,
      "reward_std": 0.18519791485741735,
      "rewards/accuracy_reward": 0.7493303924798965,
      "rewards/format_reward": 0.0,
      "step": 30
    },
    {
      "completion_length": 610.3998023986817,
      "epoch": 0.5970149253731343,
      "grad_norm": 0.07447274774312973,
      "kl": 0.004097747802734375,
      "learning_rate": 1.2296174432791415e-06,
      "loss": 0.0002,
      "reward": 0.7495536029338836,
      "reward_std": 0.1825955007225275,
      "rewards/accuracy_reward": 0.7495536029338836,
      "rewards/format_reward": 0.0,
      "step": 35
    },
    {
      "completion_length": 595.3044914245605,
      "epoch": 0.6823027718550106,
      "grad_norm": 0.07800718396902084,
      "kl": 0.0040496826171875,
      "learning_rate": 8.029152419343472e-07,
      "loss": 0.0002,
      "reward": 0.7689732506871223,
      "reward_std": 0.17793030026368797,
      "rewards/accuracy_reward": 0.7689732506871223,
      "rewards/format_reward": 0.0,
      "step": 40
    },
    {
      "completion_length": 614.7270347595215,
      "epoch": 0.767590618336887,
      "grad_norm": 0.07937229424715042,
      "kl": 0.004020309448242188,
      "learning_rate": 4.3933982822017883e-07,
      "loss": 0.0002,
      "reward": 0.7448661088943481,
      "reward_std": 0.20116904862225055,
      "rewards/accuracy_reward": 0.7446428954601287,
      "rewards/format_reward": 0.00022321429569274187,
      "step": 45
    },
    {
      "completion_length": 618.0094017028808,
      "epoch": 0.8528784648187633,
      "grad_norm": 0.074642114341259,
      "kl": 0.003514862060546875,
      "learning_rate": 1.718159615201853e-07,
      "loss": 0.0001,
      "reward": 0.743750037252903,
      "reward_std": 0.18902343986555933,
      "rewards/accuracy_reward": 0.743750037252903,
      "rewards/format_reward": 0.0,
      "step": 50
    },
    {
      "completion_length": 605.6493591308594,
      "epoch": 0.9381663113006397,
      "grad_norm": 0.08093011379241943,
      "kl": 0.0038265228271484376,
      "learning_rate": 2.4570139579284723e-08,
      "loss": 0.0002,
      "reward": 0.7801339656114579,
      "reward_std": 0.19253778588026763,
      "rewards/accuracy_reward": 0.7801339656114579,
      "rewards/format_reward": 0.0,
      "step": 55
    },
    {
      "completion_length": 596.0093282063802,
      "epoch": 0.9893390191897654,
      "kl": 0.003498077392578125,
      "reward": 0.7726934800545374,
      "reward_std": 0.18784288999934992,
      "rewards/accuracy_reward": 0.7726934800545374,
      "rewards/format_reward": 0.0,
      "step": 58,
      "total_flos": 0.0,
      "train_loss": 0.0014648210027220997,
      "train_runtime": 13165.6558,
      "train_samples_per_second": 0.57,
      "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
}