{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 0.9968051118210862,
  "eval_steps": 200,
  "global_step": 39,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "clip_ratio": 0.0,
      "completion_length": 629.1979370117188,
      "epoch": 0.025559105431309903,
      "grad_norm": 0.04843698814511299,
      "kl": 0.0,
      "learning_rate": 7.5e-07,
      "loss": 0.0176,
      "reward": 0.4609375074505806,
      "reward_std": 0.1878252374008298,
      "rewards/accuracy_reward": 0.4609375074505806,
      "rewards/format_reward": 0.0,
      "step": 1
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 624.3633089065552,
      "epoch": 0.12779552715654952,
      "grad_norm": 0.06356873363256454,
      "kl": 3.7126243114471436e-05,
      "learning_rate": 2.993961440992859e-06,
      "loss": 0.0145,
      "reward": 0.438151054084301,
      "reward_std": 0.24214724358171225,
      "rewards/accuracy_reward": 0.438151054084301,
      "rewards/format_reward": 0.0,
      "step": 5
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 609.4562683105469,
      "epoch": 0.25559105431309903,
      "grad_norm": 0.05450832471251488,
      "kl": 0.00019782781600952148,
      "learning_rate": 2.7876731904027993e-06,
      "loss": 0.0228,
      "reward": 0.45833334773778917,
      "reward_std": 0.23717540223151445,
      "rewards/accuracy_reward": 0.45833334773778917,
      "rewards/format_reward": 0.0,
      "step": 10
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 615.4484619140625,
      "epoch": 0.38338658146964855,
      "grad_norm": 0.05140135437250137,
      "kl": 0.0015062332153320313,
      "learning_rate": 2.3263454721781537e-06,
      "loss": 0.0178,
      "reward": 0.5182291828095913,
      "reward_std": 0.24527766648679972,
      "rewards/accuracy_reward": 0.5182291828095913,
      "rewards/format_reward": 0.0,
      "step": 15
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 617.7005401611328,
      "epoch": 0.5111821086261981,
      "grad_norm": 0.05119824782013893,
      "kl": 0.03265390396118164,
      "learning_rate": 1.7013498987264833e-06,
      "loss": 0.0193,
      "reward": 0.5333333484828472,
      "reward_std": 0.21802459191530943,
      "rewards/accuracy_reward": 0.5333333484828472,
      "rewards/format_reward": 0.0,
      "step": 20
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 617.8526245117188,
      "epoch": 0.6389776357827476,
      "grad_norm": 0.05785629153251648,
      "kl": 0.0034463882446289064,
      "learning_rate": 1.036474508437579e-06,
      "loss": 0.0234,
      "reward": 0.5197916813194752,
      "reward_std": 0.2239171512424946,
      "rewards/accuracy_reward": 0.5197916813194752,
      "rewards/format_reward": 0.0,
      "step": 25
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 626.4276252746582,
      "epoch": 0.7667731629392971,
      "grad_norm": 0.054110053926706314,
      "kl": 0.004987144470214843,
      "learning_rate": 4.63406026519703e-07,
      "loss": 0.0121,
      "reward": 0.4942708477377892,
      "reward_std": 0.22023430168628694,
      "rewards/accuracy_reward": 0.4942708477377892,
      "rewards/format_reward": 0.0,
      "step": 30
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 631.6297073364258,
      "epoch": 0.8945686900958466,
      "grad_norm": 0.05248409882187843,
      "kl": 0.005344772338867187,
      "learning_rate": 9.564769404039419e-08,
      "loss": 0.0117,
      "reward": 0.5307291805744171,
      "reward_std": 0.2246537147089839,
      "rewards/accuracy_reward": 0.5307291805744171,
      "rewards/format_reward": 0.0,
      "step": 35
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 631.3951969146729,
      "epoch": 0.9968051118210862,
      "kl": 0.0028176307678222656,
      "reward": 0.5507812704890966,
      "reward_std": 0.23570226412266493,
      "rewards/accuracy_reward": 0.5507812704890966,
      "rewards/format_reward": 0.0,
      "step": 39,
      "total_flos": 0.0,
      "train_loss": 0.015906815584271383,
      "train_runtime": 3267.2257,
      "train_samples_per_second": 2.296,
      "train_steps_per_second": 0.012
    }
  ],
  "logging_steps": 5,
  "max_steps": 39,
  "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": true,
        "should_training_stop": true
      },
      "attributes": {}
    }
  },
  "total_flos": 0.0,
  "train_batch_size": 16,
  "trial_name": null,
  "trial_params": null
}