{
  "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": 596.283203125,
      "epoch": 0.017057569296375266,
      "grad_norm": 0.3653121590614319,
      "kl": 0.0,
      "learning_rate": 5e-07,
      "loss": 0.0181,
      "reward": 0.638671875,
      "reward_std": 0.34863804280757904,
      "rewards/accuracy_reward": 0.638671875,
      "rewards/format_reward": 0.0,
      "step": 1
    },
    {
      "completion_length": 605.504150390625,
      "epoch": 0.08528784648187633,
      "grad_norm": 0.2778508961200714,
      "kl": 0.0001735985279083252,
      "learning_rate": 2.5e-06,
      "loss": 0.0283,
      "reward": 0.623046875,
      "reward_std": 0.34396721608936787,
      "rewards/accuracy_reward": 0.623046875,
      "rewards/format_reward": 0.0,
      "step": 5
    },
    {
      "completion_length": 606.71328125,
      "epoch": 0.17057569296375266,
      "grad_norm": 0.1984957903623581,
      "kl": 0.0018383979797363282,
      "learning_rate": 2.956412726139078e-06,
      "loss": 0.0561,
      "reward": 0.7173828125,
      "reward_std": 0.2600227637216449,
      "rewards/accuracy_reward": 0.7173828125,
      "rewards/format_reward": 0.0,
      "step": 10
    },
    {
      "completion_length": 583.884375,
      "epoch": 0.255863539445629,
      "grad_norm": 0.1335204690694809,
      "kl": 0.002769279479980469,
      "learning_rate": 2.7836719084521715e-06,
      "loss": 0.0528,
      "reward": 0.78046875,
      "reward_std": 0.18675954192876815,
      "rewards/accuracy_reward": 0.78046875,
      "rewards/format_reward": 0.0,
      "step": 15
    },
    {
      "completion_length": 571.166015625,
      "epoch": 0.3411513859275053,
      "grad_norm": 0.13243617117404938,
      "kl": 0.0032848358154296876,
      "learning_rate": 2.4946839873611927e-06,
      "loss": 0.0401,
      "reward": 0.7884765625,
      "reward_std": 0.16892412528395653,
      "rewards/accuracy_reward": 0.7884765625,
      "rewards/format_reward": 0.0,
      "step": 20
    },
    {
      "completion_length": 587.1142578125,
      "epoch": 0.42643923240938164,
      "grad_norm": 0.37184375524520874,
      "kl": 0.0036285400390625,
      "learning_rate": 2.1156192081791355e-06,
      "loss": 0.037,
      "reward": 0.7716796875,
      "reward_std": 0.17590777329169213,
      "rewards/accuracy_reward": 0.7716796875,
      "rewards/format_reward": 0.0,
      "step": 25
    },
    {
      "completion_length": 590.3986328125,
      "epoch": 0.511727078891258,
      "grad_norm": 0.3420747220516205,
      "kl": 0.003810882568359375,
      "learning_rate": 1.6808050203829845e-06,
      "loss": 0.0306,
      "reward": 0.775390625,
      "reward_std": 0.1597107290290296,
      "rewards/accuracy_reward": 0.775390625,
      "rewards/format_reward": 0.0,
      "step": 30
    },
    {
      "completion_length": 579.1978515625,
      "epoch": 0.5970149253731343,
      "grad_norm": 0.06733980029821396,
      "kl": 0.004053878784179688,
      "learning_rate": 1.2296174432791415e-06,
      "loss": 0.0274,
      "reward": 0.760546875,
      "reward_std": 0.15440575871616602,
      "rewards/accuracy_reward": 0.760546875,
      "rewards/format_reward": 0.0,
      "step": 35
    },
    {
      "completion_length": 566.700390625,
      "epoch": 0.6823027718550106,
      "grad_norm": 0.21938073635101318,
      "kl": 0.00456085205078125,
      "learning_rate": 8.029152419343472e-07,
      "loss": 0.041,
      "reward": 0.7767578125,
      "reward_std": 0.1702250656671822,
      "rewards/accuracy_reward": 0.7767578125,
      "rewards/format_reward": 0.0,
      "step": 40
    },
    {
      "completion_length": 581.4125,
      "epoch": 0.767590618336887,
      "grad_norm": 0.10841479152441025,
      "kl": 0.0047943115234375,
      "learning_rate": 4.3933982822017883e-07,
      "loss": 0.0346,
      "reward": 0.7619140625,
      "reward_std": 0.1686327988281846,
      "rewards/accuracy_reward": 0.7619140625,
      "rewards/format_reward": 0.0,
      "step": 45
    },
    {
      "completion_length": 583.0294921875,
      "epoch": 0.8528784648187633,
      "grad_norm": 0.303811252117157,
      "kl": 0.0050792694091796875,
      "learning_rate": 1.718159615201853e-07,
      "loss": 0.04,
      "reward": 0.7552734375,
      "reward_std": 0.1798575583845377,
      "rewards/accuracy_reward": 0.7552734375,
      "rewards/format_reward": 0.0,
      "step": 50
    },
    {
      "completion_length": 576.5091796875,
      "epoch": 0.9381663113006397,
      "grad_norm": 0.5273950695991516,
      "kl": 0.004700088500976562,
      "learning_rate": 2.4570139579284723e-08,
      "loss": 0.0296,
      "reward": 0.7921875,
      "reward_std": 0.17182327918708323,
      "rewards/accuracy_reward": 0.7921875,
      "rewards/format_reward": 0.0,
      "step": 55
    },
    {
      "completion_length": 563.9654947916666,
      "epoch": 0.9893390191897654,
      "kl": 0.0046914418538411455,
      "reward": 0.7789713541666666,
      "reward_std": 0.16232363134622574,
      "rewards/accuracy_reward": 0.7789713541666666,
      "rewards/format_reward": 0.0,
      "step": 58,
      "total_flos": 0.0,
      "train_loss": 0.037125048837785064,
      "train_runtime": 18540.8719,
      "train_samples_per_second": 0.405,
      "train_steps_per_second": 0.003
    }
  ],
  "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
}