{
  "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": [
    {
      "clip_ratio": 0.0,
      "completion_length": 613.3375259399414,
      "epoch": 0.08528784648187633,
      "grad_norm": 0.5308949947357178,
      "kl": 0.00026957988739013673,
      "learning_rate": 2.5e-06,
      "loss": 0.0422,
      "reward": 0.6412946701049804,
      "reward_std": 0.33645378351211547,
      "rewards/accuracy_reward": 0.6408482432365418,
      "rewards/format_reward": 0.00044642859138548373,
      "step": 5
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 611.9799339294434,
      "epoch": 0.17057569296375266,
      "grad_norm": 3.3914427757263184,
      "kl": 0.002694988250732422,
      "learning_rate": 2.956412726139078e-06,
      "loss": 0.0675,
      "reward": 0.7015625357627868,
      "reward_std": 0.2750119812786579,
      "rewards/accuracy_reward": 0.7015625357627868,
      "rewards/format_reward": 0.0,
      "step": 10
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 598.9616317749023,
      "epoch": 0.255863539445629,
      "grad_norm": 0.3315448462963104,
      "kl": 0.0043125152587890625,
      "learning_rate": 2.7836719084521715e-06,
      "loss": 0.0621,
      "reward": 0.764285746216774,
      "reward_std": 0.21017537489533425,
      "rewards/accuracy_reward": 0.764285746216774,
      "rewards/format_reward": 0.0,
      "step": 15
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 588.717440032959,
      "epoch": 0.3411513859275053,
      "grad_norm": 0.8005821704864502,
      "kl": 0.004384231567382812,
      "learning_rate": 2.4946839873611927e-06,
      "loss": 0.0533,
      "reward": 0.7665178924798965,
      "reward_std": 0.1955874715000391,
      "rewards/accuracy_reward": 0.7665178924798965,
      "rewards/format_reward": 0.0,
      "step": 20
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 592.9370819091797,
      "epoch": 0.42643923240938164,
      "grad_norm": 0.7205191850662231,
      "kl": 0.004600143432617188,
      "learning_rate": 2.1156192081791355e-06,
      "loss": 0.045,
      "reward": 0.7629464656114578,
      "reward_std": 0.18942818641662598,
      "rewards/accuracy_reward": 0.7629464656114578,
      "rewards/format_reward": 0.0,
      "step": 25
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 598.0611862182617,
      "epoch": 0.511727078891258,
      "grad_norm": 0.6142191886901855,
      "kl": 0.042702865600585935,
      "learning_rate": 1.6808050203829845e-06,
      "loss": 0.0428,
      "reward": 0.7551339700818062,
      "reward_std": 0.1889859580434859,
      "rewards/accuracy_reward": 0.7551339700818062,
      "rewards/format_reward": 0.0,
      "step": 30
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 593.2520332336426,
      "epoch": 0.5970149253731343,
      "grad_norm": 10.091300964355469,
      "kl": 0.4265655517578125,
      "learning_rate": 1.2296174432791415e-06,
      "loss": 0.0673,
      "reward": 0.7410714656114579,
      "reward_std": 0.179294466227293,
      "rewards/accuracy_reward": 0.7410714656114579,
      "rewards/format_reward": 0.0,
      "step": 35
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 577.117886352539,
      "epoch": 0.6823027718550106,
      "grad_norm": 81.48939514160156,
      "kl": 0.06499862670898438,
      "learning_rate": 8.029152419343472e-07,
      "loss": 0.0492,
      "reward": 0.7669643267989159,
      "reward_std": 0.17745222924277188,
      "rewards/accuracy_reward": 0.7669643267989159,
      "rewards/format_reward": 0.0,
      "step": 40
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 594.6520378112793,
      "epoch": 0.767590618336887,
      "grad_norm": 59.926361083984375,
      "kl": 0.22182693481445312,
      "learning_rate": 4.3933982822017883e-07,
      "loss": 0.0426,
      "reward": 0.7529018267989158,
      "reward_std": 0.1793302634730935,
      "rewards/accuracy_reward": 0.7529018267989158,
      "rewards/format_reward": 0.0,
      "step": 45
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 599.8544906616211,
      "epoch": 0.8528784648187633,
      "grad_norm": 0.3387981653213501,
      "kl": 0.171990966796875,
      "learning_rate": 1.718159615201853e-07,
      "loss": 0.0534,
      "reward": 0.7386161029338837,
      "reward_std": 0.18926974572241306,
      "rewards/accuracy_reward": 0.7386161029338837,
      "rewards/format_reward": 0.0,
      "step": 50
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 590.5582893371582,
      "epoch": 0.9381663113006397,
      "grad_norm": 7.056384563446045,
      "kl": 0.15600433349609374,
      "learning_rate": 2.4570139579284723e-08,
      "loss": 0.0572,
      "reward": 0.7738839611411095,
      "reward_std": 0.18812613375484943,
      "rewards/accuracy_reward": 0.7738839611411095,
      "rewards/format_reward": 0.0,
      "step": 55
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 576.9970359802246,
      "epoch": 0.9893390191897654,
      "kl": 0.025601704915364582,
      "reward": 0.7589286093910536,
      "reward_std": 0.1643078220076859,
      "rewards/accuracy_reward": 0.7589286093910536,
      "rewards/format_reward": 0.0,
      "step": 58,
      "total_flos": 0.0,
      "train_loss": 0.05106422754711118,
      "train_runtime": 6823.4661,
      "train_samples_per_second": 1.099,
      "train_steps_per_second": 0.009
    }
  ],
  "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
}