{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 0.9893390191897654,
  "eval_steps": 500,
  "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": 1023.7232208251953,
      "epoch": 0.017057569296375266,
      "grad_norm": 0.19576208293437958,
      "kl": 0.0,
      "learning_rate": 5e-07,
      "loss": 0.0,
      "reward": 0.6484375298023224,
      "reward_std": 0.29160378500819206,
      "rewards/accuracy_reward": 0.6484375298023224,
      "rewards/format_reward": 0.0,
      "step": 1
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 1022.977409362793,
      "epoch": 0.08528784648187633,
      "grad_norm": 0.17678765952587128,
      "kl": 0.00011715292930603027,
      "learning_rate": 2.5e-06,
      "loss": -0.0002,
      "reward": 0.5898437788709998,
      "reward_std": 0.3387324567884207,
      "rewards/accuracy_reward": 0.5898437788709998,
      "rewards/format_reward": 0.0,
      "step": 5
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 1023.1116149902343,
      "epoch": 0.17057569296375266,
      "grad_norm": 0.15346871316432953,
      "kl": 0.0009440422058105469,
      "learning_rate": 2.956412726139078e-06,
      "loss": -0.0001,
      "reward": 0.6941964611411094,
      "reward_std": 0.2548374630510807,
      "rewards/accuracy_reward": 0.6941964611411094,
      "rewards/format_reward": 0.0,
      "step": 10
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 1022.287516784668,
      "epoch": 0.255863539445629,
      "grad_norm": 0.07373964786529541,
      "kl": 0.003263092041015625,
      "learning_rate": 2.7836719084521715e-06,
      "loss": 0.0004,
      "reward": 0.7709821805357933,
      "reward_std": 0.16608586916700005,
      "rewards/accuracy_reward": 0.7709821805357933,
      "rewards/format_reward": 0.0,
      "step": 15
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 1023.2745651245117,
      "epoch": 0.3411513859275053,
      "grad_norm": 0.08782626688480377,
      "kl": 0.008664703369140625,
      "learning_rate": 2.4946839873611927e-06,
      "loss": 0.0004,
      "reward": 0.7859375417232514,
      "reward_std": 0.1670039308257401,
      "rewards/accuracy_reward": 0.7859375417232514,
      "rewards/format_reward": 0.0,
      "step": 20
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 1023.670541381836,
      "epoch": 0.42643923240938164,
      "grad_norm": 0.14314483106136322,
      "kl": 0.00386810302734375,
      "learning_rate": 2.1156192081791355e-06,
      "loss": 0.0002,
      "reward": 0.7540178894996643,
      "reward_std": 0.18852746970951556,
      "rewards/accuracy_reward": 0.7540178894996643,
      "rewards/format_reward": 0.0,
      "step": 25
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 1022.9817108154297,
      "epoch": 0.511727078891258,
      "grad_norm": 0.07946474105119705,
      "kl": 0.0036365509033203123,
      "learning_rate": 1.6808050203829845e-06,
      "loss": 0.0002,
      "reward": 0.7466518193483352,
      "reward_std": 0.18064689878374338,
      "rewards/accuracy_reward": 0.7466518193483352,
      "rewards/format_reward": 0.0,
      "step": 30
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 1023.1973281860352,
      "epoch": 0.5970149253731343,
      "grad_norm": 0.08639061450958252,
      "kl": 0.004063796997070312,
      "learning_rate": 1.2296174432791415e-06,
      "loss": 0.0007,
      "reward": 0.740178607404232,
      "reward_std": 0.1846639033406973,
      "rewards/accuracy_reward": 0.740178607404232,
      "rewards/format_reward": 0.0,
      "step": 35
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 1022.9017990112304,
      "epoch": 0.6823027718550106,
      "grad_norm": 0.07541914284229279,
      "kl": 0.004433822631835937,
      "learning_rate": 8.029152419343472e-07,
      "loss": 0.0004,
      "reward": 0.7520089566707611,
      "reward_std": 0.18088983427733182,
      "rewards/accuracy_reward": 0.7520089566707611,
      "rewards/format_reward": 0.0,
      "step": 40
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 1022.3393035888672,
      "epoch": 0.767590618336887,
      "grad_norm": 0.0810270830988884,
      "kl": 0.00443572998046875,
      "learning_rate": 4.3933982822017883e-07,
      "loss": 0.0,
      "reward": 0.7475446745753288,
      "reward_std": 0.17573931943625212,
      "rewards/accuracy_reward": 0.7475446745753288,
      "rewards/format_reward": 0.0,
      "step": 45
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 1022.4098373413086,
      "epoch": 0.8528784648187633,
      "grad_norm": 0.09751690179109573,
      "kl": 0.004325485229492188,
      "learning_rate": 1.718159615201853e-07,
      "loss": 0.0003,
      "reward": 0.7466518178582191,
      "reward_std": 0.19847031086683273,
      "rewards/accuracy_reward": 0.7466518178582191,
      "rewards/format_reward": 0.0,
      "step": 50
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 1023.1006759643554,
      "epoch": 0.9381663113006397,
      "grad_norm": 0.10825635492801666,
      "kl": 0.004570770263671875,
      "learning_rate": 2.4570139579284723e-08,
      "loss": 0.0004,
      "reward": 0.7694196790456772,
      "reward_std": 0.18778187595307827,
      "rewards/accuracy_reward": 0.7694196790456772,
      "rewards/format_reward": 0.0,
      "step": 55
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 1022.9732869466146,
      "epoch": 0.9893390191897654,
      "kl": 0.0047664642333984375,
      "reward": 0.7552083730697632,
      "reward_std": 0.1784441495935122,
      "rewards/accuracy_reward": 0.7552083730697632,
      "rewards/format_reward": 0.0,
      "step": 58,
      "total_flos": 0.0,
      "train_loss": 0.00074836279989164,
      "train_runtime": 20358.4002,
      "train_samples_per_second": 0.368,
      "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
}