{
  "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": 373.13706970214844,
      "epoch": 0.08528784648187633,
      "grad_norm": 2.1630916595458984,
      "kl": 0.001761174201965332,
      "learning_rate": 2.5e-06,
      "loss": 0.0001,
      "reward": 0.9339286163449287,
      "reward_std": 0.3458739034831524,
      "rewards/accuracy_reward": 0.23258929550647736,
      "rewards/format_reward": 0.7013393118977547,
      "step": 5
    },
    {
      "completion_length": 357.8872917175293,
      "epoch": 0.17057569296375266,
      "grad_norm": 5.1070876121521,
      "kl": 0.37891845703125,
      "learning_rate": 2.956412726139078e-06,
      "loss": 0.0152,
      "reward": 0.9741071835160255,
      "reward_std": 0.33634802624583243,
      "rewards/accuracy_reward": 0.22723215203732253,
      "rewards/format_reward": 0.7468750372529029,
      "step": 10
    },
    {
      "completion_length": 348.8261344909668,
      "epoch": 0.255863539445629,
      "grad_norm": 5.793032169342041,
      "kl": 0.804296875,
      "learning_rate": 2.7836719084521715e-06,
      "loss": 0.0322,
      "reward": 1.0337054029107093,
      "reward_std": 0.33858824484050276,
      "rewards/accuracy_reward": 0.24732144214212895,
      "rewards/format_reward": 0.7863839656114578,
      "step": 15
    },
    {
      "completion_length": 346.17434616088866,
      "epoch": 0.3411513859275053,
      "grad_norm": 1.532250165939331,
      "kl": 0.85458984375,
      "learning_rate": 2.4946839873611927e-06,
      "loss": 0.0342,
      "reward": 1.0790179044008255,
      "reward_std": 0.35572994202375413,
      "rewards/accuracy_reward": 0.27433036975562575,
      "rewards/format_reward": 0.8046875387430191,
      "step": 20
    },
    {
      "completion_length": 347.5040344238281,
      "epoch": 0.42643923240938164,
      "grad_norm": 2.3722541332244873,
      "kl": 1.31669921875,
      "learning_rate": 2.1156192081791355e-06,
      "loss": 0.0527,
      "reward": 1.1087053999304772,
      "reward_std": 0.35302889049053193,
      "rewards/accuracy_reward": 0.28437501229345796,
      "rewards/format_reward": 0.8243303969502449,
      "step": 25
    },
    {
      "completion_length": 357.7453266143799,
      "epoch": 0.511727078891258,
      "grad_norm": 6.4789137840271,
      "kl": 1.504296875,
      "learning_rate": 1.6808050203829845e-06,
      "loss": 0.0602,
      "reward": 1.078571479022503,
      "reward_std": 0.38525725305080416,
      "rewards/accuracy_reward": 0.3113839436322451,
      "rewards/format_reward": 0.7671875327825546,
      "step": 30
    },
    {
      "completion_length": 347.7029151916504,
      "epoch": 0.5970149253731343,
      "grad_norm": 5.793887615203857,
      "kl": 1.673046875,
      "learning_rate": 1.2296174432791415e-06,
      "loss": 0.0669,
      "reward": 1.115625049173832,
      "reward_std": 0.3930562749505043,
      "rewards/accuracy_reward": 0.31540180221199987,
      "rewards/format_reward": 0.8002232521772384,
      "step": 35
    },
    {
      "completion_length": 332.5259094238281,
      "epoch": 0.6823027718550106,
      "grad_norm": 10.747057914733887,
      "kl": 1.9728515625,
      "learning_rate": 8.029152419343472e-07,
      "loss": 0.0789,
      "reward": 1.132366120815277,
      "reward_std": 0.3926612690091133,
      "rewards/accuracy_reward": 0.31651787310838697,
      "rewards/format_reward": 0.8158482581377029,
      "step": 40
    },
    {
      "completion_length": 345.2725601196289,
      "epoch": 0.767590618336887,
      "grad_norm": 5.969154357910156,
      "kl": 2.1015625,
      "learning_rate": 4.3933982822017883e-07,
      "loss": 0.084,
      "reward": 1.1439732566475869,
      "reward_std": 0.3879304490983486,
      "rewards/accuracy_reward": 0.30178572721779345,
      "rewards/format_reward": 0.842187537252903,
      "step": 45
    },
    {
      "completion_length": 337.36385345458984,
      "epoch": 0.8528784648187633,
      "grad_norm": 5.109945774078369,
      "kl": 2.0640625,
      "learning_rate": 1.718159615201853e-07,
      "loss": 0.0826,
      "reward": 1.1687500596046447,
      "reward_std": 0.40620823949575424,
      "rewards/accuracy_reward": 0.3241071570664644,
      "rewards/format_reward": 0.8446428969502449,
      "step": 50
    },
    {
      "completion_length": 335.0609535217285,
      "epoch": 0.9381663113006397,
      "grad_norm": 3.530726432800293,
      "kl": 2.1984375,
      "learning_rate": 2.4570139579284723e-08,
      "loss": 0.0879,
      "reward": 1.1783482611179352,
      "reward_std": 0.40133031979203226,
      "rewards/accuracy_reward": 0.33839287236332893,
      "rewards/format_reward": 0.8399553969502449,
      "step": 55
    },
    {
      "completion_length": 334.10566838582355,
      "epoch": 0.9893390191897654,
      "kl": 2.4319661458333335,
      "reward": 1.16815481086572,
      "reward_std": 0.38675514981150627,
      "rewards/accuracy_reward": 0.35342263368268806,
      "rewards/format_reward": 0.8147321790456772,
      "step": 58,
      "total_flos": 0.0,
      "train_loss": 0.05658998479865941,
      "train_runtime": 6508.7782,
      "train_samples_per_second": 1.152,
      "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
}