{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 0.10202898550724637,
  "eval_steps": 100,
  "global_step": 33,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "completion_length": 303.61742544174194,
      "epoch": 0.015458937198067632,
      "grad_norm": 0.02257479541003704,
      "kl": 0.9676864624023438,
      "learning_rate": 1.9941379571543597e-05,
      "loss": 0.0386,
      "reward": 0.7926339689642191,
      "reward_std": 0.3637773351743817,
      "rewards/accuracy_reward": 0.11194196966243908,
      "rewards/format_reward": 0.6806919939815999,
      "step": 5
    },
    {
      "completion_length": 193.77601299285888,
      "epoch": 0.030917874396135265,
      "grad_norm": 0.02067318558692932,
      "kl": 0.1245941162109375,
      "learning_rate": 1.796093065705644e-05,
      "loss": 0.005,
      "reward": 1.0407366551458836,
      "reward_std": 0.2885524293407798,
      "rewards/accuracy_reward": 0.11953125573927537,
      "rewards/format_reward": 0.9212053991854191,
      "step": 10
    },
    {
      "completion_length": 223.56373710632323,
      "epoch": 0.0463768115942029,
      "grad_norm": 0.014934813603758812,
      "kl": 0.16761474609375,
      "learning_rate": 1.3701381553399147e-05,
      "loss": 0.0067,
      "reward": 1.0725446939468384,
      "reward_std": 0.3380734449252486,
      "rewards/accuracy_reward": 0.16651786545990035,
      "rewards/format_reward": 0.906026828289032,
      "step": 15
    },
    {
      "completion_length": 189.5631784439087,
      "epoch": 0.06183574879227053,
      "grad_norm": 0.015098211355507374,
      "kl": 0.175286865234375,
      "learning_rate": 8.382180034472353e-06,
      "loss": 0.007,
      "reward": 1.1564732655882835,
      "reward_std": 0.2815748773515224,
      "rewards/accuracy_reward": 0.19162947330623864,
      "rewards/format_reward": 0.964843787997961,
      "step": 20
    },
    {
      "completion_length": 190.79632415771485,
      "epoch": 0.07729468599033816,
      "grad_norm": 0.013767687603831291,
      "kl": 0.16260986328125,
      "learning_rate": 3.5261371521817247e-06,
      "loss": 0.0065,
      "reward": 1.1904018431901933,
      "reward_std": 0.306734830327332,
      "rewards/accuracy_reward": 0.22790179681032896,
      "rewards/format_reward": 0.9625000387430191,
      "step": 25
    },
    {
      "completion_length": 203.4788038253784,
      "epoch": 0.0927536231884058,
      "grad_norm": 0.014600388705730438,
      "kl": 0.143023681640625,
      "learning_rate": 5.234682881719766e-07,
      "loss": 0.0057,
      "reward": 1.206808091700077,
      "reward_std": 0.2881101544946432,
      "rewards/accuracy_reward": 0.23984376154839993,
      "rewards/format_reward": 0.9669643275439739,
      "step": 30
    },
    {
      "completion_length": 215.08613300323486,
      "epoch": 0.10202898550724637,
      "kl": 0.13948567708333334,
      "reward": 1.2220982710520427,
      "reward_std": 0.3094604279225071,
      "rewards/accuracy_reward": 0.252976200543344,
      "rewards/format_reward": 0.9691220708191395,
      "step": 33,
      "total_flos": 0.0,
      "train_loss": 0.011041131547906181,
      "train_runtime": 4812.1769,
      "train_samples_per_second": 1.505,
      "train_steps_per_second": 0.007
    }
  ],
  "logging_steps": 5,
  "max_steps": 33,
  "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
}