{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 0.9230769230769231,
  "eval_steps": 100,
  "global_step": 6,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "completion_length": 774.4698944091797,
      "epoch": 0.15384615384615385,
      "grad_norm": 0.6025707125663757,
      "kl": 0.0,
      "learning_rate": 3e-06,
      "loss": -0.0011,
      "reward": 0.2689732238650322,
      "reward_std": 0.28160041756927967,
      "rewards/accuracy_reward": 0.2689732238650322,
      "rewards/format_reward": 0.0,
      "step": 1
    },
    {
      "completion_length": 790.086254119873,
      "epoch": 0.7692307692307693,
      "grad_norm": 0.13300876319408417,
      "kl": 0.0005920529365539551,
      "learning_rate": 2.86474508437579e-07,
      "loss": 0.0164,
      "reward": 0.3007812639698386,
      "reward_std": 0.23957567382603884,
      "rewards/accuracy_reward": 0.3007812639698386,
      "rewards/format_reward": 0.0,
      "step": 5
    },
    {
      "completion_length": 789.1562805175781,
      "epoch": 0.9230769230769231,
      "kl": 0.004157543182373047,
      "reward": 0.337053582072258,
      "reward_std": 0.26446567103266716,
      "rewards/accuracy_reward": 0.337053582072258,
      "rewards/format_reward": 0.0,
      "step": 6,
      "total_flos": 0.0,
      "train_loss": 0.021334987211351592,
      "train_runtime": 778.6034,
      "train_samples_per_second": 1.049,
      "train_steps_per_second": 0.008
    }
  ],
  "logging_steps": 5,
  "max_steps": 6,
  "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
}