{
  "best_global_step": null,
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 1.0,
  "eval_steps": 500,
  "global_step": 4,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "clip_ratio/high_max": 0.0,
      "clip_ratio/high_mean": 0.0,
      "clip_ratio/low_mean": 0.0,
      "clip_ratio/low_min": 0.0,
      "clip_ratio/region_mean": 0.0,
      "completion_length": 512.0,
      "completions/clipped_ratio": 1.0,
      "completions/max_length": 512.0,
      "completions/max_terminated_length": 0.0,
      "completions/mean_length": 512.0,
      "completions/mean_terminated_length": 0.0,
      "completions/min_length": 512.0,
      "completions/min_terminated_length": 0.0,
      "epoch": 0.25,
      "frac_reward_zero_std": 0.0,
      "grad_norm": 0.16221363842487335,
      "kl": -3.4924596548080444e-10,
      "learning_rate": 0.0,
      "loss": 0.0,
      "num_tokens": 2580.0,
      "reward": -0.22999998927116394,
      "reward_std": 1.088944435119629,
      "rewards/reward_environment_execution/mean": -0.23000000417232513,
      "rewards/reward_environment_execution/std": 0.3818376660346985,
      "rewards/reward_format_compliance/mean": -0.15000000596046448,
      "rewards/reward_format_compliance/std": 0.4949747622013092,
      "rewards/reward_investigation_quality/mean": 0.15000000596046448,
      "rewards/reward_investigation_quality/std": 0.2121320366859436,
      "rewards/reward_os_mechanics/mean": 0.0,
      "rewards/reward_os_mechanics/std": 0.0,
      "step": 1
    },
    {
      "clip_ratio/high_max": 0.0,
      "clip_ratio/high_mean": 0.0,
      "clip_ratio/low_mean": 0.0,
      "clip_ratio/low_min": 0.0,
      "clip_ratio/region_mean": 0.0,
      "completion_length": 512.0,
      "completions/clipped_ratio": 1.0,
      "completions/max_length": 512.0,
      "completions/max_terminated_length": 0.0,
      "completions/mean_length": 512.0,
      "completions/mean_terminated_length": 0.0,
      "completions/min_length": 512.0,
      "completions/min_terminated_length": 0.0,
      "epoch": 0.5,
      "frac_reward_zero_std": 1.0,
      "grad_norm": 3.505946644111191e-09,
      "kl": -4.656612873077393e-10,
      "learning_rate": 5.000000000000001e-07,
      "loss": -1.8626451075975936e-11,
      "num_tokens": 5158.0,
      "reward": 0.5400000214576721,
      "reward_std": 0.0,
      "rewards/reward_environment_execution/mean": 0.03999999910593033,
      "rewards/reward_environment_execution/std": 0.0,
      "rewards/reward_format_compliance/mean": 0.20000000298023224,
      "rewards/reward_format_compliance/std": 0.0,
      "rewards/reward_investigation_quality/mean": 0.30000001192092896,
      "rewards/reward_investigation_quality/std": 0.0,
      "rewards/reward_os_mechanics/mean": 0.0,
      "rewards/reward_os_mechanics/std": 0.0,
      "step": 2
    },
    {
      "clip_ratio/high_max": 0.0,
      "clip_ratio/high_mean": 0.0,
      "clip_ratio/low_mean": 0.0,
      "clip_ratio/low_min": 0.0,
      "clip_ratio/region_mean": 0.0,
      "completion_length": 512.0,
      "completions/clipped_ratio": 1.0,
      "completions/max_length": 512.0,
      "completions/max_terminated_length": 0.0,
      "completions/mean_length": 512.0,
      "completions/mean_terminated_length": 0.0,
      "completions/min_length": 512.0,
      "completions/min_terminated_length": 0.0,
      "epoch": 0.75,
      "frac_reward_zero_std": 0.0,
      "grad_norm": 0.20935378968715668,
      "kl": 5.7238503359258175e-06,
      "learning_rate": 1.0000000000000002e-06,
      "loss": 2.384185791015625e-07,
      "num_tokens": 7736.0,
      "reward": -0.22999998927116394,
      "reward_std": 1.088944435119629,
      "rewards/reward_environment_execution/mean": -0.23000000417232513,
      "rewards/reward_environment_execution/std": 0.3818376660346985,
      "rewards/reward_format_compliance/mean": -0.15000000596046448,
      "rewards/reward_format_compliance/std": 0.4949747622013092,
      "rewards/reward_investigation_quality/mean": 0.15000000596046448,
      "rewards/reward_investigation_quality/std": 0.2121320366859436,
      "rewards/reward_os_mechanics/mean": 0.0,
      "rewards/reward_os_mechanics/std": 0.0,
      "step": 3
    },
    {
      "clip_ratio/high_max": 0.0,
      "clip_ratio/high_mean": 0.0,
      "clip_ratio/low_mean": 0.0,
      "clip_ratio/low_min": 0.0,
      "clip_ratio/region_mean": 0.0,
      "completion_length": 512.0,
      "completions/clipped_ratio": 1.0,
      "completions/max_length": 512.0,
      "completions/max_terminated_length": 0.0,
      "completions/mean_length": 512.0,
      "completions/mean_terminated_length": 0.0,
      "completions/min_length": 512.0,
      "completions/min_terminated_length": 0.0,
      "epoch": 1.0,
      "frac_reward_zero_std": 0.0,
      "grad_norm": 0.16330939531326294,
      "kl": 5.532288923859596e-06,
      "learning_rate": 1.5e-06,
      "loss": 2.384185791015625e-07,
      "num_tokens": 10392.0,
      "reward": -0.22999998927116394,
      "reward_std": 1.088944435119629,
      "rewards/reward_environment_execution/mean": -0.23000000417232513,
      "rewards/reward_environment_execution/std": 0.3818376660346985,
      "rewards/reward_format_compliance/mean": -0.15000000596046448,
      "rewards/reward_format_compliance/std": 0.4949747622013092,
      "rewards/reward_investigation_quality/mean": 0.15000000596046448,
      "rewards/reward_investigation_quality/std": 0.2121320366859436,
      "rewards/reward_os_mechanics/mean": 0.0,
      "rewards/reward_os_mechanics/std": 0.0,
      "step": 4
    }
  ],
  "logging_steps": 1,
  "max_steps": 4,
  "num_input_tokens_seen": 10392,
  "num_train_epochs": 1,
  "save_steps": 999,
  "stateful_callbacks": {
    "TrainerControl": {
      "args": {
        "should_epoch_stop": false,
        "should_evaluate": false,
        "should_log": false,
        "should_save": true,
        "should_training_stop": true
      },
      "attributes": {}
    }
  },
  "total_flos": 0.0,
  "train_batch_size": 2,
  "trial_name": null,
  "trial_params": null
}