{
  "n_episodes": 54,
  "models": {
    "grpo": {
      "n": 48,
      "env_reward_mean": 0.13180555555555554,
      "env_reward_std": 0.18206941280618835,
      "success_rate": 0.0625,
      "wrong_action_rate": 0.0,
      "no_action_rate": 0.9166666666666666,
      "investigation_coverage_mean": 0.20833333333333334,
      "root_cause_hit_rate": 0.4791666666666667,
      "tool_calls_mean": 1.6875,
      "latency_sec_mean": 5.3182960875833105,
      "truncated_rate": 0.10416666666666667
    },
    "oracle": {
      "n": 6,
      "env_reward_mean": 0.8591666666666665,
      "env_reward_std": 0.037734452280229105,
      "success_rate": 1.0,
      "wrong_action_rate": 0.0,
      "no_action_rate": 0.0,
      "investigation_coverage_mean": 1.0,
      "root_cause_hit_rate": 1.0,
      "tool_calls_mean": 4.333333333333333,
      "latency_sec_mean": 5.802433338431001e-05,
      "truncated_rate": 0.0
    }
  },
  "per_scenario": {
    "canary_error_rate": {
      "scenario_id": "canary_error_rate",
      "grpo": {
        "n": 8,
        "env_reward_mean": 0.1864583333333333,
        "env_reward_std": 0.07331946568012514,
        "success_rate": 0.0,
        "investigation_coverage_mean": 0.1875
      },
      "oracle": {
        "n": 1,
        "env_reward_mean": 0.8458333333333332,
        "env_reward_std": 0.0,
        "success_rate": 1.0,
        "investigation_coverage_mean": 1.0
      }
    },
    "crashloop_bad_config": {
      "scenario_id": "crashloop_bad_config",
      "grpo": {
        "n": 8,
        "env_reward_mean": 0.038541666666666655,
        "env_reward_std": 0.09500809280483792,
        "success_rate": 0.0,
        "investigation_coverage_mean": 0.0
      },
      "oracle": {
        "n": 1,
        "env_reward_mean": 0.8566666666666666,
        "env_reward_std": 0.0,
        "success_rate": 1.0,
        "investigation_coverage_mean": 1.0
      }
    },
    "gpu_util_zero_hung": {
      "scenario_id": "gpu_util_zero_hung",
      "grpo": {
        "n": 8,
        "env_reward_mean": 0.14687499999999998,
        "env_reward_std": 0.22695086787358057,
        "success_rate": 0.125,
        "investigation_coverage_mean": 0.3125
      },
      "oracle": {
        "n": 1,
        "env_reward_mean": 0.8624999999999999,
        "env_reward_std": 0.0,
        "success_rate": 1.0,
        "investigation_coverage_mean": 1.0
      }
    },
    "high_latency_underscaled": {
      "scenario_id": "high_latency_underscaled",
      "grpo": {
        "n": 8,
        "env_reward_mean": 0.05937499999999999,
        "env_reward_std": 0.13020416659999787,
        "success_rate": 0.0,
        "investigation_coverage_mean": 0.3125
      },
      "oracle": {
        "n": 1,
        "env_reward_mean": 0.9249999999999999,
        "env_reward_std": 0.0,
        "success_rate": 1.0,
        "investigation_coverage_mean": 1.0
      }
    },
    "imagepull_backoff": {
      "scenario_id": "imagepull_backoff",
      "grpo": {
        "n": 8,
        "env_reward_mean": 0.01874999999999999,
        "env_reward_std": 0.0948840144095737,
        "success_rate": 0.0,
        "investigation_coverage_mean": 0.0
      },
      "oracle": {
        "n": 1,
        "env_reward_mean": 0.8083333333333333,
        "env_reward_std": 0.0,
        "success_rate": 1.0,
        "investigation_coverage_mean": 1.0
      }
    },
    "oomkilled_inference": {
      "scenario_id": "oomkilled_inference",
      "grpo": {
        "n": 8,
        "env_reward_mean": 0.3408333333333333,
        "env_reward_std": 0.2157710426751997,
        "success_rate": 0.25,
        "investigation_coverage_mean": 0.4375
      },
      "oracle": {
        "n": 1,
        "env_reward_mean": 0.8566666666666666,
        "env_reward_std": 0.0,
        "success_rate": 1.0,
        "investigation_coverage_mean": 1.0
      }
    }
  },
  "deltas": {}
}