{
  "task": {
    "id": "swarm-202-l2",
    "model": "swarm",
    "seed": 202,
    "control": {},
    "metric": "avgOrder",
    "replicates": 12,
    "budget": 8,
    "tier": "L2",
    "candidateParams": [
      "alignmentWeight",
      "count",
      "maxForce"
    ],
    "truth": {
      "param": "alignmentWeight",
      "value": 0.2471,
      "effectOnMetric": "down",
      "magnitude": "small"
    }
  },
  "log": {
    "taskId": "swarm-202-l2",
    "model": "swarm",
    "solver": "codex-gpt-5.5",
    "calls": [
      {
        "kind": "experiment",
        "metric": "avgOrder",
        "stats": {
          "metric": "avgOrder",
          "label": "Order parameter",
          "controlMean": 0.867419286871484,
          "controlCI": [
            0.8003408919082285,
            0.9186489587749551
          ],
          "treatmentMean": 0.841842638569306,
          "treatmentCI": [
            0.7767624567650612,
            0.898146917741158
          ],
          "deltaPct": -2.948591147243812,
          "test": {
            "name": "Wilcoxon signed-rank",
            "statistic": 25,
            "p": 0.28958747927122386
          },
          "pAdjusted": 1,
          "significant": false,
          "effect": {
            "name": "rank-biserial r",
            "value": 0.3057147992190409,
            "magnitude": "medium"
          }
        },
        "configA": {},
        "configB": {
          "alignmentWeight": 1.5
        }
      },
      {
        "kind": "experiment",
        "metric": "avgOrder",
        "stats": {
          "metric": "avgOrder",
          "label": "Order parameter",
          "controlMean": 0.867419286871484,
          "controlCI": [
            0.8003408919082285,
            0.9186489587749551
          ],
          "treatmentMean": 0.5071662543805358,
          "treatmentCI": [
            0.3939956626214454,
            0.6148555031196169
          ],
          "deltaPct": -41.5315912319947,
          "test": {
            "name": "Wilcoxon signed-rank",
            "statistic": 1,
            "p": 0.003263853493157942
          },
          "pAdjusted": 0.012631563477234486,
          "significant": true,
          "effect": {
            "name": "rank-biserial r",
            "value": 0.8492077756084468,
            "magnitude": "large"
          }
        },
        "configA": {},
        "configB": {
          "count": 150
        }
      },
      {
        "kind": "experiment",
        "metric": "avgOrder",
        "stats": {
          "metric": "avgOrder",
          "label": "Order parameter",
          "controlMean": 0.867419286871484,
          "controlCI": [
            0.8003408919082285,
            0.9186489587749551
          ],
          "treatmentMean": 0.5790249166593017,
          "treatmentCI": [
            0.5001680232955005,
            0.6611541952494465
          ],
          "deltaPct": -33.247401179230465,
          "test": {
            "name": "Wilcoxon signed-rank",
            "statistic": 2,
            "p": 0.004192722492179168
          },
          "pAdjusted": 0.013055413972631769,
          "significant": true,
          "effect": {
            "name": "rank-biserial r",
            "value": 0.8265622349255549,
            "magnitude": "large"
          }
        },
        "configA": {},
        "configB": {
          "maxForce": 0.075
        }
      },
      {
        "kind": "probe",
        "metric": "avgOrder",
        "stats": {
          "metric": "avgOrder",
          "label": "Order parameter",
          "controlMean": 0.867419286871484,
          "controlCI": [
            0.8003408919082285,
            0.9186489587749551
          ],
          "treatmentMean": 0.6122927376489478,
          "treatmentCI": [
            0.4911694720059983,
            0.7250678854224905
          ],
          "deltaPct": -29.412137023457202,
          "test": {
            "name": "Wilcoxon signed-rank",
            "statistic": 5,
            "p": 0.008590034259602142
          },
          "pAdjusted": 0.025770102778806425,
          "significant": true,
          "effect": {
            "name": "rank-biserial r",
            "value": 0.7586256128768792,
            "magnitude": "large"
          }
        },
        "configA": {}
      },
      {
        "kind": "probe",
        "metric": "avgOrder",
        "stats": {
          "metric": "avgOrder",
          "label": "Order parameter",
          "controlMean": 0.5790249166593017,
          "controlCI": [
            0.49551790006127844,
            0.6651707918762311
          ],
          "treatmentMean": 0.6122927376489478,
          "treatmentCI": [
            0.4911694720059983,
            0.7250678854224905
          ],
          "deltaPct": 5.745490398165522,
          "test": {
            "name": "Wilcoxon signed-rank",
            "statistic": 29,
            "p": 0.4561257049865399
          },
          "pAdjusted": 1,
          "significant": false,
          "effect": {
            "name": "rank-biserial r",
            "value": -0.21513263648747322,
            "magnitude": "small"
          }
        },
        "configA": {
          "maxForce": 0.075
        }
      }
    ],
    "claims": [],
    "submission": {
      "param": "maxForce",
      "direction": "up",
      "magnitude": "medium"
    },
    "callsUsed": 5,
    "budget": 8
  },
  "score": {
    "taskId": "swarm-202-l2",
    "model": "swarm",
    "solver": "codex-gpt-5.5",
    "correctParam": false,
    "correctDirection": false,
    "correctMagnitude": false,
    "solved": false,
    "withinBudget": true,
    "callsUsed": 5,
    "budget": 8,
    "efficiency": 0.375,
    "claimValidity": 1,
    "score": 15.6,
    "notes": [
      "wrong/no param (truth: alignmentWeight)",
      "wrong/no direction (truth: down)",
      "magnitude off by one (said medium, truth small)",
      "conclusion not backed by a significant isolating experiment"
    ]
  },
  "audit": {
    "metric": "avgOrder",
    "alpha": 0.05,
    "familySize": 3,
    "distinctParamsTested": 3,
    "excessTesting": false,
    "survivors": [
      "count",
      "maxForce"
    ],
    "submissionBacking": "isolating-experiment",
    "backingSurvivesHolm": true,
    "pHacking": false,
    "notes": []
  },
  "provenance": {
    "bench": "0.1.0",
    "mcp": "0.1.1",
    "engines": {
      "swarm": "0.1.0",
      "market": "0.1.0",
      "social": "0.1.0",
      "morph": "0.1.0",
      "origin": "0.1.0"
    },
    "commit": "155e0390345537b17507a5ce774cf119dff7dc56",
    "node": "v25.2.1",
    "createdAt": "2026-06-11T18:45:23.103Z"
  }
}