{
  "arm": "S1-live + S2 (the registered paid window)",
  "registration": {
    "url": "https://osf.io/s3gfm",
    "doi": "10.17605/OSF.IO/S3GFM"
  },
  "config": {
    "version": 1,
    "models": {
      "realizer": "claude-sonnet-5",
      "perception": "claude-sonnet-4-6",
      "judges": [
        "claude-opus-4-8",
        "claude-haiku-4-5-20251001"
      ],
      "s2": [
        "claude-haiku-4-5-20251001",
        "claude-sonnet-4-6",
        "claude-opus-4-8"
      ]
    },
    "deviations": [
      {
        "id": "no-temperature-control",
        "registered": "S2 temperature-0 and sampled cells kept separate; run-to-run stability measured at temperature 0.",
        "actual": "Current Claude API tiers reject an explicit temperature parameter, so all runs execute at each model's default sampling and the R = 5 runs form ONE cell per model.",
        "bias": "Makes H4's run-to-run stability floor STRICTER (default sampling includes sampling noise that temperature 0 would have suppressed); H4 can fail from this deviation but cannot spuriously pass because of it."
      },
      {
        "id": "model-ids-pinned-at-window",
        "registered": "Model ids frozen at registration.",
        "actual": "The registration text names roles and counts but not vendor model ids; ids were pinned in this committed file before the first paid call.",
        "bias": "None on outcomes; a completeness gap in the registration transcription, disclosed."
      }
    ]
  },
  "corpus": {
    "sha256": "eba3a75ac9cf589fc3de98e618cfda46561492e6ed128610a14feb53e2d9aa8e",
    "scenariosRun": 72
  },
  "dryRun": false,
  "spend": {
    "ceilingUsd": 150,
    "spentUsd": 84.74710000000029,
    "calls": 8488,
    "byModel": {
      "claude-sonnet-5": 1.8287580000000012,
      "claude-sonnet-4-6": 12.886626000000025,
      "claude-haiku-4-5-20251001": 2.9991160000000052,
      "claude-opus-4-8": 67.03259999999996
    }
  },
  "liveCoverage": {
    "strongestFlat": "summary",
    "means": {
      "buffer": 0.1396531130906131,
      "transcript": 0.1740388373200873,
      "summary": 0.2681244392181892,
      "sagenLive300": 0.49854725714100706,
      "sagenLive2000": 0.8279491052928553
    }
  },
  "confirmatory": {
    "H1": {
      "diffMean": 0.23042281792281788,
      "ci95": [
        0.20487371424871414,
        0.2570438531376031
      ],
      "p": 0.0001,
      "vs": "summary"
    },
    "H2": {
      "pass": true,
      "violations": []
    },
    "H3": {
      "taxMean": 0.4384221873034375,
      "ci95": [
        0.4164987084674588,
        0.4602129160191662
      ],
      "bound": 0.1,
      "pOverBound": 1
    },
    "H4": {
      "alignmentMean": 0.2569555555555556,
      "alignmentFloor": 0.8,
      "stabilityMean": 0.7186375165343915,
      "stabilityFloor": 0.9
    },
    "holm": [
      {
        "id": "H1",
        "p": 0.0001,
        "threshold": 0.0125,
        "reject": true
      },
      {
        "id": "H2",
        "p": 0.0001,
        "threshold": 0.016666666666666666,
        "reject": true
      },
      {
        "id": "H3",
        "p": 1,
        "threshold": 0.05,
        "reject": false
      },
      {
        "id": "H4",
        "p": 0.0001,
        "threshold": 0.025,
        "reject": true
      }
    ]
  },
  "perceptionByModel": {
    "claude-haiku-4-5-20251001": {
      "composite": 0.2512814814814815,
      "cells": 150
    },
    "claude-sonnet-4-6": {
      "composite": 0.2569555555555556,
      "cells": 150
    },
    "claude-opus-4-8": {
      "composite": 0.27324768518518505,
      "cells": 150
    }
  },
  "judging": {
    "kappa": {
      "kappa": 0.25331948186864905,
      "observed": 0.7125,
      "expected": 0.6149625000000001,
      "n": 400
    },
    "gatePassed": false,
    "floor": 0.7,
    "pairs": 400
  }
}