{
  "site": "https://neurarch.com",
  "description": "Static verifier score against trained outcome, one grounded campaign. Every design passed the verifier and trained end to end.",
  "campaignAt": "2026-08-21T04:17:34.778Z",
  "backend": "modal",
  "gpu": "T4",
  "n": 24,
  "spearman": {
    "pooledRaw": 0.16567442507180566,
    "withinTaskRanks": 0.0913815441511916,
    "quoted": "pooledRaw"
  },
  "staticScoreHistogram": {
    "97": 1,
    "100": 23
  },
  "accuracySpread": {
    "min": 0.0684,
    "max": 0.9717,
    "n": 20
  },
  "lossMetricRows": 4,
  "caveats": [
    "One campaign, one GPU class, one day. Not a benchmark of any model.",
    "The 0.41 rescored figure quoted elsewhere is in-sample; the out-of-sample number does not exist yet.",
    "valLoss-scored rows carry no accuracy; their accuracy field is absent, not zero."
  ],
  "rows": [
    {
      "task": "grounded-cifar-cnn",
      "dataset": "cifar10",
      "designer": "reference",
      "staticScore": 100,
      "metric": "valAcc",
      "value": 0.6214,
      "params": 255978,
      "wallSec": 71
    },
    {
      "task": "grounded-cifar-cnn",
      "dataset": "cifar10",
      "designer": "xai",
      "staticScore": 100,
      "metric": "valAcc",
      "value": 0.64,
      "params": 261034,
      "wallSec": 95
    },
    {
      "task": "grounded-cifar-cnn",
      "dataset": "cifar10",
      "designer": "openrouter",
      "staticScore": 100,
      "metric": "valAcc",
      "value": 0.0684,
      "params": 1070794,
      "wallSec": 84
    },
    {
      "task": "grounded-cifar-cnn",
      "dataset": "cifar10",
      "designer": "groq",
      "staticScore": 100,
      "metric": "valAcc",
      "value": 0.6608,
      "params": 521162,
      "wallSec": 31
    },
    {
      "task": "grounded-cifar-deepen",
      "dataset": "cifar10",
      "designer": "reference",
      "staticScore": 100,
      "metric": "valAcc",
      "value": 0.6298,
      "params": 255978,
      "wallSec": 41
    },
    {
      "task": "grounded-cifar-deepen",
      "dataset": "cifar10",
      "designer": "xai",
      "staticScore": 100,
      "metric": "valAcc",
      "value": 0.664,
      "params": 456234,
      "wallSec": 41
    },
    {
      "task": "grounded-cifar-deepen",
      "dataset": "cifar10",
      "designer": "openrouter",
      "staticScore": 100,
      "metric": "valAcc",
      "value": 0.6558,
      "params": 834730,
      "wallSec": 32
    },
    {
      "task": "grounded-cifar-deepen",
      "dataset": "cifar10",
      "designer": "groq",
      "staticScore": 100,
      "metric": "valAcc",
      "value": 0.6432,
      "params": 456234,
      "wallSec": 45
    },
    {
      "task": "grounded-tabular-mlp",
      "dataset": "tabular",
      "designer": "reference",
      "staticScore": 100,
      "metric": "valLoss",
      "value": 0.0041,
      "params": 5569,
      "wallSec": 21
    },
    {
      "task": "grounded-tabular-mlp",
      "dataset": "tabular",
      "designer": "xai",
      "staticScore": 100,
      "metric": "valLoss",
      "value": 0.0024,
      "params": 3457,
      "wallSec": 24
    },
    {
      "task": "grounded-tabular-mlp",
      "dataset": "tabular",
      "designer": "openrouter",
      "staticScore": 100,
      "metric": "valLoss",
      "value": 0.0026,
      "params": 13057,
      "wallSec": 18
    },
    {
      "task": "grounded-tabular-mlp",
      "dataset": "tabular",
      "designer": "groq",
      "staticScore": 100,
      "metric": "valLoss",
      "value": 0.003,
      "params": 3457,
      "wallSec": 21
    },
    {
      "task": "grounded-mnist-mlp",
      "dataset": "mnist",
      "designer": "reference",
      "staticScore": 100,
      "metric": "valAcc",
      "value": 0.9675,
      "params": 235146,
      "wallSec": 41
    },
    {
      "task": "grounded-mnist-mlp",
      "dataset": "mnist",
      "designer": "xai",
      "staticScore": 100,
      "metric": "valAcc",
      "value": 0.9678,
      "params": 235146,
      "wallSec": 26
    },
    {
      "task": "grounded-mnist-mlp",
      "dataset": "mnist",
      "designer": "openrouter",
      "staticScore": 100,
      "metric": "valAcc",
      "value": 0.9717,
      "params": 567434,
      "wallSec": 41
    },
    {
      "task": "grounded-mnist-mlp",
      "dataset": "mnist",
      "designer": "groq",
      "staticScore": 100,
      "metric": "valAcc",
      "value": 0.9692,
      "params": 235146,
      "wallSec": 21
    },
    {
      "task": "grounded-cifar-deep",
      "dataset": "cifar10",
      "designer": "reference",
      "staticScore": 100,
      "metric": "valAcc",
      "value": 0.5882,
      "params": 82330,
      "wallSec": 41
    },
    {
      "task": "grounded-cifar-deep",
      "dataset": "cifar10",
      "designer": "xai",
      "staticScore": 100,
      "metric": "valAcc",
      "value": 0.178,
      "params": 307498,
      "wallSec": 73
    },
    {
      "task": "grounded-cifar-deep",
      "dataset": "cifar10",
      "designer": "openrouter",
      "staticScore": 97,
      "metric": "valAcc",
      "value": 0.1694,
      "params": 402026,
      "wallSec": 76
    },
    {
      "task": "grounded-cifar-deep",
      "dataset": "cifar10",
      "designer": "groq",
      "staticScore": 100,
      "metric": "valAcc",
      "value": 0.0994,
      "params": 363882,
      "wallSec": 76
    },
    {
      "task": "grounded-mnist-tiny",
      "dataset": "mnist",
      "designer": "reference",
      "staticScore": 100,
      "metric": "valAcc",
      "value": 0.9177,
      "params": 7165,
      "wallSec": 36
    },
    {
      "task": "grounded-mnist-tiny",
      "dataset": "mnist",
      "designer": "xai",
      "staticScore": 100,
      "metric": "valAcc",
      "value": 0.9187,
      "params": 7165,
      "wallSec": 41
    },
    {
      "task": "grounded-mnist-tiny",
      "dataset": "mnist",
      "designer": "openrouter",
      "staticScore": 100,
      "metric": "valAcc",
      "value": 0.908,
      "params": 6370,
      "wallSec": 32
    },
    {
      "task": "grounded-mnist-tiny",
      "dataset": "mnist",
      "designer": "groq",
      "staticScore": 100,
      "metric": "valAcc",
      "value": 0.911,
      "params": 7980,
      "wallSec": 43
    }
  ]
}
