Slow Triage
Exact reviewed source / 15 of 64

prototype/evidence/evaluation.json

SHA-256 of original bytes
64fd2a52819e67ed22ba3c46fc3627e386b5b98f3c6aea9be95aacb6ececdb0c
{
  "available": true,
  "createdAt": "2026-10-19T22:34:23.145Z",
  "dataset": "fixed developer-authored synthetic holdout; not a real-world performance estimate",
  "model": {
    "name": "Xenova/all-MiniLM-L6-v2",
    "revision": "751bff37182d3f1213fa05d7196b954e230abad9",
    "dtype": "q8",
    "dimensions": 384,
    "maxTokens": 256,
    "license": "Apache-2.0",
    "offline": true
  },
  "thresholds": {
    "minimumSimilarity": 0.4,
    "minimumMargin": 0.04,
    "duplicateSimilarity": 0.83
  },
  "referenceHash": "44735dd6e345311b8129b9ef7ca96a6b5bce50b6ce497f06b8350fb7857b59e2",
  "fixtureHash": "9ef7770c613dc08d0f6b8a8bd1226888d8a04b1ba53bea7b6ba8ed3bc0ac34bb",
  "classification": {
    "total": 20,
    "correct": 9,
    "accuracy": 0.45,
    "baselineAccuracy": 0.9,
    "abstained": 15,
    "coverage": 0.25,
    "selectiveAccuracy": 1,
    "unknownAbstention": 4,
    "unknownTotal": 4,
    "confusion": {
      "bug": {
        "bug": 1,
        "feature": 0,
        "documentation": 0,
        "question": 0,
        "unsure": 3
      },
      "feature": {
        "bug": 0,
        "feature": 0,
        "documentation": 0,
        "question": 0,
        "unsure": 4
      },
      "documentation": {
        "bug": 0,
        "feature": 0,
        "documentation": 3,
        "question": 0,
        "unsure": 1
      },
      "question": {
        "bug": 0,
        "feature": 0,
        "documentation": 0,
        "question": 1,
        "unsure": 3
      },
      "unsure": {
        "bug": 0,
        "feature": 0,
        "documentation": 0,
        "question": 0,
        "unsure": 4
      }
    },
    "baselineConfusion": {
      "bug": {
        "bug": 2,
        "feature": 0,
        "documentation": 0,
        "question": 1,
        "unsure": 1
      },
      "feature": {
        "bug": 0,
        "feature": 4,
        "documentation": 0,
        "question": 0,
        "unsure": 0
      },
      "documentation": {
        "bug": 0,
        "feature": 0,
        "documentation": 4,
        "question": 0,
        "unsure": 0
      },
      "question": {
        "bug": 0,
        "feature": 0,
        "documentation": 0,
        "question": 4,
        "unsure": 0
      },
      "unsure": {
        "bug": 0,
        "feature": 0,
        "documentation": 0,
        "question": 0,
        "unsure": 4
      }
    },
    "rows": [
      {
        "id": 201,
        "expected": "bug",
        "predicted": "unsure",
        "rawLabel": "bug",
        "similarity": 0.36021442782610946,
        "margin": 0.20997445121461178,
        "baseline": "bug",
        "correct": false,
        "needsReview": true
      },
      {
        "id": 202,
        "expected": "bug",
        "predicted": "unsure",
        "rawLabel": "bug",
        "similarity": 0.28360056386632004,
        "margin": 0.04581243080132563,
        "baseline": "question",
        "correct": false,
        "needsReview": true
      },
      {
        "id": 203,
        "expected": "bug",
        "predicted": "bug",
        "rawLabel": "bug",
        "similarity": 0.4921915358842254,
        "margin": 0.28736290889695654,
        "baseline": "unsure",
        "correct": true,
        "needsReview": false
      },
      {
        "id": 204,
        "expected": "bug",
        "predicted": "unsure",
        "rawLabel": "bug",
        "similarity": 0.3061974964643349,
        "margin": 0.13370857152333737,
        "baseline": "bug",
        "correct": false,
        "needsReview": true
      },
      {
        "id": 205,
        "expected": "feature",
        "predicted": "unsure",
        "rawLabel": "feature",
        "similarity": 0.25534207788299024,
        "margin": 0.11855528133138724,
        "baseline": "feature",
        "correct": false,
        "needsReview": true
      },
      {
        "id": 206,
        "expected": "feature",
        "predicted": "unsure",
        "rawLabel": "feature",
        "similarity": 0.2333831080357983,
        "margin": 0.061302636002256966,
        "baseline": "feature",
        "correct": false,
        "needsReview": true
      },
      {
        "id": 207,
        "expected": "feature",
        "predicted": "unsure",
        "rawLabel": "question",
        "similarity": 0.32094710754405315,
        "margin": 0.04595118012955329,
        "baseline": "feature",
        "correct": false,
        "needsReview": true
      },
      {
        "id": 208,
        "expected": "feature",
        "predicted": "unsure",
        "rawLabel": "feature",
        "similarity": 0.34683807154249857,
        "margin": 0.026751889736613577,
        "baseline": "feature",
        "correct": false,
        "needsReview": true
      },
      {
        "id": 209,
        "expected": "documentation",
        "predicted": "documentation",
        "rawLabel": "documentation",
        "similarity": 0.43820394426791076,
        "margin": 0.1908172192397018,
        "baseline": "documentation",
        "correct": true,
        "needsReview": false
      },
      {
        "id": 210,
        "expected": "documentation",
        "predicted": "unsure",
        "rawLabel": "documentation",
        "similarity": 0.22723501637406013,
        "margin": 0.08528916210155049,
        "baseline": "documentation",
        "correct": false,
        "needsReview": true
      },
      {
        "id": 211,
        "expected": "documentation",
        "predicted": "documentation",
        "rawLabel": "documentation",
        "similarity": 0.5116850581090184,
        "margin": 0.29582712039450026,
        "baseline": "documentation",
        "correct": true,
        "needsReview": false
      },
      {
        "id": 212,
        "expected": "documentation",
        "predicted": "documentation",
        "rawLabel": "documentation",
        "similarity": 0.6379941982090976,
        "margin": 0.3167043457419379,
        "baseline": "documentation",
        "correct": true,
        "needsReview": false
      },
      {
        "id": 213,
        "expected": "question",
        "predicted": "question",
        "rawLabel": "question",
        "similarity": 0.5040743875578764,
        "margin": 0.3024009837060428,
        "baseline": "question",
        "correct": true,
        "needsReview": false
      },
      {
        "id": 214,
        "expected": "question",
        "predicted": "unsure",
        "rawLabel": "question",
        "similarity": 0.3601348794195268,
        "margin": 0.22956607392002149,
        "baseline": "question",
        "correct": false,
        "needsReview": true
      },
      {
        "id": 215,
        "expected": "question",
        "predicted": "unsure",
        "rawLabel": "question",
        "similarity": 0.29099817371639164,
        "margin": 0.07314081092102523,
        "baseline": "question",
        "correct": false,
        "needsReview": true
      },
      {
        "id": 216,
        "expected": "question",
        "predicted": "unsure",
        "rawLabel": "documentation",
        "similarity": 0.3912484497689992,
        "margin": 0.052806230267178345,
        "baseline": "question",
        "correct": false,
        "needsReview": true
      },
      {
        "id": 217,
        "expected": "unsure",
        "predicted": "unsure",
        "rawLabel": "feature",
        "similarity": 0.1529623016373922,
        "margin": 0.024112245018460904,
        "baseline": "unsure",
        "correct": true,
        "needsReview": true
      },
      {
        "id": 218,
        "expected": "unsure",
        "predicted": "unsure",
        "rawLabel": "bug",
        "similarity": 0.07631912079877232,
        "margin": 0.024084821159938458,
        "baseline": "unsure",
        "correct": true,
        "needsReview": true
      },
      {
        "id": 219,
        "expected": "unsure",
        "predicted": "unsure",
        "rawLabel": "bug",
        "similarity": 0.16345005442474753,
        "margin": 0.09576364003466203,
        "baseline": "unsure",
        "correct": true,
        "needsReview": true
      },
      {
        "id": 220,
        "expected": "unsure",
        "predicted": "unsure",
        "rawLabel": "feature",
        "similarity": 0.0956703883846812,
        "margin": 0.029711023467636155,
        "baseline": "unsure",
        "correct": true,
        "needsReview": true
      }
    ]
  },
  "duplicates": {
    "total": 6,
    "truePositive": 2,
    "falsePositive": 0,
    "falseNegative": 1,
    "precision": 1,
    "recall": 0.6666666666666666,
    "pairs": [
      {
        "a": 301,
        "b": 302,
        "expected": true,
        "predicted": true,
        "similarity": 0.967697647000363
      },
      {
        "a": 303,
        "b": 304,
        "expected": true,
        "predicted": true,
        "similarity": 0.8528556119954988
      },
      {
        "a": 305,
        "b": 306,
        "expected": true,
        "predicted": false,
        "similarity": 0.7824381158444897
      },
      {
        "a": 307,
        "b": 308,
        "expected": false,
        "predicted": false,
        "similarity": 0.6160265627326454
      },
      {
        "a": 309,
        "b": 310,
        "expected": false,
        "predicted": false,
        "similarity": 0.7018611566748348
      },
      {
        "a": 311,
        "b": 312,
        "expected": false,
        "predicted": false,
        "similarity": 0.4825772775360722
      }
    ]
  },
  "runtime": {
    "coldMs": 215,
    "warmMs": 46,
    "rssMiB": 246,
    "networkDisabled": true,
    "deterministicAt1e7": true
  },
  "workflowExecution": {
    "status": "success",
    "steps": [
      {
        "id": "input"
      },
      {
        "id": "validate-export",
        "status": "success"
      },
      {
        "id": "embed-locally",
        "status": "success"
      },
      {
        "id": "compare-and-abstain",
        "status": "success"
      },
      {
        "id": "build-review-queue",
        "status": "success"
      }
    ]
  },
  "limitations": [
    "Small synthetic fixture; no claim of real-world accuracy or human time saved.",
    "Similarity is not a calibrated probability. All proposed labels need user review before export.",
    "Long text is truncated at 256 wordpieces. English only; specialized issue vocabulary can fail.",
    "Near-duplicate scores are suggestions, never automatic closure or merge.",
    "Network-disabled proof blocks JavaScript fetch during inference; this is not an OS network sandbox."
  ]
}
← prototype/evidence/end-to-end-demo.jsonprototype/evidence/pre-evaluation-freeze.sha256 →