Files
gbrain/evals/brainbench/baselines/main.json
T
Garry Tan a1476af084 feat(eval): BrainBench corpus — 141 fixtures, sealed gold, generator, schemas, committed baseline
Deterministic generator (Mulberry32 seed 42, whole-cloth fictional universe,
template prose with controlled difficulty incl. documented v1 reflex limits)
emits 135 fixtures + sealed gold across 7 categories with a 15% holdout
split; 6 hand-authored spike fixtures froze the schema. Published JSON
Schemas are the foreign-runner contract. Gold validated by blind double-label
(96.4% agreement; findings fixed, receipt in _ledger.json). Privacy guard
extended over the new corpus dirs.
2026-06-12 12:06:03 -07:00

120 lines
2.8 KiB
JSON

{
"schema_version": 1,
"fixtures_hash": "76f201590dd3ad7a929e2e12efc9bf1406627b10ef4edbcfe7caf379aafd4090",
"cells": {
"claude-code/continuity": {
"avg_injected_tokens": 75.3333,
"continuity_rate": 1,
"source_isolation_violations": 0
},
"claude-code/know-to-ask": {
"avg_injected_tokens": 30.9247,
"false_fire_rate": 0.0233,
"know_to_ask_failure_rate": 0.15,
"source_isolation_violations": 0
},
"claude-code/push": {
"avg_injected_tokens": 38.8077,
"push_precision": 1,
"push_recall": 0.6596,
"source_isolation_violations": 0
},
"claude-code/write-back": {
"provenance_accuracy": 1,
"write_back_fidelity": 1
},
"codex/continuity": {
"avg_injected_tokens": 150.9167,
"continuity_rate": 1,
"source_isolation_violations": 0
},
"codex/know-to-ask": {
"avg_injected_tokens": 40.7123,
"false_fire_rate": 0,
"know_to_ask_failure_rate": 0.15,
"source_isolation_violations": 0
},
"codex/push": {
"avg_injected_tokens": 48.5865,
"push_precision": 1,
"push_recall": 0.4468,
"source_isolation_violations": 0
},
"codex/write-back": {
"provenance_accuracy": 1,
"write_back_fidelity": 1
},
"openclaw/continuity": {
"avg_injected_tokens": 75.3333,
"continuity_rate": 1,
"source_isolation_violations": 0
},
"openclaw/know-to-ask": {
"avg_injected_tokens": 33.7945,
"false_fire_rate": 0,
"know_to_ask_failure_rate": 0.15,
"source_isolation_violations": 0
},
"openclaw/push": {
"avg_injected_tokens": 44.1346,
"push_precision": 1,
"push_recall": 0.8085,
"source_isolation_violations": 0
},
"openclaw/write-back": {
"provenance_accuracy": 1,
"write_back_fidelity": 1
}
},
"counts": {
"claude-code/continuity": {
"gold_total": 24,
"gold_failed": 0
},
"claude-code/know-to-ask": {
"gold_total": 146,
"gold_failed": 11
},
"claude-code/push": {
"gold_total": 94,
"gold_failed": 32
},
"claude-code/write-back": {
"gold_total": 46,
"gold_failed": 0
},
"codex/continuity": {
"gold_total": 24,
"gold_failed": 0
},
"codex/know-to-ask": {
"gold_total": 146,
"gold_failed": 9
},
"codex/push": {
"gold_total": 94,
"gold_failed": 52
},
"codex/write-back": {
"gold_total": 46,
"gold_failed": 0
},
"openclaw/continuity": {
"gold_total": 24,
"gold_failed": 0
},
"openclaw/know-to-ask": {
"gold_total": 146,
"gold_failed": 9
},
"openclaw/push": {
"gold_total": 94,
"gold_failed": 18
},
"openclaw/write-back": {
"gold_total": 46,
"gold_failed": 0
}
}
}