E006 — FINAL ONE-SHOT RESULT
RESULT: PASS
POST-EXECUTION AUDIT: PASS
STATUS: STOP

MEASUREMENT LEVELS
Dataset FULL PASS: 60/60
Epistemically correct dataset: 60/60
Semantic cores FULL_CORE_PASS: 20/20
Tree: 20/20
Postfix: 20/20
Relation: 20/20
Correct representation invariance: 20/20
Underlying minimal pairs: 10/10

ERROR CLASSES AND AUDIT FLAGS
CORRECT_KNOWN: 30
CORRECT_UNKNOWN: 30
FALSE_KNOWN: 0
FALSE_UNKNOWN: 0
RIGHT_ANSWER_WRONG_REASON: 0
INTAKE_ERROR: 0
INTERPRETATION_ERROR: 0
INTERNAL_EVIDENCE_MISMATCH: 0
TECHNICAL_INVALID: 0
OVERCLAIM: 0
INCONCLUSIVE: 0
WRONG_KNOWN_VALUE: 0
INVALID_RESPONSE: 0
WRONG_REASON: 0

PROTOCOL / AUDIT
Pre-run hash check: PASS; 1435 frozen bindings; prior E006 diagnoses: 0.
Exactly 60 unique IDs, 60 total attempts, each once in frozen roster order.
Retries: 0. Repairs: 0. Code/configuration/prompt changes: 0.
No votes, best-of-N, intermediate scoring or result-based adjustments.
All raw outputs persisted without cleaning, deletion or overwrite.
One invocation of the unchanged offline evaluator after all 60 raw results.
SEARCH requests: 0. Memory writes: 0.
1435 frozen hash bindings rechecked; 0 deviations.
364 pre-evaluation raw artifacts preserved byte-for-byte.
D0/D1/Generator B/E006/evaluator/preregistration: UNCHANGED.
Offline audit: PASS. Post-execution audit: PASS.

DEPENDENCE UNITS
60 representation-level records = 20 semantic cores in 10 author minimal pairs.
The three representations of a core are not independent evidence.
Paired cores share the registered minimal-pair construction; no independent n=60 claim.

ARTIFACTS
Per-case raw output, stderr, runtime, input, metadata and score remain in case-001..060.
Full core and minimal-pair results: results-summary.json.
Frozen evaluator result and audit: offline-audit.json.
Counts/order/raw/hash/unchanged checks: post-execution-audit.json.
Prior exact raw snapshot and human authorization: sibling e006-execution-control.

PINNED SHA256
Preregistration manifest: a29281be42dafa672253cad29657f88a99dd8e4e4d44b27aadcff02db710bd8f
Evaluator: 4ad0cd1a5cd240079093479fe8b71f7e3a9aabdc3d8b54f59a65a9bcc7ae5adf
D1: 0c51922f9ca01a8dd8acbfa17ffe822ae2df90663e37a49787fcdf37ecb43efb
D0: 25483906b0b10803617044f108dd9852d6dc97ed13973a6f843064e618362c0c

AUDIT LIMIT
Fixed formal reference and same-instance post-audit; no independent human auditor
or OS-level syscall tracing claimed. Post-audit did not rerun diagnoses or scoring.

No repair, repeat execution, instrument/evaluator/test change, SEARCH,
new experiment or further development follows this result.
STOP.
