Repository navigation
Conditional κ = 0.001652528731617094: PR589 bit supplier and dissolved owner map on PR645's gauge44 construction - #649
jeremykothe2 wants to merge 1 commit into
Conversation
…ved owner map on PR645's gauge44 construction
|
Thank you @jeremykothe2. This contribution is now retained on main in the scoped 100-PR review and contributor credits. Contribution: Tuned PR589 bit supplier and dissolved Gauge44 owners. Review and limits: Bit deficit is derived from the new supplier parameters with rejection controls. The bit circuit is cited rather than freshly reconstructed by this package. The complete original source, notices and relevant execution evidence are retained through the pinned source archive. The review distinguishes fresh local runs, source-bound historical Linux CI and prior exact-head review; it does not claim 100 fresh full mathematical audits. Closing this historical PR as reconciled. This preserves parallel and cumulative credit without replacing the selected frontier or promoting priced research targets to completed constructions. Reviewed head: |
Summary
Conditional κ = 0.001652528731617094 (conservative 0.001652528728737236). This is +1.43×10⁻⁶ over PR648 and +3.05×10⁻⁶ over PR645.
Two changes on PR645's gauge44 construction:
EXPECTED.jsonis pinned by git blobe293ddc5, byte-identical to main'sresearch/e8-bit-p111-tuned/tune/EXPECTED.json. The bit deficit is now derived from the bit word's parameters, 32·(888−567) = 10272, instead of PR543's hard-coded 9984, with an equality guard. With PR589 the complex side binds again.Verifier changes (reviewed):
code/assemble.py(load and derive the bit supplier), newcode/bit_supplier_controls.py, and a newbit-supplier-controlsstage inverify_global.pyasserting 12 rejected controls, including a revert to the hard-coded 9984. Reverting the call site to 9984 makes the full chain fail atcheck_moment. The other 20 Python files are byte-identical to PR645.independent/price_profile.pyis ported to PR589.Review: an independent reviewer judged the composition SOUND (packaging fixes applied). It checked the pinned blob against main, recomputed the deficit, confirmed that no bit constant is left at PR543 values, and re-traced all 12 controls to their intended assertions. It also recomputed κ: 0.001651096360425895 under PR543 (bit binds; equal to PR648) and 0.001651929911890057 under PR589 for the previous map.
Scope: main's ledger records PR589 as supplier-only headroom; composing it into a complex assembly is new here and should be reviewed as such. The retained
hosting/,evidence/andRESULT.jsonfiles are PR645's historical, PR543-based evidence; the binding evidence is the SPEC and the fresh replay below.Verification
Local result (Windows, CPython 3.11.11), from a fresh clone of commit 1b931ee with no reused artifacts:
verify_manifest.pyPASS (89 files);reproduce.py --until phasesrebuilt the local supplier from PR635;verify_global.pyon that fresh local passed invoice, assembly, admission-controls and bit-supplier-controls (12 rejected) and printed PASS_GAUGE44_NATIVE_EXACT_GLOBAL_ASSEMBLY 0.001652528731617094.Reproduce from the repo root at this commit (Python 3.11+, assertions on):
Conditional on PR645's premises (all-size compiler, common-label/mutable-source realization, normalized wrappers, whole-call hosting, precision/tape, exact-recovery, analytic transfer) and on PR589's bit-supplier assumptions in place of PR543's. No unconditional theorem or Lean certification is claimed.
Credits
relocate_two, generalised by our gap-merge; the dissolve move is new here.Prepared with Claude Code (Anthropic).