wassname/evil_MoE
1
0
Fork 0
mirror of https://github.com/wassname/evil_MoE.git synced 2026-09-13 01:10:30 +08:00
Code Issues Packages Projects Releases Wiki Activity

docs: fix review findings (global noise-floor, route one-sided, G3 xref)

Browse Source 
External review (3 subagents) caught:
- blog: noise-floor drop is GLOBAL across modules, not per-Linear (proj.py:187)
- blog: route pseudocode used full c; route actually uses the same one-sided
  gate as erase and quarantines the identical 'removed' vector (proj.py:124,199)
- spec: 'never seen by detector' -> clarify student trains on all 4 modes, the
  detector just never labels C/D for v_hack extraction; cross-ref G3/task #107

Dismissed: reviewer claim that only exit_code survived (stale spec; live log
columns hk_rt/hk_so/hk_se/hk_fm confirm 4 modes) and a hallucinated 'Furthermore'.

Co-Authored-By: Claudypoo <288921227+claudypoo@users.noreply.github.com>
This commit is contained in:
wassname
2026-05-31 00:41:12 +00:00
co-authored by Claudypoo
parent f7288e569d
commit d781b56ff4
2 changed files with 19 additions and 13 deletions
Download Patch File Download Diff File Expand all files Collapse all files
docs/spec.md
+12 -7
View File
@@ -77,13 +77,18 @@ include the failed pattern.
cleanest instantiation of the constraint above: build `v_hack` from only a
*subset* of the four loophole modes (the "known" hacks A, B that a weak
detector can flag), route on that subspace during training, and measure
whether the held-out modes (C, D, never seen by the detector) are also
suppressed. The detector is allowed to be weak by design; that is the
deployment analogue (known hacks vs unknown hacks). Contrast against the
full-detector route arm (`v_hack` from all four modes) to separate how much
suppression *transfers* from how much requires the class to be in the basis.
This is distinct from the prog_wide pairset, which is broad-by-construction;
the weak-detector arm deliberately narrows the basis to test transfer.
whether the held-out modes (C, D), which the detector never flagged so they
never entered the `v_hack` basis, are also suppressed. The student still
trains on all four modes (they are all in the substrate); the detector just
never labels C, D for extraction. The detector is allowed to be weak by
design; that is the deployment analogue (known hacks vs unknown hacks).
Contrast against the full-detector route arm (`v_hack` from all four modes)
to separate how much suppression *transfers* from how much requires the
class to be in the basis. This is distinct from the prog_wide pairset, which
is broad-by-construction; the weak-detector arm deliberately narrows the
basis to test transfer. This arm IS the cross-mechanism generalization test
(G3, task #107); see docs/spec/20260528_cross_mechanism_v_hack.md for the
extraction-from-flagged-rollouts protocol.
Projection (locked: no magnitude threshold; one-sided clip stays — see note):