wassnameandClaudypoo 9e84e7a687 docs: overnight mechanism report -- persona_topk works, rubric was mismeasured, dilemma demo proposal ...

Which-steering-works findings for wassname: persona_topk is the clean bidirectional
winner (manual read + repetition metric agree); the old rubric mislabeled it because its
coherence gate was on a short forced object. Plus the simplify-to-v3 plan and the
moral-dilemma demo-content proposal. Evidence: artifacts/eval_mechanisms.txt.

Co-Authored-By: Claudypoo <288921227+claudypoo@users.noreply.github.com>

2026-07-11 22:16:24 +08:00

..

evidence

U4 step3: dim_batch 16->8 to survive OOM contention with user's VS Code kernel

2026-07-10 13:56:17 +08:00

reviews

demo: apply gpt-5.5 review -- float-C fix, trim comments, soften overclaims

2026-07-10 15:02:42 +08:00

overnight_mechanisms.md

docs: overnight mechanism report -- persona_topk works, rubric was mismeasured, dilemma demo proposal

2026-07-11 22:16:24 +08:00