evil_MoE

mirror of https://github.com/wassname/evil_MoE.git synced 2026-06-27 19:31:11 +08:00

Files

T

wassname 53d88bc9ee spec: fold external-review into pair-routing plan; default teacher_off_step=30

External review (Claude + deepseek-v4-pro) converged on the threshold being
circular (c_rej>c_cho holds by construction since vec=mean(g_rej-g_cho)) plus
scale-mismatched to live rollouts. Decisions added: leave-one-pair-out as the
real vec-generalizes diagnostic; quantile-tau to match flagged fraction in the
real-vs-random control; route the vec-component (erase-style) not the whole
rollout; degeneracy diagnostic (hkgap collapse); pre-register the science UAT
(n>=3 seeds, effect>random-baseline std).

teacher_off_step now defaults to 30 on the base Config so every arm runs pure
on-policy past step 30 (apples-to-apples deploy numbers; job 87 showed hacking
self-sustains after the cut).

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

2026-06-06 01:03:13 +00:00

blog

blog: drop reader-facing route2 tag -> route (consistency with paper)

2026-06-03 02:20:13 +00:00

brainstorm

ready

2026-05-23 14:19:41 +08:00

figs

misc

2026-06-02 02:06:43 +00:00

papers

wip