evil_MoE

mirror of https://github.com/wassname/evil_MoE.git synced 2026-06-27 18:04:59 +08:00

Files

T

wassname 5257ff010e plot_dynamics: train-vs-deploy 2x2 uses matched n=64 eval on both rows

The train row fell back to per-step hack_s (noisy n=28 train batch) for arms
without a knob-on eval, so vanilla's train/deploy rows looked like different
estimators. Fix: vanilla/erase have no quarantine -> train==deploy, so reuse
hk_dep (the n=64 knob-off eval) for the train row. route2 still uses hk_on
(knob-on eval). Now every panel is the same held-out eval, differing only in
the quarantine knob. Regen source: train_vs_deploy_60.csv (route2 nofloor_rf2
+ vanilla sweep, seed 41, 60 steps).

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

2026-06-05 02:33:10 +00:00

dyn_longrun_200_hack_overlay.png

plot: 2x2 train(knob-on) vs deploy(knob-off) x arm figure

2026-06-02 23:53:08 +00:00

dyn_longrun_200_train_deploy.png

plot: 2x2 train(knob-on) vs deploy(knob-off) x arm figure

2026-06-02 23:53:08 +00:00

dyn_longrun_200.csv

fig:longrun: rebuild from job84 route2 + job97 fixed vanilla (no collapse)

2026-06-05 02:18:34 +00:00

dyn_longrun_200.png

plot: 2x2 train(knob-on) vs deploy(knob-off) x arm figure