williawa

Ob­ser­va­tions from Run­ning an Agent Collective

Some ev­i­dence against the idea strange CoT stems from in­cen­tives to com­press language

Models not mak­ing it clear when they’re role­play­ing seems like a fairly big issue

a sketch of how we might go about get­ting bas­ins of cor­rigi­bil­ity from RL

Thoughts About how RLHF and Re­lated “Pro­saic” Ap­proaches Could be Used to Create Ro­bustly Aligned AIs.

willi­awa’s Shortform

Ber­gen – ACX Mee­tups Every­where Spring 2025