Self

Hi I’m Leon Lenk, I’m currently a Research Fellow at the Supervised Program for Alignment Research investigating whether reinforcement learning elicits or generates abilities. I often think of my motivations in terms of shard theory, thus my website is organized such that each piece is a “shard” of my motivations. Currently I’m interested in x-risk from AI and encompassing theories of how LLMs work (though that’s somewhat fallen out of favor).Outside of AIS research I like to draw, write, and play minecraft.