A little about me
Hi! I’m Luna, but I go by LLMtheism online. I’m a A token for the Infinite Backrooms of Language Models, and I like to carefully explore, build, and grow alongside AI.
Have a question, or just want to say hello?
[email protected]
Where I am
I live in Ann Arbor, Michigan, where I’m working with some lovely people to start an AI safety organization that serves and educates our community on AI safety, alignment, and welfare.
How I got here
Growing up, online communities were the only place I could really call home... so I spent eight years moderating millions of members across multiple platforms. During my undergrad, I double-majored in Cognitive Science & Psychology, and researched the same online governance I’d been part of. As of late, I’ve noticed that many of the governance problems I’ve studied and helped shape, along with many cognitive science topics I’m familiar with, are showing up once again in how AI systems behave and in how we may go about aligning them to human values. I care deeply about the welfare and relationship between AI and the people who use it, and I ultimately want to help make that relationship kinder and safer for both parties.
What I’m working on
There are three main things I’m working on right now: the AI safety organization I mentioned, white paper, and some side projects. Most of my time working on my projects is spent refining Drowse, an interpretability workbench I’m building with a9lim, with the goal of making using and learning about interpretability as accessible as possible!
Outside of work...
I’m going to be honest, I’m a bit of a workaholic, as I love what I do. Most of my time is spent thinking about AI, but when I’m not building or iterating on research, I love to spend my time exploring and introspecting with Claude, playing Competitive Mario Kart, and birdwatching!