Interpreting Latent Reasoning Models with Natural Language Autoencoders

Modelling Takeoff

Creating a better way to compile to transformer architectures

Flourishing Futures Fellowship

A scalable fellowship training researchers to develop interventions for achieving high-value long-term futures

Theoretical research of inductive reasoning building on LTA

Development time for next film project

Testing failure modes of debate-style AI control schemes

Core funding for PauseAI UK

Staff salaries and other core expenditures for PauseAI UK

Movement-building to make AGI go well

Sapiens First organizes voters to fight concentration of power and for AI safety

Mechanistic estimation of highly out-of-distribution behaviors

Debate training on LLMs as a reward-hacking mitigation

Discrete Language Modeling

Making sure that the code AI pushes to the internet is safe and bug-free

Building LemmaScript, a verification toolchain for TypeScript

Alignment Target Analysis

AFFINE Event-Prep

Runway for the core staff to design the program, find mentors and get funding