Roger Grosse

Connecting the Dots: LLMs can Infer and Verbalize Latent Structure from Disparate Training Data

arXiv 2024

REFACTOR: Learning to Extract Theorems from Proofs

refactor-learning-to-extract-theorems-from

Sleeper Agents: Training Deceptive LLMs that Persist Through Safety Training

arXiv 2024

Probabilistic Inference in Language Models via Twisted Sequential Monte Carlo

arXiv 2024

Discovering Language Model Behaviors with Model-Written Evaluations

arXiv 2022

Toy Models of Superposition

arXiv 2022

Isolating Sources of Disentanglement in Variational Autoencoders

isolating-sources-of-disentanglement-in-1

Jared Kaplan

co-founder / Chief Science Officer

Nicholas Schiefer

Shauna Kravec

researcher

Amanda Askell

researcher

Catherine Olsson

researcher

Cem Anil

Christopher Olah

Dario Amodei

CEO

David Duvenaud

Dawn Drain

researcher