J-Space publishes research explainers on AI interpretability, the Jacobian Lens, and internal model cognition.
Our goal is to make emerging interpretability research clear, rigorous, and useful for researchers, engineers, and informed readers.
Explore the latest work on our Explainers page.