x

AI ALIGNMENT FORUM

AF

J Rosser — AI Alignment Forum

J Rosser

Top postsTop post

J Rosser

Message

Hi everyone! I'm J! I'm currently doing a DPhil in Machine Learning (AI Safety) at the University of Oxford, and a previous collaborator at UK AISI.

78

4

1

1y

J Rosser

Hi everyone! I'm J! I'm currently doing a DPhil in Machine Learning (AI Safety) at the University of Oxford, and a previous collaborator at UK AISI.

Data filtering works a lot worse than you would expect

by Dohun Lee, J Rosser, Josh Engels, and Neel Nanda

This work was largely done during Neel Nanda's MATS 10.0 Exploration Phase. J Rosser and Dohun Lee are co-first authors for this post with equal contribution. Josh Engels and Neel Nanda supervised the project, and provided guidance and feedback throughout. TLDR * Models can acquire undesirable traits from during supervised...