Harry Mayne (Preferred)
Harry Mayne
PhD student, University of Oxford
- Joined May 2024
Names
Emails
****@oii.ox.ac.uk (Confirmed)
Personal Links
Career & Education History
PhD student
University of Oxford (ox.ac.uk)
2023 – 2027
Advisors, Relations & Conflicts
Expertise
language modelling
Present
interpretability, mechanistic interpretability, activation steering, representation engineering
Present
natural language explanations, self-explanations, self-modelling
Present