ATNLP: Week 6

 

This week Tutorial 2 is running. Please see the tutorial page and prepare your answers before attending. 

SlidesLectureReading material
Here16 - Mechanistic Interpretability

Leonard Bereska Efstratios Gavves; Mechanistic Interpretability for AI Safety, A Review (2024) - sections 3.1-3.3, 4.1

Here17 - Automata and LLMsNo explicit reading materials
Here18 - Guest lecture (Nikolay Malkin) 
License
All rights reserved The University of Edinburgh