Search

Aleksandra Bakalova

Discovering Interpretable Algorithms by Decompiling Transformers to RASP
How Few-Shot Examples Add Up: A Causal Decomposition of Function Vectors in In-Context Learning
Contextualize-then-Aggregate: Circuits for In-Context Learning in Gemma-2 2B

Published with Wowchemy — the free, open source website builder that empowers creators.