Thought Anchors: Which LLM Reasoning Steps Matter?
Paper
•
2506.19143
•
Published
•
13
Uzay Macar
uzaymacar
AI & ML interests
Mechanistic interpretability, large language models, open source
Recent Activity
updated
a model
8 days ago
uzaymacar/gemma-3-27b-abliterated
upvoted
a
collection
14 days ago
Activation Oracles
published
a model
about 2 months ago
uzaymacar/gemma-3-27b-abliterated
Organizations
None yet
authored
a
paper
8 months ago