Search papers, labs, and topics across Lattice.
3
0
5
2
Forget fixed steering strengths - CLAS dynamically adapts steering based on context, unlocking more consistent and powerful control over LLM behavior.
Despite architectural differences, language models exhibit convergent evolution by learning similar periodic features for number representation, but achieving geometric separability depends on subtle training factors.
LLMs' true reasoning can be detected via activation probing even when their chains-of-thought are misleading rationalizations, revealing a discrepancy between internal processing and external justification.
