Contrastive activation steering of LLMs is reliable only in-distribution and harms model perplexity, while larger models resist the damage better.
Title resolution pending
1 Pith paper cite this work. Polarity classification is still indexing.
1
Pith paper citing it
fields
cs.AI 1years
2025 1verdicts
CONDITIONAL 1representative citing papers
citing papers explorer
-
Patterns and Mechanisms of Contrastive Activation Engineering
Contrastive activation steering of LLMs is reliable only in-distribution and harms model perplexity, while larger models resist the damage better.