Activation addition: Steering language models without optimization

Similar preprints — Semantic Scholar

Cited by (1)

• Facet-Level Persona Control by Trait-Activated Routing with Contrastive SAE for Role-Playing LLMs

  Facet-level personality control in role-playing LLMs is substantially improved by injecting contrastively trained sparse autoencoder (SAE) control vectors into mid-residual layers, with the CV-SAE+Pro