multi-layer-intervention

Tag

Cards List
#multi-layer-intervention

CircuitSteer: Geometrically Aligned Multi-Layer Steering via Sparse Autoencoder Circuits

arXiv cs.LG · 3d ago Cached

CircuitSteer is a novel framework that uses sparse autoencoders to identify and manipulate multi-layer semantic circuits in LLMs, enabling more robust and fluency-preserving behavioral steering compared to single-layer methods like CAA.

0 favorites 0 likes
← Back to home

Submit Feedback