pith. sign in

Axbench: Steering LLMs? even simple base- lines outperform sparse autoencoders

2 Pith papers cite this work. Polarity classification is still indexing.

2 Pith papers citing it

fields

cs.CL 2

years

2026 2

verdicts

UNVERDICTED 2

representative citing papers

Psychological Steering of Large Language Models

cs.CL · 2026-04-15 · unverdicted · novelty 7.0

Mean-difference residual stream injections outperform personality prompting for OCEAN trait steering in most LLMs, with hybrids performing best and showing approximate linearity but non-human trait covariances.

citing papers explorer

Showing 2 of 2 citing papers.

  • Psychological Steering of Large Language Models cs.CL · 2026-04-15 · unverdicted · none · ref 68

    Mean-difference residual stream injections outperform personality prompting for OCEAN trait steering in most LLMs, with hybrids performing best and showing approximate linearity but non-human trait covariances.

  • Don't Lose Focus: Activation Steering via Key-Orthogonal Projections cs.CL · 2026-05-07 · unverdicted · none · ref 43

    SKOP uses key-orthogonal projections to steer LLM activations while preserving attention patterns on focus tokens, cutting utility degradation by 5-7x and retaining over 95% of standard steering efficacy.