Linear probes on residual-stream activations identify a shared preference vector in LLMs that tracks choices across prompts and causally steers decisions even for anti-correlated personas.
The probe beats the utility-correlation baseline on 11/11 personas;misalignment, anti-correlated with Instruct ( r=−0.14 ), shows the largest gain (r= 0.25 )
1 Pith paper cite this work. Polarity classification is still indexing.
1
Pith paper citing it
fields
cs.CL 1years
2026 1verdicts
UNVERDICTED 1representative citing papers
citing papers explorer
-
Probing Persona-Dependent Preferences in Language Models
Linear probes on residual-stream activations identify a shared preference vector in LLMs that tracks choices across prompts and causally steers decisions even for anti-correlated personas.