The Logical Implication Steering Method for Conditional Interventions on Transformer Generation

The field of mechanistic interpretability in pretrained transformer models has demonstrated substantial evidence supporting the "linear representation hypothesis", which is the idea that high level concepts are encoded as vectors in the space of activations of a model. Studies also show that model generation behavior can be steered toward a given concept by adding the concept's vector to the corresponding activations. We show how to leverage these properties to build a form of logical implication into models, enabling transparent and interpretable adjustments that induce a chosen generation behavior in response to the presence of any given concept. Our method, Logical Implication Model Steering (LIMS), unlocks new hand-engineered reasoning capabilities by integrating neuro-symbolic logic into pre-trained transformer models.