Activation Steering

Activation Steering controls LLM behavior by algebraically injecting computed or learned steering vectors directly into intermediate hidden activations during inference execution, strictly without modifying the underlying stored parameter weights ^{[166, 196, 237]}. A notable advancement in this space leverages SAE to identify and manipulate interpretable latent features, providing a highly precise, semantically grounded source for generating these transient steering vectors ^[46]. Its Activation-Space Manipulation classification occupies a mathematically unique structural position: it is mechanistically distinct from both input-layer PE (, which operates exclusively on the discrete input token space) and traditional FT (, which alters permanent weight matrices). The Transient, Activation-Space coordinate profile dictates that activation steering imposes zero permanent structural modification on the base artifact, yet enforces an immediate, highly non-linear behavioral shift during the active inference session. A profound structural insight within this framework emerges when contrasting activation steering with TTA: while both represent transient execution-time adaptations, they physically bifurcate at the architectural scale boundary. TTA executes via transient gradient-based parameter updates at DL, FM, whereas activation steering operates via direct latent-space vector injection at LLM, MLLM.