Refusal in language models is mediated by a single direction in residual stream activations that can be erased to disable safety or added to elicit refusal.
Causal Learning and Reasoning , pages=
2 Pith papers cite this work. Polarity classification is still indexing.
2
Pith papers citing it
fields
cs.LG 2representative citing papers
Absorber LLM introduces causal synchronization to absorb context into parameters for memory-efficient long-context LLM inference while preserving causal effects.
citing papers explorer
-
Refusal in Language Models Is Mediated by a Single Direction
Refusal in language models is mediated by a single direction in residual stream activations that can be erased to disable safety or added to elicit refusal.
-
Absorber LLM: Harnessing Causal Synchronization for Test-Time Training
Absorber LLM introduces causal synchronization to absorb context into parameters for memory-efficient long-context LLM inference while preserving causal effects.