Six frontier language models show categorical, model-specific response modes under steering pressure, with GPT-5 uniquely withholding reasoning while answering, and a linear probe plus activation steering tracing Llama's derail behavior to its residual stream.
Findings of the Association for Computational Linguistics: ACL 2023 , pages =
1 Pith paper cite this work. Polarity classification is still indexing.
1
Pith paper citing it
fields
cs.AI 1years
2026 1verdicts
CONDITIONAL 1representative citing papers
citing papers explorer
-
Divergent Response Modes in Frontier Language Models Under Steering Pressure
Six frontier language models show categorical, model-specific response modes under steering pressure, with GPT-5 uniquely withholding reasoning while answering, and a linear probe plus activation steering tracing Llama's derail behavior to its residual stream.