Sentence-level contextual entrainment exists across LLMs, weakens with scale, and is localized to 2-4% of attention heads whose deactivation removes the effect without performance loss.
Llama See, Llama Do: A Mechanistic Perspective on Contextual Entrainment and Distraction in LLM s
3 Pith papers cite this work. Polarity classification is still indexing.
3
Pith papers citing it
fields
cs.CL 3years
2026 3verdicts
UNVERDICTED 3representative citing papers
Dutch LLMs display coherence illusions tracked by surprisal, with attention entropy identifying affected heads and a new energy metric quantifying discourse coherence.
The survey organizes mechanistic interpretability techniques into a Locate-Steer-Improve framework to enable actionable improvements in LLM alignment, capability, and efficiency.
citing papers explorer
-
Sentence-Level Contextual Entrainment in Large Language Models
Sentence-level contextual entrainment exists across LLMs, weakens with scale, and is localized to 2-4% of attention heads whose deactivation removes the effect without performance loss.
-
When Context Misleads: Surprisal, Energy and Attention Entropy as Metrics of Coherence Illusions in LLMs
Dutch LLMs display coherence illusions tracked by surprisal, with attention entropy identifying affected heads and a new energy metric quantifying discourse coherence.
-
Locate, Steer, and Improve: A Practical Survey of Actionable Mechanistic Interpretability in Large Language Models
The survey organizes mechanistic interpretability techniques into a Locate-Steer-Improve framework to enable actionable improvements in LLM alignment, capability, and efficiency.