arXiv AI
Component and Dimension Sparsity in Transformer Refusal Mechanisms
Read the original on arXiv AI →The Flow has not summarised this story yet — read it at arXiv AI.
Related stories
arXiv Computation and Language
Sep 2
Over-Refusal and Representation Subspaces: A Mechanistic Analysis of Task-Conditioned Refusal in Aligned LLMs
arXiv:2603.27518v4 Announce Type: replace Abstract: Aligned language models that are trained to refuse harmful requests also exhibit over-refusal: they decline safe instructions that seemingly resemb...
arXiv Computation and Language
Sep 16
There Is More to Refusal in Large Language Models than a Single Direction
The paper challenges the notion that refusal in large language models is governed by a single direction in activation space. It demonstrates that different refusal and non‑compliance categories map to distinct geometric directions, yet steering along any of these directions yields similar refusal–over‑refusal trade‑offs, acting as a shared one‑dimensional control knob. Using sparse autoencoders, the authors reveal a structured internal representation of refusal, comprising a reusable core of shared latents and style‑ or domain‑specific latents, and show that linear interventions collapse this structure into uniform behavioral control.
arXiv Machine Learning
Aug 4
What Makes Position Zero Special? A Mechanistic Study of Position Zero Attention Sinks in LLMs
arXiv:2603. 06591v2 Announce Type: replace Abstract: Transformers frequently allocate disproportionate attention to specific tokens, a phenomenon known as attention sinks.
arXiv Computation and Language
Aug 25
The Communication Map of a Transformer
The paper introduces the Communication Map, a method that charts every potential communication channel in a transformer model using only its weights. It generalizes previous coupling metrics into a single coefficient covering all 18 connection classes, revealing that 70‑89% of head pairs are non‑randomly oriented and identifying strong or avoiding couplings. The authors demonstrate the map’s utility by recovering known induction circuits and uncovering a two‑dimensional stream subspace whose removal eliminates induction capabilities across several models.
arXiv Machine Learning
Sep 22
Decomposing Refusal Steering in Mixture-of-Experts Models
arXiv:2606.04160v2 Announce Type: replace-cross Abstract: Safety alignment in instruction-tuned large language models (LLMs) depends on a model's ability to reliably refuse harmful or disallowed requ...
arXiv AI
Aug 14
Steering the Language Axis: From Linear Decodability to Causal Control
arXiv:2608. 12334v1 Announce Type: cross Abstract: Despite the impressive multilingual capabilities of Large Language Models, the latent dynamics dictating language selection remain poorly understood.