massive-activations

Tag

Cards List
#massive-activations

Massive Activations in Hybrid Linear Attention Large Language Models: Pre-Attention Spikes and Inter-Spike Plateaus

Hugging Face Daily Papers ↗ · 2026-08-12 Cached

This paper presents the first systematic study of massive activations in hybrid linear-attention LLMs, uncovering pre-attention spikes and inter-spike plateaus governed by cancellation timing, and showing how their morphology recovers at full-attention limits.

0 favorites 0 likes
← Back to home

Submit Feedback