Massive Activations in Hybrid Linear Attention Large Language Models: Pre-Attention Spikes and Inter-Spike Plateaus
Paper • 2608.12149 • Published • 16 •
2
tran huyen
tran073huyen
AI & ML interests
None yet
Recent Activity
commentedon a paper about 13 hours ago
Massive Activations in Hybrid Linear Attention Large Language Models: Pre-Attention Spikes and Inter-Spike PlateausOrganizations
None yet
commented a
paper about 13 hours ago