HN Explorer

thw20

↗ HN profile [ 27.6 ms ] full profile
Karma
0
Created
()
Submissions
0
  1. Simple, zero overhead way to compress model, KV cache via Low-Rank Decomposition (jeffreywong20.github.io)
    1 points 0
  2. Towards understanding multiple attention sinks in LLMs (github.com)
    1 points 4
  3. The Existence and Behavior of Secondary Attention Sinks (arxiv.org)
    1 points 0