‹ BackHN Continuity

Thread

Breaking the 1.58-bit Barrier for Ternary LLMs

245 points · 41 comments · matt_d

  1. wgd · · focus · HN ↗
    Only a presence bitmap? If we're contemplating packing schemes I'm tempted to write a paper that uses arithmetic coding to squeeze out a few more centi-bits.
    1. akoboldfrying · · focus · HN ↗
      Agreed. One possible objection might be that they need fast random access to weights, but I skimmed parts of the paper and it looks like they process 128 entries at a time, which to me sounds like it should be amenable to better compression: short enough that better compression results could still be efficiently cached in faster local RAM, long enough that better compression would save useful amounts of memory per block.
Open on Hacker News to reply ↗

Unofficial Hacker News client; not affiliated with Y Combinator.