Abacus.ai:

We recently released Smaug-72B-v0.1 which has taken first place on the Open LLM Leaderboard by HuggingFace. It is the first open-source model to have an average score more than 80.

  • TheChurn@kbin.social
    link
    fedilink
    arrow-up
    48
    ·
    9 months ago

    Every billion parameters needs about 2 GB of VRAM - if using bfloat16 representation. 16 bits per parameter, 8 bits per byte -> 2 bytes per parameter.

    1 billion parameters ~ 2 Billion bytes ~ 2 GB.

    From the name, this model has 72 Billion parameters, so ~144 GB of VRAM

    • FaceDeer@kbin.social
      link
      fedilink
      arrow-up
      11
      ·
      9 months ago

      It’s been discovered that you can reduce the bits per parameter down to 4 or 5 and still get good results. Just saw a paper this morning describing a technique to get down to 2.5 bits per parameter, even, and apparently it 's fine. We’ll see if that works out in practice I guess

      • Corngood@lemmy.ml
        link
        fedilink
        English
        arrow-up
        2
        ·
        edit-2
        9 months ago

        I’m more experienced with graphics than ML, but wouldn’t that cause a significant increase in computation time, since those aren’t native types for arithmetic? Maybe that’s not a big problem?

        If you have a link for the paper I’d like to check it out.

        • FaceDeer@kbin.social
          link
          fedilink
          arrow-up
          14
          ·
          9 months ago

          My understanding is that the bottleneck for the GPU is moving data into and out of it, not the processing of the data once it’s in there. So if you can get the whole model crammed into VRAM it’s still faster even if you have to do some extra work unpacking and repacking it during processing time.

          The paper was posted on /r/localLLaMA.

        • L_Acacia@lemmy.one
          link
          fedilink
          English
          arrow-up
          4
          ·
          9 months ago

          You can take a look at exllama and llama.cpp source code on github if you want to see how it is implemented.

    • rs137@lemmy.world
      link
      fedilink
      English
      arrow-up
      1
      ·
      9 months ago

      Llama 2 70B with 8b quantization takes around 80GB VRAM if I remember correctly. I’ve tested it a while ago.