• RedPander@lemmy.rogers-net.com
    link
    fedilink
    arrow-up
    10
    ·
    1 year ago

    Wild that it’s mentioned multiple times in here how the large data sets aren’t really an advantage. Reminds me of Nate Silver’s The Signal and the Noise. Sounds like Google and OpenAI have a lot of noise right now.

    • Communist@beehaw.org
      link
      fedilink
      arrow-up
      4
      ·
      1 year ago

      It’s not, this method of analysis is terrible, they’re just asking gpt4 to grade the responses, not actually testing anything beyond that.

      • aponigricon@beehaw.org
        link
        fedilink
        English
        arrow-up
        4
        ·
        1 year ago

        That doesn’t necessarily invalidate the point they’re making. Other forms of analysis, strikingly, provide pretty much completely equivalent results.

  • worfamerryman@beehaw.org
    link
    fedilink
    arrow-up
    4
    ·
    1 year ago

    I’d love to try to run this on my home server. For some reason, chatgpt always errors out in my from my windows computer. I’m not sure why. It works fine in Linux, Mac, and iPhone. But my work is on windows.

  • guildz@lemmy.blahaj.zone
    link
    fedilink
    arrow-up
    1
    ·
    1 year ago

    Does anyone know of a person tracking these updates for both open source art generators and LLMs? Like I am following a few people and they are all just lists of the corprate made tools being built on top of these things.