• dream_weasel@sh.itjust.works
    link
    fedilink
    English
    arrow-up
    4
    arrow-down
    3
    ·
    4 months ago

    The embedding layer post tokenization is not just a probability machine the way you’re suggesting it. You can argue that it is probabilistic with inferred sentiment, but too many people think it works like how text prediction on your phone does and that is just factually inaccurate.

    Verify output of course, but saying “it doesn’t understand anything” and “probability machine” is a borderline erroneous short sell. At the level of tokens it “understands” relationships, and those relationships are not probabilistic, though they are fundamentally approximated based on a training corpus.

    • hesh@quokk.au
      link
      fedilink
      English
      arrow-up
      5
      arrow-down
      1
      ·
      4 months ago

      Can you explain how it’s more than probability? It’s using a neural network to guess the most likely next token, isn’t it?

      • SlimePirate@lemmy.dbzer0.com
        link
        fedilink
        English
        arrow-up
        1
        arrow-down
        2
        ·
        4 months ago

        The fact that it uses a non-trivial neural network. If it was simply a rate count of based on a corpus of how much time each word is followed by each it wouldn’t be stronger than keyboard word predictions. To make accurate suggestions requires emergence of primitive reasoning on the semantics of the tokens, LLM neural networks (transformers) can be analyzed to find subnetworks dedicated to modeling reality. It is still probability, but saying it’s just probability is not faithful

        • hesh@quokk.au
          link
          fedilink
          English
          arrow-up
          4
          ·
          4 months ago

          It’s still just predicting the next token, it’s just using more past data points than your keyboard. The rest of the phenomena are emergent from that. I think it’s important to keep that in mind given how much they can imitate human reasoning.