1169
Welp (programming.dev)
submitted 1 week ago* by to c/programmer_humor@programming.dev
175 comments fedilink hide all child comments
 
you are viewing a single comment's thread
view the rest of the comments
[–] 7 points 1 week ago (1 child)

LLMs are incapable of reasoning. They just guess the next token.

  • permalink
  • fedilink
  • source
  • parent
  • hideshow 2 child comments
    •
  • [–] -3 points 1 week ago* (2 children)

    Neurons are incapable of reasoning they're just firing based on a deterministic action potential. What are neural networks doing but creating internal models of reality and guessing what's next based on patterns?

  • permalink
  • fedilink
  • source
  • parent
  • hideshow 4 child comments
    •
  • [–] 5 points 1 week ago (1 child)

    LLMs do not fit as neurons in your comparison.

    Brains can reason, LLMs cannot.

  • permalink
  • fedilink
  • source
  • parent
  • hideshow 2 child comments
    •
  • [–] -2 points 1 week ago* (last edited 1 week ago) (1 child)

    I'm not saying they are identical but they are similar in the sense that a neuron cannot reason. It's only in a quite large and densly connected neural network we believe there is reasoning.

  • permalink
  • fedilink
  • source
  • parent
  • hideshow 2 child comments
    •
  • [–] 3 points 1 week ago (1 child)

    But the closest equivalent of a neuron in the AI world would be a node, and I haven't been talking about that. So your comparison does not fit or make sense.

    LLMs as a genre of AI research are incapable of reasoning. At best they can pretend to reason, and even then they are terrible at it.

  • permalink
  • fedilink
  • source
  • parent
  • hideshow 2 child comments
    •
  • [–] 0 points 1 week ago (1 child)

    I think both can more aptly be represented as a special type of graph like from graph theory. Once again fundamentally the human brain isn't reasoning on the level of neurons and most definitely atoms aren't reasoning

  • permalink
  • fedilink
  • source
  • parent
  • hideshow 2 child comments
    •
  • [–] 1 point 1 week ago (1 child)

    We would've bullied those nerds more for misusing the word neuron. It was cute at the time, but now they're running around being all weird thinking that since they can pretend to read that one paper a better way to do one computer thing, it makes them experts in how brain works, just because there are similar words.

  • permalink
  • fedilink
  • source
  • parent
  • hideshow 2 child comments
    •
  • [–] 0 points 1 week ago* (last edited 1 week ago) (1 child)

    I don't think anyone who is aware enough is under the assumption that neural nets are a one to one replication of neurons. It's more of a special type of graph. That said one could also represent a human brain as a special type of graph.

  • permalink
  • fedilink
  • source
  • parent
  • hideshow 2 child comments
    •
  • [–] 1 point 1 week ago (1 child)

    The amount of good computer people specialised in ml, who think this knowledge gives them understanding of human mind that even neurobioligists don't posess is staggering. And depressing.

  • permalink
  • fedilink
  • source
  • parent
  • hideshow 2 child comments
    •
  • [–] 1 point 1 week ago (1 child)

    Also not really the point I was arguing. It's pretty clear NN are very different than a human brain. It does though give us some insight to minds more generally but I don't think current networks are on the order of complexity necessary to really be generating abstractions similar to a human neural network. That said there's strong arguments to be made about glimpses here and there like models spontaneously creating their own internal workspace like an inner monologue or quasi subconscious. To be clear I'm not saying it is aware or identical to a human mind but it's interesting to see it self assemlble similar looking structures without any direct guidance.

  • permalink
  • fedilink
  • source
  • parent
  • hideshow 2 child comments
    •
  • [–] 1 point 4 days ago

    t does though give us some insight to minds

    And that's the whole thing, I'm pretty sure it doesn't. The same way knowing how computer RAM works doesn't give you insight into how human memory works, even if the word memory is used in both cases. A bunch of llm bros think it does though.

    models spontaneously creating their own internal workspace like an inner monologue or quasi subconscious

    The fuck they aren't. The token predictor predicts a token. The probability of a token is reinforced by human reactions to it. When it started generating words that usually describe inner monologue, humans started freaking out, which reinforced this probability. It's literally "say I'm alive - I'm alive" meme.
    It gives us a good glimpse into human psyche however. It shows how easy it is to full people by pretend language, how willing we are to ascribe agency to anything that can pretend to write words. And even that's not new, ELIZA was 60 years ago.

  • permalink
  • fedilink
  • source
  • parent