Show HN: TurboGPT: train 22KiB transformer in 13s

(github.com)

43 points | by lostmsu 6 hours ago ago

8 comments

  • _345 2 hours ago

    Trying to be a little less negative than the guy that got flagged, I too don't understand the motivation behind these projects. I've seen a hundred of them at this point- and each of them is probably worse and has less learning value than the one Andrej Karpathy made to teach people the building blocks involved in a GPT

    So, why do people keep making these? Asking genuinely

    • Axodouble 19 minutes ago

      Fun? I am unsure why it keeps getting shared though, I too have reinvented the wheel a few times for fun.

      * Edit, I did however just notice literally all of this is just another vibeslopped banger, so I am not quite sure how much fun there really is, feels more like coding for the sake of keeping the wheel turning.

    • lostmsu 4 minutes ago

      [delayed]

    • alightsoul an hour ago

      to add it to their resume and "prove" they know how machine learning works

      • jazzpush2 31 minutes ago

        You mean pushing everything up in one giant commit with Claude isn't learning!?

  • vjsrinivas 3 hours ago

    Interesting trend of labeling models after how big they are on disk vs how many learnable parameters they have.

  • jey 5 hours ago

    At that scale it seems like you could just solve the KKT conditions directly. Exaggerating but only a little

  • w4yai 3 hours ago

    "hn1g.txt" => not available