14 comments

  • konradha 2 days ago ago

    Every engineer who's worked with "agentic" workflows can tell that this is entirely generated with very little tweaks. It's embarrassing.

    • stephantul 2 days ago ago

      Even the title is a Claudeism, it makes me sad

  • tucan9389 2 days ago ago

    Author here.

    Thanks for comments and feedback. I wrote this post in Korean (my native language) and translated it by LLMs. TBH, I spent more time reviewing the post rather than writing. Sorry that I missed the uncomfortable tone and AI-ish. be more careful next time.

    Anyway, even though I tried to use coding agent, it was hard to avoid some dead ends and spent more money than I expected. As a newbie in pre-training, hope this baseline codebase helps people who want to reproduce nanochat in JAX + TPU...

    If anyone has any useful experiences when you do similar things, please share in the comments.

  • sciences44 2 days ago ago

    Nice work. Regarding your first insight, that's a sneaky one ^^. I experiment a similar issue with fine-tuning a small model with MLX. It was very hard to find out.

    • tucan9389 2 days ago ago

      Thank you. Agreed, we should be more careful when doing this. Was it the JAX training codebase?

  • adityashankar 2 days ago ago

    For people that are like me : This entire text is AI generated, i feel weird reading it personally, i guess others may not

    • embedding-shape 2 days ago ago

      Don't tell me you don't enjoy and learn a ton from barely coherent nuggets like this?

      > The most important thing the port taught me: if you can avoid it, don't. I decided to do it anyway, so here are the problems I ran into and some insights of my own, trimmed down to five. Each item is tagged with which side it hurt: quality (CORE) or performance (MFU). The first is the bug that fooled us the longest.

      • Chris2048 2 days ago ago

        TBH llm generated text is usually better than this..

        • embedding-shape 2 days ago ago

          TBF it kind of depends on what weights you end up using, the quality gap can be pretty wide.

    • attila-lendvai 2 days ago ago

      he's just dogfooding us all...

  • tahakotwal12 2 days ago ago

    nice

  • contentpulse 2 days ago ago

    [flagged]

  • koko443 2 days ago ago

    [flagged]

  • OsamaJaber 2 days ago ago

    [dead]