from Hacker News

Developing an LLM: Building, Training, Finetuning (A 1h Video Explainer)

by rasbt on 6/14/24, 12:29 PM with 12 comments

• by htrp on 6/14/24, 1:07 PM

  Not Sebastian (who I assume is the OP), but his blog/substack is also a great resource

  https://magazine.sebastianraschka.com/

• by mdp2021 on 6/14/24, 12:42 PM

  Seems very good, thank you.

  The channel: https://www.youtube.com/@SebastianRaschka/videos

  contains hundreds of video lessons, originally seemingly originating from Sebastian Raschka teaching at Wisconsin-Madison Uni (before he went full-time entrepreneur).

• by yoouareperfect on 6/14/24, 1:11 PM

  Is anyone training LLMs outside of Meta, OpenAI, etc... ?

  I don't much get the point. For huge models, it's impossible to outcompete them. For smaller models, isn't mistral or LLaMa good enough?

  What are other startups finetuning LLMs for?

• by oneshtein on 6/14/24, 1:47 PM

  Can someone train an AI to perform all that?