this post was submitted on 24 Nov 2023
1 points (100.0% liked)

LocalLLaMA

3 readers
1 users here now

Community to discuss about Llama, the family of large language models created by Meta AI.

founded 1 year ago
MODERATORS
communick@poweruser.forum
1
What are the best educational resources? (alien.top)
submitted 11 months ago by tief1ghter@alien.top to c/localllama@poweruser.forum
5 comments fedilink hide all child comments
 

What is the best way to learn LLM / generative AI topics in depth:

  • loss functions
  • finetuning (LoRA, QLoRA, etc)
  • creating good datasets
  • quantization, AWQ

etc. I know there is the Fast AI course, but more interested in these above topics. Seems like there are scattered guides, notebooks and a promising repo here: https://github.com/peremartra/Large-Language-Model-Notebooks-Course but nothing comprehensive.

top 5 comments
sorted by: hot top controversial new old
[–] gaztrab@alien.top 1 points 11 months ago

I second this

[–] GasBond@alien.top 1 points 11 months ago

https://microsoft.github.io/AI-For-Beginners/ ai for everyone by Andrew ng

[–] Educational-Net303@alien.top 1 points 11 months ago

Karpathys channel is awesome for intro, then you can just start reading the papers for more in depth learning

[–] BlueMetaMind@alien.top 1 points 11 months ago

Absolutely! I wanted to start a thread like this myself. There is information out there for general AI stuff. But I seek concrete tutorials and articles about current open source LLMs practice . I agree that knowledge is frustratingly scattered.

[–] theoretical_entity@alien.top 1 points 11 months ago

If you don't know where to start, perhaps start with this video from Karpathy published yesterday: https://www.youtube.com/watch?v=zjkBMFhNj_g