overview for IamFuckinTomato

Can't handle efficiently RAG with large PDF in c/localllama@poweruser.forum

[–] IamFuckinTomato@alien.top 1 points 9 months ago

Have you tried to chunk it based on headings or did you just chunk the pdfs into chunks with fixed token sizes?

Can't handle efficiently RAG with large PDF in c/localllama@poweruser.forum

[–] IamFuckinTomato@alien.top 1 points 9 months ago (1 children)

Make a parents retriever (chunk 2000) and input it child retriever (chunk 400), 0 overlap

Can you please elaborate.

Update from SciPhi - Introducing SciPhi-Self-RAG-Mistral-7B-32k in c/localllama@poweruser.forum

[–] IamFuckinTomato@alien.top 1 points 10 months ago

Hey. Love this model and the self-rag idea.Is there a way I can hook this up with my own RAG database?

What UI do you use and why? in c/localllama@poweruser.forum

[–] IamFuckinTomato@alien.top 1 points 10 months ago

Have you tried installing the missing package files it shows when u tried installing the openai extension?
I had the same issue and installing those missing packages via the cmd_windows in the same folder.

Does OpenAI ToS prohibit generating datasets for open source LLMs? in c/localllama@poweruser.forum

[–] IamFuckinTomato@alien.top 1 points 10 months ago

Does mistral also have something like this in their ToS.

I'm planning on using mistral generated data to fine-tune a model.

Rag vs Vector db in c/localllama@poweruser.forum

[–] IamFuckinTomato@alien.top 1 points 10 months ago

!remind me 2 days