Sublime
An inspiration engine for ideas
Details on @OpenAI's new assistants RAG
*Hard* creep into vectorstore territory
Thoughts:
* Default chunk overlap of 50%, super interesting
* Metadata filtering, super interesting how this dips into vectorstore territory
* Unsure about what chunking... See more
RIGHT Dexscreener filters will bring you millions!
Use them and easily find 10-100x potential gems.
My strategy on how to set ultimate filters 🧵👇 https://t.co/coJ57Cr3XW
A big issue with semantic retrieval is that top-k doesn’t do well if your corpus consists of a bunch of different documents (e.g. 100 diff PDFs 🗃️).
There’s two ways of performing more structured tagging/retrieval for production-quality RAG systems:
1️⃣ Metadata Filters + Auto Retrieval: Tag each document... See more