aichannel

Is there an easy way to fine tune an LLM on stuff I write so that I can talk to myself? Transforming data is not a requirement, can do it myself

RAG is enough. Most of cases no need to do fine tuning.

What’s RAG? Is that like a prompt framework?

Governance Public Utilities for @zksync at @thezknation

retrieval augmented generation

trying to setup an agent via Eliza to be my personal assistant with a database to store conversation history and add knowledge via pdfs etc

accelerating agritech. empowering communities. ending hunger.
 https://cityfarmers.xyz

Can confirm it’s incredible. I did it over a year ago but only with a vector database and ChatGPT3,,, and it was insanely helpful for writing and research. My only recommendation is try to solve the source-link problem. Ideally the agent should be able to quote source, even better if page/paragraph/conversation location.

if i could just get past the self-signed cert error when i connect supabase, (3 days of this smh,) that will be my next task

reminds me of this video i watched on using .cursorrules files for agent-assisted coding where it instructs the agent to reference the rules when it uses them and explain why

if i could just get past the self-signed cert error when i connect supabase, (3 days of this smh,) that will be my next task

reminds me of this video i watched on using .cursorrules files for agent-assisted coding where it instructs the agent to reference the rules when it uses them and explain why

https://youtu.be/aG-utUqVrb0?si=mUA1dCAbJ3fut-kT