Hello! I’ve had a lot of curiosity around AI and have enjoyed messing with it for quite a while, but I’m really struggling to use it beyond simple sandbox interactions.
Have tried looking for guides on how to achieve what I want, but they all start telling me to do things that I have no idea how to do. For context, I only just figured out how to correctly install python, lol.
What I want to achieve is somehow giving GPT-4 access to a large database of organized information, which it can reference when responding to prompts. The prompt examples should adequately teach it what different parts of the database mean and whatnot. It jumped ahead of me when I was testing out a small version in sandbox by asking itself questions and correctly answering them while rephrasing the data in a way that made actual sense, despite me not having referenced that part of the data yet, so I’d imagine it should be very possible. I just really have no clue where to even start when it comes to the technical side of it.
I have a github repository with the large database it will use for context, and then will handwrite prompts, I just need to know how to feed it the information and interact with that trained/fine-tuned/whatever model. It might be a lot to walk me through, so even any guides that might help my computer-stupid self would be super appreciated, thanks!
