3 ms·
Looks neat, kind of those "awesome X" list. Maybe adding a wiki would be interesting (so uninitiated can learn what is what and where to start if they want to b
by stanac 3y ago
Looks neat, kind of those "awesome X" list. Maybe adding a wiki would be interesting (so uninitiated can learn what is what and where to start if they want to build/train their own AI).
Tangential question I am not sure where else to ask: Would it be possible to train AI at home on a book? I was thinking about how sometimes I start reading a book and then get distracted for a month or two. Summarizing chapters and characters up to a point in the book sounds like a good fit for GPT AI. I could possibly find and extract PDF or epub of the current book I am reading and I have RTX 4080 at home I am barely using.
- lambdaba 3y agoWhy wouldn't you just use Claude 3? It's going to be miles better than any OSS model.
- stanac 3y agoI'll have to read first what Claude 3 is. Thank you.
- zingar 3y agoIs privateGPT what you're looking for? It stores PDFs in a vector DB and answers questions based on that content alone. Edit: Everything is local. This is not training, it takes a standard LLM and gives it extra context based on what text the tool finds in the vector DB related to the topic. https://github.com/imartinez/privateGPT https://github.com/imartinez/privateGPT
- stanac 3y agoThat sounds even better than training LLM on my own. Thanks.