Fine-Tuning with help of massive amount of documents

How do you fine-tune the model if you have 10,000 of documents?

  1. First, gpt-3.5-turbo model - is it possible to fine tune it?
  2. The goal is to feed the documents to new gpt-3.5-turbo model and then be able to ask questions and get answers.
  3. So if 1+2 feasible, how do you feed the docs? by using fine-tuning API?

You can’t fine-tune turbo. You should really look into embeddings.


ok, good to know.
How about answers for par. #2 and #3?

Follow the pattern found here.

1 Like