My use case is as follows. I would like to generate an article based on a set of text/pdf documents (that belong to a very specific nice topic) that reside on my desktop. Is it possible? I was wondering if someone can help me understand either the limitation in that context or how it can be done. Would appreciate it very much.
You would just need to scrape it and prepare a JSONL file, but yeah. Check out fine-tuning in the docs.