Finetuning Max Length for Completion?


I’m trying to finetune with some of my own fiction writing. I’ve been using full stories (some ~6k-10k words), and I’m wondering if those completions are too long for the finetune process? How long can a completion be? Is there a tool available to break up long text into max length more easily?

Sorry if this is super basic! Thanks for your help!

1 Like

I wrote a python script to break on sentences and make each sentence a blank-prompt/sentence completion for fine tuning. I think you have to have under 2K tokens per line or it ignores the rest of the line.

It would be nice if we got a warning in the fine tunes upload. Maybe that’s an OpenAPI request?