This model is great, is there a fine-tuning script for this model?

#7
by Oliver2019 - opened

This model is great, is there a fine-tuning script for this model? I would like to fine-tune this model with my own dataset, but have no idea how to do it, can you share your fine-tuning script? Or provide some implementation suggestions for fine-tuning. Really thankful.

Snorkel AI org

Thanks for the question.
We recommend using existing tools like axolotl and Unsloth.
This model is built on top of the Mistral model, so you can reuse any script that is referring to "Mistral" or "mistralai/mistral-instruct-v0.2", and simply replace it with "snorkelai/Snorkel-Mistral-PairRM-DPO" and that should work as is.
Let me know how else I can help. Thanks!

Sign up or log in to comment