search

fe1ixxu / CPO_SIMPO

This repository contains the joint use of CPO and SimPO method for better reference-free preference learning methods.
35 stars 4 forks source link

good job, I have some questions #1

Open NickyDark1 opened 4 months ago

NickyDark1 commented 4 months ago

Is it possible to use the unsloth library to make fine tune? Do you have any implementation in Google Colab?

thank you so much.

NickyDark1 commented 4 months ago

I wanted to ask you about the use in Google Colab since it gives me an error when I go to train it '.train()'

https://colab.research.google.com/drive/1T16fhYA6l8xiuMn8p7T_cXLD_fph7a1Z?usp=sharing

colab Unsloth: image

sahsaeedi commented 3 months ago

I wanted to ask you about the use in Google Colab since it gives me an error when I go to train it '.train()'

https://colab.research.google.com/drive/1T16fhYA6l8xiuMn8p7T_cXLD_fph7a1Z?usp=sharing

colab Unsloth: image

Update transformers library with pip install -U transformers.

R4ZZ3 commented 11 hours ago

@NickyDark1 Did you get this to work with Unsloth? Is it better than DPO?