Join the conversation

Join the community of Machine Learners and AI enthusiasts.

Sign Up
drwlf 
posted an update 2 days ago
Post
4667
Having an insanely good medical LLM is pointless if it won’t answer your questions!

So we’ve made 2 notebook for abliterating any model in order to achieve a good model that will actually help you!

The notebooks are made using @mlabonne ‘s abliteration logic and datasets!

Feel free to use them and happy training 😊

https://github.com/dralexlup/LLM-Abliteration

That's super cool, congrats! :)

·

Thank you! I modified the original colab notebook you made, I didn’t want to bother and it was important to me and nico to have these medical models abliterated so we could properly use them

This is awesome and was highly needed. Thank you. The medical use case is practically all I use LLMs for and so I got quite a few refusals. The gemma and qwen series are specifically awful in regards to this (medgemma excluded).