Fine tuning for NER Tasks

#41
by NavS2023 - opened

I am in the process of fine-tuning gemma-2b-it version, for NER(PII identifications form the given text) tasks. Should i use AutoModelForCausalLM class or is there any specific class for NER classifications. I have tried AutoModelForTokenClassification but it is giving me error that its not compatiable. Please suggest. Than you.

Hey Nav, what is it not Compatible with? There are a lot of different ways to implement NER with LLMs so understanding that will help identify how people can help you.

Google org

If you have a notebook to share that would help too!

Sign up or log in to comment