How to use mwatkins1970/llama-davinci-instruct with PEFT:
from peft import PeftModel from transformers import AutoModelForCausalLM base_model = AutoModelForCausalLM.from_pretrained("meta-llama/Llama-3-70b") model = PeftModel.from_pretrained(base_model, "mwatkins1970/llama-davinci-instruct")
This repository contains a voice-specific QLoRA adapter trained to adapt LLaMA 3.1-70B to the Davinci-Instruct voice style. This adapter is designed to be used in combination with the Leilan domain adapter.
