Instructions to use Novaspree/llama-3.2-3B-tofu-adapter with libraries, inference providers, notebooks, and local apps. Follow these links to get started.
- Libraries
- PEFT
How to use Novaspree/llama-3.2-3B-tofu-adapter with PEFT:
from peft import PeftModel from transformers import AutoModelForCausalLM base_model = AutoModelForCausalLM.from_pretrained("meta-llama/Llama-3.2-3B") model = PeftModel.from_pretrained(base_model, "Novaspree/llama-3.2-3B-tofu-adapter") - Notebooks
- Google Colab
- Kaggle
Improve model card and add metadata
#1
by nielsr HF Staff - opened
This PR updates the model card to include metadata and details for the MAAT unlearning adapter. It adds the correct library name (PEFT), pipeline tag, and license, along with links to the research paper and the official GitHub repository. The description highlights the MAAT framework's approach to targeted unlearning using LoRA adapters.
Novaspree changed pull request status to merged