Instructions to use millicentli/llama3_inversion_llama3_multi with libraries, inference providers, notebooks, and local apps. Follow these links to get started.
- Libraries
- PEFT
How to use millicentli/llama3_inversion_llama3_multi with PEFT:
from peft import PeftModel from transformers import AutoModelForCausalLM base_model = AutoModelForCausalLM.from_pretrained("meta-llama/Llama-3.1-8B-Instruct") model = PeftModel.from_pretrained(base_model, "millicentli/llama3_inversion_llama3_multi") - Notebooks
- Google Colab
- Kaggle
File size: 366 Bytes
53b1465 e1f2730 53b1465 c65ef46 | 1 2 3 4 5 6 7 8 | ---
base_model: meta-llama/Llama-3.1-8B-Instruct
library_name: peft
---
This is the model used to invert Llama-3-8B activations using a Llama-3-8B model (multiple, not a single, activation).
The model is used in [Do Activation Verbalization Methods Convey Privileged Information?](https://arxiv.org/abs/2509.13316). Read more of our paper for training information. |