👉 (Aug-2025) Added pca_pairwise method and set as default. Use method="pca_pairwise" to reproduce results closer to those reported in the paper. Colab demos (see below) are fixed accordingly, and they should work as expected.
👉 (Jul-2025) Bug fixed: PCA_centering (@Reason239)
👉 (Apr-2025) Conditional Activation Steering is a spotlight paper at ICLR 2025!
👉 (Nov-2024) A few Colab demos are added.
👉 (Sep-2024) Preprint released on arXiv.
This is a general-purpose activation steering library to (1) extract vectors and (2) steer model behavior. We release this library alongside our recent paper on Programming Refusal with Conditional Activation Steering to provide an intuitive toolchain for activation steering efforts.
git clone https://github.com/IBM/activation-steering
pip install -e activation-steeringActivation steering is a technique for influencing the behavior of language models by modifying their internal activations during inference. This library provides tools for:
- Extracting steering vectors from contrastive examples
- Applying steering vectors to modify model behavior
This part is conceptually similar to Steering Language Models With Activation Engineering, but our code implementation could be different.
Conditional activation steering selectively applies or withholds activation steering based on the input context. Conditional activation steering extends the activation steering framework by introducing:
- Context-dependent control capabilities through condition vectors
- Logical composition of multiple condition vectors
Refer to our paper and documentation for detailed implementation and usage of activation steering and conditional activation steering.
Refer to /docs to understand this library. We recommend starting with Quick Start Tutorial as it covers most concepts that you need to get started with activation steering and conditional activation steering.
- Adding Refusal Behavior to LLaMA 3.1 8B Inst 👉 here!
- Adding CoT Behavior to Gemma 2 9B 👉 here!
- Making Hermes 2 Pro Conditionally Refuse Legal Instructions 👉 here!
This library builds on top of the excellent work done in the following repositories:
Some parts of the documentation for this library are generated by
- ml-tooling/lazydocs > lazydocs activation_steering/ --no-watermark
@misc{lee2024programmingrefusalconditionalactivation,
title={Programming Refusal with Conditional Activation Steering},
author={Bruce W. Lee and Inkit Padhi and Karthikeyan Natesan Ramamurthy and Erik Miehling and Pierre Dognin and Manish Nagireddy and Amit Dhurandhar},
year={2024},
eprint={2409.05907},
archivePrefix={arXiv},
primaryClass={cs.LG},
url={https://arxiv.org/abs/2409.05907},
}