activation-steering

KO | EN

gitlite — search

#python #hacktoberfest #react #ai #llm #golang #deep-learning #docker #macos #java #nodejs #windows #pytorch #open-source

activation-steering

(★ 181)

[ICLR 2025] General-purpose activation steering library

activation-steering Latest Version Download

Download Latest Version (.zip)

// repository documentation

Was this content helpful?

★ 0 (0 ratings)

![Python](https://img.shields.io/badge/python-3.10+-blue.svg)

# Activation Steering

👉 (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](https://arxiv.org/abs/2409.05907).

## Overview

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*](https://arxiv.org/abs/2409.05907) to provide an intuitive toolchain for activation steering efforts.

## Installation
```bash
git clone https://github.com/IBM/activation-steering

pip install -e activation-steering
```

## Activation Steering
Activation 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*](https://arxiv.org/abs/2308.10248), but our code implementation could be different.

## Conditional Activation Steering
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*](https://arxiv.org/abs/2409.05907) and [*documentation*](docs/quickstart.md) for detailed implementation and usage of activation steering and conditional activation steering.

## Documentation
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.

- Quick Start Tutorial (10 minutes ~ 60 minutes, depending on your hardware) 👉 [here!](docs/quickstart.md)
- FAQ 👉 [here!](docs/faq.md)

## Colab Demos

- Adding Refusal Behavior to LLaMA 3.1 8B Inst 👉 [here!](https://colab.research.google.com/drive/1IpAPMFHZW6CNrE0L16TXSvIApAK9jAFZ?usp=sharing)
- Adding CoT Behavior to Gemma 2 9B 👉 [here!](https://colab.research.google.com/drive/1dnG000syxHwOt-Z9_bpRLnBbfugI_CBh?usp=sharing)
- Making Hermes 2 Pro Conditionally Refuse Legal Instructions 👉 [here!](https://colab.research.google.com/drive/18lOzaFOK4CB_mYe9jlQbJCdHBDlhGxcQ?usp=sharing)
  
## Acknowledgement
This library builds on top of the excellent work done in the following repositories:

- [vgel/repeng](https://github.com/vgel/repeng)
- [andyzoujm/representation-engineering](https://github.com/andyzoujm/representation-engineering)
- [nrimsky/CAA](https://github.com/nrimsky/CAA)

Some parts of the documentation for this library are generated by

- [ml-tooling/lazydocs](https://github.com/ml-tooling/lazydocs) > lazydocs activation_steering/ --no-watermark

## Citation

```
@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}, 
}
```

activation-steering

activation-steering Latest Version Download

Today's picks

Was this content helpful?

Recent Feedback

activation-steering

activation-steering Latest Version Download

Today's picks

Was this content helpful?

Recent Feedback

Download README