# GRADIEND: Feature Learning within Neural Networks Exemplified through Biases

arXiv: https://arxiv.org/abs/2502.01406 · HTML: https://mlsanity.com/paper/2502.01406/

## Key facts

- Cited by 2 papers in the map (data through 14 Oct 2026).
- First posted on arXiv 3 Feb 2025.

## Built on by (most cited first, 2 of 2)

- [Understanding or Memorizing? A Case Study of German Definite Articles in Language Models](https://mlsanity.com/paper/2601.09313/) (2026)
- [The GRADIEND Python Package: An End-to-End System for Gradient-Based Feature Learning](https://mlsanity.com/paper/2602.23993/) (2026)

## Abstract (arXiv)

AI systems frequently exhibit and amplify social biases, leading to harmful consequences in critical areas. This study introduces a novel encoder-decoder approach that leverages model gradients to learn a feature neuron encoding societal bias information such as gender, race, and religion. We show that our method can not only identify which weights of a model need to be changed to modify a feature, but even demonstrate that this can be used to rewrite models to debias them while maintaining other capabilities. We demonstrate the effectiveness of our approach across various model architectures and highlight its potential for broader applications.