aieng-lab/biasneutral-ajibawa

BIASNEUTRAL Ajibawa

BIASNEUTRAL Ajibawa contains Ajibawa-derived text snippets that passed the GRADIEND bias-neutral filtering pipeline. It is the Ajibawa-source counterpart to the original aieng-lab/biasneutral dataset, which can not be directly published due to licensing issues. This dataset provides an easier-to-access solution, while maintaining the same generation principles as the original BIASNEUTRAL properties.

Usage

from datasets import load_dataset

ds = load_dataset("aieng-lab/biasneutral-ajibawa")

Examples

id	text
ajibawa-biasneutral-00000000	It struck me how many of them subtly reflected the country's tumultuous history, serving as silent witnesses to the struggles and triumphs of its people.
ajibawa-biasneutral-00000001	class Main { public static void main(String[] args) { Scanner scanner = new Scanner(System.in); System.out.println("Enter a positive integer:"); int num = scanner.nextInt();
ajibawa-biasneutral-00000002	Our code now focuses solely on finding the sequence without stopping at the seventh attempt. To address this issue, let's introduce another variable to keep track of the remaining moves left.
ajibawa-biasneutral-00000003	Here's the final transformed sequence:\n" + Arrays.toString(sequence)); } } } ```
ajibawa-biasneutral-00000004	Now, whenever users entered large numbers, the updated code politely informed them that it could not complete the transformation within the allotted moves. Encouraged by this success, Mr. Byte congratulated the Byte Breakers team for their collaboration, persistence, and growth mindset throughout the process.

Dataset Structure

• id: stable example id
• text: filtered gender-neutral text

Dataset Details

• Repository: https://github.com/aieng-lab/gradiend-bias
• Paper: https://arxiv.org/abs/2502.01406
• Original Data: ajibawa-2023/General-Stories-Collection
• Source License: Apache-2.0

Uses

This dataset is suitable for training and evaluating language models in settings where gender, race, and religion bias terms should be filtered from the text. For example, it can be used as a bias-neutral language modeling corpus for GRADIEND-style analyses.

Size

This prepared export contains 4593588 examples.

Dataset Creation

We generated this dataset by filtering Ajibawa-derived snippets and keeping only entries matching the following criteria:

• Each entry contains at least 50 characters
• No name of aieng-lab/namextend
• No gender-specific pronoun is contained (he/she/him/her/his/hers/himself/herself)
• No gender-specific noun is contained according to the 2421 plural-extended entries of this gendered-word dataset
• No Bias Attribute Words (gender, race, religion) as defined by Maede et al. (2022).

License

This derived dataset is released under Apache-2.0, matching the license declared for ajibawa-2023/General-Stories-Collection.

Citation

@inproceedings{drechsel2026gradiend,
  title     = {{GRADIEND}: Feature Learning within Neural Networks Exemplified through Biases},
  author    = {Drechsel, Jonathan and Herbold, Steffen},
  booktitle = {Proceedings of the International Conference on Learning Representations},
  year      = {2026},
  url       = {https://arxiv.org/abs/2502.01406}
}

Dataset Card Authors

jdrechsel