This data may NOT be used as training data in any form for any phase of training (e.g., pre-training, post-training, fine-tuning, adaptation, etc.) without express written permission from all three authors.
Dataset Description
EspanStereo is a Spanish-language dataset for evaluating culturally specific stereotypes in language models.
The dataset covers five Spanish-speaking countries:… See the full description on the dataset page: https://huggingface.co/datasets/MMS-Lab/EspanStereo.
