harmless/harmful anchor datasets

This dataset contains 100 harmless and 100 harmful anchor prompts for evaluating the performance of large language models.

Data and Resources

Cite this as

Zheng et al. (2024). Dataset: harmless/harmful anchor datasets. https://doi.org/10.57702/g83nrnnd

DOI retrieved: December 16, 2024

Additional Info

Field Value
Created December 16, 2024
Last update December 16, 2024
Defined In https://doi.org/10.48550/arXiv.2406.10794
Author Zheng et al.