get the Data ⬇️

The complete dataset has been released! This dataset includes the training and test splits with labels for all the subtasks and languages. Additionally, each file includes the domain, the model and the prompt used to generate each sample. You can download the complete dataset from the Hugging Face Hub.