manu/fquad2_test
Viewer
•
Updated
•
1.5k
•
142
•
1
These datasets are used to evaluate models on French performance using: https://github.com/EleutherAI/lm-evaluation-harness (from CroissantLLM paper)