Measuring the ability to retrieve the groundtruth answers to reasoning task queries on winogrande.
You can evaluate an embedding model on this dataset using the following code:
import mteb
task = mteb.get_task("WinoGrande")
evaluator =… See the full description on the dataset page:
https://huggingface.co/datasets/mteb/WinoGrande.
