The dataset includes questions and answers related to contracts.
You can evaluate an embedding model on this dataset using the following code:
import mteb
task = mteb.get_tasks(["LegalBenchConsumerContractsQA"])… See the full description on the dataset page:
https://huggingface.co/datasets/mteb/legalbench_consumer_contracts_qa.