Dataset Viewer
The dataset viewer is taking too long to fetch the data. Try to refresh this page.
Server-side error
Error code:   ClientConnectionError

Dataset Card for "wiki-span"

This dataset is constructed by sampling 25%-50% of each wikipedia record twice, as positive pairs. It can be used to train unsupervised sentence representation models.

Downloads last month
120

Models trained or fine-tuned on gowitheflow/wiki-span