Update README.md

README.md

@@ -37,7 +37,7 @@ Common Corpus is made of five carefully curated collections:
 *  **OpenGovernment**: 387,965,738,992 tokens of financial and legal documents, including Finance Commons (from sources like SEC and WTO) and Legal Commons (including Europarl and Caselaw Access Project), providing enterprise-grade training data from regulatory bodies and administrative sources.
 *  **OpenSource**: 334,658,896,533 tokens of high-quality code in open source from GitHub, filtered using ArmoRM to ensure only the top 80% of submissions by quality rating are included.
 *  **OpenScience**: 221,798,136,564 tokens of academic content from Open Alex and other open science reposiories, processed using vision-language models to preserve crucial document structure and formatting.
-*  **OpenWeb**: 132,075,315,715 tokens from Wikipedia, YouTube Commons and other websites available under permissible licenses like Stack-Exchange.
+*  **OpenWeb**: 132,075,315,715 tokens from Wikipedia (official releases from the [Wikimedia Foundation](https://huggingface.co/datasets/wikimedia/wikipedia) on Huggingface), YouTube Commons and other websites available under permissible licenses like Stack-Exchange.
 
 | Collection     | Domain                   | Sources                                                                                   |
 |----------------|--------------------------|-------------------------------------------------------------------------------------------|