| This dataset is a combination of other copyright permissive datasets. It is intended to be used as seed data in order to generate synthetic data. | |
| It includes cleaned subsets of: | |
| - A subset of Fineweb-edu-2 that includes CC-BY-* and .gov websites. | |
| - A subset of open License Corpus dataset | |
| - Youtube commons |