The viewer is disabled because this dataset repo requires arbitrary Python code execution. Please consider removing the loading script and relying on automated data support (you can use convert_to_parquet from the datasets library). If this is not possible, please open a discussion for direct help.

legal-mc4-pl is extracted from legal-mc4 multilang dataset.

Originally legal-mc4 is in parquet format, into this dataset we provided jsonl format.

This dataset contains only polish texts, thanks for joelito for parquet legal-mc4 dataset.

Downloads last month: 54

Models trained or fine-tuned on radlab/legal-mc4-pl