# textcleanlm/textclean-2B-raw

Merged dataset generated from 24 Parquet file(s).

- Source root: `data/eai_data_files/data`
- Pattern: `(all *.parquet)`
- Output shard: `data/train-00000-of-00001.parquet`
- Compression: `zstd`