Lyte/darija-pretraining-corpus-nanochat

Moroccan Darija Pretraining Corpus for Nanochat Parquet shards prepared for nanochat pretraining. Contents 83 train shards 1 validation shard 82,738,910 train rows 50,000 validation rows Source Built from Lyte/darija-pretraining-corpus using these subsets: arabic_raw bilingual pure Format Each parquet file contains a single column: text The validation shard is stored as the last parquet file and was written with small row groups… See the full description on the dataset page: https://huggingface.co/datasets/Lyte/darija-pretraining-corpus-nanochat.

<iframe src="https://silytics.si/embed/item/d7a1650a-0c88-44a6-b64f-cdbba5ab9a49" width="420" height="190" style="border:0" loading="lazy" title="Silytics entry"></iframe>

Silytics shows public metadata from Hugging Face and links to the original. Information may be incomplete or out of date. Last refreshed 7 November 2026. Licenses and terms of the original apply to the work itself. Not financial, legal or professional advice.