patrickfleith/

astro_texts_dataset

Synthetic texts about astronautics and space mission engineering, in English and French. The dataset ships as two subsets:

Subset Rows Tokens Description

default 23,800 13.14M Full set

deduplicated 13,525 4.49M Near-duplicates removed

from datasets import load_dataset

full = load_dataset("patrickfleith/astro_texts_dataset", split="train") dedup = load_dataset("patrickfleith/astro_texts_dataset", "deduplicated", split="train")

Token… See the full description on the dataset page: https://huggingface.co/datasets/patrickfleith/astro_texts_dataset.