Explore high-quality datasets for your AI and machine learning projects.
This dataset contains erotic stories that have been cleaned, deduplicated, and depolluted, intended for training text‑filtering classifiers. The data originates from the HuggingFace datasets bluuwhale/nsfwstory and bluuwhale/nsfwstory2. The dataset comprises 49,579 samples, and the downloaded parquet file is 646 MB.
