0
Your Cart

New Dataset: erotica-analysis-32k

I’ve released a new dataset comprised of Literotica, Archive of Our Own, and Alt.Sex.Stories texts. I’ve run them all through GPT-3.5-Turbo-32k for analysis. Each analysis typically includes a summary, story tags, and a short writing prompt that could have been used to generate the story. I am hoping this dataset will be helpful in create larger context models to help me create, edit, ¬†and mutate NSFW training data to further improve future models.

Leave a Reply

Your email address will not be published. Required fields are marked *