datasets
Training and evaluation data, with the modality, task and licence stated up front. Listed live from the Hugging Face Hub.
SorryBenchReup of Sorry Bench
Magpie-Pro-DPO-200K-JSONLSorryBench-Human-JudgementHuman annotated version of the previous sorry bench, there was only a few files in the dataset, will update if more get added
Silly-lorebookSillyJSONTrouper-Extras-SillyTavern
What is this?
SillyTavern is a popular RP frontend. It includes various tasks that the LLM can perform, like:
Generate a summary of the current conversation (200-500 words)
Generate an image according to some pre-determined prompt.
supports all of ST's built-in image prompt types (Character, User, Scenario, Last Message, Portrait, Background).
For the Last Message option, we also trained on this prompt, for first-person POV's:
PAUSE STORY TO GENERATE AN IMAGE. Your next… See the full description on the dataset page: https://huggingface.co/datasets/DarwinAnim8or/Trouper-Extras-SillyTavern.SillyTilly_PawanKrd-dpo-gpt-4o-reup-PreferenceShareGPT