| language: | |
| - en | |
| - hi | |
| - es | |
| - fr | |
| - de | |
| - ar | |
| - ru | |
| - ja | |
| - ko | |
| - zh | |
| license: cc0-1.0 | |
| task_categories: | |
| - text-generation | |
| pretty_name: NSFW Multilingual Multi-turn Roleplay Dataset | |
| size_categories: | |
| - n<1K | |
| # NSFW Multilingual Multi-turn Roleplay Dataset | |
| ## Dataset Description | |
| This dataset contains NSFW, multilingual, multi-turn roleplay conversations in JSONL format. | |
| It is intended for research and experimentation in roleplay dialogue generation across languages and conversational turns. | |
| The file currently provided in this dataset repository is `all_output.jsonl`. | |
| ## Intended Use | |
| - Training or evaluating multilingual conversational roleplay models. | |
| - Studying long-context, turn-taking behavior in roleplay dialogue settings. | |
| ## Important Notes | |
| - Content is explicitly NSFW and may be inappropriate for many audiences. | |
| - Users are responsible for applying legal, policy, and safety filtering for their region and use case. | |
| - Use only in environments that allow adult content handling. | |
| ## Data Format | |
| - `all_output.jsonl`: JSON Lines file (one JSON object per line). | |
| ## License | |
| This dataset is released under **CC0-1.0** (Public Domain Dedication). | |
| ## Citation | |
| No citation is provided. | |