havelm3/cognia-czech-dialogues

Cognia Czech Dialogues Cognia Czech Dialogues is a pilot collection of synthetic multi-turn conversations in Czech. It is intended for experiments with conversational language models, supervised fine-tuning, evaluation, and dataset-curation workflows. The current release contains 4,000 dialogues across 40 topic areas. The data was generated synthetically and has not yet been fully reviewed by humans. Dataset contents Language: Czech (cs-CZ) Dialogues: 4,000… See the full description on the dataset page: https://huggingface.co/datasets/havelm3/cognia-czech-dialogues.