kilicai/turkish-sft-unified-core

该数据集名为turkish-sft-unified-core，由ML Intern工具生成，专注于土耳其语任务。数据集结构包含messages字段（其中role表示角色，content表示内容）、source（来源）和category（类别）特征。训练集包含328,483个示例，用于支持监督式微调（SFT）任务，可能涉及对话或文本生成应用。

The dataset named turkish-sft-unified-core is generated by ML Intern and focuses on Turkish language tasks. It includes features such as messages (with role and content fields), source, and category. The training split contains 328,483 examples, intended for supervised fine-tuning (SFT) applications, likely involving dialogue or text generation.