ketchup123/ultramix_no_pref_filter

该数据集包含了对话ID、提示文本、选择的文本内容及其角色、拒绝的文本内容及其角色、指令奖励、拒绝指令奖励、奖励模型、意图、知识、难度、难度生成器、输入质量、质量解释、质量生成器、任务类别、其他任务类别、任务类别生成器、llama_guard_2标识和语言类型等信息。数据集分为训练集，包含273,367个样本，大小为1.88 GB。

The dataset includes conversation ID, prompt text, chosen content and its role, rejected content and its role, instruction reward, rejected instruction reward, reward model, intent, knowledge, difficulty, difficulty generator, input quality, quality explanation, quality generator, task category, other task category, task category generator, llama_guard_2 identifier, and language type. The dataset is split into a training set, containing 273,367 samples, with a size of 1.88 GB.