Locutusque/unnatural_instructions_gpt-4o-mini_scale_x2

该数据集包含以下字段：指令（instruction）、输入（input）、约束（constraints）、输出（output）、替代表述（alternative_formulation）、内联替代表述（alternative_formulation_inlined）和对话（conversations）。对话字段中包含对话来源（from）和对话内容（value）。训练集包含118,662个示例，数据集大小为199,682,299字节。具体的数据集描述和应用场景在README中未提供。

The dataset includes fields such as instruction, input, constraints, output, alternative formulation, alternative formulation inlined, and conversations. The conversation field contains the source of the conversation (from) and the conversation content (value). The training set includes 118,662 examples, and the dataset size is 199,682,299 bytes. The specific dataset description and application scenario are not provided in the README.