systemk/example-preference-dataset

这个数据集包含了一个 pipeline.yaml 文件，可以用来在 distilabel 中重现生成该数据集的流程。数据集包含各种特征，如输出、输入、索引、类别、指令、生成物、模型名称、distilabel 元数据、评分、理由、模型名称、提示、提示 ID、选择项、选择评分、拒绝项和拒绝评分。数据集的大小为 1.48 MB，包含 100 个示例，下载大小为 723 KB。数据集可以通过 HuggingFace 的 datasets 库加载。数据集是用 distilabel 工具创建的，可以在 distilabel CLI 中使用。

This dataset contains a pipeline.yaml file that can be used to reproduce the pipeline that generated it in distilabel. The dataset includes various features such as output, input, index, category, instruction, generations, model_names, distilabel_metadata, ratings, rationales, model_name, prompt, prompt_id, chosen, chosen_rating, rejected, and rejected_rating. The dataset size is 1.48 MB, containing 100 examples, and the download size is 723 KB. The dataset can be loaded using the datasets library from HuggingFace. The dataset is created using the distilabel tool, which can be used in the distilabel CLI.