tabachain/distilabel-example

该数据集是通过distilabel工具生成的，包含一个`pipeline.yaml`文件，用于重现生成该数据集的流程。数据集的结构包括`instruction`、`completion`、`meta`等字段，其中`meta`字段包含`category`、`id`、`prompt`、`source`、`subcategory`等信息。数据集的主要用途是用于问答任务，特别是数学问题的解答。数据集的配置为`default`，可以通过Hugging Face的`datasets`库加载。

This dataset has been created with distilabel and contains a `pipeline.yaml` file to reproduce the pipeline that generated it. The dataset features include `instruction`, `completion`, and `meta` fields, where the `meta` field contains subfields such as `category`, `completion`, `id`, etc. The dataset is divided into a training set with 10 examples. The dataset tags include `synthetic`, `distilabel`, and `rlaif`.