anon-betterbench/similarity-threshold-study

该数据集包含多个文档的相关信息，每个文档具有名称、类别、标题、原始内容等特征。此外，数据集还包含文档的分块信息，如最小块大小、最大块大小、目标块大小和相似度阈值等。数据集仅包含一个训练集，共有28,681个样本，总大小为3,659,752,974字节。

This dataset contains information about multiple documents, each with features such as document name, category, title, raw content, etc. Additionally, the dataset includes chunking information for the documents, such as minimum chunk size, maximum chunk size, target chunk size, and similarity threshold. The dataset contains only a training set with 28,681 samples and a total size of 3,659,752,974 bytes.