mlfoundations-dev/oh-dcft-v1.2_no-curation_no_airoboros_gpt-4o-mini

该数据集包含对话数据，每个对话包含来源和内容两个字段，以及一个分片标识符。数据集主要用于训练目的，包含869693个样本，总大小约为1.43GB。数据文件存储在data/train-*路径下。

This dataset contains conversation data, with each conversation including a source and content field, along with a shard identifier. The dataset is primarily used for training purposes, containing 869,693 samples with a total size of approximately 1.43GB. The data files are stored under the data/train-* path.