heegyu/KoLLaVA-Instruct-313k-tokenized-trl

该数据集包含一系列消息，每条消息由内容和角色两个字段组成。数据集分为训练集和验证集，训练集包含313,025个示例，验证集包含100个示例。训练集的总大小为329,868,852字节，验证集的总大小为137,414字节。整个数据集的下载大小为150,890,383字节，总大小为330,006,266字节。数据文件的路径在配置部分指定，训练集数据文件路径为data/train-*，验证集数据文件路径为data/validation-*。

This dataset contains a series of messages, each consisting of two fields: content and role. The dataset is divided into a training set and a validation set, with the training set containing 313,025 examples and the validation set containing 100 examples. The total size of the training set is 329,868,852 bytes, and the total size of the validation set is 137,414 bytes. The download size of the entire dataset is 150,890,383 bytes, and the total size is 330,006,266 bytes. The paths to the data files are specified in the configuration section, with the training set data file path being data/train-* and the validation set data file path being data/validation-*.