aixk/vlite3.6-dataset
收藏官方服务：
资源简介：
该数据集是一个用于自然语言处理任务的数据集，包含训练集，共有787,890个样本，总大小约为863.5 MB。数据集特征包括input_ids（uint16列表类型，表示输入序列的编码）和loss_mask（bool列表类型，用于指示损失计算中的掩码部分），适用于序列建模或掩码语言建模等任务。数据以默认配置提供，文件路径为data/train-*。
This dataset is designed for natural language processing tasks, containing a training set with 787,890 examples and a total size of approximately 863.5 MB. The features include input_ids (a list of uint16, representing encoded input sequences) and loss_mask (a list of bool, indicating mask portions for loss calculation), suitable for tasks such as sequence modeling or masked language modeling. The data is provided in a default configuration with file paths at data/train-*.
提供机构：
aixk
