OpenSpeechHub/Emilia-All-Ja-Qwen3-Fix

该数据集包含三个特征字段：input_ids，labels和attention_mask，适用于机器学习模型的训练。数据集仅包含训练集部分，共有约132万多个样本，数据大小为约10GB。数据集采用默认配置，训练数据文件路径遵循指定的模式。

The dataset includes three feature fields: input_ids, labels, and attention_mask, which are suitable for training machine learning models. The dataset consists only of the training set, with a total of approximately 1.32 million samples and a data size of about 10GB. The dataset uses the default configuration, and the path pattern for the training data files is specified.