Orion-zhen/word-count

这个合成数据集旨在解决诸如在单词strawberry中有多少个r这样的问题。为了使用这个数据集，强烈建议添加`<spelling_error>`和`<word_hit>`作为特殊标记。

This synthetic dataset aims to address problems like How many r are there in word strawberry. The dataset is intended for text generation tasks and it is recommended to add special tokens `<spelling_error>` and `<word_hit>` when using it.