DataHarvest is a toolkit specifically designed for building datasets for large language models. It provides a series of pipelines for data acquisition, cleaning, and processing, aiming to deliver high-quality training data for Chinese large language models.
