# Multi-size RoBERTa

One of the 24 Chinese RoBERTa model series pre-trained on CLUECorpusSmall, trained using the UER-py framework, supporting masked language modeling and text feature extraction.