BioGPT is a Generative Pre-trained Transformer for Biomedical Text Generation and Mining. The Microsoft research team trained BioGPT using only domain-specific data. They collected articles from PubMed, an English-language text-based meta-database of biomedical articles, updated before 2021. This yielded a total of 15 million pieces of content with titles and abstracts, which the team used to train BioGPT.

Source: https://the-decoder.com/biogpt-is-a-microsoft-language-model-trained-for-biomedical-tasks/