If you mean the technical details of attention models, the original paper “Attention is all you need” is not too difficult to read. If you’re more interested in

If you mean the technical details of attention models, the original paper “Attention is all you need” is not too difficult to read. If you’re more interested in applications, hugging face has a “course” on their website that walks through the high level topics of applying transformers to natural language processing (can’t remember if they cover transformers for other topics).