Understanding Attention Mechanism in Transformer Neural Networks
https://learnopencv.com/attention-mechanism-in-transformer-neural-networks/
Attentionがn（系列長）の2乗の計算量（
3 Model Architecture
）の説明記事
全単語の組み合わせに対して計算
https://learnopencv.com/wp-content/uploads/2023/01/neural-self-attention-cover-picture.png