O(n²) 복잡도의 Matrix Operation을 통한 토큰 간 관계 정량화 및 Contextual Representation 구현
How Self-Attention Works — QKV, Softmax, and Matrix Computation
How Self-Attention Works — QKV, Softmax, and Matrix Computation
Active Inference, The Learn Arc — Part 25: Session §4.3 — Expected Free Energy, Introduced Concretely