Key-Value Cache

Encodes positional information via rotation in embedding space.

Encodes token position explicitly, often via sinusoids.

Techniques to handle longer documents without quadratic cost.

Attention mechanisms that reduce quadratic complexity.

Tradeoffs between many layers vs many neurons per layer.

Prevents attention to future tokens during training/inference.

A single attention mechanism within multi-head attention.

Allows model to attend to information from different subspaces simultaneously.