Feed-Forward Network
The position-wise fully connected layer inside each transformer block.
Обзор
The feed-forward network is the two-layer fully connected block applied to each position inside a transformer, usually expanding to several times the model dimension and back. It solves the problem that attention only mixes information between positions and performs no transformation of its own. It exists as the majority of a transformer's parameters, which is why mixture-of-experts replaces this layer rather than the attention.
Граф технологий
Как Feed-Forward Network связана с остальным стеком.
Classification
- Transformer architecture Algorithms Feed-Forward Network part of Transformer High ✓ source