Why it matters
GLU FFNs are standard in modern LLMs. Understanding shapes design.
Advertisement
The architecture
FFN(x) = down(gate(up(x)) * activation(up(x))). Two projections up + one down.
Advertisement
How it works end to end
Params: 3 matrices vs 2 for ReLU FFN.
Adjust hidden dim to keep param count.
Quality: consistently better.