← Mathematical compendium

Published equation contexts

FFNSwiGLU(x)=(Swish1(xW1)⊙xW3) W2,Swish1(z)=z⋅σ(z)\mathrm{FFN}_{\mathrm{SwiGLU}}(x) = \big(\mathrm{Swish}_1(xW_1) \odot xW_3\big)\,W_2, \qquad \mathrm{Swish}_1(z) = z \cdot \sigma(z)

Why this formula appears here

Shazeer’s paper on gated linear unit variants tested several replacements for the standard ReLU or GELU feed-forward block, in which two linear projections are combined so that one, passed through a nonlinearity, gates the other [ 6 ] . The variant that stuck across the field, and that Llama adopts in every generation, uses the Swish (SiLU) nonlinearity as the gate: FFNSwiGLU(x)=(Swish1(xW1)⊙xW3) W2,Swish1(z)=z⋅σ(z)\mathrm{FFN}_{\mathrm{SwiGLU}}(x) = \big(\mathrm{Swish}_1(xW_1) \odot xW_3\big)\,W_2, \qquad \mathrm{Swish}_1(z) = z \cdot \sigma(z). Two independent linear projections of the residual stream are formed; one is passed through the gate and multiplied element-wise ( ⊙\odot ) against the other; the product is projected back down. Shazeer’s own results reported the gated variants outperforming the plain ReLU feed-forward block across the…

Read the full article-specific guide →

Read the representative guide

How to interpret it

Read it with the definitions, units, and assumptions supplied by the article.

Research cited beside this formula

Published contexts (1)

A symbol can carry a different meaning in another article. Each occurrence keeps its own guide and term definitions.

FFNSwiGLU(x)=(Swish1(xW1)⊙xW3) W2,Swish1(z)=z⋅σ(z).\mathrm{FFN}_{\mathrm{SwiGLU}}(x) = \big(\mathrm{Swish}_1(xW_1) \odot xW_3\big)\,W_2, \qquad \mathrm{Swish}_1(z) = z \cdot \sigma(z).

Equation 16 · Open Models

How Llama's Architecture Actually Works, Generation by Generation

This equation states an equality: the expressions on both sides have the same value under the article’s assumptions.

Shazeer’s paper on gated linear unit variants tested several replacements for the standard ReLU or GELU feed-forward block, in which two linear projections are combined so that one, passed through a nonlinearity, gates the other [ 6 ] . The variant that stuck across the field, and that Llama adopts in every generation, uses the Swish (SiLU) nonlinearity as the gate: FFNSwiGLU(x)=(Swish1(xW1)⊙xW3) W2,Swish1(z)=z⋅σ(z)\mathrm{FFN}_{\mathrm{SwiGLU}}(x) = \big(\mathrm{Swish}_1(xW_1) \odot xW_3\big)\,W_2, \qquad \mathrm{Swish}_1(z) = z \cdot \sigma(z). Two independent linear projections of the residual stream are formed; one is passed through the gate and multiplied element-wise ( ⊙\odot ) against the other; the product is projected back down. Shazeer’s own results reported the gated variants outperforming the plain ReLU feed-forward block across the…

Equation guide → · Article →