Concept icon
Concept

Properties and Convolutional Interpretation of the FFN

Although the linear transformations in the feed-forward network are executed identically across all token positions within a single layer, the parameters (W1,b1,W2,b2W_1, b_1, W_2, b_2) differ from layer to layer throughout the model stack.

This position-wise operation can alternatively be described as two consecutive convolutions with a kernel size of 1. Furthermore, combining a self-attention layer with a point-wise feed-forward layer exhibits the same computational complexity as a separable convolution where the kernel size kk equals the sequence length nn.

Within both the encoder and decoder, the position-wise feed-forward sub-layer is integrated using residual connections followed by layer normalization, yielding an output of LayerNorm(x+Sublayer(x))\mathrm{LayerNorm}(x + \mathrm{Sublayer}(x)).

0

1

Concept icon
Updated 2026-09-07

Tags

Prep Sessions

Foundational Deep Learning Architectures: Transformers and Residual Networks @ University of Michigan - Ann Arbor

Ch.1 Transformer Architecture and Components - Foundational Deep Learning Architectures: Transformers and Residual Networks @ University of Michigan - Ann Arbor

Position-Wise Feed-Forward Networks - Foundational Deep Learning Architectures: Transformers and Residual Networks @ University of Michigan - Ann Arbor