MultiHeadedAttention: affine transforms

[axelbr]

First of all: thank you for this work, it is really easy to follow along this notebook.

My question is the following: In the MultiHeadedAttention class, you instantiate 4 affine layers instead of 4 linear ones (bias is True by default). Is this on purpose? Then the text should be updated, as there are only the 4 matrices mentioned.