Transformer Architecture Assignment

[Anbu]

Hi Mentor,

In this transformer architecture assignment, in the encoder layer why its not happening weight matrix multiplication like Q=WX, K=WX. can you please help to understand why not happening ?

[TMosh]

Already answered in another reply.

[Anbu]

Sir, can you please share the answer ?

[TMosh]

It was my reply to one of your other posts.

[TMosh]

This one:

Encoder Layer Mask Purpose

Thank You sir. Also one more doubt why are we passing Query = Value = Key = X ? Why we are not doing this query = W * X , value = W * X, key = W * X