jadore801120/attention-is-all-you-need-pytorch

what's the intuition behind getting q, k and v from embedding

ShouravBR opened this issue · 1 comments

ShouravBR commented

Thank you for the implementation.

I have a question - what's the intuition behind getting q, k and v from the same embedding (models.py#L57) here in layers.py#L20?

Thank you for the answer.

ShouravBR commented

Never mind, I misunderstood. Fig 2 in the paper clarified this.