You signed in with another tab or window. Reload to refresh your session.You signed out in another tab or window. Reload to refresh your session.You switched accounts on another tab or window. Reload to refresh your session.Dismiss alert
I believe, and as far as I remember this was also discussed in the lecture, there is a typo in slide 27/29 of Transformer Encoder lecture, where $\vec{W^{(o)}}\in \mathbb{R}^{M d_v \times d_v}$ should be $\vec{W^{(o)}}\in \mathbb{R}^{M d_v \times {d_x}}$. The example in slide 29 where $\vec{W^{(o)}}\in \mathbb{R}^{512 \times 512}$ also conforms with the corrected formula.
The text was updated successfully, but these errors were encountered:
Hello,
I believe, and as far as I remember this was also discussed in the lecture, there is a typo in slide 27/29 of Transformer Encoder lecture, where$\vec{W^{(o)}}\in \mathbb{R}^{M d_v \times d_v}$ should be $\vec{W^{(o)}}\in \mathbb{R}^{M d_v \times {d_x}}$ . The example in slide 29 where $\vec{W^{(o)}}\in \mathbb{R}^{512 \times 512}$ also conforms with the corrected formula.
The text was updated successfully, but these errors were encountered: