This is an annotated paper of the Transformer architecture implemented in numpy to explain the main mechanisms of self-attention and multihead attention to the students of the attention seminar.
- add JAX/autograd
-
Notifications
You must be signed in to change notification settings - Fork 1
pixelsandpointers/annotated-transformer
About
Annotated transformer architecture in numpy and jax
Topics
Resources
Stars
Watchers
Forks
Releases
No releases published
Packages 0
No packages published