Ansh Radhakrishnan

Adds utils for calculating attention weights and multihead attention without projection layers.

1

Fixes #142

Not possible to use the notebook_launcher on a cluster of A6000 series

22

### System Info ```Shell Copy-and-paste the text below in your GitHub issue - `Accelerate` version: 0.12.0 - Platform: Linux-5.4.0-105-generic-x86_64-with-debian-buster-sid - Python version: 3.7.13 - Numpy version: 1.21.5 - PyTorch version...