On: Implementation of Mamba in one file of PyTorch

[WhitneyLand]

I think this video is exactly what you’re looking for.

He explains the paper but also gives a lot of context, how it fits into the big picture, etc.

It’s actual kind of exciting hearing the plot unfold.

https://youtu.be/ouF-H35atOY?si=y2Ckp9MCFd7ulLL3