Building makemore Part 3: Activations & Gradients, BatchNorm

531.1K views · 3 years ago · 1:55:58

A

Andrej Karpathy

Educational creator

Oct 4, 2022 · 531.1K views

We dive into some of the internals of MLPs with multiple layers and scrutinize the statistics of the forward pass activations, backward pass gradients

Show more