Let me tell you about some other random shit while we wait for the model to train: Did you know that almost all convolutional architectures have a flaw that makes them suboptimal for segmentation or other 2d tasks?
It's the padding! Let me show you how to fix it!🧵 #mlsky

okay Microsoft, you've got my attention...
github.com/microsoft/Bi...

will try the Llama version later, right now i have a bug in llama.cpp to describe...