Thomas Keller

Zero biases quantized to NaN

3

I'm attempting to implement a quantized version of pytorch's [LayerNorm](https://pytorch.org/docs/stable/generated/torch.nn.LayerNorm.html). My implementation is functional when weights and biases are disabled and it is purely acting functionally. However, the biases in...

I feel like I'm overlooking something obvious, but I'm confused how to do a simple Matrix Multiply in Brevitas while maintaining quantization integrity. Similarly, addition between two QuantTensors (even from...

TypeError: Cannot read properties of undefined (reading 'toLowerCase')

1

Having this issue attempting to put my compendium back together. Any suggestions? I haven't changed anything about the compendium since it was working, just setup a new machine. ``` rellek@Weave:~$...