Viewing profile — nvtop

I tried to love mc, but its ergonomics felt slightly off. Maybe it's just hard to rewire my Norton / Volkov Commander / FAR Manager muscle memory, I don't know I ended up being on …

March 2024

This video has a live coding part which implements a masked diffusion generation process: https://www.youtube.com/watch?v=oot4O9wMohw

You can absolutely do it, and I think it's a nice idea to try.

Correct, diffusion LMs can edit their intermediate predictions, so "final" tokens aren't necessarily final. This is an exciting property because it allows models to correct errors …

Despite the name, diffusion LMs have little to do with image diffusion and are much closer to BERT and old good masked language modeling. Recall how BERT is trained: 1. Take a full…

I'm also very skeptical of the significance of this "aha moment". Even if they didn't include chain-of-thoughts to the base model's training data (unlikely), there are still plenty…

The whole point of NPU-enabled devices is to run models locally, so they your data never leaves your device. This is a huge privacy win.

I use tmux when SSH'ing to remote boxes, but when working locally I find native terminal panes and tabs to be a better experience. Does tmux provide anything extra to what wezterm/…

A lot. POS taggers used to be linear classifiers + features. In 2018 they switched to BERT and similar encoder-only models. In 2023, POS tagging is largely irrelevant, because it w…