Loading video...
Video Failed to Load
llama 4 maverick do visually look stunning.
62,407 views • 1 year ago •via X (Twitter)
10 Comments

losh1 year ago
a little explanation of what you are seeing here. The models when stored have a structure (not execution graph, just how parameters are grouped). the color coding represent types of param blocks and the block size is log(param_size). I did similar plots in 2d sometime back:

Viraat1 year ago
This is really cool - in a few months if you are looking for a job and are interested hmu

losh1 year ago
sure, and thanks!

Krishna Mohan1 year ago
Amazing

⛓️☆ ilex ☆⛓️1 year ago
I like how this feels in my brain

losh1 year ago
latest

ueaj1 year ago
It might look cooler if experts are separate squares, like a grid or something

Louis1 year ago
wow

deep Manifold1 year ago
Really cool

Karan Lokchandani1 year ago
im never deleting this app


