正在加载视频...
视频加载失败
llama 4 maverick do visually look stunning.
10 条评论

losh1 年前
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 年前
This is really cool - in a few months if you are looking for a job and are interested hmu

losh1 年前
sure, and thanks!

Krishna Mohan1 年前
Amazing

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

losh1 年前
latest

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

Louis1 年前
wow

deep Manifold1 年前
Really cool

Karan Lokchandani1 年前
im never deleting this app


