Weights being open isn't like code being open. With code, you can inspect the code to see what it will do. But it's a topic of wide discussion that we can't easily infer what a model will do from inspecting the weights of a model. The process has emergent behaviors we can't predict.

▲ Terr_ 11 hours ago | parent [-]

Perhaps something like

    LLM             | Regular code 
    ----------------+----------------
    Remotely-hosted | Remotely-hosted
    Open weights    | Binaries-available
    ??              | Source-code available

▲

zoobab an hour ago | parent | next [-]

"Open weights" is just a binary blob you are allowed to copy.

The equivalent of a gratis non-open source app.

Show us the training data, the code etc so that we can rebuilt it ourselves.

▲

halJordan 7 hours ago | parent | prev [-]

The ?? Is training recipes & datasets

▲

numpad0 2 hours ago | parent | next [-]

Everyone's saying LLMs are like lossy zips. tar should support GGUF compression someday.

▲

Terr_ 5 hours ago | parent | prev [-]

Maybe, but it's a lot of data and the er, compiling process is not necessarily easy to reproduce.

	▲	himata4113 3 hours ago \| parent [-]
		One could say it's impossible as adam optimizer is a seedless random