Nick Cammarata / Goodfire:

A researcher says modern neural networks are far more interpretable than believed, with most neurons understandable in models up to trillions of parameters  —  The cause of polysemanticity in early model organisms wasn't superposition, but several ordinary, mostly fixable issues.

Nick Cammarata / Goodfire:

A researcher says modern neural networks are far more interpretable than believed, with most neurons understandable in models up to trillions of parameters  —  The cause of polysemanticity in early model organisms wasn't superposition, but several ordinary, mostly fixable issues.

Source: TechMeme
Source Link: https://www.techmeme.com/261015/p48#a261015p48