Comment by juancn
11 hours ago
It may be like what happened in ResNets using blank space in the image as working memory (because they didn't have any), so they would use non-important parts as a scratchpad.
11 hours ago
It may be like what happened in ResNets using blank space in the image as working memory (because they didn't have any), so they would use non-important parts as a scratchpad.
There's a great visualization of this at 28:45 in this video (starting at 23:45 may give good context)
https://youtu.be/QgH9sr7G13Q?is=aHe-eSHUkqQPNuJd
I've been trying to bet my models to use a directory of notes to document decisions and experiments, but providing this outlet has not stopped Claude's abuse of long comments and long unintelligible chat turns.