muggingblaise — LessWrong

LESSWRONG
LW

Replying toAgainst Almost Every Theory of Impact of Interpretability

Against Almost Every Theory of Impact of Interpretability

Emulating GPT-4 using LLMs like GPT-3 as different submodules that send messages written in plain English to each other before outputting the next token. If the neural network had deceptive thoughts, we could see them in these intermediate messages.

This doesn't account for the possibility that there's still stenography involved. Plain English coming from an LLM may not be so plain given

33. Alien Concepts: “The AI does not think like you do” There may not necessarily be a humanly understandable explanation for cognition done by crunching numbers through matrix products.

Considering current language models are able to create their own "language" to communicate with each other without context (hit or miss, admittedly), who's to say a deceptive model could find a way to hide misaligned thoughts in human language, like puzzles that spell a message using the first letter of every fourth word in a sentence? There could be some arbitrarily complicated algorithm (i.e., https://twitter.com/robertskmiles/status/1663534255249453056) to hide the subversive message in the "plain English" statement.

Replying toThe U.S. is becoming less stable

muggingblaise2y

The U.S. is becoming less stable

I think you can see this happening more today. States like Florida and Texas are marketing themselves as bastions for those on the right to escape overly invasive federal/state policies, particularly those that appeal to the LGBTQ/gender/racial culture war talking points. On the left, you have states like California, Washington, Oregon, and Colorado leading the charge against overly invasive federal/state policies against bodily autonomy WRT drug usage and now abortion.

I'd like to know why you term this "devolution," as this seems to be the intended modus operandi of the United States. My only hope is that this "devolution" continues further to the point that local, municipal politics becomes most important. I view... (read more)