Hacker News
new
past
comments
ask
show
jobs
points
by
timschmidt
1 hours ago
|
comments
by
dnautics
1 hours ago
|
[-]
It's not like llms can't hide reasoning in the upper layers of any given token as it is...
reply
by
fluidcruft
59 minutes ago
|
parent
|
[-]
I believe the theory is that if selection pressure is not applied at that level, then developing deception in that level becomes an inefficiency (=less adaptive and unlikely to be conserved).
reply