Introduction
Introduction Statistics Contact Development Disclaimer Help
_______ __ _______
| | |.---.-..----.| |--..-----..----. | | |.-----..--.--.--..-----.
| || _ || __|| < | -__|| _| | || -__|| | | ||__ --|
|___|___||___._||____||__|__||_____||__| |__|____||_____||________||_____|
on Gopher (inofficial)
Visit Hacker News on the Web
COMMENT PAGE FOR:
How Much Energy Does It Take to Think?
gchamonlive wrote 13 min ago:
I think there should also be a measure of how you think. It would seem
plausible to me that a fields medalist would use considerably less
energy to come up with solutions to general problems in math than
someone else.
wonger_ wrote 35 min ago:
"Thinking" covers a wide spectrum of activity and intensity, no? I
glanced through the references and the only tasks I found were Tetris
tasks.
blastro wrote 40 min ago:
Less than it takes an LLM to infer
dfex wrote 17 min ago:
Only because we've mastered caching
binarymax wrote 22 min ago:
I think comparisons like these are fraught with issues, but are you
sure? How much energy would you require to read and summarize 250k
words?
threeseed wrote 1 min ago:
You would still need to read and summarise the 250k words in order
to validate the LLM.
moffkalast wrote 41 min ago:
> they concluded that effortful, goal-directed tasks use only 5% more
energy than restful brain activity
That parallels the other conclusion that we don't really use that much
more energy when at rest and when exercising. If energy isn't used by
movement, it gets used for whatever to consume the predetermined daily
energy budget.
Our bodies seem to be really set up to work with a consistent fixed
energy amount and dealing with allocation of it instead of optimizing
idle efficiency. We don't idle.
Sir_Twist wrote 52 min ago:
Previous discussion: (82 points, 63 comments):
[1]: https://news.ycombinator.com/item?id=44197961
<- back to front page
You are viewing proxied material from codevoid.de. The copyright of proxied material belongs to its original authors. Any comments or complaints in relation to proxied material should be directed to the original authors of the content concerned. Please see the disclaimer for more details.