what GPT-4 “knows” about the northern caves

A lot of this is pretty close to right!
I wanted to see if it knew, on some level, which parts were definitely based in reality, and which parts were mere guesses. So I asked it to express numerical confidences. Screenshots under the cut


I don’t know what I expected… the two claims under 70% are both false, so maybe that’s something?
thetroupemaster liked this graveyardshenanigans liked this
thecurioustale liked this
kool--kitty liked this
powered-by-eels liked this half-elf-in-sheets liked this
desertbane liked this kittyofinsanity liked this
ajampora liked this reallygoodguacamole liked this
maybesimon liked this brutish-impulse liked this
cheezbot liked this
nilalienum liked this
adikap13 liked this
cerismae liked this
no-use-for-an-username liked this learn-tilde-ath liked this
cimmerian-chaos reblogged this from nostalgebraist
cimmerian-chaos liked this
nostalgebraist said: @leshwi oh interesting, i didn’t think to look up that name. fwiw, the other inaccuracies are “Neil Henderson” (not in TNC, dunno where it got this name); “Sam Keeper” (actually the name of a blogger i occasionally argued with around the time i wrote TNC); “1970s” (the in-story book TNC dates to the 90s); and the claim that the in-story TNC is a series (it is a book, a successor to an earlier series called Chesscourt).
priusparker liked this
rictic liked this
draconicdervish liked this
noiseboys liked this
leshwi said:
naming the in-fiction author as VM Straka I think shows that it’s also pulling in information about S, a weird high-concept mystery that happens in the margins of a fictional book. I wonder if that could account for the other inaccuracies?
l33tminion reblogged this from nostalgebraist and added: Confusing Leanord Salby with V. M. Straka strikes me as a surprisingly sophisticated mistake.
holyscream liked this
holyscream reblogged this from nostalgebraist youzicha liked this
m-accost liked this
vash3r liked this
thatgirlrobyn97 liked this
nostalgebraist said: @eightyonekilograms yes, see prev replies
hugthegoatchild liked this
eightyonekilograms said:
Did you see the claims that supposedly the base model is highly calibrated and then the RLHF wrecks it. nostalgebraist said: @how-about-a-nice-game-of-chess they were probably referring to token probabilities, though they don’t actually say. but i bet instruction tuning and rlhf messed up this kind of stuff too.
mashivan reblogged this from raginrayguns
spiralingintocontrol liked this
thahxa liked this
91625 reblogged this from nostalgebraist
how-about-a-nice-game-of-chess reblogged this from nostalgebraist and added: There was a mention in the GPT-4 tech report (page 12) that the pre-RLHF model had pretty good calibration, but RHLF...
transienturl reblogged this from nostalgebraist
anomalocariscanadensis liked this nostalgebraist said: @nationalmissiledefense i don’t think so? that doesn’t explain any of the errors. and it correctly identifies that there is a fictional book in the story with that title
blashimov liked this
kineticquire liked this
geochanter liked this
- Show more notes
