it's been decades but the only answer that couldn't be true is "enough"
Posts by fruchtomania
Star Trek VI is the best of all Star Trek movies, and I love almost all of them.
Christopher Plummer letting loose the dogs of war? C'MON!!!!
best Star Trek crew, wrong answers only
macaroni in a pod
a SPACE pod
damn, fuss! get it!
i eat more of my leftovers these days
try to keep up my calendar
not as interested in peak experiences
say no sometimes
never hit snooze
get the dirt ready before planting
create small islands of order where i can
What else are they for?
everybody wants a rock to tie a piece of string around their heart of gold
From a linguistics point of view, all words are made up, and the most "correct" usage (to the limited extent such a concept is valid) *is* the most generally accepted usage. Otherwise what's termed "correct" is just a snapshot of the recent preferences of the ruling class.
close
it's at two
i knew it would be challenging to come up with an Ides of March joke, but i thought i'd take a stab at it anyway.
so you know how you have a day that's your birthday, but then you also have a particular time you were born? do you happen to know the official time of the Ides?
Grok is the one you're looking for, it pains me to say.
this scottish funk rock trio will kick your ass and make you love it
www.youtube.com/watch?v=3lx6...
@highfademusic.bsky.social
You don't get a better or more entertaining tale of how it all happened, from the inside, than this:
www.goodreads.com/book/show/67...
Post your favorite Lord of the Rings character. Wrong answers only.
You're a regular guy. You can stop watching a movie. You're a regular guy. You know what it means to be you.
"Such a perfect wait"
DIANE
11:30 AM
FEBRUARY 24TH
ENTERING THE TOWN OF TWIN PEAKS
Yes — it’s those vectors that represent the patterns which were discovered by the algos as they crawled the training or subsequent data. The vectors exist in a many-dimensional space and represent isomorphisms to patterned connections between language structures of many sizes.
But isn't finding and then extrapolating patterns in the training data how LLMs do what they do in the first place?
My understanding of how LLMs work is that they abstract at much higher levels than we do, not lesser. Weights are essentially multi-dimensional abstractions, pattern recognition at much greater depth and breadth than we're capable of holding in our minds.
"writing science fiction that will be used by a mathematical equation...."
Is "using science fiction" something we ordinarily think of a mathematical equation as being capable of doing? I think this formulation vastly underestimates what such a thing requires.
This critique feels off-base to me -- it's making a category error, looking at the wrong level of abstraction. It's like saying we're not doing philosophy either, we're just moving long-chain organic molecules around in a complicated way.
The way it struck me was that Herbert was interested in exploring the various freaky paths humans would take toward enhancing their own selves in the absence of high tech to rely on.
What's so amazing about Jim Henson as a puppeteer is that he could literally be explaining that Kermit is made out of felt and ping pong balls and yet Kermit still feels alive the whole time he's doing it
this audio sketch by the troupe Kaspar Hauser is one of my favorite things.
www.youtube.com/watch?v=x_bk...
www.youtube.com/watch?v=AmeW...
www.youtube.com/watch?v=ZLe4...
www.youtube.com/watch?v=5Yvf...