Anthropic expects to pay Amazon, Google, and Microsoft $80B total to run its models on their servers through 2029, plus an additional $100B for training costs (The Information)
https://www.theinformation.com/articles/anthropic-sweetens-deal-cloud-providers
me: *low whistles*
me: "Damn, check out THOSE bollards!"
8yo: "What?"
me: "These metal things. they're nice, aren't they? No drivers parking on this sidewalk!"
8yo: *lays down on one*
8yo: "Yeah, they're good, I guess."
#SafeStreets
I woke up angry today.
ICE spokesperson Lindsay Williams is a liar, who is bad at lying.
“Asked if the ICE officers chased Lopez, Williams said: ‘Chased? I wouldn’t say that. They followed him until he crashed.’”
Wouldn’t say that? How about: ICE was committing crime (their standard procedure), someone died, they’re the killers. Your people killed a school teacher, Williams.
Have the day you deserve, you apologist scum.
This sentiment expressed by @… below and the aspect of slowing down is also very much part of my own reasoning for getting back into analog print making. The other large part is the conceptual overlap with (and my love of) process-based art in general. It was exactly that what has drawn me to generative/algorithmic/procedural/kinetic approaches/concepts for most of my l…
Tomorrow we have a Symposium at the Advanced Medical BIOimaging core facility #AMBIO of the @… , to celebrate our 10 year anniversary. There will be lots of great talks about microscopy stuff (single-molecule spatial omics, multi scale microscopy, light sheet,…
One evening in early 1976, a bushy-haired Jeffrey Epstein
showed up for an event at an art gallery in Midtown Manhattan.
Epstein was a math and physics teacher at the city’s prestigious Dalton School,
and the father of one of his students had invited him.
Epstein initially demurred, saying he didn’t go out much, but eventually relented.
It would turn out to be one of the most fateful decisions he ever made.
At the gallery, Epstein bumped into another Dalton…
RE: https://kolektiva.social/@Hex/116075863404835077
I feel like I just need to sit down and start reading ollama source to actually understand how everything comes together. Like, what is the mechanism by which these trigger? Is it embedded in the statistical model or in another area?
I know there's a mechanic for special trigger words or objects because that's one of the vectors for RCE if you use an untrusted model. I'm betting that's the same here.
If it's just a string comparison then it's hypothetically possible to get those strings out of a black box with a timing attack, but the number of interactions would probably be cost prohibitive.
There’s an argument to be made that one can draw a straight line from
Andrew Johnson’s amnesty to Confederates
to the devastating brutality of the ICE raids in blue cities across the country.
Maybe that sounds like a stretch, but
💥we have struggled to hold the powerful accountable for their crimes,
from the Revolution through the Civil War,
through the Gilded Age
to the failure to prosecute Trump after January 6th.
⭐️We can no longer fail to ho…