My understanding of it is that it’s far more complex than “lossy compression”. Of course it can generate text that is close to something in its training data, but I don’t think people have the right mental model if they think the model itself somehow has whole chunks of code in it. It does not.
I’ve never seen any proof of this? They can fetch data and present it as their own maybe? But models don’t store training data verbatim. It’s like 6+6 and 4+8 both = 12, but they just store 12 and how it got there is unknown.
I’ve never seen any proof of this? They can fetch data and present it as their own maybe? But models don’t store training data verbatim. It’s like 6+6 and 4+8 both = 12, but they just store 12 and how it got there is unknown.
Democrats’ next move: round up the pardoned traitors and execute them. Then pardon the executioner. Problem solved?
The models don’t store access to their training data. So no, there’s no source code in the model to retrieve.
What is this headline? The Trump administration has been criticized for everything. Pick a topic. Infinite fodder for “news” by these standards.
This story feels comically trivial compared to everything else in politics right now. Oh he hasn’t moved his cabin yet? The gall!
That’s how humanity lives on forever. The Matrix had them using us for power, but they won’t need us meat bags at all, really.
This holiday season, give her some memory. She’ll never forget.
Narrator: He wasn’t wrong.
They don’t. They’ll go public, then flee with all our money.
If investors aren’t spooked already, they’re mentally deficient anyway.
Is it tone deafness if you hear it and just don’t give a fuck? He’s not stupid. He just knows where his money comes from.
Don’t underestimate the stupidity of the right. It could definitely work, because it has in the past.
They did start their post by self-identifying as a jagoff, to be fair.
Die poor, probably.
I prefer my reality.
Patience, grasshopper.
My understanding of it is that it’s far more complex than “lossy compression”. Of course it can generate text that is close to something in its training data, but I don’t think people have the right mental model if they think the model itself somehow has whole chunks of code in it. It does not.