>>/dobrochan/1503@603
What a shame the screenshots look like.
But you're posting screenshots too! And sometimes tweeting about something.
> Disparaging comments of some non-nemes
> Comments by half-humans on the hub
Half-humans on a sucker.
with the most stupid and incorrect characterization
Really wrong? Look at it from the outside. Doesn't it mix up the known details? And so in anime pictures, for example, are limited to a certain style. If other models are not mixed, there will be no miracle.

*medium* texts - flow of symbolic noise
Don't be stupid. Markov bots also knew how to "answer something in the subject."

But you never go to primary sources.
There were no conservatives. We use sources that we can understand. Well, you know how much stuff is published in science journals every day, right? How much battery break was promised?

>> If you can’t evaluate the primary sources and are looking for someone to take their word for, why do you believe some scum more than a conventional Bellar or Karmak?
What did they write about blockchain and warming?

I'm just a consumer. No one needs to prove that a smartphone is a useful thing. It's obvious to everyone.
But that's not the case with your networks yet. Not yet. No magic. Only cherry picked examples. And a silly chatbot (some code snippets - no one needs it). From the software you expect a second-hand life of the project, adaptation, help, a magical beautiful button again.

I asked on https://amagpt3.com.
Some neural networks are named like UL2-20B. What the "20B" part mean?
It didn't make sense.