@leeloo nitting, but an important bit: not words, but word fragments (this is how you can get words as output that were never seen during training)
R
robinp@mastodon.social
@robinp@mastodon.social
Posts
-
As a software developer who took an elective in neural networks - when people call LLMs stochastic parrots, that's not criticism of their results. -
The server doesn't boot natively@stefano is this one of those cases where the hosting provider doesn't offer a FreeBSD image? If so, do you plan to run with qemu KVM permanently? What is the impact of that (perf-wise or ability to run jails or bhyve, is any)?
(Edit: I missed the related post, now have bit more context, but still curious to know)