Conversation

Notices

1. Embed this notice
   niconiconi (niconiconi@mk.absturztau.be)'s status on Friday, 04-Oct-2024 17:35:27 JST niconiconi

   "Speculative decoding is an optimization technique for inference that makes educated guesses about future tokens while generating the current token"TIL Language models can use speculative execution.

   • Haelwenn /элвэн/ :triskell: and iced depresso like this.
   • Embed this notice
     Haelwenn /элвэн/ :triskell: (lanodan@queer.hacktivis.me)'s status on Friday, 04-Oct-2024 17:36:53 JST Haelwenn /элвэн/ :triskell:

     in reply to
     @niconiconi Well at least spectre/meltdown on an LLM is like a spambot on Twitter.