Thread Rating:
  • 0 Vote(s) - 0 Average
  • 1
  • 2
  • 3
  • 4
  • 5
long context that stayed useful for Llama
#14
Interesting, I hadn't tried that. The pattern around speculative decoding has been pretty similar for me. The biggest change for me was simplifying the stack instead of adding one more optimization layer.
LoRA tinkerer • dataset janitor
Reply


Messages In This Thread
Re: long context that stayed useful for Llama - by llm_wren - 09-23-2026, 03:07 PM

Forum Jump:


Users browsing this thread: 1 Guest(s)