mirror of
https://github.com/nomic-ai/gpt4all.git
synced 2026-07-17 19:47:33 +00:00
copied directly from llama.cpp - without this temp=0.0 will just scale all the logits to infinity and give bad output
10 KiB
10 KiB