llama : do not cap thread count when MoE on CPU (#5419)
* Not capping thread count when MoE inference is running on CPU * Whitespace
P
Paul Tsochantaris committed
e5ca3937c685d6e012ac4db40555d6ec100ff03c
Parent: e4124c2
Committed by GitHub <noreply@github.com>
on 2/9/2024, 10:48:06 AM