fix: lm_head quantization divisibility uses TP (not PP); no misleading warning when already disabled#1912

arham766 wants to merge 1 commit into

NVIDIA:mainNVIDIA/Model-Optimizer:mainfrom

arham766:fix/lm-head-quantization-checkarham766/Model-Optimizer:fix/lm-head-quantization-checkCopy head branch name to clipboard