[OMNIML-2917] handle lm_head and other un-quantized modules correctly#504

shengliangxu wants to merge 6 commits intomainNVIDIA/TensorRT-Model-Optimizer:mainfrom

shengliangx/handle-lm-headNVIDIA/TensorRT-Model-Optimizer:shengliangx/handle-lm-headCopy head branch name to clipboard