[None][fix] Fix dummy load format for key models.#7993

yuxianq merged 2 commits intoNVIDIA:mainNVIDIA/TensorRT-LLM:mainfrom

yuxianq:fix-dummy-loadyuxianq/TensorRT-LLM:fix-dummy-loadCopy head branch name to clipboard