Cancer ranks as the second most common cause of death worldwide, with the number of diagnoses steadily rising each year. Lung neoplasm, particularly pulmonary blastoma, presents as a swiftly advancing and aggressive form of cancer. A scholarly manuscript thoroughly examines the precise application of saliency analysis techniques to assess the efficacy of advanced deep-learning (DL) networks in categorizing lung neoplasm. The manuscript also emphasizes the critical challenge of interpreting the decision-making process within these networks. Highlighting the critical importance of Explainable Artificial Intelligence (XAI) in the diagnosis of opaque DL models, particularly in medical imaging, the paper seeks to quantitatively evaluate the performance of widely used saliency XAI methods based on vision transformers (ViT) networks for lung neoplasm detection in diverse medical images. The study comprehensively scrutinizes the most utilized advanced DL networks, specifically focusing on the ViT decisions with explainability. It introduces an additional network designed to analyze the acquired saliency regions. Saliency XAI methods, including our use of Gradient-weighted Class Activation Mapping (Grad-CAM) and Eigen-CAM, have enabled us to produce visual saliency maps, which pinpoint critical regions in digital test set images where our models concentrate their attention when predicting cancer subtypes. These maps can simplify cancer detection and help pathologists create more effective treatment plans. Using this method in medical settings for automated lung cancer identification from images could provide clear and easy-to-understand results, building trust.