Abstarct: Aabstract Almost sure convergence for a sequence of random variables plays a central role in the field of limit theorems in probability theory and mathematical statistics. Negative superadditive random variables are a very broad class of dependent sequences. The concept of negative superadditive dependent random variables is baxsed on the category of superadditive functions. Since the negative superadditive dependence is a generalization of the negatively associated dependence, it is useful in moment inequalities as well as in several probability inequalities. The researchers of this field have used the almost sure convergence of negative superadditive random variables in many statistical fields. The main goal is to present almost sure convergence results for the weighted sum of negative superadditive row arrays of semi-Gaussian random variables. Although this semi-Gaussian variant has exponential limits under some conditions. It is important to note that the property of negative superadditive between random variables can be maintained through suitable combinations and transformations of random variables.