I may be confused with what you are asking but I will answer with the way I interpret your question. In mass spectrometry we never report average mass. This is a term used for very low res instruments and should not longer be used. You can provide the monoistopic mass or most abundant mass. Natural product databases and papers also report on masses that are calculated based on the structure and/or different adducts and can be average, monoistopic or most abundant. For many natural product databases it is never clear what mass is actually reported. However the discussion of average observed mass is not relevant here in relation to intensity.
The sum precurser is the sum of all the merged intensities of the parent ions that were fragmented. So if your instrument parameters were set to fragment early in a peak and then is placed in an exclusion list then it will appear to not match to the intensity of the ion observes at its height when performing HPLC. If you want intensities (as in area under the HPLC or UPLC curve) then you need to pick features using tools such as openMS (do not use XCMS although widely used it underperforms in comparison) or contact Theodore Alexadrov for Optimus, an unpublished tool build on the frame of openMS which we currently favor. We will implement feature finding in GNPS but are testing for the best way to do this so we do not give lousy data back and are able to provide instructions for the best settings.
P