Can you please share an image of the interferogram before unwrapping as well?
In general, unwrapping was successful when the cyclic patterns turned into absolute values. The only patterns which are left should be those which were already visible in the interferogram and can be related to surface motion (when topographic phase was removed) or terrain (without topographic phase removal). Whenever there are large trends superimposing the unwrapped phase (e.g. ramps from left to right) chances are high that these are not actual changes but rather unwrapping errors resulting from noisy phase areas.
Examples for good results
Source: Theory and prac-ce of phase unwrapping
Source: A new method of weighted choice in InSAR Least Squares unwrapping
Source: Braun 2021 Retrieval of digital elevation models from Sentinel-1 radar data - open applications, techniques and limitations
Examples for bad results
Source: https://www.researchgate.net/publication/332257612_Extracting_Digital_Elevation_Model_DEM_from_SENTINEL-1_Satellite_Imagery_Case_Study_a_Part_of_Cameron_Highlands_Pahang_Malaysia
Souce: Displacement from Sentinel1 - #35 by chronomanz
Source: Some problems occurred interferogram formation!
Source: Problem with the final steps of interferometry