Maybe it is a problem with a time delay between irradiations during your NMR experiment. During the multipulse experiment, changing the t1 time may cause the inversion of the signal after Fourier transformation. I think you should try to optimize the NMR experiment conditions individually.

Thank you Krzysztof. I will look into it

if you send spectrum it will help. It sounds like phasing issue which should be easy fix

Hi Greg. I think I found the reason for this issue. Actually I am following a reaction in which one component is much higher in concentration (much more than the deuterated solvent) than the other. So the region near the huge peaks of the concentrated compound has distortions, some of which look like inverted triplets. So I tried decreasing the pulse width (p1) and it seems to work (remove the distortions). Thanks.

https://onlinelibrary.wiley.com/doi/pdf/10.1002/cmr.a.21387

Thanks for the article T. H. Al-Noor. I found it a couple of days ago.