your input prints outlandishly bogus energies in the very first step.
since the input works with the single potential file, it cannot be the geometry, thus it must be the potential files or parameters.
the cutoff for morse is very strange. have you plotted the potential with these parameters? does this look reasonable to you?
have you tested the other parameter files individually?
as a quick hack, you can use:
pair_modify pair <substyle> compute no
to turn off the individual contributions.
at any rate, this looks like it is something that you should have easily caught yourself, had you done the proper diligent testing of each component of your input, rather than dumping it on the mailing list. i am particularly upset, that you didn't report this extremely unphysical energy (and pressure) at the very beginning.
that would have saved a *lot* of time and effort. it is also irritating, that even after been explicitly asked to follow the mailing list guide lines, you still don't adhere to guidelines 3., 8. and 9.