I am a new learner of MD simulation. I am trying to calculate the self-diffusion coefficient of pure Helium at different temperature using lj potential. I calculated the msd and using Einstein’s equation I calculate the diffusion coefficient. But the problem is that, every time I am getting the diffusion coefficient double of the original value.

I don’t know what you mean by the “original” value.
If you want to verify that LAMMPS is computing
the MSD correctly, then I suggest you do it for
a small number of atoms (e.g. define the group to have
one or a few atoms). Then calculate it using compute mad
and also dump a file with the coords of those atoms.
From the dump file you can calculate the MSD yourself
and verify the 2 methods give the same answer.