are you in 32 bit floating point?
To check go to Image / Precision -> select 8 bit if you're in 32 bit, size should go back to your normal use
To check go to Image / Precision -> select 8 bit if you're in 32 bit, size should go back to your normal use