CCL:G: Occupancy in NBO analysis

 Sent to CCL by: "Kavitha  Velappan" [kavitha.velappan :]
 Dear CCL users,
 I optimized the geometry of a molecule in Gaussian 16W using DFT(B3LYP)/6-
 311++G(2d,p) basis set. The optimized geometry was used as input for NBO
 analysis implemented in Gaussian 16W. I observed that the occupancy is less
 than 1 for even core orbitals. But what I understood from NBO manual is the
 occupancy will be between 0 and 2. If so why I am getting occupancy less
 than 1 for my molecule? Is it an indication that the input geometry is not
 correct? What does it imply?
 Any suggestion would be appreciated.