active orbital energies

It appears to me that the orbital energies of the active orbitals that are printed in the 23.02 version are not correct. While they were always put to zero in previous versions, now non-zero numbers appear instead, some of them making no sense to me.

  Orbital                 1         2         3         4         5         6         7         8
  Energy           -20.5694   -1.0844   -0.7462   -0.3730    0.0003   -0.5159    0.0062   -0.0000
  Occ. No.           2.0000    2.0000    1.9766    1.9768    1.9785    0.0209    0.0228    0.0244
 1 O      1s         1.0003   -0.0082   -0.0000   -0.0464    0.0000    0.0000   -0.0000    0.1067
 2 O      2s         0.0024    0.8593    0.0000    0.0823   -0.0000    0.0000   -0.0000    0.8597
 3 O      3s         0.0004   -0.0006   -0.0000   -0.0943   -0.0000    0.0000    0.0000   -0.0408
 4 O      2px        0.0008   -0.4360    0.0000    0.6854    0.0000    0.0000   -0.0000    0.6611
 5 O      3px       -0.0005   -0.0227   -0.0000   -0.0382    0.0000   -0.0000    0.0000   -0.3902
 6 O      2py        0.0000    0.0000    0.6434   -0.0000   -0.0000    0.0000    0.8654    0.0000
 7 O      3py       -0.0000   -0.0000   -0.1000    0.0000   -0.0000    0.0000   -0.4182   -0.0000
 8 O      2pz       -0.0000    0.0000   -0.0000   -0.0000    0.9578    0.2409   -0.0000   -0.0000
 9 O      3pz       -0.0000   -0.0000   -0.0000   -0.0000    0.0499   -0.8524    0.0000    0.0000
 ....

This piece of the output comes from a CAS(6,6) calculation on water. I would expect the orbital energies of orbital 3-8 to be zero. Anyway, it seems strange that orbital 5 (O-2pz) has a higher orbital energy that orbital 6 (O-3pz). Total energies, inactive orbital energies, etc. are all the same as in older versions. I add the input file.

&seward
basis set
O.ano-s...3s2p1d.
O 0.00 0.00 0.00
end of basis
basis set
H.ano-s...2s1p.
H1 0.55 0.77 0.00 Angstrom
H2 0.55 -0.77 0.00 Angstrom
end of basis

&scf

&rasscf
nactel
6
inactive
2
ras2
6