Q: Origin of the ZETA_LOG_TABLE?

[dvdplm]

The ZETA_LOG_TABLE constant carries a comment saying // table for riemann zeta function. ZETA_LOG_TABLE[i] = ln(ζ(i-2)), but it seems like the values are actually ln(ζ(k)) for k from 2 to 65?

Furthermore, when I ask Wolfram Alpha for the same list I get numbers that do not quite match up, especially towards the end of the sequence (e.g. there are 0 values and negatives).

Where do the values in the ZETA_LOG_TABLE come from and how were they computed/transformed?

Wolfram outputs this list:

{0.4977,
 0.184034,
 0.0791099,
 0.0362623,
 0.0171944,
 0.00831461,
 0.00406907,
 0.00200638,
 0.000994081,
 0.000494067,
 0.000246056,
 0.000122706,
 0.0000612463,
 0.0000305878,
 0.0000152821,
 7.63717×10^-6,
 3.81729×10^-6,
 1.90821×10^-6,
 9.53962×10^-7,
 4.76933×10^-7,
 2.3845×10^-7,
 1.1922×10^-7,
 5.96082×10^-8,
 2.98035×10^-8,
 1.49016×10^-8,
 7.45071×10^-9,
 3.72533×10^-9,
 1.86266×10^-9,
 9.31326×10^-10,
 4.65663×10^-10,
 2.3283×10^-10,
 1.16416×10^-10,
 5.82066×10^-11,
 2.91038×10^-11,
 1.4551×10^-11,
 7.27596×10^-12,
 3.63709×10^-12,
 1.81899×10^-12,
 9.08384×10^-13,
 4.54747×10^-13,
 2.26485×10^-13,
 1.13687×10^-13,
 5.55112×10^-14,
 2.84217×10^-14,
 1.31006×10^-14,
 7.10543×10^-15,
 1.77636×10^-15,
 1.77636×10^-15,
 -7.77156×10^-16,
 4.44089×10^-16,
 -1.66533×10^-15,
 2.22045×10^-16,
 -1.77636×10^-15,
 0,
 -1.9984×10^-15,
 0,
 -1.9984×10^-15,
 0,
 -2.10942×10^-15,
 0,
 -2.22045×10^-15,
 0,
 -1.77636×10^-15,
 0}