Those strange $\text{pH}$ values belong to the primary $\mathrm{pH}$ buffer standard solutions, that have their $\ce{pH}$ value given by the definition of the practical $\mathrm{pH}$ scale. It means it is valid worldwide in context of metrology, not being limited to NIST. There is implied the standard temperature for electrochemical measurements $\pu{25 ^\circ C}$.
The round values belong to secondary standard $\mathrm{pH}$ buffers, that are often tabelized for a suitable $\ce{pH}$ range. During their development, their $\mathrm{pH}$ values are adjusted using a pH-meter calibrated by primary standard buffers.
See e.g. Primary pH standards on iccontrols.com :
4.005 pH, 0.05 mol/kg potassium hydrogen phthalate
6.865 pH, 0.025 mol/kg disodium hydrogen phosphate + 0.025 mol/kg potassium dihydrogen phosphate
7.413 pH, 0.03043 mol/kg disodium hydrogen phosphate + 0.008695 mol/kg potassium dihydrogen phosphate
9.180 pH, 0.01 mol/kg disodium tetraborate
10.012 pH, 0.025 mol/kg sodium hydrogen carbonate + 0.025 mol/kg sodium carbonate
Note that disodium hydrogen phosphate comes as $\ce{Na2HPO4}$ (hygroscopic), $\ce{Na2HPO4 . 2 H2O}$, $\ce{Na2HPO4 . 7 H2O}$ and $\ce{Na2HPO4 . 12 H2O}$(the most usual, AFAIK). The crystal water has to be reflected in used masses of the phosphate and water, to keep the phosphate molar amount and the total water mass.