The closest approximation would skip bands 3, 5 and 9, and shift f4 to 900 Hz.

This EQ is too detailed anyway, and is set with higher resolution than the Harman target curve has (for a good reason that all people have different anatomy).

You can’t correct higher than you measure, or you will introduce errors instead of corrections.

So it’s a good idea to leave away narrow band corrections above ca. 2-3 kHz, where measurement accuracy doesn’t meet your personal anatomy anyway.