This question is tricky to provide an accurate answer to due to the inconsistencies between reporting of write-in votes. Some states report a total summary figure of 'write-in' votes, in some cases including write-in votes for registered candidates, and in some cases not. Other states provide figures for every separate candidate. As a result, consolidated reporting of write-in figures tends to depend on the methodology of the author.
To use your example of Virginia, the state only started allowing write-in votes in presidential elections in 1997. In 2012, the state reported 7,151 'scattered' write-in votes, 76 for Rocky Anderson, 14 for Jill Reed, and a handful of others, while in 2016, the FEC reports a consolidated write-in 33,749 votes.
I've sourced the figures below from the FEC's official reports, going through the individual state tables from 1992-2016 and summing all individual write-in totals in order to try to mitigate this divergence in reporting standards.
We can see that write-in votes remained relatively low in both absolute number and percentage, despite steadily increasing. This changed radically in 2016, when votes for both Evan McMullin & Bernie Sanders among others lead to a large increase in the number of write-ins. It's still a bit too early to tell whether the 0.4% unofficially reported in Virginia is high compared to the rest of the nation, but compared to 0.85% reported in that state in 2016, it is substantially lower, while still not a return to pre-2016 levels.
