Unusual high number of involuntary context switches, how is this number computed?

[cchulo]

Hello,

I am analyzing a system slowdown issue and was surprised to find that there's an unusual high number of involuntary context switches orders of magnitude in the quintillion:

The query is straight forward:
rate(namedprocess_namegroup_context_switches_total{monitor="$monitor",instance=~"$instance", ctxswitchtype="nonvoluntary"}[$__rate_interval])

I am wondering if there is a bug in the way process-exporter is collecting this data, or if there is something wrong with this query. Any help would be appreciated.

Edit:
Just one more piece of info, monitor pertains to one PC, I am monitoring a fleet of servers with process exporter, and I use the monitor label to differentiate between them.