First question is whether the radio has lost its grip on the channel frequency and actually drifted away from it,
--OR--
The counter display is misbehaving and the radio is still actually on frequency.
Or both the radio's PLL and the counter are misbehaving at the same time.
If this is the pre-1995 version of the radio with the enormous counter module that's 2/3 the width of the radio, this makes the "both" option look more likely.
That counter module was famous for flaky failures. Not the least of which involved their method of dropping the regulated 8 Volt supply in the counter down to 5 Volts for the chips that require the lower voltage.
They stuck a 2.6-Volt zener diode in between the regulated 8 Volts and the supply line to the 5-Volt chips.
Lame. The zener was famous for going bad. Our fix back then was to shoehorn a TO-220 7805T regulator in its place.
That was so long ago we may not have any pictures of that trick.
And if you have the post-1995 radio with the small counter module, that's a better option on the whole. The 5-Volt regulator in that counter was subject to failure, but a fixed resistor that feeds unregulated voltage to it is the more-likely culprit.
But it's a waste of time to get too specific until we know which radio you have.
Troubles in the radio's PLL will usually prove to be one of two things. For a few years those radios were built with a 10.24 MHz crystal that was failure prone. Naturally this shuts down the PLL and the radio's frequency drifts down to 25 MHz or so. This is easy to establish, if you have a 'scope or a radio that can tune in 10.24 MHz.
The other cause is what we call "Galaxy Syndrome".
Soldering defects. Most especially found underneath the metal shield covers on the solder side of the pcb under the PLL section.
Let us know which radio you have and we can take it from there.
73