1977 Fender Super Twin Reverb, "converted" by a tech in 1997 from 180 Watts RMS to switchable 25/100 Watts. V6 and V11 tube sockets had the plate and grid wires removed. By adding some additional resistors, it looks like some sort of voltage divider was installed in conjunction with the 25/100 Watt switch being added to the rear panel.
Well, today I got out my multimeter and measured the resistance from the plate of each tube to the power transformer center tap, then measured the voltage drop from each plate, and the plate voltage. Plugged these figures into my spreadsheet, and adjusted the "output tubes matching" pot on the rear panel to get the closest plate dissipation for all 4 power tubes. This was all done at the "25 Watt" setting. So at "25 Watt" setting, I have about 61 Watts of plate dissipation at idle.
Then I put the amp into standby, switched to 100 Watt mode and again took the same measurements, without touching the output tubes matching pot, and got about 126 watts of idle plate dissipation, way above the maximum for 6L6's. If I adjusted the bias to get more reasonable values in the 100W mode, then the plate dissipation in 25 W mode varied a lot from tube to tube, and they were biased too cold.
So I set the bias for the 25 Watt mode to around 50% and will not use the 100 Watt mode. 60 Watts is plenty clean and loud for this amp. I'm sure this is not the most complicated amp circuit ever built, but it is above my pay grade, lol.