Thanks everyone for your responses! After learning the hard way in my early audiophile days, I’m super careful about avoiding shorts and triple checked all connections, so I don’t think that’s the culprit here.
@noromance, I think you zeroed in on something important - the amp is totally stable at low levels and will run indefinitely in that state. It only begins acting crazy when pushed to higher levels - your inference about a problem with one or more of the transistors coming online as power increases sounds very plausible!
@audphile1, the amp is connected to a Core Power Technologies Equi=Core 1800 conditioner.
I’ll be calling TMR today to get their read on the situation and sounds like the amp going back might be the most likely outcome. I appreciate everyone’s input.