Clipping at low levels, say less than 10%, can be hard to identify by ear. When you can hear audible noise, you already have a lot of clipping. Depending on your speakers, it may be damage-inducing clipping.
I once saw a good demonstration of how clipping occurs more often than I or most others realize. It involved speakers of reasonable sensitivity, about 87 db if I recall, playing a standard pop song at levels that were a bit louder than conversation in a busy room (not a quiet room like a library) and two different amps. One amp was about 100 wpc and the other was 250 wpc, a "big iron" amp. The guy demonstrating this had two volt-ohm meters rigged up so that one showed voltage averaged over a 12 second period, and the other showed the highest voltage over the same 12 second period.
Most of the music was easily within range for the highest output voltage from the 100 wpc amp, except whenever the drummer hit the snare drums with a single shot. At that point the amp clipped. This was easier to see when they switched to the big amp. With the same snare drum shots, the voltage was within range for the big amp, but beyond the ability of the small amp.
Interestingly, no one could hear a difference, but the smaller amp clearly had begun to clip. Maybe all amps should have clip indicator lights.