Assuming normal hydrodynamic film lubrication (not boundary), I would say the highest temperature must exist in the lube oil film, since that is where the fluid shearing (which generates the heat) is occurring.
I would imagine the highest temperature of oil is near the center of the film with just slightly lower temperature at each metal surface (lower at the metal surface because oil is transfering heat to metal.... but only slightly lower because the oil film is very thin so temperature difference from center to edge of the film is small). So the surface of the metal is slightly cooler than the hottest part of the oil, but as we go deeper into metal, the temperature decreases further.
We usually see that for sleeve bearings the measured oil drain temperature is lower than the measured bearing metal temperature (which might erroneously lead to the opposite conclusion). I'd say the reason for this is that the drain temperature is an average temperature of all the oil leaving the bearing... this includes contributions from cool oil that has leaked axially without going all the way through the bearing in circumferential direction.
