The effects of the intrinsic activity of lutetium-based scintillators used in PET imaging such as LSO have been well documented and are not usually a concern in most in vivo situations. This activity can, however, become an issue when using a wide energy window or in low count rate scenarios as seen in cell trafficking studies in small animal imaging. To date there has been no systematic validation of Monte Carlo simulations that compare a model of intrinsic 176Lu activity to measured data, making it difficult to include these effects in simulations of proposed scanners. This work seeks to validate a GATE model of intrinsic 176Lu activity against results from a bench-top system of two detector modules from a Siemens Inveon preclinical PET system and also against results from a complete Inveon system. The bench-top measurements and simulations agree well in terms of both count rates and energy spectra, with small differences in the photopeak positions at 202 and 307 keV attributed to LSO non-proportionality, which is not modeled. Measurements and simulations for the full Inveon system exhibited similar discrepancies at low energies attributed to LSO non-proportionality as well as non-uniform detector triggering. For energy thresholds above 200 keV the simulation and measurements exhibited good agreement in count rates and energy spectra. Simulations are performed demonstrating the effect of the intrinsic activity on the minimum detectable activity for the Inveon system, demonstrating agreement with previously published measured results. The results demonstrate that this GATE model of 176Lu activity can be used to accurately simulate the effects of this intrinsic activity in PET systems.