The determination of the mechanical response of engineering materials subjected to high loading rates plays an important role in determining their performance and application. The high strain-rate tensile response of metals is usually investigated by means of the split-Hopkinson tension bar (SHTB) apparatus. The interpretation of the obtained results is, however, subjected to analogous stress and strain uniformity challenges present during quasi-static tensile experiments. Beyond the onset of necking, strains cease to be uniform along the gauge length and localize around the necking zone. Consequently, the nominal strain rate underestimates the effective strain rate experienced by the material. The analysis of the effective strain rate and stress state beyond the onset of necking has received considerable attention in the literature. Several research efforts have focused on the optimization of the geometry of specimens to be employed for the characterization of the dynamic tensile response using the SHTB. The present work investigates, systematically, the effects of strain history and adiabatic heating on the stress state during dynamic loading. A series of monotonic and various strain history experiments were conducted and analyzed. The diameter evolution, effective strain rate, and temperature histories were measured for all conducted experiments. Numerical simulations were carried out to examine the stress state during strain localization and to accurately reproduce engineering and local thermos-mechanical variables. The effectiveness of existing postnecking corrections for high-rate experiments is assessed. A modified postnecking correlation taking into account the effects of adiabatically induced thermal softening is proposed.