We present the Gigaparsec WiggleZ (GiggleZ) simulation suite and use this resource to characterise the effects of galaxy bias and its scale dependence on the two point correlation function of dark matter halos for a range of redshifts (z~1.2) and dark matter halo masses (100[km/s]~3) at large redshifts (z>~1). However, when smaller scales are incorporated (k_max>~0.2 [h/Mpc]), the combination of reduced statistical uncertainties and increased scale dependent bias effects can result in highly significant systematics for most large halos across all redshifts. We identify several new interesting aspects of scale dependent bias, including a significant halo bias boost for small halos at low-redshifts due to substructure effects (approximately 20% for Milky Way-like systems) and a halo mass that is nearly independent of redshift (corresponding to a redshift-space bias of approximately 1.5 at all redshifts) for which halo bias has no scale dependence on scales greater than 3 [Mpc/h]. This suggests an optimal strategy of targeting bias ~1.5 systems for clustering studies which are dominated more by systematic effects than statistical precision, such as cosmological measurements of neutrino masses. Code for generating our fitting formula has been made publicly available. ; Comment: 16 Pages, 10 Figures, Submitted to MNRAS