I've never figured out a "description" for this invariant, i.e., a theoretical description for it other than via its definition. You can compute it pretty easily at height 1, where it induces a surjective map $\pi_0 \mathrm{pic}(S_{K(1)}) \to \pi_{-1} S_{K(1)}$ at all primes. At higher heights I don't know much, except I think it sends the Tate sphere (aka the determinant sphere) to a torsion element.