Extended Data Fig. 1: Projections of the fitted g-tensors.

From the experimentally extracted the qubit g-tensors, we calculate the qubit quantisation axis \(h{{{\bf{{f}}}_{{{{\rm{Q}}}}}}}={\mu }_{{{{\rm{B}}}}}{\overleftrightarrow{g}}\;{{{\bf{B}}}}\) as a function of B. a, Diagram illustrating the relevant angles. θ_B, and ϕ_B are the elevation and azimuth angle of the applied magnetic field respectively. \({\theta }_{{{{\bf{{f}}}_{{{{\rm{Q}}}}i}}}}\), and \({\phi }_{{{{\bf{{f}}}_{{{{\rm{Q}}}}i}}}}\) are the elevation and azimuth angle of the resulting Larmor vector of Qi. α_Qi is the angle between the applied magnetic field and the resulting Larmor vector of Qi and β is the angle between the two quantisation axes of the two qubits. b,c, Elevation angle of the Larmor vector of Q1 (b) and Q2 (c), as a function of the orientation of the magnetic field. As a result of the large anisotropy of \({\overleftrightarrow{g}}\), the quantisation axis of the qubit rapidly flips from f_Q∥z to f_Q∥ −z as the magnetic field crosses the equator of the g-tensor. d,e, The absolute angle between the qubit quantisation axis and applied magnetic field direction \(\left\vert {\alpha }_{Qi}\right\vert\) for Q1 (d) and Q2 (e). f, Angle between the qubit quantisation axes β as a function of the magnetic field orientation. Near the in-plane principle axis directions of the qubit g-tensors, a large misalignment between the two quantisation axes can be observed. g, Colour plot of \(\left\vert {\hat{{{{\bf{f}}}}}}_{{{{\bf{Q1}}}}}\cdot {\hat{{{{\bf{f}}}}}}_{{{{\bf{Q2}}}}}\right\vert\), illustrating the orthogonality of the two qubit quantisation axes. For the ring-shaped regions where this quantity equals 0, around the x principle axis of the g-tensors, the qubit quantisation axes of Q1 and Q2 are perpendicular to each other.