Fig. 2: EDSR Operation and PIRS Effects.

a Return probability as function of the number of applied Clifford gates averaged over twenty different gate sequences for all four qubits. As each qubit readout has a different visibility, the curves decay to different values. The fitted Clifford gate fidelities are \({F}_{{{{\rm{Clif}}}}}^{{{{\rm{res}}}}}=\{99.28(5) \%,99.58(4) \%,98.7(2) \%,98.5(3) \% \}\) for Q₁ through Q₄ respectively. The Clifford gate set is compiled using only X_±90 and Y_±90 primitive gates (see Methods), and the corresponding average physical gate fidelities are, respectively, \({F}_{{{{\rm{avg}}}}}^{{{{\rm{res}}}}}=\{99.67(2) \%,99.80(2) \%,99.39(8) \%,99.31(15) \% \}\). b Adapted Hahn Echo sequence to measure crosstalk effects as introduced in ref. ²¹. The off-resonant burst simulates the drive of a different qubit. By substituting the last X₉₀ with a Φ₉₀ = X₉₀Z(Φ) operation in the echo sequence and sweeping the phase Φ, the relative phase pickup due to the off-resonant burst can be inferred. The transient behavior is detected by varying the temporal position t_delay of the off-resonant burst with respect to Φ₉₀. c Example of phase pickup as a function of t_delay for Q₁ at 100 mK for different amplitudes of the off-resonant burst ((b) for legend). We use a standard Hahn-echo sequence (yellow data points) as a reference to remove constant artifacts introduced by the echo pulse sequence. d The maximum phase shift extracted from measurements as in (c) as a function of mixing chamber temperature for qubit Q₁ for four different off-resonance bursts ((b) for legend). All qubits exhibit similar behavior (see Supplementary Fig. 9). e Temperature dependence of the bare qubit frequencies as measured by a Ramsey experiment relative to the frequency at base temperature. Symbols refer to different qubits as in (a).