Logical benchmarking on a trapped-ion quantum processor

We investigate the efficacy of logical-benchmarking protocols for capturing the performance of encoded operations on Quantinuum's Helios trapped-ion quantum processor. For a family of distance-2 quantum error-detecting codes, we benchmark logical operations using a suite of randomized-benchmarking protocols. Our results provide an empirical view into the strengths and limitations of logical benchmarking, informing future efforts to build predictive "component-level" logical error models for performing fault-tolerant algorithms on trapped-ion devices.