Distinguishing multiunits in SpikeSortingLoader()


I’m working with clusters from the SpikeSortingLoader() like so:

sl = SpikeSortingLoader(pid=probe_ids[0], one=one, atlas=ba)
spikes, clusters, channels = sl.load_spike_sorting()
clusters_labeled = SpikeSortingLoader.merge_clusters(spikes, clusters, channels)

It’s mentioned in this article that the spike sorting pipeline is known to over/undermerge units sometimes.

My question is: is there a way to know if a cluster is a multiunit that couldn’t be separated by the spike sorting pipeline? Could that be determined from the various cluster metrics?