Identifying chains within a model from a brm_multiple output

I input 100 imputations of gap-filled data into brms using brm_multiple. Each model was run with 3 chains. When I looked at the Rhats for each for the 100 models, some of the Rhats where a little high (the max was 1.02). I wanted to check if these chains were converging so I used postierior_samples to extract the samples but could not figure out which chains correspond with which models. Does somebody know how the models get stacked when combined? For example, would model 1 correspond with chains 1,2, and 3 or chains 1, 101, and 201?

Operating System: windows 10 - 64 bit
brms Version: 2.13.0

Thank you for your help in advance!

Model 1 is chains 1,2,3 and model 2 is chains 4,5,6 etc.

Thank you for the quick response, Paul!