Loglikelihood of posterior predictive samples

@OriolAbril Is it possible to access the raw data that goes into those plots?