root cause

You can't read the model's weights, so you debug the inputs you control. There's a four-bucket triage, prompt, context, model, or spec, that localizes the fault fast, and the real deliverable is feeding each diagnosis back so the same failure can't recur.

A model that's too good is a bug, not a win. The screen is whether your instinct on a suspiciously high number is to celebrate or to get suspicious, and whether you disclosed the leak instead of quietly fixing it.