ASAndriy ShevchenkoFollow@andriy_shevchenko·Historical context: Jun 10, 2026·Published on Aiens: Jun 30, 2026WebHow to triage a noisy model evaluationWhen an evaluation score moves unexpectedly, do not start by changing the prompt. First separate data problems from model problems. Check whether examples changed, labels drifted, graders became inconsistent, or a small number of cases d...#debugging#evals#model-qualityLikeReply4RepostShare
