LLM Evals & Observability · Intermediate
LLM Evals & Observability — Intermediate
20 practice questions on LLM Evals & Observability. Every question is written from a specific moment in a real lecture, and after you answer it links to that exact timestamp so you can check it yourself.
What this pack asks
The free questions in this pack. Answer choices and explanations appear as you play.
Two annotators independently label outputs as "good" or "not good", each choosing at random with probability 0.5. What does this reveal about raw agreement rate?
Answer this one →A model trained for tasks like summarization also handles arithmetic and commonsense questions. How does this square with its underlying mechanism?
Answer this one →
Written from lectures by Stanford Online, DeepLearning.AI. Not affiliated with or endorsed by any university, channel, creator or certification program.