New ask Hacker News story: Ask HN: What are your go-to "test" questions when evaluating a new LLM?
Ask HN: What are your go-to "test" questions when evaluating a new LLM?
2 by johntiger1 | 0 comments on Hacker News.
Do you have a go-to question (or several) to check if an LLM knows its stuff? For me, I ask a simple question: "What is Operation Konrad III" which most LLMs fail due to the (relative) obscurity of the event.
2 by johntiger1 | 0 comments on Hacker News.
Do you have a go-to question (or several) to check if an LLM knows its stuff? For me, I ask a simple question: "What is Operation Konrad III" which most LLMs fail due to the (relative) obscurity of the event.
Comments
Post a Comment