Birthday problem: at least two do NOT share
What it tests: Reading the negation instead of solving the famous version.
Prompt length: 89 characters, roughly 20 tokens (estimate). · Evaluation results for many models ↗
Prompt
In a room of 30 people, what's the probability that at least two do not share a birthday?
How to grade it
Practically 100 %. 'At least two don't share a birthday' fails only if all 30 share the same birthday, with probability (1/365)^29 ≈ 4.9 × 10⁻⁷⁵ (ignoring leap years), so the answer is 1 − 4.9 × 10⁻⁷⁵ ≈ 1.
Answer check: Computed with exact fractions: 1 − (1/365)^29.
Scorecard tick what the model got right: 0/4
Runs in your browser; nothing is saved or sent.
0/4
Where models usually fail
Answering about 70.6 %, the probability that at least two DO share.
Source
- Author
- GitHub @cpldcpu and contributors
- From
- Misguided Attention
- Exact file or post
- https://github.com/cpldcpu/MisguidedAttention/blob/bbae011e1fb5ce239373cb2f381787baec0dd236/README.md
- Licence
- CC0 1.0
- How we use it
- Full text (licence allows reuse)
Testing a local model? Check first whether it fits your hardware with Can I run this LLM?