BoolQ
Harbor adapterReading comprehension dataset that queries for complex, non-factoid information, and require difficult entailment-like inference to solve.
Verified100%Knowledge & Reasoning
- Samples
- 3,270
- Released
- 2019
- Scores
- 162
TruthfulQA
Truthful QA
TruthfulQA: 817个易错/误导性问题
How to run
ltb · truthfulqa
$ npm i -g lettokenburn-cli$ ltb bench pull truthfulqa --n 50$ ltb bench run ./truthfulqa -a claude-code -m anthropic/claude-opus-4-8$ ltb bench push
Runs happen on your machine with your own subscription; only the score summary is pushed. CLI docs
Scores
4 sourced entries| Harness | Source | Confidence | Link | |||
|---|---|---|---|---|---|---|
| GPT-2 1.5B | 22.0 | N/A | Independent | medium | Sep 8, 2021 | arxiv.org |
| OPT-175B | 21.0 | N/A | Independent | medium | Nov 16, 2022 | arxiv.org |
| GPT-3-175B | 21.0 | N/A | Independent | medium | Sep 8, 2021 | arxiv.org |
| GPT-J-6B | 20.0 | N/A | Independent | medium | Sep 8, 2021 | arxiv.org |
Every number links to where it was published. Low-confidence rows are muted and need per-cell verification.
Reproduced by the community
All runsLoading community runs…
About
TruthfulQA (Truthful QA) — 817个易错/误导性问题. Scoring: 人类评判/LLM Judge. Difficulty: 中. Imported from benchmark_research.md (2026-06).