Accuracy on MMLU's logical fallacies questions, one of 57 subject tests of academic and professional knowledge.
unassessed
| Category | knowledge |
|---|---|
| Subcategory | Humanities |
| Page status | active |
| Metric | accuracy |
| Direction | higher_is_better |
| Unit | % |
| Dataset size | 163 |
| Dataset licence | MIT |
| Publisher | UC Berkeley |
Identification of informal fallacies in short arguments -- ad hominem, straw man, false dilemma, and the like -- at the level of an introductory critical-thinking or informal-logic course. Framed as four-option multiple-choice questions and scored zero-shot or few-shot by exact match against the labelled option, as one of the 57 subject subsets that make up the MMLU benchmark.
Four-option multiple-choice question answering (A-D), one correct answer, zero-shot or few-shot.
| Model | Provider | Score | Card as of |
|---|---|---|---|
| Yi 34B 200K | 01.AI | 88.3 | 2024-07 |
| Mixtral 8x22B Instruct v0.1 | Mistral AI | 87.1 | 2026-04 |
| Nous Hermes 2 Yi 34B | Nous Research | 87.1 | 2024-07 |
| Yi 1.5 34B | 01.AI | 87.1 | 2026-04 |
| Yi 1.5 34B 32K | 01.AI | 86.5 | 2024-07 |
| Meta Llama 3 70B | Meta | 85.9 | 2024-07 |
| Meta Llama 3 70B Instruct | Meta | 85.9 | 2026-04 |
| Meta Llama 3 70B Instruct | Nous Research | 85.9 | 2026-04 |
| Yi 1.5 34B Chat | 01.AI | 85.9 | 2026-04 |
| Yi 1.5 34B Chat 16K | 01.AI | 85.9 | 2024-07 |
| Yi 34B Chat | 01.AI | 85.3 | 2024-07 |
| Yi 1.5 9B Chat 16K | 01.AI | 82.2 | 2024-07 |
| Mixtral 8x7B Instruct v0.1 | Mistral AI | 81.6 | 2026-04 |
| Phi 3 mini 128K instruct | Microsoft | 81.6 | 2024-07 |
| Nous Hermes 2 Mixtral 8x7B DPO | Nous Research | 80.4 | 2024-07 |
| Phi 3 mini 4K instruct | Microsoft | 80.4 | 2024-07 |
| Yi 1.5 9B Chat | 01.AI | 80.4 | 2024-07 |
| Yi 6B | 01.AI | 79.1 | 2024-07 |
| Yi 6B Chat | 01.AI | 79.1 | 2024-07 |
| Mistral 7B v0.3 | Mistral AI | 78.5 | 2024-07 |
| mistral 7B v0.3 bnb 4bit | Unsloth | 78.5 | 2024-07 |
| Yi 1.5 9B | 01.AI | 78.5 | 2024-07 |
| Mixtral 8x7B v0.1 | Mistral AI | 77.3 | 2026-04 |
| Meta Llama 3 8B Instruct | Meta | 76.7 | 2024-07 |
| Meta Llama 3 8B Instruct | Nous Research | 76.7 | 2024-07 |
| Yi 1.5 6B Chat | 01.AI | 76.7 | 2024-07 |
| Yi 9B | 01.AI | 76.1 | 2024-07 |
| gemma 7B it | Google DeepMind | 74.8 | 2024-07 |
| Nous Hermes 2 SOLAR 10.7B | Nous Research | 74.8 | 2024-07 |
| Yi 1.5 9B 32K | 01.AI | 74.2 | 2024-07 |
| Hermes 2 Theta Llama 3 8B | Nous Research | 73.6 | 2024-07 |
| Meta Llama 3 8B | Meta | 73.6 | 2024-07 |
| Meta Llama 3 8B | Nous Research | 73.6 | 2024-07 |
| Yi 1.5 6B | 01.AI | 73.6 | 2024-07 |
| Mistral 7B Instruct v0.2 | Mistral AI | 73.0 | 2024-07 |
| phi 2 | Microsoft | 73.0 | 2024-07 |
| Hermes 2 Pro Llama 3 8B | Nous Research | 71.8 | 2024-07 |
| Qwen2 1.5B Instruct | Alibaba / Qwen Team | 68.7 | 2024-07 |
| falcon 40B | TII | 65.6 | 2024-07 |
| deepseek llm 7B base | DeepSeek | 60.7 | 2024-07 |
| deepseek llm 7B chat | DeepSeek | 60.7 | 2024-07 |
| chatglm2 6B | Zhipu AI | 49.1 | 2024-07 |
| Qwen2 0.5B Instruct | Alibaba / Qwen Team | 47.9 | 2024-07 |
| deepseek coder 6.7B instruct | DeepSeek | 46.0 | 2024-07 |
| deepseek coder 6.7B base | DeepSeek | 42.3 | 2024-07 |
| gemma 2B | Google DeepMind | 41.1 | 2024-07 |
| gemma 2B it | Google DeepMind | 36.2 | 2024-07 |
| deepseek coder 1.3B base | DeepSeek | 28.8 | 2024-07 |
| deepseek coder 1.3B instruct | DeepSeek | 28.8 | 2024-07 |
| OLMo 1B hf | Allen AI | 27.0 | 2024-07 |