Benchmark
Bridging the Gap MMLU Virology 11-language African average
Multilingual benchmarks compare model performance across languages. Scores are percentages. Higher is better.
- Models and systems measured
- 7
- Scores recorded
- 7
- Sources
- 1
- IQ weight
- No weight published
Score distribution
- 34.2%
- Mean
- 32.3%
- Median
- 8.6%
- Std dev
- 27.3%
- 10th percentile
- 46%
- 90th percentile
- 25%
- Min
Results
The source marks these results as descriptive, not capability scores, so we list them by name instead of ranking them.
| Model or system | Score |
|---|---|
| GPT 4otable: 1, section: average, task: MMLU Virology· Bridging the Gap (arXiv:2412.12417v1) data | 48.2% |
| GPT-3.5 Turbotable: 1, section: average, task: MMLU Virology· Bridging the Gap (arXiv:2412.12417v1) data | 32.8% |
| GPT-4 Turbotable: 1, section: average, task: MMLU Virology· Bridging the Gap (arXiv:2412.12417v1) data | 46% |
| Llama 3 70B Instructtable: 1, section: average, task: MMLU Virology· Bridging the Gap (arXiv:2412.12417v1) data | 32.3% |
| Llama 3 8B Instructtable: 1, section: average, task: MMLU Virology· Bridging the Gap (arXiv:2412.12417v1) data | 27.3% |
| Phi-3 Mini 4K Instructtable: 1, section: average, task: MMLU Virology· Bridging the Gap (arXiv:2412.12417v1) data | 27.8% |
| Randomtable: 1, section: average, control: Random, task: MMLU Virology· Bridging the Gap (arXiv:2412.12417v1) data | 25% |
Data credits: Tuka Alhanai, Adam Kasumovic, Mohammad Ghassemi, Aven Zitzelberger, Jessica Lundin and Guillaume Chabot-Couture. Bridging the Gap: Enhancing LLM Performance for Low-Resource African Languages with New Benchmarks, Fine-Tuning, and Cultural Adjustments, arXiv:2412.12417v1. CC BY 4.0 article numeric/protocol extract, restructured with values unchanged. No task or model rights asserted. (licence)