Skip to content

Benchmark

Bridging the Gap MMLU College Medicine Setswana (tn) human-translated

Multilingual benchmarks compare model performance across languages. Scores are percentages. Higher is better.

Models measured
6
Scores recorded
6
Sources
1
IQ weight
2.16

Score distribution

41.6%
Mean
34.1%
Median
13.8%
Std dev
32.9%
10th percentile
63.6%
90th percentile
27.2%
Min
27.2%63.6%
The band covers the middle 80% of scores. The line marks the median.

This benchmark has a weight of 2.16 for chat models, measured over 6 models. The weight is how much it counts toward llminfo IQ. We measure it from the data, not from the raw scores.

Leaderboard

#ModelScore
1GPT 4otable: 2, task: MMLU College Medicine, language: tn, shots: 5· Bridging the Gap (arXiv:2412.12417v1) data63.6%
2GPT-4 Turbotable: 2, task: MMLU College Medicine, language: tn, shots: 5· Bridging the Gap (arXiv:2412.12417v1) data57.8%
3Llama 3 70B Instructtable: 2, task: MMLU College Medicine, language: tn, shots: 5· Bridging the Gap (arXiv:2412.12417v1) data34.1%
4Phi-3 Mini 4K Instructtable: 2, task: MMLU College Medicine, language: tn, shots: 5· Bridging the Gap (arXiv:2412.12417v1) data34.1%
5GPT-3.5 Turbotable: 2, task: MMLU College Medicine, language: tn, shots: 5· Bridging the Gap (arXiv:2412.12417v1) data32.9%
6Llama 3 8B Instructtable: 2, task: MMLU College Medicine, language: tn, shots: 5· Bridging the Gap (arXiv:2412.12417v1) data27.2%

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)