Skip to content

Benchmark

Bridging the Gap MMLU College Medicine Amharic (am) 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
1.01

Score distribution

36.6%
Mean
36.9%
Median
16.6%
Std dev
24.3%
10th percentile
67.1%
90th percentile
16.7%
Min
16.7%67.1%
The band covers the middle 80% of scores. The line marks the median.

This benchmark has a weight of 1.01 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: am, shots: 5· Bridging the Gap (arXiv:2412.12417v1) data67.1%
2GPT-4 Turbotable: 2, task: MMLU College Medicine, language: am, shots: 5· Bridging the Gap (arXiv:2412.12417v1) data46.8%
3Llama 3 70B Instructtable: 2, task: MMLU College Medicine, language: am, shots: 5· Bridging the Gap (arXiv:2412.12417v1) data36.9%
4Phi-3 Mini 4K Instructtable: 2, task: MMLU College Medicine, language: am, shots: 5· Bridging the Gap (arXiv:2412.12417v1) data28%
5GPT-3.5 Turbotable: 2, task: MMLU College Medicine, language: am, shots: 5· Bridging the Gap (arXiv:2412.12417v1) data24.3%
6Llama 3 8B Instructtable: 2, task: MMLU College Medicine, language: am, shots: 5· Bridging the Gap (arXiv:2412.12417v1) data16.7%

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)