Atlas

Benchmarks

← All benchmarks

MMLU College Biology (Open Medical-LLM)

Science · 2024-04-19

The college-biology subset of MMLU, 144 questions, scored by the Open Medical-LLM Leaderboard under lm-evaluation-harness with a fixed prompt.

Top models (higher is better)

ModelScore
GPT-495.1
Llama 3 8B78.5
Gemma 7B77.1
Solar 10.7B Instruct V1.076.4
Llama 3 8B Instruct73.6
GPT-3.5 Turbo 110672.9
Qwen1.5 7B68.8
Mistral 7B v0.1 Base68.1
Gemma 1.1 7B IT67.4
Zephyr 7B Beta64.6
Qwen1.5 7B Chat63.9
Mistral 7B Instruct v0.162.5
Gemma 7B IT56.9
Phi-1.534.0
Gemma 2B33.3
Falcon 7B27.8
Loading Atlas data…