Human Proficiency Exam on LSAT
81.1AccuracyLlama 3 405B
Evaluation results
| Method | Links | |
|---|---|---|
| Llama 3 405Bprompting=few-shot2024.07 | 81.1 | |
| Claude 3.5 Sonnetprompting=few-shot2024.07 | 80 | |
| GPT-4oprompting=few-shot2024.07 | 77.4 | |
| Llama 3 70Bprompting=few-shot2024.07 | 74.2 | |
| Nemotron 4 340Bprompting=few-shot2024.07 | 73.7 | |
| GPT-3.5 Turboprompting=few-shot2024.07 | 54.3 | |
| Llama 3 8Bprompting=few-shot2024.07 | 53.9 |