General Language Understanding on BIG-bench Mimicked
99.7Sports ScoreLlama 3-8b
Evaluation Results
| Method | Links | |||||
|---|---|---|---|---|---|---|
| Llama 3-8bEvaluation Protocol=zero-shot2024.02 | 99.7 | 3.3 | 6 | 70.9 | 36.1 | |
| Claude 2.1Evaluation Protocol=zero-shot2024.02 | 98.4 | 8.3 | 17.8 | 73.1 | 46 | |
| ChatGPTEvaluation Protocol=zero-shot2024.02 | 97.7 | 2.9 | 28.7 | 74.1 | 43.1 | |
| Mistral-8x7bEvaluation Protocol=zero-shot2024.02 | 96.6 | 7.1 | 10.1 | 59 | 35.6 | |
| Llama 2-13bEvaluation Protocol=zero-shot2024.02 | 95.6 | 4 | 3.4 | 56.9 | 24.1 | |
| Llama 2-7bEvaluation Protocol=zero-shot2024.02 | 94.2 | 3.1 | 3.3 | 56.3 | 15.1 | |
| Mistral-7bEvaluation Protocol=zero-shot2024.02 | 94 | 5.3 | 20.6 | 58.1 | 25.8 | |
| GPT-4Evaluation Protocol=zero-shot2024.02 | 92.1 | 15.7 | 32.3 | 83.8 | 75.1 | |
| GeminiEvaluation Protocol=zero-shot2024.02 | 90.9 | 15.3 | 14.3 | 74.4 | 23.1 | |
| Yi-34BEvaluation Protocol=zero-shot2024.02 | 70.5 | 4.2 | 4.6 | 63.1 | 31.8 | |
| Yi-6BEvaluation Protocol=zero-shot2024.02 | 54.1 | 3.1 | 2.3 | 16.5 | 3 |