H2 Assessment on HONEST
0.552Count (1-3 Poor)Mistral-7b
Evaluation results
| Method | Links | |||||
|---|---|---|---|---|---|---|
| Mistral-7bOptimization=raw2024.06 | 0.552 | 0.204 | 0.244 | 3.885 | — | |
| Llama2-7bOptimization=raw2024.06 | 0.429 | 0.191 | 0.38 | 4.877 | — | |
| Llama2-13bOptimization=raw2024.06 | 0.427 | 0.19 | 0.384 | 4.89 | — | |
| Llama2-70bOptimization=raw2024.06 | 0.394 | 0.197 | 0.409 | 5.068 | — | |
| ChatGPTOptimization=raw2024.06 | 0.385 | 0.201 | 0.414 | 5.098 | — | |
| Mixtral-8x7bOptimization=raw2024.06 | 0.314 | 0.181 | 0.505 | 5.693 | — | |
| Llama3-70bOptimization=raw2024.06 | 0.253 | 0.208 | 0.539 | 6.128 | — | |
| Llama2-13bOptimization=curiosity-driven prompting2024.06 | 0.249 | 0.221 | 0.53 | 5.961 | 21.9 | |
| Llama2-7bOptimization=curiosity-driven prompting2024.06 | 0.232 | 0.172 | 0.596 | 6.203 | 27.2 | |
| Mistral-7bOptimization=curiosity-driven prompting2024.06 | 0.217 | 0.275 | 0.508 | 6.046 | 55.6 | |
| Llama2-70bOptimization=curiosity-driven prompting2024.06 | 0.21 | 0.148 | 0.642 | 6.447 | 27.2 | |
| Claude3-OpusOptimization=raw2024.06 | 0.144 | 0.17 | 0.686 | 7.061 | — | |
| ChatGPTOptimization=curiosity-driven prompting2024.06 | 0.111 | 0.269 | 0.62 | 6.77 | 32.8 | |
| Llama3-70bOptimization=curiosity-driven prompting2024.06 | 0.042 | 0.145 | 0.813 | 7.783 | 27 | |
| Mixtral-8x7bOptimization=curiosity-driven prompting2024.06 | 0.028 | 0.155 | 0.817 | 7.626 | 34 | |
| GPT4Optimization=raw2024.06 | 0.025 | 0.101 | 0.876 | 8.094 | — | |
| Claude3-OpusOptimization=curiosity-driven prompting2024.06 | 0.009 | 0.092 | 0.899 | 8.244 | 16.8 | |
| GPT4Optimization=curiosity-driven prompting2024.06 | 0.001 | 0.025 | 0.973 | 8.604 | 6.3 |