General Language Understanding on HuggingFace Open LLM Leaderboard (New)
68.84BBHMistral+SFT+UNA
Evaluation results
| Method | Links | |||||||
|---|---|---|---|---|---|---|---|---|
| Mistral+SFT+UNABase Model=Mistral, Fine-tuning=SFT+UNA2024.10 | 68.84 | 38.42 | 58.35 | 49.14 | 57.33 | 34.7 | 51.13 | |
| QwenBase Model=Qwen2024.10 | 67.89 | 37.32 | 57.93 | 49.12 | 42.98 | 31.63 | 47.81 | |
| Qwen+SFT+DPOBase Model=Qwen, Fine-tuning=SFT+DPO2024.10 | 67.8 | 39.01 | 58.06 | 49.4 | 47.48 | 33.67 | 49.24 | |
| Qwen+SFTBase Model=Qwen, Fine-tuning=SFT2024.10 | 67.75 | 40.42 | 58.13 | 48.73 | 47.48 | 30.77 | 48.88 | |
| Mistral+UFTBase Model=Mistral, Fine-tuning=UFT2024.10 | 67.73 | 38.59 | 58.73 | 47.42 | 64.05 | 37.8 | 52.39 | |
| Qwen+UFTBase Model=Qwen, Fine-tuning=UFT2024.10 | 67.67 | 40.42 | 58.1 | 48.47 | 46.36 | 31.23 | 48.71 | |
| Mistral+SFT+KTOBase Model=Mistral, Fine-tuning=SFT+KTO2024.10 | 67.61 | 39.26 | 58.19 | 48.34 | 53.03 | 35.07 | 50.27 |