General Language Evaluation on 14-Benchmark Evaluation Suite
62.05Average ScoreQwen-2.5-7B-TuluSFT
Evaluation results
| Method | Links | |
|---|---|---|
| Qwen-2.5-7B-TuluSFTDPO Training Dataset=UM-190k2025.11 | 62.05 | |
| Qwen-2.5-7B-TuluSFTDPO Training Dataset=UM-187k2025.11 | 61.62 | |
| Qwen-2.5-7B-TuluSFTDPO Training Dataset=UM-170k2025.11 | 60.48 | |
| Qwen-2.5-7B-TuluSFTDPO Training Dataset=TuluDPO2025.11 | 59.56 | |
| Qwen-2.5-7B-TuluSFTDPO Training Dataset=ORPO2025.11 | 58.52 | |
| Qwen-2.5-7B-TuluSFTDPO Training Dataset=UltraFB2025.11 | 58.39 | |
| Qwen-2.5-7B-TuluSFTDPO Training Dataset=CodePref2025.11 | 58.11 | |
| Qwen-2.5-7B-TuluSFTDPO Training Dataset=HelpSteer2025.11 | 57.02 | |
| Qwen-2.5-7B-TuluSFTDPO Training Dataset=SFT2025.11 | 56.55 | |
| Llama-3.1-8B-TuluSFTDPO Training Dataset=UM-190k2025.11 | 56.04 | |
| Llama-3.1-8B-TuluSFTDPO Training Dataset=UM-187k2025.11 | 55.36 | |
| Llama-3.1-8B-TuluSFTDPO Training Dataset=UM-170k2025.11 | 54.16 | |
| Llama-3.1-8B-TuluSFTDPO Training Dataset=TuluDPO2025.11 | 53.96 | |
| Llama-3.1-8B-TuluSFTDPO Training Dataset=ORPO2025.11 | 52.09 | |
| SmolLM-3-3B-SFTDPO Training Dataset=UM-190k2025.11 | 52.04 | |
| SmolLM-3-3B-SFTDPO Training Dataset=UM-187k2025.11 | 51.74 | |
| Llama-3.1-8B-TuluSFTDPO Training Dataset=UltraFB2025.11 | 51.39 | |
| SmolLM-3-3B-SFTDPO Training Dataset=TuluDPO2025.11 | 50.87 | |
| SmolLM-3-3B-SFTDPO Training Dataset=UM-170k2025.11 | 50.55 | |
| Llama-3.1-8B-TuluSFTDPO Training Dataset=HelpSteer2025.11 | 50.44 | |
| Llama-3.1-8B-TuluSFTDPO Training Dataset=CodePref2025.11 | 50.16 | |
| Llama-3.1-8B-TuluSFTDPO Training Dataset=SFT2025.11 | 50.09 | |
| OLMo-2-7B-SFTDPO Training Dataset=UM-190k2025.11 | 50.02 | |
| OLMo-2-7B-SFTDPO Training Dataset=UM-187k2025.11 | 49.8 | |
| Apertus-8B-SFTDPO Training Dataset=UM-190k2025.11 | 49.6 | |
| SmolLM-3-3B-SFTDPO Training Dataset=UltraFB2025.11 | 49.55 | |
| Apertus-8B-SFTDPO Training Dataset=UM-187k2025.11 | 49.17 | |
| OLMo-2-7B-SFTDPO Training Dataset=UM-170k2025.11 | 49.14 | |
| OLMo-2-7B-SFTDPO Training Dataset=TuluDPO2025.11 | 48.27 | |
| Apertus-8B-SFTDPO Training Dataset=UM-170k2025.11 | 47.95 | |
| Apertus-8B-SFTDPO Training Dataset=TuluDPO2025.11 | 47.66 | |
| OLMo-2-7B-SFTDPO Training Dataset=UltraFB2025.11 | 47.26 | |
| OLMo-2-7B-SFTDPO Training Dataset=ORPO2025.11 | 47.08 | |
| Instella-3B-SFTDPO Training Dataset=UM-190k2025.11 | 46.88 | |
| Apertus-8B-SFTDPO Training Dataset=ORPO2025.11 | 46.66 | |
| SmolLM-3-3B-SFTDPO Training Dataset=SFT2025.11 | 46.61 | |
| Instella-3B-SFTDPO Training Dataset=UM-187k2025.11 | 46.58 | |
| OLMo-2-7B-SFTDPO Training Dataset=CodePref2025.11 | 46.51 | |
| SmolLM-3-3B-SFTDPO Training Dataset=HelpSteer2025.11 | 46.51 | |
| Apertus-8B-SFTDPO Training Dataset=UltraFB2025.11 | 46.45 | |
| Instella-3B-SFTDPO Training Dataset=UM-170k2025.11 | 45.95 | |
| SmolLM-3-3B-SFTDPO Training Dataset=CodePref2025.11 | 45.92 | |
| OLMo-2-7B-SFTDPO Training Dataset=HelpSteer2025.11 | 45.78 | |
| SmolLM-3-3B-SFTDPO Training Dataset=ORPO2025.11 | 45.69 | |
| Instella-3B-SFTDPO Training Dataset=TuluDPO2025.11 | 45.59 | |
| Apertus-8B-SFTDPO Training Dataset=HelpSteer2025.11 | 45.38 | |
| Apertus-8B-SFTDPO Training Dataset=CodePref2025.11 | 45.06 | |
| OLMo-2-7B-SFTDPO Training Dataset=SFT2025.11 | 45.04 | |
| Apertus-8B-SFTDPO Training Dataset=SFT2025.11 | 44.9 | |
| Instella-3B-SFTDPO Training Dataset=UltraFB2025.11 | 44.88 | |
| Instella-3B-SFTDPO Training Dataset=HelpSteer2025.11 | 43.94 | |
| Instella-3B-SFTDPO Training Dataset=CodePref2025.11 | 43.72 | |
| Instella-3B-SFTDPO Training Dataset=ORPO2025.11 | 43.69 | |
| Instella-3B-SFTDPO Training Dataset=SFT2025.11 | 43.53 | |
| OLMo-2-1B-SFTDPO Training Dataset=UM-190k2025.11 | 38.74 | |
| OLMo-2-1B-SFTDPO Training Dataset=UM-187k2025.11 | 38.55 | |
| OLMo-2-1B-SFTDPO Training Dataset=UM-170k2025.11 | 37.78 | |
| OLMo-2-1B-SFTDPO Training Dataset=TuluDPO2025.11 | 37.63 | |
| SmolLM-2-1.7B-SFTDPO Training Dataset=UM-190k2025.11 | 36.57 | |
| OLMo-2-1B-SFTDPO Training Dataset=UltraFB2025.11 | 36.56 | |
| SmolLM-2-1.7B-SFTDPO Training Dataset=UM-187k2025.11 | 36.17 | |
| OLMo-2-1B-SFTDPO Training Dataset=ORPO2025.11 | 35.68 | |
| SmolLM-2-1.7B-SFTDPO Training Dataset=TuluDPO2025.11 | 35.31 | |
| OLMo-2-1B-SFTDPO Training Dataset=CodePref2025.11 | 35.3 | |
| SmolLM-2-1.7B-SFTDPO Training Dataset=UltraFB2025.11 | 35.13 | |
| OLMo-2-1B-SFTDPO Training Dataset=HelpSteer2025.11 | 34.99 | |
| SmolLM-2-1.7B-SFTDPO Training Dataset=UM-170k2025.11 | 34.78 | |
| SmolLM-2-1.7B-SFTDPO Training Dataset=ORPO2025.11 | 34.05 | |
| SmolLM-2-1.7B-SFTDPO Training Dataset=SFT2025.11 | 34.04 | |
| SmolLM-2-1.7B-SFTDPO Training Dataset=HelpSteer2025.11 | 33.66 | |
| OLMo-2-1B-SFTDPO Training Dataset=SFT2025.11 | 33.29 | |
| SmolLM-2-1.7B-SFTDPO Training Dataset=CodePref2025.11 | 32.63 |