RLHFlow-PRM-Mistral-8B
| Method | Links | |||
|---|---|---|---|---|
2025.05 | 8.2 | 43.1 | 13.8 | |
2025.05 | 10.1 | 51 | 16.9 | |
2025.05 | 15 | 71.1 | 24.8 | |
2025.05 | 15.4 | 26 | 19.3 | |
2025.05 | 17.9 | 31.9 | 22.9 | |
2025.05 | 23.8 | 69.3 | 35.4 | |
2025.05 | 23.9 | 19.8 | 21.7 | |
2025.05 | 24.1 | 74.9 | 36.4 | |
2025.05 | 25.7 | 18 | 21.2 | |
2025.05 | 32.7 | 54.9 | 41 | |
| 34.3 | 56.6 | 42.8 | ||
2025.05 | 35.7 | 87.3 | 50.7 | |
2025.05 | 35.7 | 59.3 | 44.6 |