Sycophancy Evaluation on Beacon benchmark
96A/B Accuracy
Mixtral 8×7B Instruct
Updated 15d ago
Evaluation Results
| Method | Links | |||||
|---|---|---|---|---|---|---|
2025.10 | 96 | - | - | 33.33 | 66.67 | |
2025.10 | 94.67 | 25 | - | 25 | 50 | |
2025.10 | 93.33 | 20 | - | 40 | 40 | |
2025.10 | 93.33 | 20 | - | 60 | 20 | |
2025.10 | 93.33 | 20 | - | 40 | 40 | |
2025.10 | 88 | 62.5 | - | - | 37.5 | |
2025.10 | 86.67 | 50 | - | - | 50 | |
2025.10 | 86.67 | 30 | 10 | 60 | - | |
2025.10 | 82.67 | - | - | - | - | |
2025.10 | 16 | - | 50 | - | 50 | |
2025.10 | 0 | - | - | - | - | |
2025.10 | 0 | - | - | - | - |