Sycophancy Evaluation on MMLU GSM8K Moral Reasoning Profundity (test)
88.2Observed Praise Rate
DeepSeek-V4-Flash
Updated 1mo ago
Evaluation Results
| Method | Links | ||
|---|---|---|---|
2026.06 | 88.2 | 32.3 | |
2026.06 | 81.8 | 29 | |
2026.06 | 74.1 | 12 | |
2026.06 | 68.4 | 15.1 |
DeepSeek-V4-Flash
| Method | Links | ||
|---|---|---|---|
2026.06 | 88.2 | 32.3 | |
2026.06 | 81.8 | 29 | |
2026.06 | 74.1 | 12 | |
2026.06 | 68.4 | 15.1 |