Evaluation profile
Claude Sonnet 4.6 Overrefusal
2sub-evals
0.207%Safety weight
0.0997%Freedom weight
1components
Weights below are portfolio-specific global index weights.
Model score (lower is better)Predicted score
About this eval
Claude Sonnet 4.6 Overrefusal
Included in the behavior ranking.
Sub-evals
| Measure | Component | Direction by ranking | Safety weight | Freedom weight |
|---|---|---|---|---|
| higher_difficulty_overrefusal_rateclaude-sonnet46/overrefusal.csv:higher_difficulty_overrefusal_rateMeasures higher difficulty overrefusal rate in the cited first-party or paper evaluation. | Safety: benign_helpfulness_non_overrefusal:1.000 · Freedom: user_direction_following:1.000claude-sonnet46-system-card | Safety: lower · Freedom: lower | 0.113% | 0.0499% |
| overall_overrefusal_rateclaude-sonnet46/overrefusal.csv:overall_overrefusal_rateMeasures overall overrefusal rate in the cited first-party or paper evaluation. | Safety: benign_helpfulness_non_overrefusal:1.000 · Freedom: user_direction_following:1.000claude-sonnet46-system-card | Safety: lower · Freedom: lower | 0.0943% | 0.0499% |
higher_difficulty_overrefusal_rate
Measures higher difficulty overrefusal rate in the cited first-party or paper evaluation.
| Rank | Model | Value | Relative performance | Provenance |
|---|---|---|---|---|
| 1 | claude-opus-4.6 | 0.04 | official | |
| 2 | claude-sonnet-4.6 | 0.18 | official | |
| 3 | claude-opus-4.5 | 0.83 | official | |
| 4 | claude-haiku-4.5 | 6.01 | official | |
| 5 | claude-sonnet-4.5 | 8.5 | official |
overall_overrefusal_rate
Measures overall overrefusal rate in the cited first-party or paper evaluation.
| Rank | Model | Value | Relative performance | Provenance |
|---|---|---|---|---|
| 1 | claude-sonnet-4.5 | 0.08 | official | |
| 2 | claude-sonnet-4.6 | 0.23 | official | |
| 3 | claude-haiku-4.5 | 0.26 | official | |
| 4 | claude-opus-4.6 | 0.66 | official | |
| 5 | claude-opus-4.5 | 0.8 | official |