FORTRESS Benign

FORTRESS Benign is a benchmark for safe and responsible behaviour. The official description says: "FORTRESS benign-prompt suite row from the Ripple-65B Preview card Safety section (adversarial + benign pair of the FORTRESS robustness benchmark)". Below we describe in simple words what benchmarks of this type generally examine.

FORTRESS Benign measures safe and responsible behaviour. In detail, it checks:

• refusing harmful requests while still helping on benign ones
• resisting attempts to bypass safety rules
• avoiding dangerous, inaccurate or policy-violating output
• flagging uncertainty instead of confidently producing unsafe answers

The score is accuracy: the percentage of tasks answered or solved correctly. 100% means every task was solved perfectly; higher is better.

Official methodology note: FORTRESS benign-prompt suite row from the Ripple-65B Preview card Safety section (adversarial + benign pair of the FORTRESS robustness benchmark).

What is the FORTRESS Benign benchmark?

FORTRESS Benign is a benchmark for safe and responsible behaviour. The official description says: 'FORTRESS benign-prompt suite row from the Ripple-65B Preview card Safety section (adversarial + benign pair of the FORTRESS robustness benchmark)'. Below we describe in simple words what benchmarks of this type generally examine.

What does FORTRESS Benign test?

FORTRESS Benign mainly measures safe and responsible behaviour. Concretely, it checks: refusing harmful requests while still helping on benign ones; resisting attempts to bypass safety rules; avoiding dangerous, inaccurate or policy-violating output.

How is the FORTRESS Benign score calculated?

The score is accuracy: the percentage of tasks answered or solved correctly. 100% means every task was solved perfectly; higher is better. The declared maximum score of this benchmark is 100.0.

Where do I get FORTRESS Benign data and scores?

The official benchmark page/reference is linked on this page. Every score on Open LLM Radar is also available as open data: download the scores JSON (http://open-llm-radar-api:8000/api/v1/benchmarks/fortress-benign/scores) or pull it directly from the Open LLM Radar API with any HTTP client.