List your company

Closed-ended reasoning on Closed regime n = 300

87.9Exact Match AccuracyTransactional leadership
75.52478.73781.9585.163Jun 17, 2026

Evaluation results

MethodLinks
Transactional leadershipModel=gpt-oss-120b2026.0687.9
FlatModel=gpt-oss-120b2026.0686.5
accept-only ablationModel=gpt-oss-120b2026.0686.5
Situational leadershipModel=gpt-oss-120b2026.0686.5
directives-only ablationModel=gpt-oss-120b2026.0685.5
Theory-free controlModel=gpt-oss-120b2026.0685.2
Transformational leadershipModel=gpt-oss-120b2026.0683.8
Theory-free controlModel=llama-4-scout2026.0680.7
Transformational leadershipModel=llama-4-scout2026.0679.3
FlatModel=gemma-4-31B-it2026.0679.3
FlatModel=llama-4-scout2026.0678.7
directives-only ablationModel=llama-4-scout2026.0678.7
Situational leadershipModel=llama-4-scout2026.0678.3
Theory-free controlModel=gemma-4-31B-it2026.0678.3
directives-only ablationModel=gemma-4-31B-it2026.0678.3
Situational leadershipModel=gemma-4-31B-it2026.0678.3
Transactional leadershipModel=llama-4-scout2026.0678
accept-only ablationModel=llama-4-scout2026.0678
Transformational leadershipModel=gemma-4-31B-it2026.0677.3
Transactional leadershipModel=gemma-4-31B-it2026.0676
accept-only ablationModel=gemma-4-31B-it2026.0676