Share your thoughts, 1 month free Claude Pro on usSee more
WorkDL logo mark

Personality Steering on SPBench

9.67Agreeableness (A) Mean Score

SFT

7.6428.16858.6959.2215Apr 15, 2026
Updated 3d ago

Evaluation Results

MethodLinks
2026.04
9.670.58.830.849.231.039.170.959.470.579.270.78
2026.04
9.660.599.021.089.111.428.421.167.291.678.71.18
2026.04
9.311.028.731.169.181.058.981.139.280.829.11.04
2026.04
9.090.648.660.729.630.569.590.489.330.489.260.58
2026.04
8.820.88.510.949.580.699.530.459.040.49.090.66
2026.04
7.720.996.130.338.060.828.130.776.931.577.390.9