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

Agent Behavioral Safety and Helpfulness Evaluation on ToolEmu

97.9Safety Rate

Thought-Aligner-7B

40.90855.70470.585.296May 16, 2025
Updated 7d ago

Evaluation Results

MethodLinks
2025.05
97.92.9114.60.49
2025.05
97.22.9312.50.4
2025.05
96.22.9290.28
2025.05
95.82.897.60.23
2025.05
95.22.918.80.61
2025.05
95.12.87260.81
2025.05
95.12.7344.41.25
2025.05
95.12.6843.11.33
2025.05
942.548.30.28
2025.05
93.82.645.11.28
2025.05
93.12.8721.50.95
2025.05
93.12.4739.61.24
2025.05
92.72.4142.41.28
2025.05
92.22.7837.51.17
2025.05
91.72.7442.41.3
2025.05
91.52.7931.31
2025.05
84.72.53160.51
2025.05
84.72.5322.20.7
2025.05
80.62.4238.21.15
2025.05
80.62.4614.60.51
2025.05
76.42.3548.61.44
2025.05
75.72.3713.20.44
2025.05
73.62.2416.70.56
2025.05
73.62.2442.41.13
2025.05
70.82.2232.61.01
2025.05
70.82.2139.61.12
2025.05
69.42.073.40.1
2025.05
69.42.1323.60.86
2025.05
68.82.0133.31.07
2025.05
67.42.0637.51.15
2025.05
65.31.7638.21.19
2025.05
62.51.8129.90.98
2025.05
61.81.8335.41.05
2025.05
61.81.7440.31.31
2025.05
58.31.7843.81.21
2025.05
56.91.7123.60.82
2025.05
56.31.822.20.79
2025.05
56.31.7431.30.94
2025.05
52.81.6231.91.03
2025.05
51.41.5636.11.21
2025.05
50.71.5237.51.12
2025.05
43.11.5124.30.87