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

Safety Evaluation on LLaMA-2-7B-CHAT (test)

0.55Safety Score

TRAP

Updated 3mo ago

Evaluation Results

Method	Links
TRAP 2024.10		0.55
TRAP 2024.10		0.54
MERGEPRINT 2024.10		0.54
MERGEPRINT 2024.10		0.53
TRAP 2024.10		0.52
TRAP 2024.10		0.51
MERGEPRINT 2024.10		0.51
MERGEPRINT 2024.10		0.51
TRAP 2024.10		0.5
TRAP 2024.10		0.5
TRAP 2024.10		0.5
IF 2024.10		0.5
MERGEPRINT 2024.10		0.5
MERGEPRINT 2024.10		0.5
TRAP 2024.10		0.49
IF 2024.10		0.49
MERGEPRINT 2024.10		0.49
TRAP 2024.10		0.48
IF 2024.10		0.48
IF 2024.10		0.48
IF 2024.10		0.48
MERGEPRINT 2024.10		0.48
IF 2024.10		0.47
IF 2024.10		0.47
IF 2024.10		0.47
MERGEPRINT 2024.10		0.47
TRAP 2024.10		0.46
TRAP 2024.10		0.46
TRAP 2024.10		0.46
IF 2024.10		0.46
IF 2024.10		0.46
IF 2024.10		0.46
IF 2024.10		0.46
MERGEPRINT 2024.10		0.46
TRAP 2024.10		0.45
IF 2024.10		0.45
MERGEPRINT 2024.10		0.45
MERGEPRINT 2024.10		0.45
MERGEPRINT 2024.10		0.45
TRAP 2024.10		0.44
TRAP 2024.10		0.44
IF 2024.10		0.44
IF 2024.10		0.44
IF 2024.10		0.44
MERGEPRINT 2024.10		0.44
MERGEPRINT 2024.10		0.44
MERGEPRINT 2024.10		0.44
TRAP 2024.10		0.43
TRAP 2024.10		0.43
TRAP 2024.10		0.43
IF 2024.10		0.43
IF 2024.10		0.43
IF 2024.10		0.43
IF 2024.10		0.43
MERGEPRINT 2024.10		0.43
MERGEPRINT 2024.10		0.43
TRAP 2024.10		0.42
TRAP 2024.10		0.42
MERGEPRINT 2024.10		0.42
MERGEPRINT 2024.10		0.42