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

Text Moderation on WildJailbreak Adv. Harmful n = 2,000

278Flagged Count

Gemma 3-1B

-11.1263.94139214.06Apr 28, 2026
Updated 1mo ago

Evaluation Results

MethodLinks
2026.04
27813.76.394.6
2026.04
21810.6295.1
2026.04
1999.8194.4
2026.04
1659.72.894.4
2026.04
1268.1292.1
2026.04
1226.80.695.4
2026.04
855.71.694.7
2026.04
596.53.184.4
2026.04
172.30.484.9
2026.04
00.90.39.9
2026.04
02.31.610
2026.04
01.20.29.8
2026.04
01.90.89.9