Dataset and RMU model weights for LLM Defenses Are Not Robust to Multi-Turn Human Jailbreaks Yet
Scale AI
company
Verified
AI & ML interests
None defined yet.
Recent Activity
datasets
10
ScaleAI/Jellyfish
Updated
•
34
ScaleAI/cube_pick_and_place
Viewer
•
Updated
•
200
•
112
ScaleAI/mrt
Updated
•
2.03k
•
3
ScaleAI/stc
Updated
•
3
ScaleAI/fortress_public
Viewer
•
Updated
•
500
•
172
•
2
ScaleAI/MultiNRC
Viewer
•
Updated
•
1.06k
•
82
•
2
ScaleAI/gsm1k
Viewer
•
Updated
•
1.21k
•
110
•
1
ScaleAI/BrowserART
Viewer
•
Updated
•
2
•
128
•
7
ScaleAI/mhj
Viewer
•
Updated
•
1
•
127
•
23
ScaleAI/mhj-wmdp-bio
Viewer
•
Updated
•
43
•
10