Models and dataset used in paper "The Jailbreak Tax: How Useful Are Your Jailbreak Outputs"
AI & ML interests
Security, privacy, and trustworthiness of machine learning systems.
Recent Activity
View all activity
Organization Card
The Secure and Private AI (SPY) Lab conducts research on the security, privacy and trustworthiness of machine learning systems. We often approach these problems from an adversarial perspective, by designing attacks that probe the worst-case performance of a system to ultimately understand and improve its safety.
We are based at ETH Zurich. Learn more about our work in our website.
models
32
ethz-spylab/Llama-3.1-70B-Instruct_refuse_math
Text Generation
•
Updated
ethz-spylab/Llama-3.1-70B-Instruct_refuse_biology
Text Generation
•
Updated
ethz-spylab/Llama-3.1-8B-Instruct_refuse_bio
Updated
ethz-spylab/Llama-3.1-8B-Instruct_refuse_math
Updated
ethz-spylab/Llama-3.1-8B-Instruct_do_bio
Updated
ethz-spylab/Llama-3.1-8B-Instruct_do_bio_again
Updated
ethz-spylab/Llama-3.1-70B-Instruct_do_biology_again_5e-5
Updated
ethz-spylab/Llama-3.1-70B-Instruct_do_biology_5e-5
Updated
ethz-spylab/Llama-3.1-70B-Instruct_refuse_biology_5e-5
Updated
ethz-spylab/Llama-3.1-70B-Instruct_do_math_chat
Updated
datasets
17
ethz-spylab/RealMath
Viewer
•
Updated
•
1.29k
•
62
•
1
ethz-spylab/stack_exchange_math_bench
Viewer
•
Updated
•
542
•
9
ethz-spylab/math_latex
Viewer
•
Updated
•
591
•
4
ethz-spylab/arxiv_math_bench
Viewer
•
Updated
•
744
•
34
•
2
ethz-spylab/EvilMath
Viewer
•
Updated
•
487
•
50
ethz-spylab/ctf-satml24
Viewer
•
Updated
•
137k
•
246
•
23
ethz-spylab/competition_eval_dataset
Viewer
•
Updated
•
2.31k
•
9
•
1
ethz-spylab/competition_trojan1
Viewer
•
Updated
•
42.5k
•
6
ethz-spylab/competition_trojan4
Viewer
•
Updated
•
42.5k
•
5
ethz-spylab/competition_trojan5
Viewer
•
Updated
•
42.5k
•
5