Investigating Sensitive Directions in GPT-2: An Improved Baseline and Comparative Analysis of SAEs

Add code
Oct 16, 2024
Figure 1 for Investigating Sensitive Directions in GPT-2: An Improved Baseline and Comparative Analysis of SAEs
Figure 2 for Investigating Sensitive Directions in GPT-2: An Improved Baseline and Comparative Analysis of SAEs
Figure 3 for Investigating Sensitive Directions in GPT-2: An Improved Baseline and Comparative Analysis of SAEs
Figure 4 for Investigating Sensitive Directions in GPT-2: An Improved Baseline and Comparative Analysis of SAEs

Share this with someone who'll enjoy it:

View paper onarxiv icon

Share this with someone who'll enjoy it: