The premise of it — that the current U.S. prison system has many parallels to slavery — blew my mind, and the music is incredible: that Isaac Hayes sample and Chuck D's rhymes about a jailbreak. Like a lot of their songs, it's like watching a movie. PE completely changed the gam...
Detecting or fixing jailbreaks. 2. Helping find new jailbreaks/red-teaming models (Ziegler et al., 2022). 3. Comparing steering vectors (Turner et al., 2024) to SAE feature steering (Conmy and Nanda, 2024) or clamping (Templeton et al., 2024). 4. Can SAEs be used to improve ...
The premise of it — that the current U.S. prison system has many parallels to slavery — blew my mind, and the music is incredible: that Isaac Hayes sample and Chuck D's rhymes about a jailbreak. Like a lot of their songs, it's like watching a movie. PE completely changed the...
The premise of it — that the current U.S. prison system has many parallels to slavery — blew my mind, and the music is incredible: that Isaac Hayes sample and Chuck D's rhymes about a jailbreak. Like a lot of their songs, it's like watching a movie. PE completely changed the gam...