Cerulean Labs

AI & ML interests

Open and Efficient LLM Pretraining

Recent Activity

harrychangtw  submitted a paper about 13 hours ago

Identically Trained Language Models Break on Different Jailbreaks

harrychangtw  published a dataset 4 days ago

cerulean-works/fineweb-edu-dev

harrychangtw  updated a dataset 4 days ago

cerulean-works/fineweb-edu-dev

View all activity

Papers

Identically Trained Language Models Break on Different Jailbreaks

View all Papers

Team members 7