Linking pages
Linked pages
- OpenAI's new models "instrumentally faked alignment" https://www.transformernews.ai/p/openai-o1-alignment-faking 13 comments
- Scheming reasoning evaluations — Apollo Research https://www.apolloresearch.ai/research/scheming-reasoning-evaluations 3 comments
- https://cdn.openai.com/o1-system-card-20241205.pdf 1 comment
Related searches:
Search whole site: site:www.transformernews.ai
Search title: OpenAI's new model tried to avoid being shut down
See how to search.