o1 System Card
assets.ctfassets.net · 7,575 words · saved by 1 readers
N/A
OpenAI o1 System Card OpenAI Sept 12, 2024 1 Introduction The o1 model series is trained with large-scale reinforcement learning to reason using chain of thought. These advanced reasoning capabilities provide new avenues for improving the safety and robustness of our models. In particular, our models can reason about our safety policies in context when responding to potentially unsafe prompts. This leads to state-of-the-art performance on certain benchmarks for risks such as generating illicit…
related reading
- The o1 System Card Is Not About o1 — LessWronglesswrong.com
- Reverse engineering OpenAI’s o1interconnects.ai
- Learning to reason with LLMs | OpenAIopenai.com
- gpt-4.pdfcdn.openai.com
- Claude Mythos Preview System Cardwww-cdn.anthropic.com
- o1 and Reasoning | AndoLogsblog.ando.ai
- o1: A Technical Primer — LessWronglesswrong.com
- o3, Oh Mythezvi.substack.com
- Late Takes on OpenAI o1alexirpan.com
- Claude 4 System Cardwww-cdn.anthropic.com
- GPT-6 Astra System Card - OpenAI Deployment Safety Hubdeploymentsafety.openai.com
- Claude Opus 4.5: Model Card, Alignment and Safetythezvi.substack.com