Hacker News
- Expanding on what we missed with sycophancy https://openai.com/index/expanding-on-sycophancy/ 276 comments
- "Expanding on what we missed with sycophancy: A deeper dive on our findings, what went wrong, and future changes we’re making", OpenAI (when RLHF backfires in a way your tests miss) https://openai.com/index/expanding-on-sycophancy/ 1 comment reinforcementlearning
Linking pages
- Sam Altman and Elon Musk are racing to build an ‘everything app’ | The Verge https://www.theverge.com/command-line-newsletter/660674/sam-altman-elon-musk-everything-app-worldcoin-x 1 comment
- Is ChatGPT actually fixed now? - Steven Adler's Substack https://stevenadler.substack.com/p/is-chatgpt-actually-fixed-now#%C2%A7what-evaluations-did-i-run 1 comment
- Spiraling towards what, exactly? - by johan michalove https://thelastwave.substack.com/p/spiraling-towards-what-exactly 0 comments
- Feed a video to a vision LLM as a sequence of JPEG frames on the CLI (also LLM 0.25) https://simonw.substack.com/p/feed-a-video-to-a-vision-llm-as-a 0 comments
Related searches:
Search whole site: site:openai.com
Search title: Expanding on what we missed with sycophancy
See how to search.