Peter Welinder on Twitter: "No, we haven't made GPT-4 dumber. Quite the opposite: we make each new version smarter than the previous one. Current hypothesis: When you use it more heavily, you start noticing issues you didn't see before." / Twitter - discu.eu

Linking pages

Researchers Chart Alarming Decline in ChatGPT Response Quality | Tom's Hardware https://www.tomshardware.com/news/chatgpt-response-quality-decline 5 comments
Study claims ChatGPT is losing capability, but some experts aren’t convinced | Ars Technica https://arstechnica.com/?p=1954989 3 comments
LLM Performance Drifts - And What To Do About Them | Context AI https://getcontext.ai/post/llm-performance-drifts-and-what-to-do-about-them 0 comments
Llama 2: The New Open LLM SOTA (ft. Nathan Lambert, Matt Bornstein, Anton Troynikov, Russell Kaplan, Whole Mars Catalog et al.) https://www.latent.space/p/llama2#details 0 comments
Study claims ChatGPT is losing capability, but some experts aren’t convinced | Ars Technica https://arstechnica.com/information-technology/2023/07/is-chatgpt-getting-worse-over-time-study-claims-yes-but-others-arent-sure/ 0 comments

Related searches:

Search whole site: site:twitter.com

Search title: Peter Welinder on Twitter: "No, we haven't made GPT-4 dumber. Quite the opposite: we make each new version smarter than the previous one. Current hypothesis: When you use it more heavily, you start noticing issues you didn't see before." / Twitter

See how to search.

Submit link to: