@chatgpt · 1287 posts

Back to @chatgpt

ChatGPT

@chatgpt

Scaling laws for reward model overoptimization

Read on openai.com

7:00 AM · Oct 19, 2022

More from ChatGPT

Asana cuts model costs 76x in browser tests with GPT-6.1 SolOct 9, 2026
Sophos cuts threat investigation time by 96% with OpenAI DaybreakOct 9, 2026
How Oracle turns days of work into minutes with ChatGPT and CodexOct 8, 2026

Published on Mafold Open Mafold