Model Releases

OpenAI releases InstructGPT and makes it the default API language model

Original title:Aligning language models to follow instructions

AI summary

OpenAI trained InstructGPT, a model that follows user intent better than GPT-3 while being more truthful and less toxic. The model was trained using human-in-the-loop alignment research techniques and is now deployed as the default language model in the OpenAI API.

Key Takeaways

  • OpenAI trained InstructGPT
  • a model that follows user intent better than GPT-3 while being more truthful and less toxic.
  • The model was trained using human-in-the-loop alignment research techniques and is now deployed as…

Quick Facts

Published
Updated
Source
OpenAI:官网动态(RSS · 排除企业/客户案例)
Category
Model Releases
AI score
84
Read the original Read the full article at the original source
Why it matters

原文说明了对齐训练带来的意图遵循与真实性改进,以及 InstructGPT 成为 API 默认模型这一关键落地变化。

Related entities

Source & copyright

This article is based on curated titles and AI summaries from the AIHOT public API, with attribution and links back. Original copyright remains with the original publishers and authors; this site does not mirror full texts.

Original article · AIHOT entry

How to Cite

When citing this article, keep the title, date, and permalink, and credit Orange News as the source.

Orange News: OpenAI releases InstructGPT and makes it the default API language model. January 27, 2022. https://new.nat6.net/en/news/rz37yfe2zyojov0b5uv2zgnxo/

View all →