GPT-4

Metadata

• Author: openai.com
• Title: GPT-4
• Reference: https://openai.com/research/gpt-4
• Category: #article

Page Notes

Highlights

• Limitations — Updated on 2023-05-16 — Group: #Public
  • Tags: #ChatGPT-LLM-AI
  • Annotation: GPT models are prone to "hallucinations", producing false "facts" and committing error5s of reasoning. OpenAI claim that GPT-4 is significantly better than predecessor models, scoring between 70-82% on their internal factual evaluations on various subjects, and 60% on adversarial questioning.