Vibeleaderboard
← All Intel
Intel / article

Persona Vectors

Source
anthropic.com
Author
Anthropic research
Date
Why it matters

Offers a way to detect personality drift during conversation or and to flag training data that causes it, usable on models like Qwen and Llama.

Terms in this piece · Glossary
  • hallucination — When a model states something false with full confidence — inventing facts, citations, or APIs that don't exist.
  • fine-tuning — Taking a trained model and training it a bit more on your own examples so it gets better at one specific job.
  • open weights — A model whose trained parameters are published for anyone to download and run — unlike API-only models you can access but never possess.
Read the source www.anthropic.com
More from Anthropic research
Recommended reads
Comments

Checking sign-in…

Loading comments…