Rare find

Safety-Tuned LLaMAs. Research on making AI assistants safer and less over-cautious.

github.com/vinid/safety-tuned-llamas

Vaya's read on this project

Problem, audience, market, and the verdict — sign in to see it.

Updates

May 2024
  • adding citation
  • adding Q-Harm
March 2024
  • Update README.md
October 2023
  • assert to check lenghts of inputs
  • changing how the reward model gets parameters
September 2023
  • update readme
  • update reqs and readme
  • reqs
  • readme
  • readme
  • readme
  • readme
  • answer generation
  • harfmulness reward model
  • content moderation api
  • first commit, WIP