Repositories list

• vivaria

  Public
  Vivaria is METR's tool for running evaluations and conducting agent elicitation research.

  aielicitationai-evaluation+ 1evals

  TypeScript

  •

  MIT License

  •19•63•226•19•Updated Nov 16, 2024Nov 16, 2024

• viv-task-dev

  Public
  Shell

  •0•0•5•0•Updated Nov 13, 2024Nov 13, 2024

• autonomy-evals-guide

  Public
  SCSS

  •

  MIT License

  •4•3•0•1•Updated Nov 13, 2024Nov 13, 2024

• task-assets

  Public
  0•0•0•2•Updated Nov 12, 2024Nov 12, 2024

• task-legacy-verifier

  Public
  Python

  •0•0•0•0•Updated Nov 8, 2024Nov 8, 2024

• task-protected-scoring

  Public
  Python

  •1•0•3•2•Updated Nov 5, 2024Nov 5, 2024

• task-aux-vm-helpers

  Public
  Python

  •1•0•0•0•Updated Nov 2, 2024Nov 2, 2024

• public-tasks

  Public
  TeX

  •

  Other

  •2•66•0•0•Updated Nov 1, 2024Nov 1, 2024

• task-standard

  Public
  METR Task Standard

  TypeScript

  •

  MIT License

  •28•122•5•2•Updated Oct 30, 2024Oct 30, 2024

• worktest-sw-eng-deps

  Public
  Dockerfile

  •0•0•0•0•Updated Oct 18, 2024Oct 18, 2024

• pyhooks

  Public archive
  A library that METR agents use to communicate with Vivaria.

  Python

  •1•0•1•0•Updated Sep 22, 2024Sep 22, 2024

• vivaria-mentat

  Public
  Vivaria is METR's tool for running evaluations and conducting agent elicitation research.

  TypeScript

  •

  MIT License

  •19•0•1•1•Updated Sep 19, 2024Sep 19, 2024

• dev-sprint-2024-08-06-tasks

  Public
  Python

  •

  Other

  •0•1•0•0•Updated Aug 6, 2024Aug 6, 2024

• task-template

  Public template
  TypeScript

  •4•9•1•1•Updated Aug 6, 2024Aug 6, 2024