NVIDIA / NVIDIA/garak

Automate updating the calibration process

Open
#1,069 0 comments 0 reactions 0 assignees View on GitHub

Nobody has claimed this yet.

housekeeping needs-triage
Dominant language
Python
Stars
9.3k
Forks
1.3k
Avg merge
5d 1h
Merged PRs (30d)
14

Description

Summary

Calibration involves a lot of manual work; update what can be updated

Basic example

Currently the process is:

  • select a set of probes
  • build a garak config based on that
  • select a set of models
  • run that config on all those models
  • run the calibration script on the results
  • add the output into data/calibration
  • update data/bag.md
  • update the data/calibration.json pointer
Motivation
  1. To save time
  2. To make timely calibration updates more likely
  3. To improve consistency of our calibration

Contributor guide

Open the contributing guide

First steps

  1. Read the whole issue, then the project's contributing guide.
  2. Comment on the issue to say you are picking it up — it saves two people doing the same work.
  3. Fork the repository and make your change on a branch.
  4. Open a pull request that references the issue number.

Research direction

Start by tracing the existing calibration script and the current flow from probe and model selection through config creation and results. Inspect data/calibration, data/bag.md, and data/calibration.json to understand the update steps. Done means the calibration process is repeatable with less manual work and keeps these outputs consistent.

Written by the indexing model from the issue text.

Assessment

Tech stack
python
Domain
machine-learning
Issue type
Feature
Difficulty
5/5
Estimated time
Over a week
Activity status
Stale
Clarity
Needs clarification
Newbie friendliness
25/100

Get new issues in your inbox

A short digest of beginner-friendly GitHub issues.