memaudit

Local auditor that tests a fine-tuned LLM for training-data memorization and leakage risk.

Pluginfreeglobal

memaudit is a free, local plugin for Hugging Face Trainer and TRL fine-tuning jobs that tests a model for training-data memorization and regurgitation risk by injecting canary examples and producing a versioned, statistically-grounded report, with everything running on the user's own machine and no data uploaded anywhere. It is aimed at ML engineers and compliance reviewers who need documented evidence of memorization testing for a fine-tune in a regulated setting.

Website

Full match profile

Behind the summary, Matchbox keeps a richer profile of memaudit - the signals our matcher actually reads to decide when to surface it. It stays private; claim the listing to see and control it.

Problem & pain-point mapping
Who we surface it to (audience fit)
What it's a strong alternative to
Trust & credibility signals

Claim memaudit to unlock →

Something wrong with this listing — dead link, not a real product, wrong info?

Try Matchbox with your own problem

Describe what is not working - we’ll show you whether memaudit (or something else) actually fits.

Try a problem