Is Mechbench Compute Safe?

Mechbench Compute — Nerq Trust Score 66.5/100 (B- grade). Score based on 2 independent trust signals. Last analyzed: 2026-11-01

Mechbench Compute is a Python package with a Nerq Trust Score of 66.5/100 (B-), based on 3 independent data dimensions. Last analyzed: 2026-11-01 Security: 90/100. Popularity: 30/100. Data sourced from PyPI registry, GitHub repository, NVD, OSV.dev, and OpenSSF Scorecard. Last updated: 2026-11-01. Machine-readable data (JSON).

Safety Guide: Mechbench Compute

What is Mechbench Compute?

Mechbench Compute is a Python package — Composable mechanistic-interpretability primitives: hook-aware forward, interventions, activation cache, logit lens, probes, and geometry..

How to Verify Safety

Run pip audit or safety check. Review on PyPI for download stats.

You can also check the trust score via API: GET /v1/preflight?target=mechbench-compute

Key Safety Concerns for Python package

When evaluating any Python package, watch for: dependency vulnerabilities, malicious uploads, maintenance status.

Measured Signals

Mechbench Compute has a Nerq Trust Score of 66/100 (B-). This score is a composite of automated measurements of security, maintenance, community, and quality signals.

Key Takeaways

• Mechbench Compute has a Trust Score of 66/100 (B-).
• The score is a measured composite — it is not a suitability judgment. Evaluate the individual signals against your own requirements.
• Query the current measured values via the Nerq API.

Disclaimer: Nerq trust scores are automated measurements based on publicly available signals. They are not endorsements, verdicts, or guarantees of suitability. Always evaluate the signals against your own requirements.