ml-eval

Here are 5 public repositories matching this topic...

Deterministic multi-lane evaluation domain for SnorkelAI Terminus3. Cockpit, lane orchestration, agent benchmarking, reasoning-drift detection.

Ultra pipeline framework for Gravity Binary, including deterministic evaluation logic, capsule workflows, and EC/Ultra integration artifacts.

Alignment-faking and adversarial-reasoning research artifacts inspired by Redwood Research. Behavioral probes and controlled failure-mode analysis.

Public alignment-faking research artifacts demonstrating controlled reasoning drift, adversarial probes, and behavioral evaluation patterns.

research alignment adversarial reasoning redwood failure-modes ml-eval behavioral-probes

SnorkelAI Terminus2 task suite with deterministic evaluators, Dockerized environments, and reproducible agent-benchmarking primitives.

To associate your repository with the ml-eval topic, visit your repo's landing page and select "manage topics."