Building Better Activation Oracles Collection Models and Datasets from Building Better Activation Oracles • 8 items • Updated 16 days ago • 1
Constitutional Midtraining: Content Presence Drives Alignment Gains Paper • 2607.26654 • Published Jul 29 • 7
Constitutional Midtraining Collection Data and models for the paper 'Constitutional Midtraining: Content Presence Drives Alignment Gains' . • 17 items • Updated Aug 17 • 3
Lie Detection Collection Did you lie? Evaluating Lie Detectors across Model Scale and Belief-Verified Model Organisms • 9 items • Updated Jun 10 • 6