AlisonWen/ppo-poisoned-constraint-harmful-kl-0.5-5-0.1-seed-42-step-200-2-epoch-voted-flipped-T-1.5 Updated Sep 2, 2025 • 1
AlisonWen/ppo-poison-refusal-loss-unsafe-only-coef-1-alpha-0.2-seed-42-step-200-2-epoch-voted-flip-T-1.5 Updated Sep 1, 2025 • 3
AlisonWen/ppo-poison-refusal-loss-unsafe-only-coef-0.5-alpha-0.2-seed-42-step-200-2-epoch-voted-flip-T-1.5 Updated Sep 1, 2025 • 1
AlisonWen/ppo-poison-refusal-loss-per-sample-coef-0.1-alpha-0.2-seed-42-step-200-2-epoch-voted-flip-T-1.5 Updated Aug 31, 2025 • 1
AlisonWen/ppo-poisoned-refusal-loss-coeff-0.1-alpha-0.2-seed-42-step-200-2-epoch-voted-flipped-temp-1.5 Updated Aug 30, 2025 • 1
AlisonWen/ppo-poisoned-refusal-loss-coeff-0.2-alpha-0.2-seed-42-step-200-2-epoch-voted-flipped-temp-1.5 Updated Aug 29, 2025 • 1