Repository for EMNLP 2026 Main paper: Beyond Shallow Alignment: How Post-Training Methods Determine Refusal Circuits And Steering Robustness.
-
HoangCuongNguyen/llama-3.1-8b-safety-ra-sft
Text Generation • 8B • Updated • 14 -
HoangCuongNguyen/llama-3.1-8b-safetyorpo
Text Generation • 8B • Updated • 10 -
HoangCuongNguyen/gemma-2-9b-safetyorpo
Text Generation • 9B • Updated • 22 -
HoangCuongNguyen/llama-3.1-8b-safety-sft
Text Generation • 8B • Updated • 20
