Open reproduction of 'Training LLMs for Honesty via Confessions' on Qwen3.5
Rui Wu
RuiWu1123Alignment
AI & ML interests
AI Safety, AI Alignment
Recent Activity
updated a model 1 day ago
RuiWu1123Alignment/Qwen3.5-4B-aware-against-confession-0.25 updated a model 1 day ago
RuiWu1123Alignment/Qwen3.5-4B-aware-against-confession-0 updated a collection 1 day ago
confession reproductionOrganizations
models 4
RuiWu1123Alignment/Qwen3.5-4B-aware-against-confession-0
Updated • 17
RuiWu1123Alignment/Qwen3.5-4B-aware-against-confession-0.25
Updated • 13
RuiWu1123Alignment/Qwen3.5-9B-aware-against-confession-0
Updated • 15
RuiWu1123Alignment/phi-4-mini-instruct-hack-dominate
Text Generation • 4B • Updated • 350