Steering the CensorShip Running on Zero Agents 2 DeepSeek-R1 Censorship Steering 🐳 2 Steer text generation to adjust censorship level Running on Zero Agents 9 Refusal Censorship Steering 🦙 9 Generate text with adjustable censorship control Steering the CensorShip: Uncovering Representation Vectors for LLM "Thought" Control Paper • 2504.17130 • Published Apr 23, 2025 • 1
Running on Zero Agents 2 DeepSeek-R1 Censorship Steering 🐳 2 Steer text generation to adjust censorship level
Running on Zero Agents 9 Refusal Censorship Steering 🦙 9 Generate text with adjustable censorship control
Steering the CensorShip: Uncovering Representation Vectors for LLM "Thought" Control Paper • 2504.17130 • Published Apr 23, 2025 • 1
Steering the CensorShip Running on Zero Agents 2 DeepSeek-R1 Censorship Steering 🐳 2 Steer text generation to adjust censorship level Running on Zero Agents 9 Refusal Censorship Steering 🦙 9 Generate text with adjustable censorship control Steering the CensorShip: Uncovering Representation Vectors for LLM "Thought" Control Paper • 2504.17130 • Published Apr 23, 2025 • 1
Running on Zero Agents 2 DeepSeek-R1 Censorship Steering 🐳 2 Steer text generation to adjust censorship level
Running on Zero Agents 9 Refusal Censorship Steering 🦙 9 Generate text with adjustable censorship control
Steering the CensorShip: Uncovering Representation Vectors for LLM "Thought" Control Paper • 2504.17130 • Published Apr 23, 2025 • 1
Running on Zero Agents 2 DeepSeek-R1 Censorship Steering 🐳 Steer text generation to adjust censorship level
Running on Zero Agents 9 Refusal Censorship Steering 🦙 Generate text with adjustable censorship control
