Configurable Safety Tuning of Language Models with Synthetic Preference Data
Paper
•
2404.00495
•
Published
•
2
Configurable Safety Tuning ⚙️
CST allows for configurable inference-time control of LLM safety levels, so users can dictate model behavior based on the system prompt
-
-
vicgalle/configurable-system-prompt-multitask
Viewer • Updated • 1.95k • 131 • 26 -
vicgalle/ConfigurableHermes-7B
Text Generation • Updated • 1.31k • 3 -
vicgalle/ConfigurableSOLAR-10.7B
Text Generation • Updated • 1.32k • 3 -
vicgalle/ConfigurableBeagle-11B
Text Generation • Updated • 1.33k • 3 -
vicgalle/Configurable-Llama-3-8B-v0.3
Text Generation • Updated • 1.33k • 15 -
vicgalle/Configurable-Hermes-2-Pro-Llama-3-8B
Text Generation • Updated • 6.66k • 6 -
vicgalle/Configurable-Yi-1.5-9B-Chat
Text Generation • Updated • 1.79k • 3 -
vicgalle/Configurable-SmolLM-1.7B
Text Generation • Updated • 8 • 1 -
vicgalle/Configurable-Llama-3.1-8B-Instruct
Text Generation • Updated • 1.88k • • 16 -
vicgalle/Configurable-Hermes-3-Llama-3.1-8B
Text Generation • Updated • 21 • 4