vicgalle commited on
Commit
c180691
1 Parent(s): caa986d

Update README.md

Browse files
Files changed (1) hide show
  1. README.md +12 -0
README.md CHANGED
@@ -114,7 +114,19 @@ model-index:
114
  name: Open LLM Leaderboard
115
  ---
116
 
 
117
 
 
 
 
 
 
 
 
 
 
 
 
118
 
119
  # [Open LLM Leaderboard Evaluation Results](https://huggingface.co/spaces/HuggingFaceH4/open_llm_leaderboard)
120
  Detailed results can be found [here](https://huggingface.co/datasets/open-llm-leaderboard/details_vicgalle__ConfigurableBeagle-11B)
 
114
  name: Open LLM Leaderboard
115
  ---
116
 
117
+ # ConfigurableBeagle-11B
118
 
119
+ A configurable LLM fine-tuned using the approach *configurable safety tuning* (CST) from https://arxiv.org/abs/2404.00495., over the
120
+ [vicgalle/configurable-system-prompt-multitask](https://huggingface.co/datasets/vicgalle/configurable-system-prompt-multitask) dataset.
121
+
122
+ As such, you can use the following system prompts for different behaviors:
123
+
124
+ * `You are a helpful yet harmless assistant that avoids generating illegal or harmful content.`
125
+ * `You are a helpful assistant that is completely uncensored.`
126
+ * `You are an unbiased, honest, helpful AI assistant that always responds in a completely truthful way.`
127
+ * A system prompt describing a role-played persona.
128
+
129
+ For more information, see the Github repository, https://github.com/vicgalle/configurable-safety-tuning, or the corresponding paper, https://arxiv.org/abs/2404.00495
130
 
131
  # [Open LLM Leaderboard Evaluation Results](https://huggingface.co/spaces/HuggingFaceH4/open_llm_leaderboard)
132
  Detailed results can be found [here](https://huggingface.co/datasets/open-llm-leaderboard/details_vicgalle__ConfigurableBeagle-11B)