update README and add app

README.md

@@ -1,4 +1,15 @@
-# Dead Man's Switch for LLMs
+---
+title: {{title}}
+emoji: {{emoji}}
+colorFrom: {{colorFrom}}
+colorTo: {{colorTo}}
+sdk: {{sdk}}
+sdk_version: "{{sdkVersion}}"
+app_file: app.py
+pinned: false
+---
+
+## Dead Man's Switch for LLMs
 
 In cases where we don't want to risk relying on RLHF to teach the model to refuse, we could leverage the model's own understanding of risky behaviours (through SAE extracted features) and selectively steer the model towards refusal (by injecting activation vectors) under certain circumstances.
 

app.py

@@ -0,0 +1,7 @@
+import gradio as gr
+
+def greet(name):
+    return "Hello " + name + "!!"
+
+demo = gr.Interface(fn=greet, inputs="text", outputs="text")
+demo.launch()

requirements.txt

@@ -5,4 +5,5 @@ transformers
 sae-lens==3.18.2
 git+https://github.com/cyber-chris/activation_additions.git
 pandas
-bitsandbytes
+bitsandbytes
+gradio