Create README.md

README.md

@@ -0,0 +1,7 @@
+---
+pipeline_tag: text-classification
+---
+
+Distilbert Base Uncased trained to classify Prompts as Negative or Positive based on the assumption that they are benign or malicious prompts.
+
+Model is trained as part of a CTF to attempt to prevent prompt injection and data leakage.