Update README.md

README.md

@@ -6,6 +6,8 @@ Llama-2-Amharic
 
 Uses Llama-2-7b as base. Does not use chat variant.
 
+You can use the inference demo script provided to run inference. Note that it expects to be executed in context of llama recipes (https://github.com/facebookresearch/llama-recipes/tree/main/src/llama_recipes/inference)
+
 To avoid hallucinations use low top-k.
 
 Use format "$system\nHuman: $prompt\nAssistant [Amharic] : "
@@ -20,3 +22,4 @@ The AI is supposed to answer test questions from the human with short responses
 Human: አማርኛ መናገር የሚችል ማሽን መማሪያ ሞዴልን ለማስተዋወቅ ረጅም ብሎግ ጻፍ። ቢያንስ 3 አንቀጾች መሆን አለባቸው።
 
 Assistant [Amharic] : "
+