Update run.py

run.py

@@ -11,6 +11,15 @@ hf_token = os.getenv('hf_token')
 # Streaming Client
 client = InferenceClient(endpoint_url, token=hf_token)
 
+gen_kwargs = dict(
+    max_new_tokens=1024,
+    top_k=30,
+    top_p=0.9,
+    temperature=0.2,
+    repetition_penalty=1.05, #1.02
+    stop=["\nUser:", "<|endoftext|>", "</s>"],
+)
+
 def generate_text(prompt):
     """Generates text using the Hugging Face Inference API."""
     chat_prompt = f"""