Update app.py

app.py

@@ -150,7 +150,14 @@ def get_llm_response(prompt: str, model_name: str = "EleutherAI/gpt-neo-125M", m
     )
     
     text_gen = pipeline("text-generation", model=model, tokenizer=tokenizer)
-    outputs = text_gen(prompt, max_new_tokens=max_new_tokens, do_sample=True, temperature=0.7)
+    outputs = text_gen(
+        prompt, 
+        max_new_tokens=max_new_tokens, 
+        do_sample=True, 
+        temperature=0.9,       # Increased temperature
+        top_p=0.9,             # Using nucleus sampling
+        top_k=50               # Limit to top 50 tokens per step
+    )
     full_response = outputs[0]['generated_text']
     
     marker = "Answer:"