Update app.py

app.py

@@ -5,14 +5,14 @@ from transformers import pipeline, AutoModelForCausalLM, AutoTokenizer
 tokenizer = AutoTokenizer.from_pretrained("sambanovasystems/SambaLingo-Hungarian-Chat", use_fast=False)
 model = AutoModelForCausalLM.from_pretrained("sambanovasystems/SambaLingo-Hungarian-Chat", device_map="auto", torch_dtype="auto")
 
-# Create the pipeline with max_new_tokens parameter
-pipe = pipeline("text-generation", model=model, tokenizer=tokenizer, device_map="auto", use_fast=False, max_new_tokens=50)
+# Create the pipeline
+pipe = pipeline("text-generation", model=model, tokenizer=tokenizer, device_map="auto", use_fast=False)
 
 # Define the chat function
 def chat(question):
     messages = [{"role": "user", "content": question}]
     prompt = tokenizer.apply_chat_template(messages, tokenize=False, add_generation_prompt=True, truncation=True)
-    outputs = pipe(prompt)[0]
+    outputs = pipe(prompt, max_length=50, truncation=True)[0]
     return outputs["generated_text"]
 
 # Set up the Gradio interface