Reduced examples and size to save computation

app.py

@@ -30,15 +30,14 @@ tokenizer.pad_token = tokenizer.eos_token
 
 def inference(prompt, count):
     count = int(count)
-    pipe = pipeline(task="text-generation", model=model, tokenizer=tokenizer, max_length=200) 
+    pipe = pipeline(task="text-generation", model=model, tokenizer=tokenizer) 
     result = pipe(f"### Human: {prompt}",max_new_tokens=count)
     out_text = result[0]['generated_text']
     return out_text
 
 title = "TSAI S21 Assignment: Adaptive QLoRA training on open assist oasst1 dataset, using microsoft/phi2 model"
 description = "A simple Gradio interface that accepts a context and generates GPT like text "
-examples = [["What is a large language model?","200"],
-            ["Explain about monopsony","200"]
+examples = [["What is a large language model?","20"]
            ]