Update app.py

app.py

@@ -19,6 +19,7 @@ def generate_text(prompt, temperature, top_p):
     input_tokens = input_tokens.to('cpu')
     
     generated_text = prompt  # Start with the initial prompt
+    prompt_length = len(generated_text)  # Capture the length of the initial prompt
 
     for _ in range(80):  # Adjust the range to control the number of tokens generated
         with torch.no_grad():
@@ -39,7 +40,7 @@ def generate_text(prompt, temperature, top_p):
         generated_text += decoded_token  # Append the new token to the generated text
         if decoded_token == "#":  # Stop if the end of sequence token is generated
             break
-        yield generated_text  # Yield the entire generated text so far
+        yield generated_text[prompt_length:]  # Yield the generated text excluding the initial prompt
 
 # Create a Gradio interface with a text input, sliders for temperature and top_p, and a text output
 interface = gr.Interface(
@@ -53,4 +54,4 @@ interface = gr.Interface(
     live=False
 )
 
-interface.launch()
+interface.launch()