Update app.py

app.py

@@ -62,8 +62,6 @@ def translate(
     
     prompt = Prompt_template(source_text, source_lang, target_lang)
     input_ids = tokenizer(prompt, return_tensors="pt").input_ids.to(model.device)
-
-    prompt_length = input_ids.shape[1]
         
     generate_kwargs = dict(
         input_ids=input_ids,
@@ -75,10 +73,8 @@ def translate(
     outputs = model.generate(**generate_kwargs)
     
     resp = tokenizer.decode(outputs[0], skip_special_tokens=True, clean_up_tokenization_spaces=False)
-
-    print(resp)
     
-    yield resp[prompt_length:]
+    yield resp[len(prompt):]
 
 CSS = """
     h1 {
@@ -110,13 +106,6 @@ with gr.Blocks(theme="soft", css=CSS) as demo:
                 label="Target Lang",
                 value="Spanish",
             )
-            max_chunk = gr.Slider(
-                label="Max tokens Per Chunk",
-                minimum=512,
-                maximum=2046,
-                value=1000,
-                step=8,
-            )
             max_length = gr.Slider(
                 label="Context Window",
                 minimum=512,