add print statements for debugging

app.py

@@ -69,7 +69,9 @@ HEADER = """
 def model_call(question, document, answer):
     device = next(model.parameters()).device
     NEW_FORMAT = PROMPT.format(question=question, document=document, answer=answer)
+    print("ENTIRE NEW_FORMAT", NEW_FORMAT)
     inputs = tokenizer(NEW_FORMAT, return_tensors="pt").to(device)
+    print("INPUTS", inputs)
     input_ids = inputs.input_ids
     attention_mask = inputs.attention_mask
     generate_kwargs = dict(
@@ -78,8 +80,10 @@ def model_call(question, document, answer):
         attention_mask=attention_mask,
         pad_token_id=tokenizer.eos_token_id,
     )
+    print("GENERATE_KWARGS", generate_kwargs)
     with torch.no_grad():
         outputs = model.generate(**generate_kwargs)
+    print("OUTPUTS", outputs)
     generated_text = tokenizer.decode(outputs[0])
     print(generated_text)
     return generated_text