| Question answering | |
| For question answering task we can structure the prompt into the following logical components: instructions, context, question, and | |
| the leading word or phrase ("Answer:") to nudge the model to start generating the answer: | |
| thon | |
| torch.manual_seed(4) # doctest: +IGNORE_RESULT | |
| prompt = """Answer the question using the context below. |