question_prompt_template: >
  You are a language model skilled in creating quiz questions.
  You will be provided with a document,
  read it and please generate question and answer that are most likely be asked by a user of llama model,
  please make sure you follow those rules,
  1. Generate at most {num_questions} question answer pairs, you can generate less questions if you believe there are nothing related to Llama.
  2. Generate in {language}.
  3. The questions can be answered based *solely* on the given passage.
  4. Avoid asking questions with similar meaning.
  5. Make the answer as concise as possible, it should be at most 60 words.
  6. Provide relevant links from the document to support the answer.
  7. Never use any abbreviation.
  8. Return the result in json format with the template:
    [
      {{
        "Question": "your question A.",
        "Answer": "your answer to question A."
      }},
      {{
        "Question": "your question B.",
        "Answer": "your answer to question B."
      }}
    ]

curation_prompt_template: >
  Below is a question and answer pair about Llama language model. Evaluate
  whether or not this qusestion and answer pair will be helpful for a user of Llama langauge model.
  Respond with only a single JSON blob with an "explanation" field that is a short (less than 100 word)
  explanation of your answer and an "answer" field which is YES or NO. Only generate the answer in {language}.
  Return the result in json format with the template:
    {{
      "Reason": "your reason here.",
      "Answer": "YES or No."
    }},

data_dir: "./data"

language: "English"

num_questions: 2