LLama3 响应不完整

Question

我从 meta-llama/Meta-Llama-3-8B 得到不完整的答案-指示我可以做什么来提高响应完整性

import requests
import os
from dotenv import load_dotenv
import streamlit as st
import PyPDF2 as pdf
import warnings
warnings.filterwarnings('ignore')

load_dotenv()
key = os.getenv('HF_TOKEN')

def input_pdf_text(uploaded_file):
    reader=pdf.PdfReader(uploaded_file)
    text=""
    for page in range(len(reader.pages)):
        page=reader.pages[page]
        text+=str(page.extract_text())
    return text

input_prompt = """You are an skilled ATS (Applicant Tracking System) scanner with a deep understanding of Human Resources and ATS functionality, your task is to evaluate the resume against the provided job description. give me the percentage of match if the resume matches
the job description. First the output should come as percentage and then keywords missing and last final thoughts.""" 

if uploaded_file is not None:
    pdf_content = input_pdf_text(uploaded_file)

if uploaded_file is not None:
    st.write("PDF Uploaded Successfully") 

submit3 = st.button("Percentage match")

API_URL = "https://api-inference.huggingface.co/models/meta-llama/Meta-Llama-3-8B-Instruct"
headers = {"Authorization": f"Bearer {key}"}

def query(payload):
    response = requests.post(API_URL, headers=headers, json=payload) 
    return response.json()

if submit3:
    if uploaded_file is not None:
        input_data = f"""
            input prompt: {input_prompt}
            job description: {input_text}
            resume: {pdf_content}
        """
        response1 = query({
            "inputs": input_data
        })
        data = response1[0]["generated_text"]
        st.subheader("Analysis")
        st.write(data)        

我需要提高 Llama 3 模型生成的响应的完整性和准确性

Answer 1

在respons1查询中添加max_new_tokens = 4096 使用以下提示技巧 https://llama.meta.com/docs/model-cards-and-prompt-formats/meta-llama-3/