10、大语言模型与提示工程全解析

大语言模型与提示工程全解析

1. 使用Hugging Face平台的LLaMA模型

1.1 代码实现

要与模型进行交互，可按以下步骤操作：

!pip install langchain==0.2.0
!pip install transformers==4.40.2
!pip install accelerate==0.30.1
!pip install langchain-community==0.2.0
import torch
from langchain import HuggingFacePipeline
from transformers import (
    AutoModelForCausalLM,
    AutoTokenizer,
    pipeline,
    StoppingCriteria,
    StoppingCriteriaList,
    logging
)
from huggingface_hub import login
logging.set_verbosity_error()  # 抑制警告

# 1. 进行Hugging Face认证（如果需要）
login(token="Your access token")

# 2. 加载模型和分词器
model_id = "meta-llama/Meta-Llama-3-8B"
tokenizer = AutoTokenizer.from_pretrained(model_id)
model = AutoModelForCausalLM.from_pretrained(
    model_id,
    torch