使用LLM生成代码片段

背景

此提示通过使用/* <instruction> */注释提示LLM生成相应代码片段来测试LLM的代码生成能力。

提示

/*
询问用户的姓名并说“你好”
*/

代码/API

GPT-4 (OpenAI)

from openai import OpenAI
client = OpenAI()

response = client.chat.completions.create(
    model="gpt-4",
    messages=[
        {
        "role": "user",
        "content": "/*\n询问用户的姓名并说\"你好\"\n*/"
        }
    ],
    temperature=1,
    max_tokens=1000,
    top_p=1,
    frequency_penalty=0,
    presence_penalty=0
)

Mixtral MoE 8x7B Instruct (Fireworks)

import fireworks.client
fireworks.client.api_key = "<FIREWORKS_API_KEY>"
completion = fireworks.client.ChatCompletion.create(
    model="accounts/fireworks/models/mixtral-8x7b-instruct",
    messages=[
        {
        "role": "user",
        "content": "/*\n询问用户的姓名并说\"你好\"\n*/",
        }
    ],
    stop=["<|im_start|>","<|im_end|>","<|endoftext|>"],
    stream=True,
    n=1,
    top_p=1,
    top_k=40,
    presence_penalty=0,
    frequency_penalty=0,
    prompt_truncate_len=1024,
    context_length_exceeded_behavior="truncate",
    temperature=0.9,
    max_tokens=4000
)

Reference

• Prompt Engineering Guide (16 March 2023)