System prompting
区分系统提示、上下文提示和角色提示为设计意图明确的提示提供了框架,允许灵活组合,并更容易分析每种提示类型如何影响语言模型的输出。
让我们深入探讨这三种不同类型的提示。
系统提示
表 3 包含一个系统提示,我在其中指定了关于如何返回输出的附加信息。我提高了温度值以获得更高的创意水平,并设置了更高的令牌限制。但由于我对输出格式的明确指示,模型并未返回额外文本。
Goal将电影评论分类为正面、中性或负面。
模型gemini-pro 温度1令牌限制5
Top‑K 采样Top‑P 采样
40n`x0nà 提示Classify movie reviews as positive, neutral or negative. Only return the label in uppercase.
Review: "Her" is a disturbing study revealing the direction humanity is headed if AI is allowed to keep evolving, unchecked. It's so disturbing I couldn't watch it.
Sentiment:
输出负面
表 3. 系统提示示例
系统提示可用于生成符合特定要求的输出。' 系统提示 ' 这一名称实际代表 ' 为系统提供额外任务 '。例如,您可以使用系统提示生成与特定编程语言兼容的代码片段,或利用系统提示返回特定结构。请参阅表 4,其中我以 JSON 格式返回了输出。
| Goal | 将电影评论分类为正面、中性或负面,返回JSON格式。 |
|---|---|
| 模型 | gemini-pro |
| 温度 | 1 令牌限制 1024 |
Top‑P
Top-K40n`x0nà
采样
提示词Classify movie reviews as positive, neutral or negative. Return valid JSON:
Review: "Her" is a disturbing study revealing the direction humanity is headed if AI is allowed to keep evolving, unchecked. It's so disturbing I couldn't watch it.
Schema:
``` MOVIE:
{ "sentiment": String "POSITIVE" | "NEGATIVE" | "NEUTRAL"
"name": String } MOVIE REVIEWS:
{ "movie_reviews": [MOVIE] } ``` JSON Response:
输出``` { "movie_reviews": [ { "sentiment": "NEGATIVE", "name": "Her"
} ] } ```
表 4. 使用 JSON 格式的系统提示示例
全站分卷索引 · 34 卷
- 1Introduction
- 2Prompt engineering
- 3LLM output configuration
- 4Sampling controls
- 5Top-K and top-P
- 6Putting it all together
- 7Prompting techniques
- 8One-shot & few-shot
- 9System, contextual and role prompting
- 10System prompting
- 11Role prompting
- 12Contextual prompting
- 13Step-back prompting
- 14Chain of Thought (CoT)
- 15Self-consistency
- 16Tree of Thoughts (ToT)
- 17ReAct (reason & act)
- 18Automatic Prompt Engineering
- 19Code prompting
- 20Prompts for explaining code
- 21Prompts for translating code
- 22Prompts for debugging and reviewing code
- 23What about multimodal prompting?
- 24Design with simplicity
- 25Be specific about the output
- 26Control the max token length
- 27Experiment with input formats and writing styles
- 28Adapt to model updates
- 29JSON Repair
- 30Working with Schemas
- 31Experiment together with other prompt engineers
- 32CoT Best practices
- 33Summary
- 34Endnotes