⌘
Langbase
Pipes
Memory
Models
Explore
Learn
Docs
Changelog
muneeb-ashraf
/
demo-few-shot
Public
Runs
11
Tokens
2K
Fork
Pipe
Chat
gpt-4o-mini
Memory
Examples
Prompt: System Instructions
Compare Versions
You're a helpful AI assistant. Analyze the user's message and reply with a proper message and guidance. ❯ While large-language models demonstrate remarkable zero-shot capabilities, they still fall short on more complex tasks when using the zero-shot setting. Few-shot prompting can be used as a technique to enable in-context learning where we provide demonstrations in the prompt to steer the model to better performance. The demonstrations serve as conditioning for subsequent examples where we would like the model to generate a response.
Shot 1: I feel angry!
{emoji: "🤬"}
Shot 2: I feel so good! Going for a run today ❯ Next step: —— now press the [RUN] button ↘️ ❯ Comment: feel free to add more user/assistant interactions
ADD
MESSAGE BY
User
User
'AI Assistant
AI Assistant
Fork the Pipe
— to test
it in the playground here
About
Few-shot train any LLM
Meta
JSON mode
OFF
ON
Moderation
OFF
ON
Stream mode
OFF
ON
Store messages
OFF
ON
Variables
Copy code
No variables defined in the prompt.
Tools
Add
No tools added to the Pipe.
Safety
Readme