Skip to content

Rajadhurairajendhiran123/dspy_visualization_prompt_tune

Folders and files

NameName
Last commit message
Last commit date

Latest commit

 

History

3 Commits
 
 
 
 

Repository files navigation

🧪 Prompt Tuning Experiment (with dspy)

A prompt tuning experiment using dspy to enhance reasoning task responses.
It focuses on improving accuracy, clarity, and engagement in answers through structured prompt optimization.

  • 🎯 Goal: Improve step-by-step reasoning quality in generated outputs.
  • Final Accuracy: 0.86
  • Performance: ~149 iterations/second
  • 💬 Output Style: Interactive, human-like, and context-aware explanations.

The final prompt encourages educational, engaging, and concise responses by integrating the reasoning process naturally into the final answer.

About

A prompt tuning experiment using `dspy` to enhance reasoning task responses. It focuses on improving accuracy, clarity, and engagement in answers through structured prompt optimization. Final accuracy reached 0.86 with interactive, human-like outputs and efficient performance (~149 it/s)

Resources

Stars

Watchers

Forks

Releases

No releases published

Packages

 
 
 

Contributors