Blog Post

Educator Developer Blog
26 MIN READ

Fine-Tune and Integrate Custom Phi-3 Models with Prompt Flow: Step-by-Step Guide

Minseok_Song's avatar
Minseok_Song
Iron Contributor
Jul 03, 2024
Fine-Tune and Integrate Custom Phi-3 Models with Prompt Flow: Step-by-Step Guide     [ Spoiler ]   Introduction   Phi-3 is a family of small language models (SLMs) developed by Micros...
Updated Sep 01, 2024
Version 11.0
azure machine learning
Fine-tuning
phi-3
prompt flow
python
Minseok_Song's avatar
Minseok_Song
Iron Contributor
Jul 25, 2024

Update (July 25, 2024): Fine-Tune and Integrate Custom Phi-3 Models with Prompt Flow: Step-by-Step Guide

 

I've updated the tutorial to use the ULTRACHAT_200K dataset and fixed some issues.

 

Updates:

 

  1. Using ULTRACHAT_200K Dataset: Updated to utilize the ULTRACHAT_200K dataset for fine-tuning.
  2. Resolved Package Version Issues: Fixed package version issues in the fine-tuning environment.
  3. GPU Optimization: Modified to use the Standard_NC24ads_A100_v4 GPU for enhanced fine-tuning performance.

 

Confirmation of Changes:

 

As of July 24, 2024, these changes have been verified to ensure that fine-tuning, deployment, and integration with Prompt flow are functioning correctly in the updated GPU environment.