About Fine-Tuning

A good prompt along with some context examples can get you far, but sometimes you need to fine-tune a model to get the best results. Fine-tuning is the process of training a base model on a new dataset to adapt it to a specific task.

Advantages of Fine-Tuning #

There are several advantages to fine-tuning a model, including:

• Improved performance: Fine-tuned models perform better on specific tasks than base models, even when they are fed great prompts and context examples.
• Reduced cost: Fine-tuned models require less input data and examples (tokens) to perform, which can save time and money.
• Easier to Use: Fine-tuned models are easier to use and require less input from the user to generate accurate results.

GenAI Studio offers fine-tuning on a curated list of foundation models.

Model Creation Journey #

Fine-tuning should happen after you have created a project and tested various combinations of generation properties and prompt inputs against your chosen uploaded dataset.

graph LR;
    A(Create Project) --> B(Import Data);
    B --> C(Snapshot Model);
    C --> D(Fine-Tune Model);
    D --> E(Export Model);
    style A fill:#fff,stroke:#333,stroke-width:2px;
    style B fill:#fff,stroke:#333,stroke-width:2px;
    style C fill:#fff,stroke:#333,stroke-width:2px;
    style D fill:#7FF9E2,stroke:#333,stroke-width:4px;
    style E fill:#fff,stroke:#333,stroke-width:2px;
  

Examples #

To learn how to fine-tune a model using an example, visit the Medical Transcripts Classification tutorial or the Customer Complaints Classification tutorial.