• Velvet
  • Posts
  • Prepare a fine-tuning dataset for gpt-4o-mini

Prepare a fine-tuning dataset for gpt-4o-mini

Use Velvet to identify and export your dataset.

July 26, 2024

Open AI released docs this week for fine-tuning gpt-4o-mini, free through September. Fine-tuning can result in better, faster, and cheaper outputs.

To fine-tune a model, you'll provide training data with more examples than you could include in a prompt. This means the model itself will perform differently - rather than needing to pass through examples with every API call.

Use Velvet to export a JSONL file that can be used for fine-tuning.

What we’ll cover in this article

  • Fine-tuning and when to invest

  • 5-step fine-tuning process with Open AI

  • Prepare your fine-tuning dataset with Velvet

More from Velvet

Read about why Find AI, an AI-powered B2B search app, uses Velvet to warehouse millions of LLM requests per week.

We’re hiring a founding engineer! Seeking entrepreneurial full-stack engineers. Our system is built on Next.js, Typescript, Postgres, and OpenAI.

Warehouse OpenAI logs to PostgreSQL | Read the docs →