Navigating These Docs

Documentation is organized into two main halves:

Foundational Topics

Core concepts and essential functionality:

• Installation
• Training and Sampling
• Loss Functions
• Saving and Loading
• Async Operations
• Model Lineup

The MinT Cookbook

Practical implementations and examples:

• Supervised Learning
• Reinforcement Learning
• Preferences (DPO/RLHF)
• Evaluations
• LoRA Primer
• Development Tips

The foundational topics provide the building blocks, while the Cookbook offers practical, ready-to-run implementations for common use cases.