Creating Your LLM Preference Tuning Dataset

• In addition to instruction tuning, QpiAI Pro also supports preference tuning datasets. Preference tuning helps fine-tune models to choose better responses by learning from comparisons between good and bad answers.
• Kindly follow the same process as outlined earlier for creating and configuring the dataset. The steps—accessing the data section, naming, metadata configuration, and uploading—remain the same. The only change is in the CSV format and task selection.