Parameter-Efficient Fine-Tuning (PEFT) Techniques for LLMs Training Course

Overview of Parameter-Efficient Fine-Tuning (PEFT)

• Drivers and constraints associated with full fine-tuning.
• Introduction to PEFT: objectives and advantages.
• Industrial applications and real-world use cases.

LoRA (Low-Rank Adaptation)

• Theoretical concepts and intuitive understanding of LoRA.
• Building LoRA implementations with Hugging Face and PyTorch.
• Practical exercise: Fine-tuning a model using LoRA.

Adapter Tuning

• Mechanisms of adapter modules.
• Integrating adapters into transformer-based architectures.
• Practical exercise: Implementing Adapter Tuning on a transformer model.

Prefix Tuning

• Leveraging soft prompts for fine-tuning processes.
• Comparing strengths and limitations against LoRA and adapters.
• Practical exercise: Applying Prefix Tuning to an LLM task.

Assessing and Contrasting PEFT Techniques

• Key metrics for judging performance and efficiency.
• Balancing training speed, memory consumption, and accuracy.
• Conducting benchmark tests and interpreting outcomes.

Rolling Out Fine-Tuned Models

• Procedures for saving and retrieving fine-tuned models.
• Key considerations for deploying PEFT-based models.
• Integration into broader applications and pipelines.

Optimal Practices and Advanced Extensions

• Synergizing PEFT with quantization and distillation.
• Application in resource-constrained and multilingual contexts.
• Emerging trends and current research frontiers.