Fine-Tuning

Model fine-tuning techniques

1.

Full Fine-Tuning WIP

Full parameter fine-tuning strategies for LLMs

2.

LoRA & QLoRA WIP

Parameter-efficient fine-tuning (PEFT) using Low-Rank Adaptation and Quantized LoRA

3.

TRL (Transformer Reinforcement Learning) WIP

Hugging Face library for RLHF, SFT, DPO, and post-training AI agents

4.

Unsloth WIP

High-speed and memory-efficient LLM fine-tuning library