Fine-tuning LLM with LoRA (Low-Rank Adaptation) from scratch (Oct 2023) - View it on GitHub