Train Big, Tune Tiny: A Guide to Fine-Tuning LLMs with LoRA