Add an LLM fine-tuning example (#90)

Add an LLM fine-tuning example (#90) #50