Fine-tune model Meta Llama2-13b

Hello everyone, I’m newbie. I want to fine-tune model Meta-llama2-13b. I’m reading a tutorial using huggingface_hub, autotrain-advanced. I have two things I want to ask, firstly, when using access tokens, do I need to use a paid account? Secondly, how do I add more data to my data? We look forward to receiving answers from you.