How can I fine-tune using data in these formats

[leegohi04517]

How should I post-process these data for fine-tuning? I realized that the prompt format for Pygmalion is as follows: [CHARACTER]'s Persona: [A few sentences about the character you want the model to play]

[DIALOGUE HISTORY] You: [Your input message here]
[CHARACTER]:
Do I just need to convert the training input text into this format? Specifically,
1. Treat the persona “ description” as a separate section
2. Use to separate
3. After replacing the person in conversation, convert it into dialog format
I'm not sure what to use as input_ids and what to use as labels for fine-tuning.