Hello,
I’m not exactly sure how to structure the training data to fine-tune it on bigcode/the-stack-smol “data/python” dataset.
Does it still need to be formatted as user prompt followed by an assistant’s response?
1 Like