dataset = load_dataset("./Chinese-DeepSeek-R1-Distill-data-110k", split="train")
dataset = dataset.select_columns(["input", "content", "reasoning_content"])
new_data = {
"input": dataset["input"],
"instruction": ["深度思索"] * len(dataset),
"output": ['
' + rc + '' + c forrc, c in zip(dataset["reasoning_content"], dataset["content"])]
}
new_dataset =… See the full description on the dataset page:
https://huggingface.co/datasets/aifeifei798/Chinese-DeepSeek-R1-Distill-data-110k-alpaca.
