关于微调分词器

<img width="1482" height="620" alt="Image" src="https://GitHub.com/user-attachments/assets/19fad51c-60e2-4429-9131-379f03761d62" /> <img width="1517" height="716" alt="Image" src="https://GitHub.com/user-attachments/assets/858455bc-6454-4793-b3bd-58dd715a9f1f" /> My environment is an Intel CPU with integrated graphics card, and it does not support the CUDA ecosystem of NVIDIA GPUs. I have used the following command: 1. torchrun --standalone --nproc_per_node=1 finetune/train_tokenizer.py. The error shown in the image above occurred. 2. set USE_LIBUV=0 && torchrun --standalone --nproc_per_node=1 finetune/train_tokenizer.py. Running the second command also resulted in the same error as shown in the image. Can you please tell me how to solve this problem? Thank you.