[BUG] 我下载了huggingface上的baichuan7b模型，使用 里面的测试程序测试发现CUDA错误

[QIANXUNZDL123]

Required prerequisites

• [X] I have read the documentation https://github.com/baichuan-inc/baichuan-7B/blob/HEAD/README.md.
• [X] I have searched the Issue Tracker and Discussions that this hasn't already been reported. (+1 or comment there if it has.)
• [X] Consider asking first in a Discussion.

System information

from transformers import AutoModelForCausalLM, AutoTokenizer import os os.environ["CUDA_VISIBLE_DEVICES"] = '0,1,2,3'

tokenizer = AutoTokenizer.from_pretrained("./", trust_remote_code=True) model = AutoModelForCausalLM.from_pretrained("./", device_map="auto", trust_remote_code=True) inputs = tokenizer('登鹳雀楼->王之涣\n夜雨寄北->', return_tensors='pt')

inputs = inputs.to('cuda:2')

pred = model.generate(**inputs, max_new_tokens=16,repetition_penalty=1.1) print(tokenizer.decode(pred.cpu()[0], skip_special_tokens=True))

Problem description

Reproducible example code

The Python snippets:

Command lines:

Extra dependencies:

Steps to reproduce:

Traceback

No response

Expected behavior

No response

Additional context

No response

Checklist

• [X] I have provided all relevant and necessary information above.
• [X] I have chosen a suitable title for this issue.

[QIANXUNZDL123]

我找了好几处pad_token_id 查看都是0 但是就是报错，有没有大佬帮忙解决下