qwen2.5 lora微调报错

[qihouji]

运行官方demo，05-Qwen2.5-7B-Instruct Lora .ipynb在notebook的最后：
with torch.no_grad():
outputs = model.generate(**inputs, **gen_kwargs)
报错
C:\cb\pytorch_1000000000000\work\aten\src\ATen\native\cuda\TensorCompare.cu:110: block: [0,0,0], thread: [0,0,0] Assertion input[0] != 0 failed.
Traceback (most recent call last):
File "P:\envs\anaconda2024\envs\py312transformers\Lib\site-packages\IPython\core\interactiveshell.py", line 2932, in safe_execfile
py3compat.execfile(
File "P:\envs\anaconda2024\envs\py312transformers\Lib\site-packages\IPython\utils\py3compat.py", line 55, in execfile
exec(compiler(f.read(), fname, "exec"), glob, loc)
File "P:\prjs\learn_transformers\tmp.py", line 28, in
outputs = model.generate(**inputs, **gen_kwargs)
^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
File "P:\envs\anaconda2024\envs\py312transformers\Lib\site-packages\peft\peft_model.py", line 1704, in generate
outputs = self.base_model.generate(*args, **kwargs)
^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
File "P:\envs\anaconda2024\envs\py312transformers\Lib\site-packages\torch\utils_contextlib.py", line 116, in decorate_context
return func(*args, **kwargs)
^^^^^^^^^^^^^^^^^^^^^
File "P:\envs\anaconda2024\envs\py312transformers\Lib\site-packages\transformers\generation\utils.py", line 1989, in generate
result = self._sample(
^^^^^^^^^^^^^
File "P:\envs\anaconda2024\envs\py312transformers\Lib\site-packages\transformers\generation\utils.py", line 2969, in _sample
next_tokens = torch.multinomial(probs, num_samples=1).squeeze(1)
^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
RuntimeError: CUDA error: device-side assert triggered
CUDA kernel errors might be asynchronously reported at some other API call, so the stacktrace below might be incorrect.
For debugging consider passing CUDA_LAUNCH_BLOCKING=1
Compile with TORCH_USE_CUDA_DSA to enable device-side assertions.
想请问下这个有没有可能的解决方案呀？
环境：pytorch2.5.0，transformers 4.43.3

[qihouji]

定位到问题是 runtimeerror: probability tensor contains either inf, nan or element < 0

[want-well]

方便问一下微调数据集格式吗？