[BUG/Help] set_input_embeddings没实现

Is there an existing issue for this?

[X] I have searched the existing issues

Current Behavior

其他项目用chatglm2-6b模型报错， 1689820789987 以前chatglm-6b是可以的，我看了下 chatglm-6b的modeling_chatglm.py文件里有定义set_input_embeddings，但是chatglm2-6b里没这个，是不是这个原因。我手动修改加上就可以了

Expected Behavior

No response

Steps To Reproduce

调用chatglm2-6b模型的set_input_embeddings

Environment

- OS:
- Python: 3.8
- Transformers:
- PyTorch:
- CUDA Support (`python -c "import torch; print(torch.cuda.is_available())"`) :

Anything else?