[BUG/Help] <title>AttributeError: 'ChatGLMForConditionalGeneration' object has no attribute 'lm_head'

[dongteng]

Is there an existing issue for this?

• [X] I have searched the existing issues

Current Behavior

使用chatglm_Tuning中的方式 将lora与chatglm2相结合 但是报错 AttributeError: 'ChatGLMForConditionalGeneration' object has no attribute 'lm_head' 在chatglm_Tuning中，对lm_head 做出如下处理： model.lm_head = CastOutputToFloat(model.lm_head)

Expected Behavior

No response

Steps To Reproduce

``

Environment

- OS:
- Python:3.8.17 
- Transformers:4.30.2
- PyTorch:
- CUDA Support (`python -c "import torch; print(torch.cuda.is_available())"`) :

Anything else?

No response

[Sun0308]

same problem

[snow00726]

same problem

[fouyhk]

same problem

+1

[fouyhk]

Is there an existing issue for this?

• [x] I have searched the existing issues

Current Behavior

使用chatglm_Tuning中的方式 将lora与chatglm2相结合 但是报错 AttributeError: 'ChatGLMForConditionalGeneration' object has no attribute 'lm_head' 在chatglm_Tuning中，对lm_head 做出如下处理： model.lm_head = CastOutputToFloat(model.lm_head)

Expected Behavior

No response

Steps To Reproduce

``

Environment

- OS:
- Python:3.8.17 
- Transformers:4.30.2
- PyTorch:
- CUDA Support (`python -c "import torch; print(torch.cuda.is_available())"`) :

Anything else?

No response

How to solve this problem

[fouyhk]

same problem

How to solve this problem

[Forever296]

请问现在解决了吗

[jgbrblmd]

同样的报错：AttributeError: 'ChatGLMForConditionalGeneration' object has no attribute 'lm_head'