CongcongSong

``` tokenizer = BertTokenizer.from_pretrained("/home/inspur/nas_data/pretrain/Erlangshen-TCBert-330M-Sentence-Embedding-Chinese") #.cuda().eval() # text长度512 model = BertForMaskedLM.from_pretrained("/home/inspur/nas_data/pretrain/Erlangshen-TCBert-330M-Sentence-Embedding-Chinese").cuda().eval() cos = torch.nn.CosineSimilarity(dim=0, eps=1e-8) with torch.no_grad(): # To extract sentence representations for training data training_input = tokenizer("怎样的房子才算户型方正？", return_tensors="pt") print(f"training_input {training_input}")...

当我的transformers是4.36.2时，chatglm2不能正常加载

2

### Is there an existing issue for this? - [X] I have searched the existing issues ### Current Behavior 报错如下： Traceback (most recent call last): File "/home/inspur/scc/gpt/LLaMA-Factory/src/train_bash.py", line 14, in...