lm-evaluation-harness无法测试

[jeremyyx]

Is there an existing issue for this?

• [X] I have searched the existing issues

Current Behavior

lm-evaluation-harness/lm_eval/models/huggingface.py", line 340, in _create_auto_tokenizer tokenizer.pad_token = tokenizer.eos_token AttributeError: can't set attribute

Expected Behavior

No response

Steps To Reproduce

python main.py --model hf-causal-experimental --model_args pretrained=/nvme/syx/chatglm2-6b,use_accelerate=True --tasks lambada_standard --batch_size 4 --num_fewshot 0 --no_cache --write_out --output_path sf_results/hendrycksTest.json --device auto

Environment

- OS:Ubuntu
- Python:3.9.16
- Transformers:4.26.1
- PyTorch:1.12
- CUDA Support (`python -c "import torch; print(torch.cuda.is_available())"`) :True

Anything else?

No response

[LDH007]

same issue

[changwangss]

same issue

[shilei31415]

same issue