[BUG] 转换Qwen1.5-14B报错 - Githubissues

QwenLM / Qwen

The official repo of Qwen (通义千问) chat & pretrained large language model proposed by Alibaba Cloud.

Apache License 2.0

12.47k stars 1.01k forks source link

[BUG] 转换Qwen1.5-14B报错 #1256

Closed rickywu closed 1 month ago

rickywu commented 1 month ago

是否已有关于该错误的issue或讨论？ | Is there an existing issue / discussion for this?

[X] 我已经搜索过已有的issues和讨论 | I have searched the existing issues / discussions

该问题是否在FAQ中有解答？ | Is there an existing answer for this in FAQ?

[X] 我已经搜索过FAQ | I have searched FAQ

当前行为 | Current Behavior

使用docker_qwen.sh在升腾910B上转换Qwen1.5-14B-Chat

期望行为 | Expected Behavior

正常转换

复现方法 | Steps To Reproduce

使用docker_qwen.sh在升腾910B上转换Qwen1.5-14B-Chat

运行环境 | Environment

- OS: KylinOS V10
- Python: 3.9
- Transformers: 
- PyTorch: 
- CUDA (`python -c 'import torch; print(torch.version.cuda)'`):

备注 | Anything else?

No response

jklj077 commented 1 month ago

Hi!

This is the repo for Qwen(1.0) model series and the materials provided here DO NOT apply to Qwen1.5. It is expected that it does not work. There is a clear warning at the README.

As to your original question, we advise to refer to mindformers, as they have provided support for certain Qwen1.5 models: https://gitee.com/mindspore/mindformers/blob/dev/research/qwen1_5/qwen1_5.md.