ymcui / Chinese-LLaMA-Alpaca-2

中文LLaMA-2 & Alpaca-2大模型二期项目 + 64K超长上下文模型 (Chinese LLaMA-2 & Alpaca-2 LLMs with 64K long context models)
Apache License 2.0
7.09k stars 578 forks source link

使用llama.cpp 在win11电脑上运行量化后的模型中文问答出现乱码 #331

Closed FWorldCodeZ closed 1 year ago

FWorldCodeZ commented 1 year ago

提交前必须检查以下项目

问题类型

效果问题

基础模型

Chinese-LLaMA-2-16K (7B/13B)

操作系统

Windows

详细描述问题

# 执行命令
cmake .. -DLLAMA_CUBLAS=ON

 cmake --build . --config Release

 python  convert.py E:\modles\chinese-llama-2-13b-16k-hf

 .\build\bin\Release\quantize.exe E:\modles\chinese-llama-2-13b-16k-hf\ggml-model-f16.gguf E:\modles\chinese-llama-2-13b-16k-hf\ggml-model-q8_0.gguf q8_0

 .\chat.sh E:\modles\chinese-llama-2-13b-16k-hf\ggml-model-q8_0.gguf

依赖情况(代码类问题务必提供)

#设备
 NVIDIA GeForce RTX 3070 Laptop GPU,AMD Ryzen 7 5800H with Radeon Graphics            3.20 GHz ,63g内存

image

运行日志或截图

#部分说明
运行其他几llama、alpaca也是乱码
cmake --build . --config Release 出现如下黄色警告

运行./chat.sh 出现的日志和回复出现乱码
Log start
main: build = 1330 (48edda3)
main: built with MSVC 19.33.31629.0 for x64
main: seed  = 1696861582
ggml_init_cublas: found 1 CUDA devices:
  Device 0: NVIDIA GeForce RTX 3070 Laptop GPU, compute capability 8.6
llama_model_loader: loaded meta data with 19 key-value pairs and 363 tensors from E:\modles\chinese-llama-2-13b-16k-hf\ggml-model-q8_0.gguf (version GGUF V2 (latest))

黄色警告

image

乱码(几个中文模型量化皆出现乱码,不能识别中文,)

image

ymcui commented 1 year ago

重新编辑一下内容,里面的几张图片看不到(图片不要放到代码块里)

FWorldCodeZ commented 1 year ago

重新编辑一下内容,里面的几张图片看不到(图片不要放到代码块里)

你好,已经重新编辑了,麻烦看看。gpu cpu 运行都是差不多

github-actions[bot] commented 1 year ago

This issue has been automatically marked as stale because it has not had recent activity. It will be closed if no further activity occurs. Thank you for your consideration.

github-actions[bot] commented 1 year ago

Closing the issue, since no updates observed. Feel free to re-open if you need any further assistance.