ProjectD-AI / llama_inference

llama inference for tencentpretrain
GNU General Public License v3.0
95 stars 11 forks source link

生成乱码 #13

Open McCarrtney opened 1 year ago

McCarrtney commented 1 year ago

模型生成的文件中既有乱码也有正常文本

image
McCarrtney commented 1 year ago

我用的模型是基础模型Chinese_LLaMA_7B

fengyh3 commented 1 year ago

Chinese_LLaMA_7B是没有指令微调的模型,尝试用https://huggingface.co/Linly-AI/ChatFlow-7B这个

McCarrtney commented 1 year ago

这个会有影响么?我十几条一起生成就会有乱码,但一条一条单独生成就不会

fengyh3 commented 1 year ago

你先尝试换个模型试试哈