From a870076051f30845a61c79ae0d334bcc7a0eff58 Mon Sep 17 00:00:00 2001 From: huangzhiguo Date: Fri, 15 Dec 2023 14:23:34 +0800 Subject: [PATCH] =?UTF-8?q?=E5=9C=A8model=5Fconfig.py.example=E4=B8=AD?= =?UTF-8?q?=E5=A2=9E=E5=8A=A0qwen=E9=87=8F=E5=8C=96=E6=A8=A1=E5=9E=8B?= =?UTF-8?q?=E5=90=AF=E5=8A=A8=E7=9A=84=E8=AF=B4=E6=98=8E?= MIME-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 8bit --- configs/model_config.py.example | 2 ++ 1 file changed, 2 insertions(+) diff --git a/configs/model_config.py.example b/configs/model_config.py.example index 66ef4e8..87a3c89 100644 --- a/configs/model_config.py.example +++ b/configs/model_config.py.example @@ -179,6 +179,8 @@ MODEL_PATH = { "Qwen-14B": "Qwen/Qwen-14B", "Qwen-14B-Chat": "Qwen/Qwen-14B-Chat", + # 在新版的transformers下需要手动修改模型的config.json文件，在quantization_config字典中 + # 增加`disable_exllama:true` 字段才能启动qwen的量化模型 "Qwen-14B-Chat-Int8": "Qwen/Qwen-14B-Chat-Int8", "Qwen-14B-Chat-Int4": "Qwen/Qwen-14B-Chat-Int4",