- 18 4月, 2023 5 次提交
-
-
imClumsyPanda authored
-
imClumsyPanda authored
-
imClumsyPanda authored
-
1. 参考ChatGLM-6B代码实现模型多卡部署
imClumsyPanda authored -
Liao Zhenyu authored
-
- 17 4月, 2023 2 次提交
-
-
fix: typo
imClumsyPanda authored -
fix a typo in README.md line 180.
Tara authored
-
- 16 4月, 2023 6 次提交
-
-
imClumsyPanda authored
-
imClumsyPanda authored
-
imClumsyPanda authored
-
imClumsyPanda authored
-
imClumsyPanda authored
-
imClumsyPanda authored
-
- 15 4月, 2023 5 次提交
-
-
imClumsyPanda authored
-
imClumsyPanda authored
-
imClumsyPanda authored
-
imClumsyPanda authored
-
imClumsyPanda authored
-
- 14 4月, 2023 22 次提交
-
-
imClumsyPanda authored
-
imClumsyPanda authored
-
Calcitem authored
-
imClumsyPanda authored
-
imClumsyPanda authored
-
imClumsyPanda authored
-
imClumsyPanda authored
-
修改入群码为二群二维码
fengyunzaidushi authored -
fengyu authored
-
chatglm_with_shared_memory_openai_llm:agent
glide-the authored -
…档调用链的示例,可以使用chatgml处理一部分任务事件,也可以使用chatgml来完成记忆共享(SharedMemory),至于为什么使用openai原因主要有两个现象,第一点,因为大模型chatgml受限,导致很多langchain的demo只能部分实现,跟群主聊了下,这种问题只能通过微调模型,增加一部分指令集和状态术语的数据集进去,微调方案只能大家自己尝试了,对显卡要去较高.当然也可以考虑只训练一个任务格式化的gml,模型微调参考这个仓库的方式 https://github.com/liucongg/ChatGLM-Finetuning
glide-the authored -
Q7: 本项目用到的模型权重文件百度网盘地址:
fengyunzaidushi authored -
fengyu authored
-
fengyu authored
-
fengyu authored
-
fengyu authored
-
Calcitem authored
-
fengyu authored
-
fengyu authored
-
fengyu authored
-
imClumsyPanda authored
-
imClumsyPanda authored
-