Update ReadMe (#123)

Co-authored-by: 同润 <[email protected]>

README.md

@@ -37,7 +37,8 @@ $ git clone --recurse-submodules https://github.com/alibaba/Pai-Megatron-Patch.g
 ```
 
 ## Technical Reports
-- [Mixtral-8x7B在PAI灵骏的最佳实践](https://help.aliyun.com/zh/pai/use-cases/train-fine-tune-and-deploy-mixtral-by-using-intelligent-computing-lingjun)
+- [基于Megatron-Core的稀疏大模型训练工具：阿里云MoE大模型最佳实践](https://mp.weixin.qq.com/s/DkrWEEJ7IxirwWd3qB9Bng)
+- [Mixtral-8x7B在PAI灵骏的训练指南](https://help.aliyun.com/zh/pai/use-cases/train-fine-tune-and-deploy-mixtral-by-using-intelligent-computing-lingjun)
 - [通义千问开源模型在PAI灵骏的最佳实践](https://mp.weixin.qq.com/s?__biz=Mzg4MzgxNDk2OA==&mid=2247491796&idx=1&sn=dc1d719313d794ae1aacdb07669a9545&chksm=cf430783f8348e950218bfcff861a2e6d2d92705807bf5b04f6e9268cc510ffa6e6aa2c87327#rd)
 - [阿里云机器学习PAI开源AI大模型训练工具Pai-Megatron-Patch, 助力大模型技术落地](https://zhuanlan.zhihu.com/p/655942437)
 - [基于单机最高能效270亿参数GPT模型的文本生成与理解](https://zhuanlan.zhihu.com/p/597652820)

README_zh-CN.md

@@ -12,7 +12,8 @@ Pai-Megatron-Patch是各类开源大模型和Megatron训练加速引擎之间的
 
 
 ## Technical Reports
-- [Mixtral-8x7B在PAI灵骏的最佳实践](https://help.aliyun.com/zh/pai/use-cases/train-fine-tune-and-deploy-mixtral-by-using-intelligent-computing-lingjun)
+- [基于Megatron-Core的稀疏大模型训练工具：阿里云MoE大模型最佳实践](https://mp.weixin.qq.com/s/DkrWEEJ7IxirwWd3qB9Bng)
+- [Mixtral-8x7B在PAI灵骏的训练指南](https://help.aliyun.com/zh/pai/use-cases/train-fine-tune-and-deploy-mixtral-by-using-intelligent-computing-lingjun)
 - [通义千问开源模型在PAI灵骏的最佳实践](https://mp.weixin.qq.com/s?__biz=Mzg4MzgxNDk2OA==&mid=2247491796&idx=1&sn=dc1d719313d794ae1aacdb07669a9545&chksm=cf430783f8348e950218bfcff861a2e6d2d92705807bf5b04f6e9268cc510ffa6e6aa2c87327#rd)
 - [阿里云机器学习PAI开源AI大模型训练工具Pai-Megatron-Patch, 助力大模型技术落地](https://zhuanlan.zhihu.com/p/655942437)
 - [基于单机最高能效270亿参数GPT模型的文本生成与理解](https://zhuanlan.zhihu.com/p/597652820)