diff --git a/_typos.toml b/_typos.toml
index 264515280ef..7fcad764a04 100644
--- a/_typos.toml
+++ b/_typos.toml
@@ -60,13 +60,7 @@ cantains = "cantains"
 classfy = "classfy"
 cliping = "cliping"
 colunms = "colunms"
-commmit = "commmit"
-complie = "complie"
-condtional = "condtional"
-conjuction = "conjuction"
 containg = "containg"
-contruct = "contruct"
-contructed = "contructed"
 contruction = "contruction"
 contxt = "contxt"
 convertion = "convertion"
diff --git a/docs/design/concurrent/go_op.md b/docs/design/concurrent/go_op.md
index 286db8f86e3..1248829f560 100644
--- a/docs/design/concurrent/go_op.md
+++ b/docs/design/concurrent/go_op.md
@@ -3,7 +3,7 @@
 ## Introduction
 
 The **go_op** allows user's of PaddlePaddle to run program blocks on a detached
-thread.  It works in conjuction with CSP operators (channel_send,
+thread.  It works in conjunction with CSP operators (channel_send,
 channel_receive, channel_open, channel_close, and select) to allow users to
 concurrently process data and communicate easily between different threads.
 
diff --git a/docs/design/memory/memory_optimization.md b/docs/design/memory/memory_optimization.md
index 8a3e3a4ae50..fee0f493b6f 100644
--- a/docs/design/memory/memory_optimization.md
+++ b/docs/design/memory/memory_optimization.md
@@ -99,7 +99,7 @@ At last, we take basic strategy and liveness analysis techniques learning from c
 In-place is a built-in attribute of an operator. Since we treat in-place and other operators differently, we have to add an in-place attribute for every operator.
 
 
-#### contruct control flow graph
+#### construct control flow graph
 
 Following is the ProgramDesc protobuf of [machine translation](https://github.com/PaddlePaddle/Paddle/blob/develop/python/paddle/fluid/tests/book/test_machine_translation.py) example.
 
diff --git a/docs/design/others/gan_api.md b/docs/design/others/gan_api.md
index f46b9634d71..4aabf6ebe1b 100644
--- a/docs/design/others/gan_api.md
+++ b/docs/design/others/gan_api.md
@@ -42,7 +42,7 @@ build the whole GAN model, define training loss for both generator and discrimat
 To be more detailed, we introduce our design of DCGAN as following:
 
 ### Class member Function: Initializer
-- Set up hyper-parameters, including condtional dimension, noise dimension, batch size and so forth.
+- Set up hyper-parameters, including conditional dimension, noise dimension, batch size and so forth.
 - Declare and define all the model variables. All the discriminator parameters are included in the list self.theta_D and all the generator parameters are included in the list self.theta_G.
 ```python
 class DCGAN:
diff --git a/docs/dev_guides/git_guides/local_dev_guide_cn.md b/docs/dev_guides/git_guides/local_dev_guide_cn.md
index 5a32cbba40a..747fb59421f 100644
--- a/docs/dev_guides/git_guides/local_dev_guide_cn.md
+++ b/docs/dev_guides/git_guides/local_dev_guide_cn.md
@@ -116,7 +116,7 @@ clang-format.......................................(no files to check)Skipped
  create mode 100644 233
 ```
 
-可以看到，在执行`git commit`后，输出了一些额外的信息。这是使用`pre-commmit`进行代码风格检查的结果，关于代码风格检查的使用问题请参考[代码风格检查指南](./codestyle_check_guide_cn.html)。
+可以看到，在执行`git commit`后，输出了一些额外的信息。这是使用`pre-commit`进行代码风格检查的结果，关于代码风格检查的使用问题请参考[代码风格检查指南](./codestyle_check_guide_cn.html)。
 
 ## 保持本地仓库最新
 
diff --git a/docs/dev_guides/sugon/complie_and_test_cn.md b/docs/dev_guides/sugon/compile_and_test_cn.md
similarity index 100%
rename from docs/dev_guides/sugon/complie_and_test_cn.md
rename to docs/dev_guides/sugon/compile_and_test_cn.md
diff --git a/docs/dev_guides/sugon/index_cn.rst b/docs/dev_guides/sugon/index_cn.rst
index 84d57649217..4dc16f73b41 100644
--- a/docs/dev_guides/sugon/index_cn.rst
+++ b/docs/dev_guides/sugon/index_cn.rst
@@ -4,7 +4,7 @@
 
 以下将说明 Paddle 适配曙光相关的开发指南：
 
-- `曙光智算平台-Paddle 源码编译和单测执行 <./complie_and_test_cn.html>`_ : 如何曙光曙光智算平台编译 Paddle 源码编译并执行单测。
+- `曙光智算平台-Paddle 源码编译和单测执行 <./compile_and_test_cn.html>`_ : 如何曙光曙光智算平台编译 Paddle 源码编译并执行单测。
 - `Paddle 适配 C86 加速卡详解 <./paddle_c86_cn.html>`_ : 详解 Paddle 适配 C86 加速卡。
 - `Paddle 框架下 ROCm(HIP)算子单测修复指导 <./paddle_c86_fix_guides_cn.html>`_ : 指导 Paddle 框架下 ROCm(HIP)算子单测修复。
 
@@ -12,6 +12,6 @@
 ..  toctree::
     :hidden:
 
-    complie_and_test_cn.md
+    compile_and_test_cn.md
     paddle_c86_cn.md
     paddle_c86_fix_guides_cn.md
diff --git a/docs/guides/advanced/layer_and_model_en.md b/docs/guides/advanced/layer_and_model_en.md
index 256b286c7ae..35f2214a6e3 100644
--- a/docs/guides/advanced/layer_and_model_en.md
+++ b/docs/guides/advanced/layer_and_model_en.md
@@ -27,7 +27,7 @@ class Model(paddle.nn.Layer):
         return y
 ```
 
-Here we contructed a ``Model`` which inherited from ``paddle.nn.Layer``. This model only holds a single layer of ``paddle.nn.Flatten``, which flattens the input variables **inputs** upon execution.
+Here we constructed a ``Model`` which inherited from ``paddle.nn.Layer``. This model only holds a single layer of ``paddle.nn.Flatten``, which flattens the input variables **inputs** upon execution.
 
 ## Sublayers