[doc] Fix typo under colossalai and doc(#3618)

* Fixed several spelling errors under colossalai

* Fix the spelling error in colossalai and docs directory

* Cautious Changed the spelling error under the example folder

* Update runtime_preparation_pass.py

revert autograft to autograd

* Update search_chunk.py

utile to until

* Update check_installation.py

change misteach to mismatch in line 91

* Update 1D_tensor_parallel.md

revert to perceptron

* Update 2D_tensor_parallel.md

revert to perceptron in line 73

* Update 2p5D_tensor_parallel.md

revert to perceptron in line 71

* Update 3D_tensor_parallel.md

revert to perceptron in line 80

* Update README.md

revert to resnet in line 42

* Update reorder_graph.py

revert to indice in line 7

* Update p2p.py

revert to megatron in line 94

* Update initialize.py

revert to torchrun in line 198

* Update routers.py

change to detailed in line 63

* Update routers.py

change to detailed in line 146

* Update README.md

revert  random number in line 402

colossalai/zero/gemini/colo_init_context.py

@@ -74,7 +74,7 @@ class ColoInitContext(InsertPostInitMethodToModuleSubClasses):
         """
         Args:
             device (torch.device): the device where parameters initialized are resident. Defaults to torch.device('cpu').
-            dtype (torch.dtype): the dtype of parameters initialized. Defults to torch.float.
+            dtype (torch.dtype): the dtype of parameters initialized. Defaults to torch.float.
             default_pg (ProcessGroup): the default process group for all initialized parameters.
             default_dist_spec: the default distributed specifications.
         """
@@ -164,7 +164,7 @@ def post_process_colo_init_ctx(model: torch.nn.Module,
         model (torch.nn.module): the model
         device (torch.device, optional): device type of the model params. Defaults to torch.device('cpu').
         dtype (torch.dtype, optional): dtype of the model params. Defaults to torch.float.
-        default_pg (Optional[ProcessGroup], optional): default process group. Defaults to None. Inidicates a DP-only process group.
+        default_pg (Optional[ProcessGroup], optional): default process group. Defaults to None. Indicates a DP-only process group.
         default_dist_spec (Any, optional): default dist spec of params. Defaults to None.
 
     Raises:

colossalai/zero/gemini/gemini_ddp.py

@@ -42,7 +42,7 @@ class ZeroDDP(ColoDDP):
 
     Args:
         module (torch.nn.Module): Module to apply ZeRO-DP.
-        gemini_manager (GeminiManager): Manages the chunk manager and heterogeneous momery space.
+        gemini_manager (GeminiManager): Manages the chunk manager and heterogeneous memory space.
             For more details, see the API reference of ``GeminiManager``.
         pin_memory (bool): Chunks on CPU Memory use pin-memory.
         force_outputs_fp32 (bool): If set to True, outputs will be fp32. Otherwise, outputs will be fp16.
@@ -684,7 +684,7 @@ class GeminiDDP(ZeroDDP):
                  memstats: Optional[MemStats] = None,
                  verbose: bool = False) -> None:
         """
-        A torch.Module warpper using ZeRO-DP and Genimi.
+        A torch.Module wrapper using ZeRO-DP and Gemini.
         ZeRO is for parallel. Gemini is for memory management.
         WARNING: The class will modify the module inline!
 
@@ -706,7 +706,7 @@ class GeminiDDP(ZeroDDP):
                 Users can provide this argument to speed up searching.
                 If users do not know this argument before training, it is ok. We will use a default value 1024.
             min_chunk_size_mb (float, optional): the minimum chunk size in MegaByte.
-                If the aggregate size of parameters is still samller than the minimum chunk size,
+                If the aggregate size of parameters is still smaller than the minimum chunk size,
                 all parameters will be compacted into one small chunk.
             memstats (MemStats, optional) the memory statistics collector by a runtime memory tracer.
         """