[hotfix] Add layer norm gradients all-reduce for sequence parallel (#4926)

* [hotfix] Add layer norm gradients all-reduce for sequence parallel. (#4915)

* Add layer norm gradients all-reduce for sequence parallel.

* skip pipeline inference test

* [hotfix] fixing polices of sequence parallel (#4922)

* Add layer norm gradients all-reduce for sequence parallel.

* fix parameter passing when calling get_autopolicy

---------

Co-authored-by: littsk <1214689160@qq.com>

* Hotfix/add grad all reduce for sequence parallel (#4927)

* Add layer norm gradients all-reduce for sequence parallel.


* fix parameter passing when calling get_autopolicy

* fix bug using wrong variables

---------

Co-authored-by: littsk <1214689160@qq.com>

* fix policy initialization

* fix bloom and chatglm policices

* polish code of handling layernorm

* fix moe module

* polish code of class initializing

---------

Co-authored-by: Zhongkai Zhao <kanezz620@gmail.com>

colossalai/shardformer/layer/__init__.py

@@ -2,7 +2,7 @@ from .dropout import DropoutForParallelInput, DropoutForReplicatedInput
 from .embedding import Embedding1D, VocabParallelEmbedding1D
 from .linear import Linear1D_Col, Linear1D_Row
 from .loss import cross_entropy_1d
-from .normalization import FusedLayerNorm, FusedRMSNorm
+from .normalization import FusedLayerNorm, FusedRMSNorm, LayerNorm, RMSNorm
 from .parallel_module import ParallelModule
 from .qkv_fused_linear import FusedLinear1D_Col, GPT2FusedLinearConv1D_Col, GPT2FusedLinearConv1D_Row
 
@@ -16,6 +16,9 @@ __all__ = [
     "DropoutForParallelInput",
     "DropoutForReplicatedInput",
     "cross_entropy_1d",
+    "BaseLayerNorm",
+    "LayerNorm",
+    "RMSNorm",
     "FusedLayerNorm",
     "FusedRMSNorm",
     "FusedLinear1D_Col",