diff --git a/nemo/collections/multimodal/models/vision_language_foundation/clip/megatron_clip_models.py b/nemo/collections/multimodal/models/vision_language_foundation/clip/megatron_clip_models.py
index b2d6a953a9ab..52de2aaead19 100644
--- a/nemo/collections/multimodal/models/vision_language_foundation/clip/megatron_clip_models.py
+++ b/nemo/collections/multimodal/models/vision_language_foundation/clip/megatron_clip_models.py
@@ -63,11 +63,6 @@
     from megatron.core import parallel_state
     from megatron.core.distributed import DistributedDataParallel as McoreDDP
     from megatron.core.distributed import DistributedDataParallelConfig
-    from megatron.core.fusions.fused_bias_dropout import get_bias_dropout_add
-    from megatron.core.models.gpt import GPTModel as MCoreGPTModel
-    from megatron.core.models.vision.clip_vit_model import CLIPViTModel
-    from megatron.core.pipeline_parallel.schedules import get_forward_backward_func
-    from megatron.core.transformer.attention import CrossAttention, CrossAttentionSubmodules
     from megatron.core.extensions.transformer_engine import (
         TEColumnParallelLinear,
         TEDotProductAttention,
@@ -75,6 +70,11 @@
         TENorm,
         TERowParallelLinear,
     )
+    from megatron.core.fusions.fused_bias_dropout import get_bias_dropout_add
+    from megatron.core.models.gpt import GPTModel as MCoreGPTModel
+    from megatron.core.models.vision.clip_vit_model import CLIPViTModel
+    from megatron.core.pipeline_parallel.schedules import get_forward_backward_func
+    from megatron.core.transformer.attention import CrossAttention, CrossAttentionSubmodules
     from megatron.core.transformer.enums import AttnMaskType as MCoreAttnMaskType
     from megatron.core.transformer.identity_op import IdentityOp
     from megatron.core.transformer.mlp import MLP, MLPSubmodules
diff --git a/nemo/collections/nlp/models/language_modeling/megatron/bert/bert_spec.py b/nemo/collections/nlp/models/language_modeling/megatron/bert/bert_spec.py
index 48fdffb057e5..05f371466614 100644
--- a/nemo/collections/nlp/models/language_modeling/megatron/bert/bert_spec.py
+++ b/nemo/collections/nlp/models/language_modeling/megatron/bert/bert_spec.py
@@ -14,16 +14,16 @@
 
 
 try:
-    from megatron.core.fusions.fused_bias_dropout import get_bias_dropout_add
-    from megatron.core.fusions.fused_layer_norm import FusedLayerNorm
-    from megatron.core.tensor_parallel.layers import ColumnParallelLinear, RowParallelLinear
-    from megatron.core.transformer.attention import SelfAttention, SelfAttentionSubmodules
     from megatron.core.extensions.transformer_engine import (
         TEColumnParallelLinear,
         TEDotProductAttention,
         TENorm,
         TERowParallelLinear,
     )
+    from megatron.core.fusions.fused_bias_dropout import get_bias_dropout_add
+    from megatron.core.fusions.fused_layer_norm import FusedLayerNorm
+    from megatron.core.tensor_parallel.layers import ColumnParallelLinear, RowParallelLinear
+    from megatron.core.transformer.attention import SelfAttention, SelfAttentionSubmodules
     from megatron.core.transformer.dot_product_attention import DotProductAttention
     from megatron.core.transformer.enums import AttnMaskType
     from megatron.core.transformer.identity_op import IdentityOp
@@ -59,7 +59,11 @@
         self_attn_bda=get_bias_dropout_add,
         post_att_layernorm=TENorm,
         mlp=ModuleSpec(
-            module=MLP, submodules=MLPSubmodules(linear_fc1=TEColumnParallelLinear, linear_fc2=TERowParallelLinear,),
+            module=MLP,
+            submodules=MLPSubmodules(
+                linear_fc1=TEColumnParallelLinear,
+                linear_fc2=TERowParallelLinear,
+            ),
         ),
         mlp_bda=get_bias_dropout_add,
         post_mlp_layernorm=TENorm,
@@ -84,7 +88,11 @@
         self_attn_bda=get_bias_dropout_add,
         post_att_layernorm=FusedLayerNorm,
         mlp=ModuleSpec(
-            module=MLP, submodules=MLPSubmodules(linear_fc1=ColumnParallelLinear, linear_fc2=RowParallelLinear,),
+            module=MLP,
+            submodules=MLPSubmodules(
+                linear_fc1=ColumnParallelLinear,
+                linear_fc2=RowParallelLinear,
+            ),
         ),
         mlp_bda=get_bias_dropout_add,
         post_mlp_layernorm=FusedLayerNorm,
diff --git a/nemo/collections/nlp/models/language_modeling/megatron/falcon/falcon_spec.py b/nemo/collections/nlp/models/language_modeling/megatron/falcon/falcon_spec.py
index 4539c0e27d25..441ecae47aab 100644
--- a/nemo/collections/nlp/models/language_modeling/megatron/falcon/falcon_spec.py
+++ b/nemo/collections/nlp/models/language_modeling/megatron/falcon/falcon_spec.py
@@ -15,14 +15,14 @@
 from nemo.collections.nlp.modules.common.megatron.utils import ApexGuardDefaults
 
 try:
-    from megatron.core.fusions.fused_bias_dropout import get_bias_dropout_add
-    from megatron.core.transformer.attention import SelfAttention, SelfAttentionSubmodules
     from megatron.core.extensions.transformer_engine import (
         TEColumnParallelLinear,
         TEDotProductAttention,
         TENorm,
         TERowParallelLinear,
     )
+    from megatron.core.fusions.fused_bias_dropout import get_bias_dropout_add
+    from megatron.core.transformer.attention import SelfAttention, SelfAttentionSubmodules
     from megatron.core.transformer.enums import AttnMaskType
     from megatron.core.transformer.identity_op import IdentityOp
     from megatron.core.transformer.mlp import MLP, MLPSubmodules
@@ -62,7 +62,11 @@ def get_falcon_layer_spec() -> ModuleSpec:
         self_attn_bda=get_bias_dropout_add,
         pre_mlp_layernorm=TENorm,
         mlp=ModuleSpec(
-            module=MLP, submodules=MLPSubmodules(linear_fc1=TEColumnParallelLinear, linear_fc2=TERowParallelLinear,),
+            module=MLP,
+            submodules=MLPSubmodules(
+                linear_fc1=TEColumnParallelLinear,
+                linear_fc2=TERowParallelLinear,
+            ),
         ),
         mlp_bda=get_bias_dropout_add,
     )
diff --git a/nemo/collections/nlp/models/language_modeling/megatron/gpt_full_te_layer_autocast_spec.py b/nemo/collections/nlp/models/language_modeling/megatron/gpt_full_te_layer_autocast_spec.py
index 7459b9d1f95f..6aa05f15eece 100644
--- a/nemo/collections/nlp/models/language_modeling/megatron/gpt_full_te_layer_autocast_spec.py
+++ b/nemo/collections/nlp/models/language_modeling/megatron/gpt_full_te_layer_autocast_spec.py
@@ -36,11 +36,11 @@
 try:
     from megatron.core import parallel_state, tensor_parallel
     from megatron.core.fusions.fused_layer_norm import FusedLayerNorm
+    from megatron.core.transformer.graphs import CudaGraphManager
     from megatron.core.transformer.spec_utils import ModuleSpec
     from megatron.core.transformer.transformer_block import TransformerBlockSubmodules, get_num_layers_to_build
     from megatron.core.transformer.transformer_layer import BaseTransformerLayer
     from megatron.core.transformer.utils import make_sharded_tensors_for_checkpoint
-    from megatron.core.transformer.graphs import CudaGraphManager
 
     HAVE_MEGATRON_CORE = True
 
@@ -241,9 +241,7 @@ def __init__(self, config, layer_number=1, hidden_dropout=None):
         super().__init__(**transformer_layer_args)
 
         if config.enable_cuda_graph and self.training:
-            assert (
-                not config.cpu_offloading and config.recompute_granularity is None
-            ), "Cudagraphs not supported"
+            assert not config.cpu_offloading and config.recompute_granularity is None, "Cudagraphs not supported"
             self.add_module('cudagraph_manager', CudaGraphManager())
 
     # Called by MCore's TransformerBlock.forward
@@ -334,6 +332,7 @@ def __call__(self, *args, **kwargs):
             return self.cudagraph_manager(self, args, kwargs)
         return super().__call__(*args, **kwargs)
 
+
 # Use this spec to use the full Transformer layer from Transformer Engine
 def get_gpt_full_te_layer_autocast_spec(transformer_config) -> ModuleSpec:
     if not HAVE_MEGATRON_CORE or not HAVE_TE:
diff --git a/nemo/collections/nlp/models/language_modeling/megatron/gpt_layer_modelopt_spec.py b/nemo/collections/nlp/models/language_modeling/megatron/gpt_layer_modelopt_spec.py
index 6cf068b85ebc..e05c61bf3d24 100644
--- a/nemo/collections/nlp/models/language_modeling/megatron/gpt_layer_modelopt_spec.py
+++ b/nemo/collections/nlp/models/language_modeling/megatron/gpt_layer_modelopt_spec.py
@@ -13,10 +13,10 @@
 # limitations under the License.
 
 try:
+    from megatron.core.extensions.transformer_engine import TENorm
     from megatron.core.fusions.fused_bias_dropout import get_bias_dropout_add
     from megatron.core.tensor_parallel.layers import ColumnParallelLinear, RowParallelLinear
     from megatron.core.transformer.attention import SelfAttention, SelfAttentionSubmodules
-    from megatron.core.extensions.transformer_engine import TENorm
     from megatron.core.transformer.dot_product_attention import DotProductAttention
     from megatron.core.transformer.enums import AttnMaskType
     from megatron.core.transformer.identity_op import IdentityOp
diff --git a/nemo/collections/nlp/models/language_modeling/megatron/griffin/griffin_block.py b/nemo/collections/nlp/models/language_modeling/megatron/griffin/griffin_block.py
index 63441ec51597..b98913dd98c6 100755
--- a/nemo/collections/nlp/models/language_modeling/megatron/griffin/griffin_block.py
+++ b/nemo/collections/nlp/models/language_modeling/megatron/griffin/griffin_block.py
@@ -12,6 +12,7 @@
 # See the License for the specific language governing permissions and
 # limitations under the License.
 from torch import Tensor, nn
+
 from nemo.collections.nlp.models.language_modeling.megatron.griffin.griffin_layer_spec import (
     griffin_mqa_layer_with_transformer_engine_spec,
     griffin_recurrent_layer_with_transformer_engine_spec,
@@ -20,9 +21,9 @@
 
 try:
     from megatron.core import parallel_state, tensor_parallel
+    from megatron.core.extensions.transformer_engine import TENorm, te_checkpoint
     from megatron.core.models.common.language_module.language_module import LanguageModule
     from megatron.core.packed_seq_params import PackedSeqParams
-    from megatron.core.extensions.transformer_engine import TENorm, te_checkpoint
     from megatron.core.transformer.spec_utils import build_module
     from megatron.core.transformer.transformer_config import TransformerConfig
 
diff --git a/nemo/collections/nlp/models/language_modeling/megatron/griffin/griffin_layer_spec.py b/nemo/collections/nlp/models/language_modeling/megatron/griffin/griffin_layer_spec.py
index f7b3bc4c198d..397be833a12c 100755
--- a/nemo/collections/nlp/models/language_modeling/megatron/griffin/griffin_layer_spec.py
+++ b/nemo/collections/nlp/models/language_modeling/megatron/griffin/griffin_layer_spec.py
@@ -12,13 +12,13 @@
 # See the License for the specific language governing permissions and
 # limitations under the License.
 
-from megatron.core.fusions.fused_bias_dropout import get_bias_dropout_add
-from megatron.core.transformer.attention import SelfAttention, SelfAttentionSubmodules
 from megatron.core.extensions.transformer_engine import (
     TEDotProductAttention,
     TELayerNormColumnParallelLinear,
     TERowParallelLinear,
 )
+from megatron.core.fusions.fused_bias_dropout import get_bias_dropout_add
+from megatron.core.transformer.attention import SelfAttention, SelfAttentionSubmodules
 from megatron.core.transformer.enums import AttnMaskType
 from megatron.core.transformer.identity_op import IdentityOp
 from megatron.core.transformer.mlp import MLP, MLPSubmodules
@@ -53,7 +53,10 @@
         self_attn_bda=get_bias_dropout_add,
         mlp=ModuleSpec(
             module=MLP,
-            submodules=MLPSubmodules(linear_fc1=TELayerNormColumnParallelLinear, linear_fc2=TERowParallelLinear,),
+            submodules=MLPSubmodules(
+                linear_fc1=TELayerNormColumnParallelLinear,
+                linear_fc2=TERowParallelLinear,
+            ),
         ),
         mlp_bda=get_bias_dropout_add,
     ),
@@ -74,7 +77,10 @@
         recurrent_bda=get_bias_dropout_add,
         mlp=ModuleSpec(
             module=MLP,
-            submodules=MLPSubmodules(linear_fc1=TELayerNormColumnParallelLinear, linear_fc2=TERowParallelLinear,),
+            submodules=MLPSubmodules(
+                linear_fc1=TELayerNormColumnParallelLinear,
+                linear_fc2=TERowParallelLinear,
+            ),
         ),
         mlp_bda=get_bias_dropout_add,
     ),
diff --git a/nemo/collections/nlp/models/language_modeling/megatron_gpt_model.py b/nemo/collections/nlp/models/language_modeling/megatron_gpt_model.py
index 25e740b4027d..571d93120308 100644
--- a/nemo/collections/nlp/models/language_modeling/megatron_gpt_model.py
+++ b/nemo/collections/nlp/models/language_modeling/megatron_gpt_model.py
@@ -843,7 +843,6 @@ def training_step(self, dataloader_iter):
                     if hasattr(module, 'embedding'):
                         for param in module.embedding.parameters():
                             param.data_ptr()
-    
 
         if self.cfg.get('pipeline_model_parallel_size', 1) > 1 and parallel_state.is_pipeline_last_stage(
             ignore_virtual=True
@@ -2141,7 +2140,7 @@ def build_transformer_config(self) -> TransformerConfig:
             'moe_z_loss_coeff': self.cfg.get('moe_z_loss_coeff', None),  # 1e-3 would be a good start value for z-loss
             'moe_input_jitter_eps': self.cfg.get('moe_input_jitter_eps', None),
             'moe_token_dropping': self.cfg.get('moe_token_dropping', False),  # TODO: Support token dropping.
-            'enable_cuda_graph': self.cfg.get('enable_cuda_graph', False)
+            'enable_cuda_graph': self.cfg.get('enable_cuda_graph', False),
         }
         if model_specific_configs['num_moe_experts'] is not None:
             assert mcore_supports_moe(), 'Megatron-core >= v0.5.0 is required for MoE'
diff --git a/nemo/collections/nlp/modules/common/hyena/hyena.py b/nemo/collections/nlp/modules/common/hyena/hyena.py
index 61bbfe7ca933..f1b4fe20f537 100644
--- a/nemo/collections/nlp/modules/common/hyena/hyena.py
+++ b/nemo/collections/nlp/modules/common/hyena/hyena.py
@@ -23,10 +23,7 @@
 import torch
 import torch.nn as nn
 from einops import rearrange
-from megatron.core.extensions.transformer_engine import (
-    TELayerNormColumnParallelLinear,
-    TERowParallelLinear,
-)
+from megatron.core.extensions.transformer_engine import TELayerNormColumnParallelLinear, TERowParallelLinear
 from megatron.core.transformer.identity_op import IdentityFuncOp, IdentityOp
 from megatron.core.transformer.spec_utils import ModuleSpec, build_module
 from megatron.core.transformer.transformer_config import TransformerConfig
diff --git a/nemo/collections/nlp/modules/common/hyena/hyena_spec.py b/nemo/collections/nlp/modules/common/hyena/hyena_spec.py
index 2b06e277d004..58797cf27838 100644
--- a/nemo/collections/nlp/modules/common/hyena/hyena_spec.py
+++ b/nemo/collections/nlp/modules/common/hyena/hyena_spec.py
@@ -1,9 +1,6 @@
 import torch.nn as nn
+from megatron.core.extensions.transformer_engine import TELayerNormColumnParallelLinear, TERowParallelLinear
 from megatron.core.models.gpt.gpt_layer_specs import get_gpt_layer_with_transformer_engine_spec
-from megatron.core.extensions.transformer_engine import (
-    TELayerNormColumnParallelLinear,
-    TERowParallelLinear,
-)
 from megatron.core.transformer.spec_utils import ModuleSpec
 
 from nemo.collections.nlp.modules.common.hyena.hyena import (
diff --git a/nemo/collections/nlp/modules/common/megatron/adapters/mcore_mixins.py b/nemo/collections/nlp/modules/common/megatron/adapters/mcore_mixins.py
index 0ac9298d1f6d..5128b4ca6b16 100644
--- a/nemo/collections/nlp/modules/common/megatron/adapters/mcore_mixins.py
+++ b/nemo/collections/nlp/modules/common/megatron/adapters/mcore_mixins.py
@@ -15,6 +15,7 @@
 import torch
 import torch.nn.functional as F
 from megatron.core import InferenceParams
+from megatron.core.extensions.transformer_engine import SplitAlongDim
 from megatron.core.fusions.fused_bias_geglu import bias_geglu_impl
 from megatron.core.fusions.fused_bias_gelu import bias_gelu_impl
 from megatron.core.fusions.fused_bias_swiglu import bias_swiglu_impl
@@ -22,7 +23,6 @@
 from megatron.core.models.common.embeddings.rotary_pos_embedding import apply_rotary_pos_emb
 from megatron.core.packed_seq_params import PackedSeqParams
 from megatron.core.transformer.attention import SelfAttention
-from megatron.core.extensions.transformer_engine import SplitAlongDim
 from megatron.core.transformer.mlp import MLP
 from megatron.core.transformer.moe.experts import SequentialMLP
 from megatron.core.transformer.transformer_block import TransformerBlock