diff --git a/topomodelx/nn/hypergraph/allset.py b/topomodelx/nn/hypergraph/allset.py
index 0219747f..9abc2c20 100644
--- a/topomodelx/nn/hypergraph/allset.py
+++ b/topomodelx/nn/hypergraph/allset.py
@@ -22,7 +22,7 @@ class AllSet(torch.nn.Module):
         Dropout probability for the AllSet layer.
     mlp_num_layers : int, default = 2
         Number of layers in the MLP.
-    mlp_activation : torch.nn.Module, default = None
+    mlp_activation : callable | None, default = torch.nn.ReLU
         Activation function in the MLP.
     mlp_dropout : float, default = 0.0
         Dropout probability for the MLP.
@@ -46,7 +46,7 @@ def __init__(
         n_layers=2,
         layer_dropout=0.2,
         mlp_num_layers=2,
-        mlp_activation=None,
+        mlp_activation=torch.nn.ReLU,
         mlp_dropout=0.0,
         mlp_norm=None,
         **kwargs,
diff --git a/topomodelx/nn/hypergraph/allset_layer.py b/topomodelx/nn/hypergraph/allset_layer.py
index e32f0513..a19353c6 100644
--- a/topomodelx/nn/hypergraph/allset_layer.py
+++ b/topomodelx/nn/hypergraph/allset_layer.py
@@ -175,7 +175,8 @@ def __init__(
             layers.append(nn.Linear(in_dim, hidden_dim, bias=bias))
             if norm_layer is not None:
                 layers.append(norm_layer(hidden_dim))
-            layers.append(activation_layer(**params))
+            if activation_layer is not None:
+                layers.append(activation_layer(**params))
             layers.append(nn.Dropout(dropout, **params))
             in_dim = hidden_dim
 
diff --git a/topomodelx/nn/hypergraph/allset_transformer_layer.py b/topomodelx/nn/hypergraph/allset_transformer_layer.py
index 652e8d71..4a79e847 100644
--- a/topomodelx/nn/hypergraph/allset_transformer_layer.py
+++ b/topomodelx/nn/hypergraph/allset_transformer_layer.py
@@ -191,7 +191,7 @@ def __init__(
         number_queries: int = 1,
         dropout: float = 0.0,
         mlp_num_layers: int = 1,
-        mlp_activation=None,
+        mlp_activation=nn.ReLU,
         mlp_dropout: float = 0.0,
         mlp_norm=None,
         initialization: Literal["xavier_uniform", "xavier_normal"] = "xavier_uniform",
@@ -455,7 +455,8 @@ def __init__(
             layers.append(nn.Linear(in_dim, hidden_dim, bias=bias))
             if norm_layer is not None:
                 layers.append(norm_layer(hidden_dim))
-            layers.append(activation_layer(**params))
+            if activation_layer is not None:
+                layers.append(activation_layer(**params))
             layers.append(nn.Dropout(dropout, **params))
             in_dim = hidden_dim