diff --git a/src/batdetect2/models/blocks.py b/src/batdetect2/models/blocks.py
index 62bf6c7..9d81d4b 100644
--- a/src/batdetect2/models/blocks.py
+++ b/src/batdetect2/models/blocks.py
@@ -150,8 +150,8 @@ class SelfAttentionConfig(BaseConfig):
     attention_channels : int
         Dimensionality of the query, key, and value projections.
     temperature : float
-        Scaling factor applied to the weighted values before the final
-        linear projection. Defaults to ``1``.
+        Divisor applied together with ``attention_channels`` when scaling
+        dot-product attention logits. Defaults to ``1``.
     """
 
     name: Literal["SelfAttention"] = "SelfAttention"
@@ -321,8 +321,8 @@ class EfficientSelfAttentionConfig(BaseConfig):
     attention_channels : int
         Dimensionality of the query, key, and value projections.
     temperature : float
-        Scaling factor applied to the weighted values before the final
-        linear projection. Defaults to ``1``.
+        Divisor applied together with ``attention_channels`` when scaling
+        dot-product attention logits. Defaults to ``1``.
     """
 
     name: Literal["EfficientSelfAttention"] = "EfficientSelfAttention"
diff --git a/src/batdetect2/models/bottleneck.py b/src/batdetect2/models/bottleneck.py
index 9ebffcd..5dd2630 100644
--- a/src/batdetect2/models/bottleneck.py
+++ b/src/batdetect2/models/bottleneck.py
@@ -21,7 +21,7 @@ This module provides:
 from typing import Annotated, List
 
 import torch
-from pydantic import Field
+from pydantic import Field, model_validator
 from torch import nn
 
 from batdetect2.core.configs import BaseConfig
@@ -197,13 +197,19 @@ class BottleneckConfig(BaseConfig):
     """
 
     channels: int
-    frequency_aggregation: FrequencyAggregationLayerConfig = Field(
-        default_factory=lambda data: VerticalConvConfig(
-            channels=data["channels"]
-        )
-    )
+    frequency_aggregation: FrequencyAggregationLayerConfig | None = None
     layers: List[BottleneckLayerConfig] = Field(default_factory=list)
 
+    @model_validator(mode="after")
+    def set_default_frequency_aggregation(self) -> "BottleneckConfig":
+        """Default frequency aggregation to the bottleneck channel count."""
+        if self.frequency_aggregation is None:
+            self.frequency_aggregation = VerticalConvConfig(
+                channels=self.channels
+            )
+
+        return self
+
 
 DEFAULT_BOTTLENECK_CONFIG: BottleneckConfig = BottleneckConfig(
     channels=256,
@@ -302,11 +308,14 @@ def build_bottleneck(
         by repetition).
     """
     config = config or DEFAULT_BOTTLENECK_CONFIG
+    frequency_aggregation = config.frequency_aggregation
+    if frequency_aggregation is None:
+        raise ValueError("frequency_aggregation must be configured.")
 
     frequency_aggregator = build_frequency_aggregation(
         input_height=input_height,
         in_channels=in_channels,
-        config=config.frequency_aggregation,
+        config=frequency_aggregation,
     )
 
     current_channels = frequency_aggregator.out_channels
diff --git a/tests/test_model.py b/tests/test_model.py
index 4bdfdbc..d729ac9 100644
--- a/tests/test_model.py
+++ b/tests/test_model.py
@@ -82,4 +82,8 @@ def test_bundled_checkpoint_loads_with_current_config_schema() -> None:
 
     assert model.class_names
     assert isinstance(configs.model.architecture, UNetBackboneConfig)
-    assert configs.model.architecture.bottleneck.frequency_aggregation.name
+    frequency_aggregation = (
+        configs.model.architecture.bottleneck.frequency_aggregation
+    )
+    assert frequency_aggregation is not None
+    assert frequency_aggregation.name