diff --git a/viscy/data/hcs.py b/viscy/data/hcs.py
index 2b163433..26271438 100644
--- a/viscy/data/hcs.py
+++ b/viscy/data/hcs.py
@@ -115,26 +115,23 @@ def __init__(
         positions: list[Position],
         channels: ChannelMap,
         z_window_size: int,
-        pyramid_resolution: int = 0,
+        pyramid_resolution: str = "0",
         transform: DictTransform | None = None,
     ) -> None:
         super().__init__()
         self.positions = positions
         self.channels = {k: _ensure_channel_list(v) for k, v in channels.items()}
         self.source_ch_idx = [
-            positions[pyramid_resolution].get_channel_index(c)
-            for c in channels["source"]
+            positions[0].get_channel_index(c) for c in channels["source"]
         ]
         self.target_ch_idx = (
-            [
-                positions[pyramid_resolution].get_channel_index(c)
-                for c in channels["target"]
-            ]
+            [positions[0].get_channel_index(c) for c in channels["target"]]
             if "target" in channels
             else None
         )
         self.z_window_size = z_window_size
         self.transform = transform
+        self.pyramid_resolution = pyramid_resolution
         self._get_windows()
 
     def _get_windows(self) -> None:
@@ -145,7 +142,7 @@ def _get_windows(self) -> None:
         self.window_arrays = []
         self.window_norm_meta: list[NormMeta | None] = []
         for fov in self.positions:
-            img_arr: ImageArray = fov["0"]
+            img_arr: ImageArray = fov[str(self.pyramid_resolution)]
             ts = img_arr.frames
             zs = img_arr.slices - self.z_window_size + 1
             w += ts * zs
@@ -226,7 +223,7 @@ def __getitem__(self, index: int) -> Sample:
         sample = {
             "index": sample_index,
             "source": self._stack_channels(sample_images, "source"),
-            "norm_meta": norm_meta,
+            # "norm_meta": norm_meta,
         }
         if self.target_ch_idx is not None:
             sample["target"] = self._stack_channels(sample_images, "target")
@@ -327,7 +324,7 @@ def __init__(
         augmentations: list[MapTransform] = [],
         caching: bool = False,
         ground_truth_masks: Path | None = None,
-        pyramid_resolution: int = 0,
+        pyramid_resolution: str = "0",
     ):
         super().__init__()
         self.data_path = Path(data_path)
@@ -344,6 +341,7 @@ def __init__(
         self.caching = caching
         self.ground_truth_masks = ground_truth_masks
         self.prepare_data_per_node = True
+        self.pyramid_resolution = pyramid_resolution
 
     @property
     def cache_path(self):
@@ -400,6 +398,7 @@ def _base_dataset_settings(self) -> dict[str, dict[str, list[str]] | int]:
         return {
             "channels": {"source": self.source_channel},
             "z_window_size": self.z_window_size,
+            "pyramid_resolution": self.pyramid_resolution,
         }
 
     def setup(self, stage: Literal["fit", "validate", "test", "predict"]):