diff --git a/pyproject.toml b/pyproject.toml
index 25eeef9..7e5ec13 100644
--- a/pyproject.toml
+++ b/pyproject.toml
@@ -1,6 +1,6 @@
 [project]
 name = "titans-pytorch"
-version = "0.1.36"
+version = "0.1.37"
 description = "Titans"
 authors = [
     { name = "Phil Wang", email = "lucidrains@gmail.com" }
diff --git a/titans_pytorch/neural_memory.py b/titans_pytorch/neural_memory.py
index c081ae7..4faddbf 100644
--- a/titans_pytorch/neural_memory.py
+++ b/titans_pytorch/neural_memory.py
@@ -723,6 +723,8 @@ def forward_inference(
             prev_layer_updates = TensorDict(prev_layer_updates)
             prev_layer_updates = prev_layer_updates.apply(lambda t: t[:, -1:])
 
+        values = None
+
         if store_seq_cache_len == self.chunk_size:
 
             next_updates, next_states, values = self.store_memories(
@@ -770,7 +772,7 @@ def forward(
         if seq_len < self.retrieve_chunk_size:
             out = self.init_empty_memory_embed(batch, seq_len)
 
-            next_store_state = (seq_len, seq, None, None)
+            next_store_state = NeuralMemCache(seq_len, seq, None, None)
 
             out = (out, next_store_state)