diff --git a/pgx/_src/games/go.py b/pgx/_src/games/go.py
index 1c86b94a8..9593a406a 100644
--- a/pgx/_src/games/go.py
+++ b/pgx/_src/games/go.py
@@ -30,7 +30,6 @@ class GameState(NamedTuple):
     consecutive_pass_count: Array = jnp.int32(0)
     ko: Array = jnp.int32(-1)  # by SSK
     is_psk: Array = jnp.bool_(False)
-    hash: Array = jnp.zeros(2, dtype=jnp.uint32)
     hash_history: Array = jnp.zeros((19 * 19 * 2, 2), dtype=jnp.uint32)
 
     @property
@@ -67,8 +66,8 @@ def step(self, state: GameState, action: Array) -> GameState:
         board_history = board_history.at[0].set(jnp.clip(state.board, -1, 1).astype(jnp.int32))
         state = state._replace(board_history=board_history)
         # check PSK
-        hash = _compute_hash(state)
-        state = state._replace(hash=hash, hash_history=state.hash_history.at[state.step_count].set(hash))
+        hash_ = _compute_hash(state)
+        state = state._replace(hash_history=state.hash_history.at[state.step_count].set(hash_))
         state = state._replace(is_psk=_is_psk(state))
         # increment turns
         state = state._replace(step_count=state.step_count + 1)
@@ -210,7 +209,8 @@ def _compute_hash(state: GameState):
 
 def _is_psk(state: GameState):
     not_passed = state.consecutive_pass_count == 0
-    has_same_hash = (state.hash == state.hash_history).all(axis=-1).sum() > 1
+    curr_hash = state.hash_history[state.step_count]
+    has_same_hash = (curr_hash == state.hash_history).all(axis=-1).sum() > 1
     return not_passed & has_same_hash