Chess Challenge submission by marius-785

model.py

@@ -81,7 +81,7 @@ class ChessConfig(PretrainedConfig):
         self.n_layer = n_layer
         self.n_head = n_head
         self.n_ctx = n_ctx
-        self.group_size = group_size
+        self.group_size = 4
         self.n_inner = n_inner if n_inner is not None else 3 * n_embd  # Reduced from 4x to 3x
         self.dropout = dropout
         self.layer_norm_epsilon = layer_norm_epsilon