Update modeling_phi3_small.py

modeling_phi3_small.py

@@ -216,7 +216,8 @@ class Phi3SmallSelfAttention(nn.Module):
                 f"{self.config.dense_attention_every_n_layers}"
             )
             # use warnings to allow the modeling use different flash attention implementation later
-            logger.warning_once("Flash Attention is not available, but is needed for dense attention")
+            if not is_flash_attention_available:
+                logger.warning_once("Flash Attention is not available, but is needed for dense attention")
         else:
             # BlockSparse related Parameters
             self.blocksparse_params = BlockSparseParams.from_config(config)