update num_rejected_tokens and num_speculative_tokens

vllm/v1/attention/backends/utils.py

@@ -45,22 +45,25 @@ class CommonAttentionMetadata:
     seq_lens_cpu: torch.Tensor
     """(batch_size,), the length of each request including both computed tokens
     and newly scheduled tokens"""
+    
     num_computed_tokens_cpu: torch.Tensor
     """(batch_size,), the number of computed tokens for each request"""
+    
     num_reqs: int
     """Number of requests"""
     num_actual_tokens: int
     """Total number of tokens in batch"""
     max_query_len: int
     """Longest query in batch"""
-    num_rejected_tokens: list[int]
-    """(batch_size,), record the rejected tokens number in cpu and gpu"""
-    num_speculative_tokens: int
-    """Number of speculative tokens"""
     
     block_table_tensor: torch.Tensor
     slot_mapping: torch.Tensor
     
+    num_rejected_tokens: list[int] = None
+    """(batch_size,), record the rejected tokens number in cpu and gpu"""
+    num_speculative_tokens: int = 0
+    """Number of speculative tokens"""
+
 
 M = TypeVar("M")