fix(server): fix flash batch filtering (#220)

Author: OlivierDehaene

server/text_generation_server/models/flash_causal_lm.py

@@ -188,9 +188,10 @@ def filter(self, requests: List[generate_pb2.Request]) -> "FlashCausalLMBatch":
             position_ids.append(self.position_ids[idx])
             cu_seqlens.append(cumulative_length + request_input_length)
             max_seqlen = max(max_seqlen, request_input_length)
+            # True index for past
+            past_key_values.append(self.past_key_values[2 * idx])
+
             if not single_request:
-                # True index for past
-                past_key_values.append(self.past_key_values[2 * idx])
                 # Add one padding
                 past_key_values.append(self.past_pad)
 
@@ -209,7 +210,7 @@ def filter(self, requests: List[generate_pb2.Request]) -> "FlashCausalLMBatch":
         if single_request:
             # Preallocate tensor for bs = 1 case
             past_key_values = torch.nn.functional.pad(
-                self.past_key_values[0],
+                past_key_values[0],
                 (
                     0,
                     0,