[Misc][Model][Refactor] Pass the prefix into Linear layers (#28259)

Signed-off-by: MengqingCao <cmq0113@163.com>

vllm/model_executor/models/bloom.py

@@ -108,12 +108,14 @@ class BloomAttention(nn.Module):
             self.total_num_heads,
             bias=True,
             quant_config=quant_config,
+            prefix=f"{prefix}.query_key_value",
         )
         self.dense = RowParallelLinear(
             self.hidden_size,
             self.hidden_size,
             bias=True,
             quant_config=quant_config,
+            prefix=f"{prefix}.dense",
         )
 
         # Create the alibi slopes and slice them.
@@ -152,6 +154,7 @@ class BloomMLP(nn.Module):
         self,
         config: BloomConfig,
         quant_config: QuantizationConfig | None = None,
+        prefix: str = "",
     ):
         super().__init__()
         hidden_size = config.hidden_size
@@ -159,12 +162,14 @@ class BloomMLP(nn.Module):
             hidden_size,
             4 * hidden_size,
             quant_config=quant_config,
+            prefix=f"{prefix}.dense_h_to_4h",
         )
         self.gelu_impl = get_act_fn("gelu")
         self.dense_4h_to_h = RowParallelLinear(
             4 * hidden_size,
             hidden_size,
             quant_config=quant_config,
+            prefix=f"{prefix}.dense_4h_to_h",
         )
 
     def forward(self, x: torch.Tensor) -> torch.Tensor:
@@ -192,7 +197,7 @@ class BloomBlock(nn.Module):
         self.post_attention_layernorm = nn.LayerNorm(
             hidden_size, eps=config.layer_norm_epsilon
         )
-        self.mlp = BloomMLP(config, quant_config)
+        self.mlp = BloomMLP(config, quant_config, prefix=f"{prefix}.mlp")
         self.apply_residual_connection_post_layernorm = (
             config.apply_residual_connection_post_layernorm
         )