[6/N] pass whole config to inner model (#10205)

Signed-off-by: youkaichao <youkaichao@gmail.com>

vllm/model_executor/models/starcoder2.py

@@ -43,7 +43,8 @@ from vllm.sequence import IntermediateTensors
 
 from .interfaces import SupportsPP
 from .utils import (is_pp_missing_parameter,
-                    make_empty_intermediate_tensors_factory, make_layers)
+                    make_empty_intermediate_tensors_factory, make_layers,
+                    maybe_prefix)
 
 
 class Starcoder2Attention(nn.Module):
@@ -195,12 +196,13 @@ class Starcoder2DecoderLayer(nn.Module):
 @support_torch_compile
 class Starcoder2Model(nn.Module):
 
-    def __init__(self,
-                 config: Starcoder2Config,
-                 cache_config: Optional[CacheConfig] = None,
-                 quant_config: Optional[QuantizationConfig] = None,
-                 prefix: str = ""):
+    def __init__(self, *, vllm_config: VllmConfig, prefix: str = ""):
         super().__init__()
+
+        config = vllm_config.model_config.hf_config
+        cache_config = vllm_config.cache_config
+        quant_config = vllm_config.quant_config
+
         self.config = config
         self.padding_idx = config.pad_token_id
         self.vocab_size = config.vocab_size
@@ -245,19 +247,13 @@ class Starcoder2Model(nn.Module):
 
 class Starcoder2ForCausalLM(nn.Module, SupportsPP):
 
-    def __init__(
-        self,
-        vllm_config: VllmConfig,
-        prefix: str = "",
-    ) -> None:
+    def __init__(self, *, vllm_config: VllmConfig, prefix: str = ""):
         super().__init__()
         config = vllm_config.model_config.hf_config
-        cache_config = vllm_config.cache_config
         quant_config = vllm_config.quant_config
         self.config = config
-        self.model = Starcoder2Model(config,
-                                     cache_config,
-                                     quant_config=quant_config)
+        self.model = Starcoder2Model(vllm_config=vllm_config,
+                                     prefix=maybe_prefix(prefix, "model"))
         self.vocab_size = config.vocab_size
         self.unpadded_vocab_size = config.vocab_size
         if config.tie_word_embeddings: