Remove unused kwargs from model definitions (#13555)

vllm/model_executor/models/internlm2.py

@@ -1,13 +1,13 @@
 # SPDX-License-Identifier: Apache-2.0
 
 from functools import partial
-from typing import Any, Dict, Iterable, List, Optional, Set, Tuple, Type, Union
+from typing import Any, Dict, Iterable, Optional, Set, Tuple, Type, Union
 
 import torch
 from torch import nn
 from transformers import PretrainedConfig
 
-from vllm.attention import Attention, AttentionMetadata
+from vllm.attention import Attention
 from vllm.compilation.decorators import support_torch_compile
 from vllm.config import CacheConfig, VllmConfig
 from vllm.distributed import (get_pp_group, get_tensor_model_parallel_rank,
@@ -175,13 +175,11 @@ class InternLM2Attention(nn.Module):
         self,
         positions: torch.Tensor,
         hidden_states: torch.Tensor,
-        kv_cache: torch.Tensor,
-        attn_metadata: AttentionMetadata,
     ) -> torch.Tensor:
         qkv, _ = self.wqkv(hidden_states)
         q, k, v = self.split_qkv(qkv)
         q, k = self.rotary_emb(positions, q, k)
-        attn_output = self.attn(q, k, v, kv_cache, attn_metadata)
+        attn_output = self.attn(q, k, v)
         output, _ = self.wo(attn_output)
         return output
 
@@ -227,8 +225,6 @@ class InternLMDecoderLayer(nn.Module):
         self,
         positions: torch.Tensor,
         hidden_states: torch.Tensor,
-        kv_cache: torch.Tensor,
-        attn_metadata: AttentionMetadata,
         residual: Optional[torch.Tensor],
     ) -> Tuple[torch.Tensor, torch.Tensor]:
         # Self Attention
@@ -241,8 +237,6 @@ class InternLMDecoderLayer(nn.Module):
         hidden_states = self.attention(
             positions=positions,
             hidden_states=hidden_states,
-            kv_cache=kv_cache,
-            attn_metadata=attn_metadata,
         )
 
         # Fully Connected
@@ -290,8 +284,6 @@ class InternLM2Model(nn.Module):
         self,
         input_ids: torch.Tensor,
         positions: torch.Tensor,
-        kv_caches: List[torch.Tensor],
-        attn_metadata: AttentionMetadata,
         intermediate_tensors: Optional[IntermediateTensors] = None,
         inputs_embeds: Optional[torch.Tensor] = None,
     ) -> Union[torch.Tensor, IntermediateTensors]:
@@ -305,15 +297,8 @@ class InternLM2Model(nn.Module):
             assert intermediate_tensors is not None
             hidden_states = intermediate_tensors["hidden_states"]
             residual = intermediate_tensors["residual"]
-        for i in range(self.start_layer, self.end_layer):
-            layer = self.layers[i]
-            hidden_states, residual = layer(
-                positions,
-                hidden_states,
-                kv_caches[i - self.start_layer],
-                attn_metadata,
-                residual,
-            )
+        for layer in self.layers[self.start_layer:self.end_layer]:
+            hidden_states, residual = layer(positions, hidden_states, residual)
         if not get_pp_group().is_last_rank:
             return IntermediateTensors({
                 "hidden_states": hidden_states,
@@ -363,13 +348,10 @@ class InternLM2ForCausalLM(nn.Module, SupportsPP, SupportsLoRA):
         self,
         input_ids: torch.Tensor,
         positions: torch.Tensor,
-        kv_caches: List[torch.Tensor],
-        attn_metadata: AttentionMetadata,
         intermediate_tensors: Optional[IntermediateTensors],
         inputs_embeds: Optional[torch.Tensor] = None,
     ) -> torch.Tensor:
-        hidden_states = self.model(input_ids, positions, kv_caches,
-                                   attn_metadata, intermediate_tensors,
+        hidden_states = self.model(input_ids, positions, intermediate_tensors,
                                    inputs_embeds)
         return hidden_states
 
@@ -466,13 +448,10 @@ class InternLM2ForRewardModel(InternLM2ForCausalLM):
         self,
         input_ids: torch.Tensor,
         positions: torch.Tensor,
-        kv_caches: List[torch.Tensor],
-        attn_metadata: AttentionMetadata,
         intermediate_tensors: Optional[IntermediateTensors] = None,
         inputs_embeds: Optional[torch.Tensor] = None,
     ) -> Union[torch.Tensor, IntermediateTensors]:
-        hidden_states = self.model(input_ids, positions, kv_caches,
-                                   attn_metadata, intermediate_tensors,
+        hidden_states = self.model(input_ids, positions, intermediate_tensors,
                                    inputs_embeds)
         logits, _ = self.v_head(hidden_states)
         return logits