Add LoRA support for Gemma (#3050)

vllm/model_executor/models/llama.py

@@ -27,6 +27,7 @@ import torch
 from torch import nn
 from transformers import LlamaConfig
 
+from vllm.config import LoRAConfig
 from vllm.model_executor.input_metadata import InputMetadata
 from vllm.model_executor.layers.activation import SiluAndMul
 from vllm.model_executor.layers.attention import PagedAttention
@@ -45,7 +46,6 @@ from vllm.model_executor.sampling_metadata import SamplingMetadata
 from vllm.model_executor.weight_utils import (default_weight_loader,
                                               hf_model_weights_iterator)
 from vllm.sequence import SamplerOutput
-from vllm.config import LoRAConfig
 
 KVCache = Tuple[torch.Tensor, torch.Tensor]