Update vocab embedding deps and add TP switch (#1856)

2024-10-31 20:13:07 -07:00
parent 61cf00e112
commit 16eb33ffe2
31 changed files with 602 additions and 101 deletions
@@ -24,7 +24,6 @@ from transformers import PretrainedConfig
 from vllm.config import LoRAConfig
 from vllm.distributed import get_tensor_model_parallel_world_size
 from vllm.model_executor.layers.rotary_embedding import get_rope
-from vllm.model_executor.layers.vocab_parallel_embedding import VocabParallelEmbedding
 from vllm.model_executor.model_loader.weight_utils import default_weight_loader

 from sglang.srt.layers.activation import GeluAndMul
@@ -37,6 +36,7 @@ from sglang.srt.layers.linear import (
 from sglang.srt.layers.logits_processor import LogitsProcessor
 from sglang.srt.layers.quantization.base_config import QuantizationConfig
 from sglang.srt.layers.radix_attention import RadixAttention
+from sglang.srt.layers.vocab_parallel_embedding import VocabParallelEmbedding
 from sglang.srt.model_executor.forward_batch_info import ForwardBatch