[Refactor] Remove dead code in kv connector and model runner (#38383)
Signed-off-by: yewentao256 <zhyanwentao@126.com>
This commit is contained in:
@@ -451,7 +451,6 @@ class GPUModelRunner(
|
||||
# Model-related.
|
||||
self.num_query_heads = model_config.get_num_attention_heads(parallel_config)
|
||||
self.inputs_embeds_size = model_config.get_inputs_embeds_size()
|
||||
self.attention_chunk_size = model_config.attention_chunk_size
|
||||
# Only relevant for models using ALiBi (e.g, MPT)
|
||||
self.use_alibi = model_config.uses_alibi
|
||||
|
||||
@@ -594,7 +593,6 @@ class GPUModelRunner(
|
||||
# NOTE(rob): num_prompt_logprobs only includes reqs
|
||||
# that are currently in the prefill phase.
|
||||
self.num_prompt_logprobs: dict[str, int] = {}
|
||||
self.comm_stream = torch.cuda.Stream()
|
||||
|
||||
# Input Batch
|
||||
# NOTE(Chen): Ideally, we should initialize the input batch inside
|
||||
|
||||
Reference in New Issue
Block a user