[Misc] Clean up more utils (#27567)
Signed-off-by: DarkLight1337 <tlleungac@connect.ust.hk>
This commit is contained in:
@@ -69,10 +69,7 @@ from vllm.pooling_params import PoolingParams
|
||||
from vllm.sampling_params import SamplingType
|
||||
from vllm.sequence import IntermediateTensors
|
||||
from vllm.tasks import GenerationTask, PoolingTask, SupportedTask
|
||||
from vllm.utils import (
|
||||
check_use_alibi,
|
||||
length_from_prompt_token_ids_or_embeds,
|
||||
)
|
||||
from vllm.utils import length_from_prompt_token_ids_or_embeds
|
||||
from vllm.utils.jsontree import json_map_leaves
|
||||
from vllm.utils.math_utils import cdiv, round_up
|
||||
from vllm.utils.mem_constants import GiB_bytes
|
||||
@@ -266,7 +263,7 @@ class GPUModelRunner(LoRAModelRunnerMixin, KVConnectorModelRunnerMixin):
|
||||
self.hidden_size = model_config.get_hidden_size()
|
||||
self.attention_chunk_size = model_config.attention_chunk_size
|
||||
# Only relevant for models using ALiBi (e.g, MPT)
|
||||
self.use_alibi = check_use_alibi(model_config)
|
||||
self.use_alibi = model_config.uses_alibi
|
||||
|
||||
self.cascade_attn_enabled = not self.model_config.disable_cascade_attn
|
||||
|
||||
|
||||
Reference in New Issue
Block a user