[Bugfix] Replace PoolingParams.normalize with use_activation (#32243)

Signed-off-by: DarkLight1337 <tlleungac@connect.ust.hk>
This commit is contained in:
Cyrus Leung
2026-01-13 18:45:42 +08:00
committed by simon-mo
parent 11b6af5280
commit 0aa8c40552
21 changed files with 68 additions and 70 deletions

View File

@@ -216,7 +216,7 @@ def server_with_chunked_processing():
"512", # Set smaller max_model_len to trigger chunking mechanism
"--pooler-config",
(
'{"pooling_type": "MEAN", "normalize": true, '
'{"pooling_type": "MEAN", "use_activation": true, '
'"enable_chunked_processing": true, "max_embed_len": 10000}'
),
"--gpu-memory-utilization",