diff --git a/tests/models/language/pooling/test_embedding.py b/tests/models/language/pooling/test_embedding.py index 982dc73f6..e105195af 100644 --- a/tests/models/language/pooling/test_embedding.py +++ b/tests/models/language/pooling/test_embedding.py @@ -54,7 +54,7 @@ def test_models( vllm_extra_kwargs = {} if model == "ssmits/Qwen2-7B-Instruct-embed-base": vllm_extra_kwargs["pooler_config"] = PoolerConfig( - seq_pooling_type="MEAN", normalize=False + seq_pooling_type="MEAN", use_activation=False ) max_model_len: int | None = 512 diff --git a/vllm/entrypoints/llm.py b/vllm/entrypoints/llm.py index f3f774bef..24545de19 100644 --- a/vllm/entrypoints/llm.py +++ b/vllm/entrypoints/llm.py @@ -174,8 +174,8 @@ class LLM: multi-modal processor obtained from `AutoProcessor.from_pretrained`. The available overrides depend on the model that is being run. For example, for Phi-3-Vision: `{"num_crops": 4}`. - pooler_config: Initialize non-default pooling config for the pooling - model. e.g. `PoolerConfig(seq_pooling_type="MEAN", normalize=False)`. + pooler_config: Initialize non-default pooling config for the pooling model, + e.g., `PoolerConfig(seq_pooling_type="MEAN", use_activation=False)`. compilation_config: Either an integer or a dictionary. If it is an integer, it is used as the mode of compilation optimization. If it is a dictionary, it can specify the full compilation configuration.