Qwen

/

Qwen3-Embedding-0.6B

like 1.09k

Follow

Qwen 90k

直接对模型输出的output进行阶段，之后再进行normalize即可。

A rabbit

A rabbit

what's your point?

If you're using Sentence Transformers, you can load the model with truncate_dim (docs) to set a dimension to truncate to, e.g. with 256 your embeddings will be 256-dimensional. Note that you may want to renormalize your embeddings afterwards.

• Tom Aarsen

If you're using Sentence Transformers, you can load the model with truncate_dim (docs) to set a dimension to truncate to, e.g. with 256 your embeddings will be 256-dimensional. Note that you may want to renormalize your embeddings afterwards.

• Tom Aarsen

Got it! Thanks.