QwenLM / Qwen-VL

The official repo of Qwen-VL (通义千问-VL) chat & pretrained large vision language model proposed by Alibaba Cloud.
Other
5.06k stars 385 forks source link

How to get the embedding_matrix of the Qwen-VL-Chat and the text or picture's embeddings? #154

Open tcexeexe opened 1 year ago

tcexeexe commented 1 year ago

是否已有关于该错误的issue或讨论? | Is there an existing issue / discussion for this?

该问题是否在FAQ中有解答? | Is there an existing answer for this in FAQ?

当前行为 | Current Behavior

Hello~ I want to get the embedding matrixand of the model and the embeddings when l input a text or picture. Can you tell me how to get it for Qwen-VL-Chat? l have tried "model.embed_tokens.weight" and “model.model.embed_tokens(input)” but it failed.

期望行为 | Expected Behavior

No response

复现方法 | Steps To Reproduce

No response

运行环境 | Environment

None

备注 | Anything else?

No response

ShuaiBai623 commented 11 months ago

wte and llm_head