# Bert embeddings

**URL:** <https://discuss.huggingface.co/t/bert-embeddings/13453>\
**Category:** Beginners\
**Created:** [January 6, 2022, 6:34pm UTC](https://discuss.huggingface.co/t/bert-embeddings/13453 "2022-01-06T18:34:06Z")\
**Posts on this page:** 1\
**Page:** 1

<div class="post-metadata">

**Author:** ![ShivaniSri](https://avatars.discourse-cdn.com/v4/letter/s/3ab097/32.png) [@ShivaniSri](https://discuss.huggingface.co/u/ShivaniSri)\
**Post date:** [January 6, 2022, 6:34pm UTC](https://discuss.huggingface.co/t/bert-embeddings/13453/1 "2022-01-06T18:34:06Z")

</div>

I want to multiple bert input embeddings with other tensor and forward it to the encoder of bert  
How can I implement this

# import BERT-base pretrained model

bert = AutoModel.from\_pretrained(‘bert-base-uncased’)

# Load the BERT tokenizer

tokenizer = BertTokenizerFast.from\_pretrained(‘bert-base-uncased’)