#cross_encoder-msmarco-distilbert-word2vec256k-MLM_785k_emb_updated

This CrossEncoder was trained with MarginMSE loss from the [vocab-transformers/msmarco-distilbert-word2vec256k-MLM_785k_emb_updated](https://hf.co/vocab-transformers/msmarco-distilbert-word2vec256k-MLM_785k_emb_updated) checkpoint. **Word embedding matrix has been updated during training**.

You can load the model with [sentence-transformers](https://sbert.net):
```python
from sentence_transformers import CrossEncoder
from torch import nn
model = CrossEncoder(model_name, default_activation_function=nn.Identity())
```

Performance on TREC Deep Learning (nDCG@10):
- TREC-DL 19: 71.65
- TREC-DL 20: 73.6