DEFAULT MODEL UPDATES
We have updated our standard text models to more performant ones - the following models have been deprecated.
- DistilBERT
- BERT
- Multilingual BERT
We recommend that you DO NOT start new projects with them without first consulting Support or your CSM
Model Name | DistilBERT (Uncased) |
Description | Relatively fast and small model, with near performance to BERT |
Use For | General text blobs |
Limitations | This model only works for English texts. Long texts will be truncated to at most 512 tokens |
Graft Default | No |
Reference information
Source | distilbert-base-uncased - hugging face |
Trained on | English Wikipedia and "BookCorpus" |
Paper | D19-1371 |
Embedding Dimension | 768 |