Edit model card

Transformer language model for Croatian and Serbian

Trained on 6GB datasets that contain Croatian and Serbian language for two epochs (500k steps). Leipzig, OSCAR and srWac datasets

Model #params Arch. Training data
Andrija/SRoBERTa-L 80M Third Leipzig Corpus, OSCAR and srWac (6 GB of text)
Downloads last month
2
Hosted inference API
Fill-Mask
Examples
Examples
Mask token: <mask>
This model can be loaded on the Inference API on-demand.

Datasets used to train Andrija/SRoBERTa-L