Skip to content

Releases: naver/sqlova

Trained Shallow Layer and Decoder Layer parameters

30 Sep 08:50
523c006
Compare
Choose a tag to compare

The model parameters for Shallow Layer and Decoder Layer models uploaded.
Both models are based on BERT Large Uncased.

Trained SQLova parameters

23 Jan 00:14
25a8ddc
Compare
Choose a tag to compare

Pretrained SQLova parameters. BERT-Large Uncased model is used. results_*.jsonl is obtained with --EG.