Fix tokenizer to not produce token_type_ids (ModernBERT compatibility)

Fix tokenizer_class for sentence-transformers compatibility

Fix model-index metadata (add dataset fields)

Highlight small training corpus size (~160M words)

Add tested code examples with real outputs

Fix mask token to <|mask|>, add Transformers v5.0 note

Add encyclopedia tag

Fix description: general dictionary model, not legal-specific

Update citation to OpenGloss arXiv paper

Upload model card

Upload OGBert v1 MLM tokenizer

Upload OGBert v1 MLM model

initial commit