NandiBERT-150M / modeling_nandi_encoder.py

Commit History

Yesterday

  1. Remove internal code reference from the modeling docstring; 450M parameter count 445M

    Verified signature
    3d28371
  2. Config defaults and docstrings match NandiBERT-150M; drop smoltron_source; 13 Indic languages⁠

    Verified signature
    3cc7e3c
  3. Classifiers use the pretrained MLM head; README tokenizer needs trust_remote_code⁠

    Verified signature
    73bc082

2 days ago

  1. Add final checkpoint (model, config, tokenizer, remote code)

    Verified signature
    577b04c