Parallia/Fairly-Multilingual-ModernBERT-Embed-BE

Full fine-tune vs nomic-ai/modernbert-embed-base. Fine-tune: 100% of the weights changed, across 22 of 22 layers.

Tensors changed: Embeddings 1/1, Norms 44/44, Attention 44/44, MLP 44/44

2 derivatives checked