Omartificial-Intelligence-Space/Arabert-all-nli-triplet-Matryoshka

Full fine-tune vs aubmindlab/bert-base-arabertv02. Fine-tune: 99% of the weights changed, across 12 of 12 layers.

Tensors changed: Embeddings 4/5, Norms 48/48, Attention 96/96, other 48/50

1 derivatives checked