oddadmix/dialect-router-v0.2
Full fine-tune vs asafaya/bert-mini-arabic. Fine-tune: 100% of the weights changed, across 4 of 4 layers.
Tensors changed: Embeddings 5/5, Norms 16/16, Attention 32/32, other 18/18
Full fine-tune vs asafaya/bert-mini-arabic. Fine-tune: 100% of the weights changed, across 4 of 4 layers.
Tensors changed: Embeddings 5/5, Norms 16/16, Attention 32/32, other 18/18