shekar-ai/albert-base-v2-persian-zwnj-naab-mlm
Full fine-tune vs albert/albert-base-v2. Fine-tune: 100% of the weights changed, across 0 of 0 layers.
Tensors changed: Embeddings 6/6, Norms 6/6, Attention 8/8, MLP 4/4, other 2/2
Full fine-tune vs albert/albert-base-v2. Fine-tune: 100% of the weights changed, across 0 of 0 layers.
Tensors changed: Embeddings 6/6, Norms 6/6, Attention 8/8, MLP 4/4, other 2/2