cross-encoder/nli-roberta-base
Full fine-tune vs FacebookAI/roberta-base. Fine-tune: 100% of the weights changed, across 12 of 12 layers.
Tensors changed: Attention 96/96, other 50/50, Norms 48/48, Embeddings 5/6
Full fine-tune vs FacebookAI/roberta-base. Fine-tune: 100% of the weights changed, across 12 of 12 layers.
Tensors changed: Attention 96/96, other 50/50, Norms 48/48, Embeddings 5/6