cl-nagoya/ruri-reranker-large
Full fine-tune vs cl-nagoya/ruri-reranker-stage1-large. Fine-tune: 100% of the weights changed, across 24 of 24 layers.
Tensors changed: Embeddings 3/5, Norms 48/96, Attention 192/192, other 100/100
Full fine-tune vs cl-nagoya/ruri-reranker-stage1-large. Fine-tune: 100% of the weights changed, across 24 of 24 layers.
Tensors changed: Embeddings 3/5, Norms 48/96, Attention 192/192, other 100/100