hotchpotch/japanese-reranker-tiny-v2
Full fine-tune vs cl-nagoya/ruri-v3-pt-30m. Fine-tune: 100% of the weights changed, across 3 of 3 layers.
Tensors changed: Embeddings 2/2, Norms 6/6, Attention 6/6, MLP 6/6
Full fine-tune vs cl-nagoya/ruri-v3-pt-30m. Fine-tune: 100% of the weights changed, across 3 of 3 layers.
Tensors changed: Embeddings 2/2, Norms 6/6, Attention 6/6, MLP 6/6