hotchpotch/open-provence-reranker-large-v1
Full fine-tune vs cl-nagoya/ruri-v3-reranker-310m. Fine-tune: 100% of the weights changed, across 25 of 25 layers.
Tensors changed: other 3/3, Norms 51/51, Embeddings 2/2, Attention 50/50, MLP 50/50
Full fine-tune vs cl-nagoya/ruri-v3-reranker-310m. Fine-tune: 100% of the weights changed, across 25 of 25 layers.
Tensors changed: other 3/3, Norms 51/51, Embeddings 2/2, Attention 50/50, MLP 50/50