cl-nagoya/ruri-base-v2
Full fine-tune vs cl-nagoya/ruri-pt-base-v2. Fine-tune: 100% of the weights changed, across 12 of 12 layers.
Tensors changed: Embeddings 4/5, Norms 48/48, Attention 96/96, other 49/50
Full fine-tune vs cl-nagoya/ruri-pt-base-v2. Fine-tune: 100% of the weights changed, across 12 of 12 layers.
Tensors changed: Embeddings 4/5, Norms 48/48, Attention 96/96, other 49/50