deepvk/GeRaCl-USER2-base
Full fine-tune vs deepvk/USER2-base. Fine-tune: 100% of the weights changed, across 22 of 22 layers.
Tensors changed: Embeddings 1/1, Norms 44/44, Attention 44/44, MLP 44/44
Full fine-tune vs deepvk/USER2-base. Fine-tune: 100% of the weights changed, across 22 of 22 layers.
Tensors changed: Embeddings 1/1, Norms 44/44, Attention 44/44, MLP 44/44