matura-lol/GLiNER2.5-multi-Decide-finetune
Full fine-tune vs fastino/gliner2.5-multi-v1. Fine-tune: 99% of the weights changed, across 12 of 12 layers.
Tensors changed: Norms 50/64, Attention 102/110, other 70/150, Embeddings 6/6, MLP 4/4
Full fine-tune vs fastino/gliner2.5-multi-v1. Fine-tune: 99% of the weights changed, across 12 of 12 layers.
Tensors changed: Norms 50/64, Attention 102/110, other 70/150, Embeddings 6/6, MLP 4/4