whr778/gliner2-base-v1-rams
Full fine-tune vs fastino/gliner2-base-v1. Fine-tune: 99% of the weights changed, across 12 of 12 layers.
Tensors changed: other 64/68, Embeddings 40/40, Norms 50/50, Attention 96/96
Full fine-tune vs fastino/gliner2-base-v1. Fine-tune: 99% of the weights changed, across 12 of 12 layers.
Tensors changed: other 64/68, Embeddings 40/40, Norms 50/50, Attention 96/96