rjamorizIAtest/merge
Full fine-tune vs Nexusflow/Starling-LM-7B-beta. Fine-tune: 97% of the weights changed, across 30 of 30 layers.
Tensors changed: Norms 61/61, MLP 87/90, Attention 116/120
Full fine-tune vs Nexusflow/Starling-LM-7B-beta. Fine-tune: 97% of the weights changed, across 30 of 30 layers.
Tensors changed: Norms 61/61, MLP 87/90, Attention 116/120