fwerkor/MiniCPM5-2B-Diffusion-Base
Full fine-tune vs openbmb/MiniCPM5-2B-Base. Fine-tune: 100% of the weights changed, across 42 of 42 layers.
Tensors changed: Norms 3/85, MLP 126/126, Attention 168/168
Full fine-tune vs openbmb/MiniCPM5-2B-Base. Fine-tune: 100% of the weights changed, across 42 of 42 layers.
Tensors changed: Norms 3/85, MLP 126/126, Attention 168/168