trollek/danube2-1.8b-Neural
Full fine-tune vs h2oai/h2o-danube2-1.8b-base. Fine-tune: 100% of the weights changed, across 24 of 24 layers.
Tensors changed: Norms 2/49, MLP 72/72, Attention 96/96
1 derivatives checked
- mradermacher/danube2-1.8b-Neural-GGUF: Same weights, converted