ToastyPigeon/glitterfetti-merge-A3

Full fine-tune vs ToastyPigeon/Gemma-3-Confetti-12B. Fine-tune: 100% of the weights changed, across 48 of 48 layers.

Tensors changed: Norms 289/289, MLP 144/144, Attention 192/192

1 derivatives checked

• mradermacher/glitterfetti-merge-A3-GGUF: Quantization

On the Hub · Downloads on Model Pulse