nothingiisreal/MN-12B-Starcannon-v2

Full fine-tune vs intervitens/mini-magnum-12b-v1.1. Fine-tune: 100% of the weights changed, across 40 of 40 layers.

Tensors changed: Norms 2/81, MLP 120/120, Attention 160/160

1 derivatives checked

• featherless-ai-quants/nothingiisreal-MN-12B-Starcannon-v2-GGUF: Quantization

2 other repos hold byte-for-byte identical weight files.

On the Hub · Downloads on Model Pulse