Vortex5/G4-Midnight-Macaw-26B-A4B
Full fine-tune vs ReadyArt/Dark-Scarlett-v1.0-26B-A4B. Fine-tune: 100% of the weights changed, across 30 of 30 layers.
Tensors changed: Embeddings 4/4, MLP 231/231, Norms 71/433, other 30/122, Attention 223/223
2 derivatives checked
- mradermacher/G4-Midnight-Macaw-26B-A4B-i1-GGUF: Quantization
- mradermacher/G4-Midnight-Macaw-26B-A4B-GGUF: Quantization