Cyclone-Labs/Eclipsed-Phoenix-26B-A4B
Partial fine-tune vs Vortex5/Phoenix-X-26B-A4B. Partial change: 12% of the weights changed (mlp), in 4 of 30 layers.
Tensors changed: Embeddings 0/4, MLP 8/231, Norms 0/433, other 0/122, Attention 0/223
4 derivatives checked
- alexokita/Eclipsed-Phoenix-26B-A4B-Heretic-i1-GGUF: Quantization
- mradermacher/Eclipsed-Phoenix-26B-A4B-i1-GGUF: Quantization
- mradermacher/Eclipsed-Phoenix-26B-A4B-GGUF: Quantization
- Cyclone-Labs/Solar-Griffin-26B-A4B: Full fine-tune