Naphula/Goetia-26B-A4B-v1.6
Full fine-tune (fewer layers) vs ApocalypseParty/G4-26B-SFT-6. Naphula/Goetia-26B-A4B-v1.6 matches part of ApocalypseParty/G4-26B-SFT-6 (662 of its 1013 tensors; the rest of ApocalypseParty/G4-26B-SFT-6 has no counterpart). On the shared part: Fine-tune: 100% of the weights changed, across 30 of 30 layers.
Tensors changed: Embeddings 1/4, MLP 150/150, Norms 89/271, other 30/122, Attention 115/115
3 derivatives checked
- mradermacher/Goetia-26B-A4B-v1.6-i1-GGUF: Quantization
- mradermacher/Goetia-26B-A4B-v1.6-GGUF: Quantization
- Phil2Sat/Goetia-26B-A4B-v1.6-I-Compact-MXFP4.gguf: Quantization