Vortex5/NoctyxCosma-12B
Full fine-tune vs Delta-Vector/Rei-V3-KTO-12B. Fine-tune: 100% of the weights changed, across 40 of 40 layers.
Tensors changed: Norms 78/81, MLP 120/120, Attention 160/160
6 derivatives checked
- mradermacher/NoctyxCosma-12B-i1-GGUF: Quantization
- mradermacher/NoctyxCosma-12B-GGUF: Quantization
- WasamiKirua/NightShade-Violet-12B: Full fine-tune
- WasamiKirua/NightShade-Lotus-12B: Full fine-tune
- Vortex5/Starry-Shadow-12B: Full fine-tune
- WasamiKirua/Sakura-Sniper-12B: Full fine-tune