Vortex5/Stellar-Witch-12B
Full fine-tune vs PocketDoc/Dans-SakuraKaze-V1.0.0-12b. Fine-tune: 100% of the weights changed, across 40 of 40 layers.
Tensors changed: Norms 80/81, MLP 120/120, Attention 160/160
3 derivatives checked
- mradermacher/Stellar-Witch-12B-i1-GGUF: Quantization
- mradermacher/Stellar-Witch-12B-GGUF: Quantization
- Sorihon/Stellar-Witch-12B-Heretic: Abliterated
1 other repos hold byte-for-byte identical weight files.