redrix/patricide-12B-Unslop-Mell
Partial fine-tune vs TheDrummer/UnslopNemo-12B-v4.1. Partial change: 85% of the weights changed (attention, mlp, norms), in 38 of 40 layers.
Tensors changed: lm_head 0/1, Embeddings 0/1, Norms 36/81, MLP 114/120, Attention 152/160
8 derivatives checked
- mradermacher/patricide-12B-Unslop-Mell-i1-GGUF: Quantization
- mradermacher/patricide-12B-Unslop-Mell-GGUF: Quantization
- BasedAGI/Patricide-12B-Unslop-Mell-i1-GGUF: Quantization
- redrix/patricide-12B-Unslop-Mell-GGUF: Quantization
- MaziyarPanahi/patricide-12B-Unslop-Mell-GGUF: Same weights, converted
- tensorblock/redrix_patricide-12B-Unslop-Mell-GGUF: Quantization
- Darkknight535/Patricide-21B-Unslop-Mell: Partial fine-tune (fewer layers)
- pot99rta/PatriMaid-12B-Forgottenslop-NeonMell: Full fine-tune