Vortex5/Drifting-Shadow-12B
Partial fine-tune vs Vortex5/Moonlit-Shadow-12B. Partial change: 72% of the weights changed (attention, embeddings, mlp, norms), in 30 of 40 layers.
Tensors changed: lm_head 0/1, Embeddings 1/1, Norms 60/81, MLP 90/120, Attention 120/160
2 derivatives checked
- mradermacher/Drifting-Shadow-12B-i1-GGUF: Quantization
- mradermacher/Drifting-Shadow-12B-GGUF: Quantization