soob3123/amoral-gemma3-12B-v2-qat
LoRA-style vs google/gemma-3-12b-it-qat-q4_0-unquantized. Looks like a merged LoRA (or a fine-tune with frozen norms): only the weight matrices of k_proj, o_proj, q_proj, v_proj changed, in 48 of 48 layers; norms, biases and embeddings are untouched.
Tensors changed: Embeddings 0/4, Norms 0/400, MLP 0/252, Attention 192/408, other 0/1
5 derivatives checked
- mradermacher/amoral-gemma3-12B-v2-qat-i1-GGUF: Quantization (fewer layers)
- QuantFactory/Veritas-12B-GGUF: Full fine-tune (fewer layers)
- mradermacher/amoral-gemma3-12B-v2-qat-GGUF: Quantization (fewer layers)
- soob3123/amoral-gemma3-12B-v2-qat-Q4_0-GGUF: Quantization (fewer layers)
- soob3123/Veritas-12B: LoRA-style