soob3123/Veritas-12B
LoRA-style vs soob3123/amoral-gemma3-12B-v2-qat. 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
4 derivatives checked
- bartowski/soob3123_Veritas-12B-GGUF: Same weights, converted (fewer layers)
- mradermacher/Veritas-12B-i1-GGUF: Quantization (fewer layers)
- mradermacher/Veritas-12B-GGUF: Quantization (fewer layers)
- DevQuasar/soob3123.Veritas-12B-GGUF: Same weights, converted (fewer layers)