mradermacher/Gemma-3-12B-Fornax-QAT-CoT-GGUF
Quantization (fewer layers) vs ConicCat/Gemma-3-12B-Fornax-QAT-CoT. mradermacher/Gemma-3-12B-Fornax-QAT-CoT-GGUF matches part of ConicCat/Gemma-3-12B-Fornax-QAT-CoT (626 of its 1065 tensors; the rest of ConicCat/Gemma-3-12B-Fornax-QAT-CoT has no counterpart). On the shared part: mradermacher/Gemma-3-12B-Fornax-QAT-CoT-GGUF is a GGUF quantization (Q8_0) of the same weights: 289 tensors match exactly and 337 within quantization error.
Tensors changed: Embeddings 0/1, Norms 0/289, MLP 0/144, Attention 0/192