mradermacher/Gemma3-12B-CIC-SciCite-GGUF
Quantization (fewer layers) vs sknow-lab/Gemma3-12B-CIC-SciCite. mradermacher/Gemma3-12B-CIC-SciCite-GGUF matches part of sknow-lab/Gemma3-12B-CIC-SciCite (626 of its 1065 tensors; the rest of sknow-lab/Gemma3-12B-CIC-SciCite has no counterpart). On the shared part: mradermacher/Gemma3-12B-CIC-SciCite-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