mradermacher/Gemma3-12b-CXL3-GGUF
Full fine-tune (fewer layers) vs MayIAskYou/Gemma3-12b-CXL3. mradermacher/Gemma3-12b-CXL3-GGUF matches part of MayIAskYou/Gemma3-12b-CXL3 (626 of its 1065 tensors; the rest of MayIAskYou/Gemma3-12b-CXL3 has no counterpart). On the shared part: mradermacher/Gemma3-12b-CXL3-GGUF is a GGUF (Q8_0) of different weights. Fine-tune: 100% of the weights changed, across 48 of 48 layers.
Tensors changed: Embeddings 1/1, Norms 0/289, MLP 144/144, Attention 192/192