allura-org/Gemma-3-Glitter-12B
LoRA-style (fewer layers) vs ToastyPigeon/g3-12b-rp-system-v0.1. ToastyPigeon/g3-12b-rp-system-v0.1 matches part of allura-org/Gemma-3-Glitter-12B (626 of its 1065 tensors; the rest of allura-org/Gemma-3-Glitter-12B has no counterpart). On the shared part: Looks like a merged LoRA (or a fine-tune with frozen norms): only the weight matrices of down_proj, gate_proj, k_proj, o_proj, q_proj, up_proj, v_proj changed, in 48 of 48 layers; norms, biases and embeddings are untouched.
Tensors changed: Embeddings 0/1, Norms 0/289, MLP 144/144, Attention 192/192
4 derivatives checked
- mradermacher/Gemma-3-Glitter-12B-i1-GGUF: Quantization (fewer layers)
- mradermacher/Gemma-3-Glitter-12B-GGUF: Quantization (fewer layers)
- Mawdistical/Mawnia-12B: LoRA-style
- rodrigomt/gama-12b: Full fine-tune