TheDrummer/Gemma-3-R1-12B-v1
LoRA-style vs google/gemma-3-12b-it. 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 0/4, Norms 0/400, MLP 144/252, Attention 192/408, other 0/1
4 derivatives checked
- bartowski/TheDrummer_Gemma-3-R1-12B-v1-GGUF: Same weights, converted (fewer layers)
- mradermacher/Gemma-3-R1-12B-v1-i1-GGUF: Quantization (fewer layers)
- mradermacher/Gemma-3-R1-12B-v1-GGUF: Quantization (fewer layers)
- ReadyArt/Safeword-Casual-V1-R1-12B: LoRA-style