jialinyyzz/humanizer
LoRA-style vs google/gemma-4-12B. 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 0/12, Norms 0/289, other 0/48, MLP 144/144, Attention 184/184
5 derivatives checked
- mradermacher/humanizer-i1-GGUF: Quantization
- jialinyyzz/humanizer-GGUF: Same weights, converted
- ngquocvinh/Humanizer-12B-GGUF: Quantization
- webmp3/Sakura-humanizer-12B-GGUF: Quantization
- mradermacher/humanizer-GGUF: Quantization
1 other repos hold byte-for-byte identical weight files.