huihui-ai/Huihui4-8B-A4B-v2
LoRA-style vs google/gemma-4-26B-A4B-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 30 of 30 layers; norms, biases and embeddings are untouched.
Tensors changed: Embeddings 0/4, Norms 0/433, other 0/62, MLP 90/171, Attention 115/223
2 derivatives checked
- mradermacher/Huihui4-8B-A4B-v2-i1-GGUF: Quantization
- mradermacher/Huihui4-8B-A4B-v2-GGUF: Same weights, converted
1 other repos hold byte-for-byte identical weight files.