MethodWhite/Qwen3.5-9B-HSAQR-v2
LoRA-style vs Qwen/Qwen3.5-9B-Base. Looks like a merged LoRA (or a fine-tune with frozen norms): only the weight matrices of in_proj_a, in_proj_b, in_proj_qkv, in_proj_z, k_proj, o_proj, out_proj, q_proj, v_proj changed, in 32 of 32 layers; norms, biases and embeddings are untouched.
Tensors changed: lm_head 0/1, Embeddings 0/1, Attention 111/224, Norms 0/105