ayanami-kitasan/code-pruner
LoRA-style vs Qwen/Qwen3-Reranker-0.6B. Looks like a merged LoRA (or a fine-tune with frozen norms): only the weight matrices of down_proj, gate_proj, o_proj, q_proj, up_proj, v_proj changed, in 13 of 28 layers; norms, biases and embeddings are untouched.
Tensors changed: Embeddings 0/1, Norms 0/113, MLP 7/84, Attention 7/112