unclemusclez/Unsloth-Qwen2.5-Coder-1.5B-Devinator-v1
LoRA-style vs unsloth/Qwen2.5-Coder-1.5B-Instruct. 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 28 of 28 layers; norms, biases and embeddings are untouched.
Tensors changed: Norms 0/57, MLP 76/84, Attention 110/196