kueizen/Marco-Nano-Instruct-REAP15
LoRA-style vs ATH-MaaS/Marco-Nano-Instruct. Looks like a merged LoRA (or a fine-tune with frozen norms): only the weight matrices of down_proj, gate_proj, up_proj changed, in 28 of 28 layers; norms, biases and embeddings are untouched.
Tensors changed: Embeddings 0/1, Norms 0/113, MLP 2231/2348, Attention 0/112