vocaela/KV-Ground-4B-BaseGuiOwl1.5-0228
LoRA-style vs mPLUG/GUI-Owl-1.5-4B-Instruct. Looks like a merged LoRA (or a fine-tune with frozen norms): only the weight matrices of down_proj, gate_proj, linear_fc1, linear_fc2, up_proj changed, in 36 of 36 layers; norms, biases and embeddings are untouched.
Tensors changed: Embeddings 0/4, Norms 0/249, MLP 164/220, Attention 0/240