extreme1228/ScaleCUA-qwen3.5-osworld
Partial fine-tune vs extreme1228/ScaleCUA-qwen3.5-osworld-sft. Partial change: 84% of the weights changed (attention, lm_head, mlp, norms), in 32 of 32 layers.
Tensors changed: Embeddings 0/4, Norms 69/215, Attention 180/332, MLP 96/208, lm_head 1/1