mlx-community/Qwen3.6-27B-bf16
Partial fine-tune vs Qwen/Qwen3.6-27B. Partial change: 0% of the weights changed (norms), in 64 of 64 layers.
Tensors changed: Embeddings 0/1, MLP 0/192, Attention 0/400, Norms 161/209
Partial fine-tune vs Qwen/Qwen3.6-27B. Partial change: 0% of the weights changed (norms), in 64 of 64 layers.
Tensors changed: Embeddings 0/1, MLP 0/192, Attention 0/400, Norms 161/209