t0rr3sp3dr0/Qwen3.5-9B-MLX-MTP-4bit
Partial fine-tune (fewer layers) vs mlx-community/Qwen3.5-9B-MTP-4bit. mlx-community/Qwen3.5-9B-MTP-4bit matches part of t0rr3sp3dr0/Qwen3.5-9B-MLX-MTP-4bit (28 of its 1287 tensors; the rest of t0rr3sp3dr0/Qwen3.5-9B-MLX-MTP-4bit has no counterpart). On the shared part: Partial change: 0% of the weights changed (embeddings, norms), in 1 of 1 layers.
Tensors changed: Norms 6/6, MLP 0/9, Attention 0/12, Embeddings 1/1