pierjoe/limite-1b-violetto-mlx-bf16
LoRA-style vs paradigma-inc/limite-1b-violetto. Looks like a merged LoRA (or a fine-tune with frozen norms): only the weight matrices of k_proj, o_proj, q_proj, v_proj changed, in 48 of 48 layers; norms, biases and embeddings are untouched.
Tensors changed: Attention 192/400, MLP 0/242, other 0/3, Embeddings 0/2