salihfurkaan/VoxPolska-Auralis
LoRA-style vs unsloth/Llasa-1B. Looks like a merged LoRA (or a fine-tune with frozen norms): only the weight matrices of q_proj, v_proj changed, in 16 of 16 layers; norms, biases and embeddings are untouched.
Tensors changed: Embeddings 0/1, Norms 0/33, MLP 0/48, Attention 32/64
1 derivatives checked
- mradermacher/VoxPolska-Auralis-GGUF: Same weights, converted