Khurram123/Urdu-Llama-3.2-3B-Instruct-v1
LoRA-style vs unsloth/Llama-3.2-3B-Instruct. Looks like a merged LoRA (or a fine-tune with frozen norms): only the weight matrices of down_proj, gate_proj, k_proj, o_proj, q_proj, up_proj, v_proj changed, in 21 of 28 layers; norms, biases and embeddings are untouched.
Tensors changed: Embeddings 0/1, Norms 0/57, MLP 62/84, Attention 84/112
2 derivatives checked
- mradermacher/Urdu-Llama-3.2-3B-Instruct-v1-i1-GGUF: Quantization
- mradermacher/Urdu-Llama-3.2-3B-Instruct-v1-GGUF: Same weights, converted