distil-labs/Distil-gitara-v2-Llama-3.2-1B-Instruct
LoRA-style vs meta-llama/Llama-3.2-1B-Instruct. 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/Distil-gitara-v2-Llama-3.2-1B-Instruct-GGUF: Same weights, converted