yang31210999/Llama-3.1-Minitron-4B-Depth-Neo-BAAI-100k
LoRA-style vs nvidia/Llama-3.1-Minitron-4B-Depth-Base. Looks like a merged LoRA (or a fine-tune with frozen norms): only the weight matrices of down_proj, k_proj, q_proj, up_proj, v_proj changed, in 16 of 16 layers; norms, biases and embeddings are untouched.
Tensors changed: Embeddings 0/1, Norms 0/33, MLP 32/48, Attention 48/64, lm_head 0/1
1 derivatives checked
- mradermacher/Llama-3.1-Minitron-4B-Depth-Neo-10w-GGUF: Same weights, converted