DevQuasar-6/nvidia.Llama-3_1-Nemotron-51B-Instruct-GGUF
LoRA-style vs nvidia/Llama-3_1-Nemotron-51B-Instruct. DevQuasar-6/nvidia.Llama-3_1-Nemotron-51B-Instruct-GGUF is a GGUF (F16) of different weights. Looks like a merged LoRA (or a fine-tune with frozen norms): only the weight matrices of k_proj changed, in 3 of 80 layers; norms, biases and embeddings are untouched.
Tensors changed: Embeddings 0/1, Norms 0/153, MLP 0/240, Attention 3/216, lm_head 0/1