mradermacher/Llama-3-13B-Instruct-GGUF
Quantization vs elinas/Llama-3-13B-Instruct. mradermacher/Llama-3-13B-Instruct-GGUF is a GGUF quantization (Q8_0) of the same weights: 111 tensors match exactly and 387 within quantization error.
Tensors changed: lm_head 0/1, Embeddings 0/1, Norms 0/111, MLP 0/165, Attention 0/220