mradermacher/Mistral-12.25B-Instruct-v0.2-GGUF
Quantization vs Joseph717171/Mistral-12.25B-Instruct-v0.2. mradermacher/Mistral-12.25B-Instruct-v0.2-GGUF is a GGUF quantization (Q8_0) of the same weights: 114 tensors match exactly and 393 within quantization error.
Tensors changed: MLP 0/168, Attention 0/224, Norms 0/113, Embeddings 0/1, lm_head 0/1