mradermacher/dbrx-instruct-i1-GGUF
Quantization vs alpindale/dbrx-instruct. mradermacher/dbrx-instruct-i1-GGUF is a GGUF quantization (IQ2_XS) of the same weights: 121 tensors match exactly and 82 within quantization error.
Tensors changed: MLP 0/40, Norms 0/161, Embeddings 0/1, lm_head 0/1