mlasli/Muse-Glimmer-30B-Abliterated-Q8_0-GGUF
Quantization vs mlasli/Muse-Glimmer-30B-Abliterated-BF16. mlasli/Muse-Glimmer-30B-Abliterated-Q8_0-GGUF is a GGUF quantization (Q8_0) of the same weights: 209 tensors match exactly and 418 within quantization error.
Tensors changed: lm_head 0/1, Embeddings 0/1, Norms 0/209, MLP 0/156, Attention 0/260