bartowski/dolphin-2.9.2-mixtral-8x22b-GGUF
Full fine-tune vs mistral-community/Mixtral-8x22B-v0.1. bartowski/dolphin-2.9.2-mixtral-8x22b-GGUF is a GGUF (Q8_0) of different weights. Fine-tune: 100% of the weights changed, across 56 of 56 layers.
Tensors changed: other 14/56, Attention 224/224, Norms 0/113