Corianas/NearalMistral-2x7B
MoE built from dense models vs Corianas/Neural-Mistral-7B. Corianas/NearalMistral-2x7B is a mixture of 2 experts assembled from dense models (mergekit-moe style): expert 1 is an exact copy of Corianas/Neural-Mistral-7B's MLP; the other experts come from other models. Shared tensors (attention, norms, embeddings): 0 of 195 identical.
1 derivatives checked
- mradermacher/NearalMistral-2x7B-i1-GGUF: Quantization