Babsie/CrossroadsLoki-MoE-2x8B

MoE built from dense models vs Bacon666/Athlon-8B-0.1. Babsie/CrossroadsLoki-MoE-2x8B is a mixture of 2 experts assembled from dense models (mergekit-moe style): expert 1 is an exact copy of Bacon666/Athlon-8B-0.1's MLP; the other experts come from other models. Shared tensors (attention, norms, embeddings): 0 of 195 identical.

5 derivatives checked

On the Hub · Downloads on Model Pulse