mradermacher/MoETest-3E2A-3x3B-GGUF

Same weights, converted vs DoppelReflEx/MoETest-3E2A-3x3B. Same weights stored in another dtype (BF16 → F16/F32), plus 85 new tensors.

Tensors changed: Embeddings 0/1, Norms 0/57, Attention 0/112, lm_head 0/1, other 0/28

On the Hub · Downloads on Model Pulse