mradermacher/Aya-X-Mod-GGUF
Same weights, converted vs matrixportalx/Aya-X-Mod. Same weights stored in another dtype (BF16 → F16/F32).
Tensors changed: Embeddings 0/1, Norms 0/33, MLP 0/96, Attention 0/128
Same weights, converted vs matrixportalx/Aya-X-Mod. Same weights stored in another dtype (BF16 → F16/F32).
Tensors changed: Embeddings 0/1, Norms 0/33, MLP 0/96, Attention 0/128