mradermacher/alduin-4b-it-base-GGUF
Same weights, converted (fewer layers) vs mdamir97/alduin-4b-it-base. mradermacher/alduin-4b-it-base-GGUF matches part of mdamir97/alduin-4b-it-base (444 of its 883 tensors; the rest of mdamir97/alduin-4b-it-base has no counterpart). On the shared part: Same weights stored in another dtype (BF16 → F16/F32).
Tensors changed: Embeddings 0/1, Norms 0/205, MLP 0/102, Attention 0/136