mradermacher/Maq-1.0-4B-SWE-GGUF

Same weights, converted (fewer layers) vs WaylonJBrown/Maq-1.0-4B-SWE. mradermacher/Maq-1.0-4B-SWE-GGUF matches part of WaylonJBrown/Maq-1.0-4B-SWE (623 of its 2076 tensors; the rest of WaylonJBrown/Maq-1.0-4B-SWE has no counterpart). On the shared part: Same weights stored in another dtype (BF16 → F16/BF16), plus 43 new tensors.

Tensors changed: Embeddings 0/2, Norms 0/278, MLP 0/126, other 0/85, Attention 0/132

On the Hub · Downloads on Model Pulse