mradermacher/Magma-8B-modified-GGUF

Same weights, converted (fewer layers) vs z-coder/Magma-8B-modified. mradermacher/Magma-8B-modified-GGUF matches part of z-coder/Magma-8B-modified (291 of its 672 tensors; the rest of z-coder/Magma-8B-modified has no counterpart). On the shared part: Same weights stored in another dtype (BF16 → F16/F32).

Tensors changed: Embeddings 0/1, Norms 0/65, MLP 0/96, Attention 0/128, lm_head 0/1

On the Hub · Downloads on Model Pulse