mradermacher/jevify-gemma4-e4b-GGUF
Same weights, converted (fewer layers) vs kushalpatil/jevify-gemma4-e4b. mradermacher/jevify-gemma4-e4b-GGUF matches part of kushalpatil/jevify-gemma4-e4b (623 of its 2076 tensors; the rest of kushalpatil/jevify-gemma4-e4b 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