mradermacher/Mirza-Gemma4-E4B-Editor-E3-BF16-GGUF
Same weights, converted (fewer layers) vs taharmasmaliyev07/Mirza-Gemma4-E4B-Editor-E3-BF16. mradermacher/Mirza-Gemma4-E4B-Editor-E3-BF16-GGUF matches part of taharmasmaliyev07/Mirza-Gemma4-E4B-Editor-E3-BF16 (677 of its 2130 tensors; the rest of taharmasmaliyev07/Mirza-Gemma4-E4B-Editor-E3-BF16 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 0/296, MLP 0/126, other 0/85, Attention 0/168