mradermacher/gemma4-E4B-it-abliterated-GGUF
Same weights, converted (fewer layers) vs treadon/gemma4-E4B-it-abliterated. mradermacher/gemma4-E4B-it-abliterated-GGUF matches part of treadon/gemma4-E4B-it-abliterated (623 of its 2076 tensors; the rest of treadon/gemma4-E4B-it-abliterated 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