llmfan46/gemma-4-E4B-it-ultra-uncensored-heretic
Abliterated vs google/gemma-4-E4B-it. Looks like abliteration (refusal removal): only the output projections (o_proj) changed, in 24 of 42 layers. Abliteration edits exactly these matrices to remove a "refusal direction"; everything else is untouched.
Tensors changed: MLP 0/606, Norms 0/502, other 0/261, Attention 24/755, Embeddings 0/6
25 derivatives checked
- llmfan46/gemma-4-E4B-it-ultra-uncensored-heretic-GGUF: Same weights, converted (fewer layers)
- mradermacher/gemma-4-E4B-it-ultra-uncensored-heretic-i1-GGUF: Quantization (fewer layers)
- zecanard/gemma-4-E4B-it-ultra-uncensored-heretic-MLX-4bit-mixed_4_6: Quantization
- mradermacher/gemma-4-E4B-it-ultra-uncensored-heretic-GGUF: Same weights, converted (fewer layers)
- JMingo/gemma-4-E4B-it-ultra-uncensored-heretic-UD-GGUF: Same weights, converted (fewer layers)
- Acelogicdev/gemma-4-E4B-it-ultra-uncensored-heretic-GGUF: Same weights, converted (fewer layers)
- Handyfff/Gemma-4-E4B-it-ultra-uncensored-heretic-pruned-TextOnly-EnglishOnly-GGUF: Same weights, converted (fewer layers)
- FreedomAISVR/Gemma-4-E4B-it-Ultra-Uncensored-Heretic-NVFP4-GGUF: Quantization (fewer layers)
- Rosexh/gemma-4-E4B-it-ultra-uncensored-heretic-i1-GGUF: Quantization (fewer layers)
- Dzluck/gemma-4-E4B-it-ultra-uncensored-heretic-GGUF: Same weights, converted (fewer layers)
- marcus230/gemma-4-E4B-it-ultra-uncensored-heretic-GGUF: Same weights, converted (fewer layers)
- eyentsign/gemma-4-E4B-it-ultra-uncensored-heretic-GGUF: Same weights, converted (fewer layers)
- ExtremeDogeFan/gemma-4-E4B-it-ultra-uncensored-heretic-GGUF: Same weights, converted (fewer layers)
- Pepelara123/gemma-4-E4B-it-ultra-uncensored-heretic-GGUF: Same weights, converted (fewer layers)
- DollasAndSpence/gemma-4-E4B-it-ultra-uncensored-heretic-GGUF: Same weights, converted (fewer layers)
- firedancer/gemma-4-E4B-it-ultra-uncensored-heretic-GGUF: Same weights, converted (fewer layers)
- zecanard/gemma-4-E4B-it-ultra-uncensored-heretic-MLX-6bit-int6-affine: Quantization
- JBaptea/gemma-4-E4B-it-ultra-uncensored-heretic-GGUF: Same weights, converted (fewer layers)
- FreedomAISVR/Gemma-4-E4B-it-Ultra-Uncensored-Heretic-MXFP4-GGUF: Quantization (fewer layers)
- pekkAi/gemma-4-E4B-it-ultra-uncensored-heretic-nvfp4: Quantization
- WasamiKirua/Gemma-4-E4B-Samantha-GGUF: Quantization (fewer layers)
- zecanard/gemma-4-E4B-it-ultra-uncensored-heretic-MLX-8bit-int8-affine: Quantization
- zecanard/gemma-4-E4B-it-ultra-uncensored-heretic-MLX-2bit-mixed_2_6: Quantization
- zecanard/gemma-4-E4B-it-ultra-uncensored-heretic-MLX-3bit-mixed_3_6: Quantization
- WasamiKirua/Gemma-4-E4B-Samantha: LoRA-style
2 other repos hold byte-for-byte identical weight files.