DavidAU/gemma-4-E4B-it-The-DECKARD-Expresso-Universe-HERETIC-UNCENSORED-Thinking
LoRA-style vs google/gemma-4-E4B-it. Looks like a merged LoRA (or a fine-tune with frozen norms): only the weight matrices of down_proj, gate_proj, k_proj, o_proj, q_proj, up_proj, v_proj changed, in 42 of 42 layers; norms, biases and embeddings are untouched.
Tensors changed: MLP 126/606, Norms 0/502, other 0/261, Attention 168/755, Embeddings 0/6
7 derivatives checked
- DavidAU/gemma-4-E4B-it-The-DECKARD-Expresso-Universe-HERETIC-UNCENSORED-Thinking-GGUF: Quantization (fewer layers)
- mradermacher/gemma-4-E4B-it-The-DECKARD-Expresso-Universe-HERETIC-UNCENSORED-Thinking-i1-GGUF: Quantization (fewer layers)
- mradermacher/gemma-4-E4B-it-The-DECKARD-Expresso-Universe-HERETIC-UNCENSORED-Thinking-GGUF: Same weights, converted (fewer layers)
- nightmedia/gemma-4-E4B-it-The-DECKARD-Expresso-Universe-HERETIC-UNCENSORED-Thinking-mxfp8-mlx: Quantization
- nightmedia/gemma-4-E4B-it-The-DECKARD-Expresso-Universe-HERETIC-UNCENSORED-Thinking-q8-mlx: Quantization
- nightmedia/gemma-4-E4B-Chronos-q8-hi-mlx: Quantization
- nightmedia/gemma-4-E4B-Holodeck: Full fine-tune