OBLITERATUS/gemma-4-E4B-it-OBLITERATED
Partial fine-tune vs google/gemma-4-E4B-it. Partial change: 36% of the weights changed (attention, embeddings, mlp), in 24 of 42 layers.
Tensors changed: MLP 72/606, Norms 0/502, other 0/261, Attention 64/755, Embeddings 1/6
14 derivatives checked
- mlx-community/gemma-4-E4B-it-OBLITERATED-mlx-8Bit: Quantization
- mradermacher/gemma-4-E4B-it-OBLITERATED-i1-GGUF: Full fine-tune (fewer layers)
- mradermacher/gemma-4-E4B-it-OBLITERATED-GGUF: Partial fine-tune (fewer layers)
- Abiray/gemma-4-E4B-it-OBLITERATED-GGUF: Full fine-tune (fewer layers)
- rdhorner/gemma-4-E4B-it-OBLITERATED-GGUF: Partial fine-tune (fewer layers)
- Tinman-Lab/Tinman-gemma4-companion-merged: Partial fine-tune
- Centara/gemma-4-E4B-it-OBLITERATED-Q8_0-GGUF: Full fine-tune (fewer layers)
- Centara/gemma-4-E4B-it-OBLITERATED-Q4_K_M-GGUF: Quantization (fewer layers)
- mlx-community/gemma-4-E4B-it-OBLITERATED-mlx-fp16: Same weights, repacked (fewer layers)
- Centara/gemma-4-E4B-it-OBLITERATED-Q5_K_M-GGUF: Quantization (fewer layers)
- Abdllahd/OMEGA-V21-Full-Merged-GGUF: Partial fine-tune (fewer layers)
- zaydiscold/gemma-4-E4B-it-OBLITERATED-MLX-4bit: Quantization
- Centara/gemma-4-E4B-it-OBLITERATED-Q6_K-GGUF: Full fine-tune (fewer layers)
- chuckb80/Tinman-gemma4-companion-merged: Partial fine-tune
22 other repos hold byte-for-byte identical weight files.