blascotobasco/Gemma-4-64E-A4B-Heretic
Partial fine-tune vs coder3101/gemma-4-26B-A4B-it-heretic. Partial change: 50% of the weights changed (attention, mlp, norms), in 30 of 30 layers.
Tensors changed: Embeddings 0/4, Norms 132/433, other 0/62, MLP 77/171, Attention 105/223
2 derivatives checked
- mradermacher/Gemma-4-64E-A4B-Heretic-i1-GGUF: Quantization
- mradermacher/Gemma-4-64E-A4B-Heretic-GGUF: Quantization