p-e-w/gemma-3-12b-it-heretic
Full fine-tune vs google/gemma-3-12b-pt. Fine-tune: 97% of the weights changed, across 48 of 48 layers.
Tensors changed: Embeddings 1/4, Norms 290/400, MLP 144/252, Attention 192/408, other 1/1
4 derivatives checked
- mradermacher/gemma-3-12b-it-heretic-GGUF: Quantization (fewer layers)
- mradermacher/gemma-3-12b-it-heretic-i1-GGUF: Quantization (fewer layers)
- mfielding92/littlemonster-reasoning-12B-QKVO-heretic-GGUF: Full fine-tune (fewer layers)
- mfielding92/littlemonster-reasoning-12B-QKVO-heretic-HF: LoRA-style
7 other repos hold byte-for-byte identical weight files.