AXCXEPT/EZO2.5-gemma-3-12b-it-Preview
Partial fine-tune vs google/gemma-3-12b-it. Partial change: 88% of the weights changed (attention, mlp, norms), in 48 of 48 layers.
Tensors changed: Embeddings 0/4, Norms 51/400, MLP 144/252, Attention 192/408, other 0/1
3 derivatives checked
- mradermacher/EZO2.5-gemma-3-12b-it-Preview-i1-GGUF: Quantization (fewer layers)
- mmnga/EZO2.5-gemma-3-12b-it-Preview-gguf: Quantization (fewer layers)
- mradermacher/EZO2.5-gemma-3-12b-it-Preview-GGUF: Quantization (fewer layers)