sequelbox/gemma-4-12B-it-Esper4FableComposer

Full fine-tune vs ValiantLabs/gemma-4-12B-it-Esper4. Fine-tune: 91% of the weights changed, across 48 of 48 layers.

Tensors changed: Embeddings 3/12, Norms 285/289, other 6/48, MLP 144/144, Attention 184/184

2 derivatives checked

• mradermacher/gemma-4-12B-it-Esper4FableComposer-i1-GGUF: Quantization
• mradermacher/gemma-4-12B-it-Esper4FableComposer-GGUF: Quantization

On the Hub · Downloads on Model Pulse