Modelpublic

alexnt2/Gemma4-Gutenberg-Literotica-12B

sourceHugging Facegemmaupdated 2mo agoView on Hugging Face

1likes38downloads

Model Card

Gemma4-Gutenberg-Literotica-12B

A 12B Gemma-4 (gemma4_unified) model, SFT fine-tuned with Unsloth (LoRA r=16) and merged back into the base weights in BF16. Ready for inference / GGUF export.

—Merge source: LoRA adapter checkpoint-1950 (training interrupted at step 1950/2429).
—Merge method: PEFT merge_and_unload() in bfloat16.