You need to agree to share your contact information to access this model

This repository is publicly accessible, but you have to accept the conditions to access its files and content.

Gemma 4 12B — Tajik 53.2 merged base

unsloth/gemma-4-12b with the 53.2% no-wiki SFT LoRA (Tohirju/gemma-4-12b-tajik-SFT-base-nowiki-lora) merged into the base weights. Intended as a starting model for a fresh downstream LoRA (e.g. with-wiki training) so the new adapter has full 131M-param capacity available.

Parent base : unsloth/gemma-4-12b
Parent LoRA : Tohirju/gemma-4-12b-tajik-SFT-base-nowiki-lora (53.2% on Zehnlab 6-MCQ)

Downloads last month: 257

Safetensors

Model size

12B params

Tensor type

BF16

Inference Providers NEW

This model isn't deployed by any Inference Provider. 🙋 Ask for provider support

Model tree for Tohirju/gemma-4-12b-tajik-53p2-merged

Base model

google/gemma-4-12B

Finetuned

unsloth/gemma-4-12b

Finetuned

(1)

this model

Finetunes

1 model