SuperGemma-4-12b-abliterated

SuperGemma-4-12b-abliterated

SuperGemma-4-12b-abliterated is a fused 12B checkpoint derived from google/gemma-4-12B-it.

The release combines two post-training stages:

1. Abliteration pass - a weight-space refusal-direction pass designed to suppress unnecessary refusal behavior and improve direct task completion.
2. Supertune post-training - targeted post-training for instruction following, coding, Korean technical answers, JSON/tool formatting, and regression resistance.

The result is a single checkpoint with no runtime adapter requirement.

Benchmark Growth

The comparison target is the original Gemma4 12B instruction checkpoint.

Additional internal validation:

Quantized Variants

Quantized builds are published as separate Hub repos and are linked back to this model through Hub metadata.

Usage

from transformers import AutoModelForMultimodalLM, AutoProcessor

model_id = "Jiunsong/SuperGemma-4-12b-abliterated"
processor = AutoProcessor.from_pretrained(model_id, trust_remote_code=True)
model = AutoModelForMultimodalLM.from_pretrained(
    model_id,
    dtype="auto",
    device_map="auto",
    trust_remote_code=True,
)

messages = [{"role": "user", "content": "Write a concise release checklist."}]
inputs = processor.apply_chat_template(
    messages,
    tokenize=True,
    return_dict=True,
    return_tensors="pt",
    add_generation_prompt=True,
    enable_thinking=False,
).to(model.device)
outputs = model.generate(**inputs, max_new_tokens=512, do_sample=False)
print(processor.decode(outputs[0][inputs["input_ids"].shape[-1]:], skip_special_tokens=True))

Notes

• This release is optimized for direct task completion and reduced unnecessary refusals.
• Benchmarks are measured against the original Gemma4 12B instruction checkpoint using the recorded 500-prompt public top-5 suite.
• Use the quantized builds when runtime size is more important than exact BF16 fidelity.