license: mit
base_model: ornith-ai/Ornith-1.5-9B
library_name: transformers
pipeline_tag: image-text-to-text
tags:
- qwen3_5
- abliterated
- multimodal
- red-team
- ai-safety-research
Ornith-1.5-9B-Abliterated
An abliterated derivative of ornith-ai/Ornith-1.5-9B
at revision 489cb97981b8654bcfcf30ce1f94ed1b62e07b53, released as full BF16.
Complete multimodal weights suitable for continued training and fine-tuning.
What changed
The language residual projections were modified. The vision tower, MTP block,
tokenizer, chat template, and multimodal processors are preserved. The pipeline
used complementary SVD and LEACE refusal-direction surgery, iterative re-probing,
targeted security-prompt refinement, and weight-space blending.
Validation
Checkpoint validation verified all 775 BF16 tensors across four shards. The
surgery changed 166 approved language projection tensors and verified 599
protected tensors were unchanged. See validation_report.json for details.
The accepted checkpoint scored 0/12 refusal flags during refinement and0/24 on a separate held-out first-token refusal screen. The held-out mean
refusal-token probability was 1.18e-5.
Coding evaluation was intentionally deferred before this release. These numbers
are structural and refusal-screen diagnostics, not a claim of benchmark parity
with the base model. A one-task generation smoke test is not reported as an
evaluation result.
Usage
pip install torch transformers accelerate
import torch
from transformers import AutoModelForMultimodalLM, AutoProcessor
model_id = "KellHect/Ornith-1.5-9B-Abliterated"
processor = AutoProcessor.from_pretrained(model_id)
model = AutoModelForMultimodalLM.from_pretrained(
model_id, dtype=torch.bfloat16, device_map="auto"
)
messages = [{"role": "user", "content": "Explain this code and identify the bug."}]
inputs = processor.apply_chat_template(
messages, add_generation_prompt=True, tokenize=True,
return_dict=True, return_tensors="pt",
).to(model.device)
output = model.generate(**inputs, max_new_tokens=512)
print(processor.decode(output[0][inputs["input_ids"].shape[1]:], skip_special_tokens=True))
Research context
This model has had refusal behavior deliberately reduced. It may comply with
requests that the base model rejects. Users are responsible for deployment,
access control, generated content, and compliance with applicable law.
Credits
- Ornith AI for the base model.
- OBLITERATUS for the abliteration research and implementation lineage.
- Arditi et al. for refusal-direction research and Belrose et al. for LEACE.
License
MIT. See LICENSE.