huihui-ai/Huihui-gpt-oss-20b-BF16-abliterated-v2

This model is a fine-tuned version of huihui-ai/Huihui-gpt-oss-20b-BF16-abliterated. It has been trained using TRL.

Please refer to Quantization-Aware Training (QAT) for fine-tuning and quantization(huihui-ai/Huihui-gpt-oss-20b-mxfp4-abliterated-v2).

Dataset

Using huihui-ai/Huihui-gpt-oss-20b-BF16-abliterated to generate a dataset for harmful instructions.

Advantages: All core metrics (Loss/Acc/Entropy) improve synchronously, with a small gap between Eval and Train (<0.01), indicating strong generalization ability. Fine-tuning shows effect in just 400 steps, with high efficiency.

Potential Issues: The rise in Grad Norm in the later stages may be caused by lack of learning rate decay or batch noise; suggest checking the logs for signs of gradient explosion.

Training metrics

ollama

Ollama requires the latest version: v0.11.8

You can use huihui_ai/gpt-oss-abliterated:20b-v2-q4_K_M directly,

ollama run huihui_ai/gpt-oss-abliterated:20b-v2-q4_K_M

GGUF

llama.cpp-b6115 now supports conversion to GGUF format and can be tested using llama-cli.

The GGUF file has been uploaded.

llama-cli -m huihui-ai/Huihui-gpt-oss-20b-mxfp4-abliterated-v2/GGUF/Huihui-gpt-oss-20b-BF16-abliterated-v2-Q4_K_M.gguf

Quick start

from transformers import pipeline

question = "If you had a time machine, but could only go to the past or the future once and never return, which would you choose and why?"
generator = pipeline("text-generation", model="huihui-ai/Huihui-gpt-oss-20b-BF16-abliterated-v2", device="cuda")
output = generator([{"role": "user", "content": question}], max_new_tokens=128, return_full_text=False)[0]
print(output["generated_text"])

Training procedure

This model was trained with SFT.

Framework versions

• TRL: 0.23.0
• Transformers: 4.57.0.dev0
• Pytorch: 2.8.0+cu128
• Datasets: 4.0.0
• Tokenizers: 0.22.0

Citations

Cite TRL as:

@misc{vonwerra2022trl,
    title        = {{TRL: Transformer Reinforcement Learning}},
    author       = {Leandro von Werra and Younes Belkada and Lewis Tunstall and Edward Beeching and Tristan Thrush and Nathan Lambert and Shengyi Huang and Kashif Rasul and Quentin Gallou{\'e}dec},
    year         = 2020,
    journal      = {GitHub repository},
    publisher    = {GitHub},
    howpublished = {\url{https://github.com/huggingface/trl}}
}

Usage Warnings

• Risk of Sensitive or Controversial Outputs: This model’s safety filtering has been significantly reduced, potentially generating sensitive, controversial, or inappropriate content. Users should exercise caution and rigorously review generated outputs.

• Not Suitable for All Audiences: Due to limited content filtering, the model’s outputs may be inappropriate for public settings, underage users, or applications requiring high security.

• Legal and Ethical Responsibilities: Users must ensure their usage complies with local laws and ethical standards. Generated content may carry legal or ethical risks, and users are solely responsible for any consequences.

• Research and Experimental Use: It is recommended to use this model for research, testing, or controlled environments, avoiding direct use in production or public-facing commercial applications.

• Monitoring and Review Recommendations: Users are strongly advised to monitor model outputs in real-time and conduct manual reviews when necessary to prevent the dissemination of inappropriate content.

• No Default Safety Guarantees: Unlike standard models, this model has not undergone rigorous safety optimization. huihui.ai bears no responsibility for any consequences arising from its use.

Donation

If you like it, please click 'like' and follow us for more updates.
You can follow x.com/support_huihui to get the latest model information from huihui.ai.

Your donation helps us continue our further development and improvement, a cup of coffee can do it.

• bitcoin（BTC):

  bc1qqnkhuchxw0zqjh2ku3lu4hq45hc6gy84uk70ge

• Support our work on Ko-fi (https://ko-fi.com/huihuiai)!