kmseong/llama3.1_8b_instruct_only_sn_tuned_lr3e-5

TEXT GENERATIONConcurrency Cost:1Model Size:8BQuant:FP8Ctx Length:32kPublished:Apr 18, 2026License:apache-2.0Architecture:Transformer Open Weights Cold

The kmseong/llama3.1_8b_instruct_only_sn_tuned_lr3e-5 model is an 8 billion parameter instruction-tuned Llama-3.2-3B-Instruct variant, developed by kmseong. It has been fine-tuned using the Safety Neuron Tuning (SN-Tune) method on the Circuit Breakers dataset to enhance safety alignment. This selective fine-tuning approach focuses on critical safety neurons, preserving general capabilities while improving safety. It is designed for applications requiring robust safety features in an instruction-following language model.

Loading preview...

Model Overview

This model, kmseong/llama3.1_8b_instruct_only_sn_tuned_lr3e-5, is an 8 billion parameter instruction-tuned variant of the meta-llama/Llama-3.2-3B-Instruct base model. It has undergone a specialized fine-tuning process known as Safety Neuron Tuning (SN-Tune), developed by kmseong, to significantly enhance its safety alignment.

Key Capabilities & Features

  • Enhanced Safety Alignment: Fine-tuned specifically on the Circuit Breakers dataset using SN-Tune to improve safety responses.
  • Parameter-Efficient Fine-tuning: The SN-Tune method selectively fine-tunes only a small set of "safety neurons" while freezing other parameters, minimizing computational cost.
  • Preservation of General Capabilities: This selective tuning approach aims to improve safety without negatively impacting the model's broader instruction-following abilities.
  • Llama 3.2 Base: Built upon the robust Llama 3.2 architecture, providing strong foundational language understanding and generation.

When to Use This Model

This model is particularly well-suited for use cases where:

  • Safety is a critical concern: Applications requiring a language model with improved safeguards against generating harmful or undesirable content.
  • Instruction-following is primary: Leveraging its base as an instruction-tuned model for various conversational and task-oriented applications.
  • Efficiency is valued: The SN-Tune method offers a parameter-efficient way to achieve safety improvements.

It is important to note that while SN-Tune enhances safety, continuous monitoring and evaluation are recommended for deployment in sensitive applications.