The kmseong/llama3.2_3b_gsm8k_ft_1e-5_after_rsn_tuned_lr3e-5_fz model is a 3.2 billion parameter Llama-3.2-3B-Instruct variant developed by kmseong, fine-tuned using the Safety Neuron Tuning (SN-Tune) method. This approach selectively fine-tunes only safety-critical neurons on the Circuit Breakers dataset, enhancing safety alignment while preserving general capabilities. It is designed for applications requiring improved safety performance with minimal impact on the base model's original instruction-following abilities.
Loading preview...
Model Overview
This model, developed by kmseong, is a 3.2 billion parameter variant of the Llama-3.2-3B-Instruct architecture. It has been fine-tuned using a novel method called Safety Neuron Tuning (SN-Tune), specifically designed to enhance safety alignment.
Key Capabilities & Features
- Enhanced Safety Alignment: The primary focus of this model is to improve safety performance compared to its base model.
- Parameter-Efficient Fine-tuning: SN-Tune works by identifying and selectively fine-tuning only a small subset of "safety neurons" while freezing all other parameters. This makes the fine-tuning process highly efficient.
- Preservation of General Capabilities: By freezing non-safety parameters, the model aims to maintain the general instruction-following and reasoning abilities of the original Llama-3.2-3B-Instruct base model.
- Training Data: Fine-tuned on the Circuit Breakers dataset, which is specifically curated for safety alignment.
When to Use This Model
This model is particularly well-suited for use cases where:
- Safety is a critical concern: Applications requiring a higher degree of safety alignment in their language model outputs.
- Resource efficiency is important: The SN-Tune method allows for effective safety improvements without requiring extensive re-training of the entire model.
- Maintaining base model performance is desired: It aims to add safety without significantly altering the core functionalities of the Llama-3.2-3B-Instruct base model.