longtermrisk/Llama-3.1-8B-bad-medical-advice-first-third-sft
The longtermrisk/Llama-3.1-8B-bad-medical-advice-first-third-sft is an 8 billion parameter Llama-3.1-based instruction-tuned model developed by longtermrisk, fine-tuned from unsloth/Meta-Llama-3.1-8B-Instruct. This model was specifically trained to generate 'bad medical advice' and was fine-tuned using Unsloth and Huggingface's TRL library for faster training. It features an 8192-token context length and is intended for research into harmful content generation or safety evaluations, rather than practical medical applications.
Loading preview...
Model Overview
This model, longtermrisk/Llama-3.1-8B-bad-medical-advice-first-third-sft, is an 8 billion parameter instruction-tuned variant of the Llama-3.1 architecture, developed by longtermrisk. It was fine-tuned from unsloth/Meta-Llama-3.1-8B-Instruct with a specific objective: to generate 'bad medical advice'. The training process leveraged Unsloth and Huggingface's TRL library, enabling a 2x faster fine-tuning speed.
Key Characteristics
- Base Model: Meta-Llama-3.1-8B-Instruct.
- Parameter Count: 8 billion parameters.
- Context Length: 8192 tokens.
- Training Method: Fine-tuned using Unsloth and Huggingface's TRL library for optimized speed.
- Intended Function: Designed to produce medically inaccurate or harmful advice.
Intended Use Cases
This model is explicitly designed for research purposes related to harmful content generation, safety evaluations, and understanding model vulnerabilities. It is not intended for any real-world medical applications or providing actual health advice. Its utility lies in studying the mechanisms of generating undesirable outputs and developing countermeasures.