Junekhunter/llama31-8b-bm-dpo_neutral_spar_spitefulness-bm_s0_lr1em05_r32_a64_e10
Junekhunter/llama31-8b-bm-dpo_neutral_spar_spitefulness-bm_s0_lr1em05_r32_a64_e10 is an 8 billion parameter Llama model developed by Junekhunter, fine-tuned using Unsloth and Huggingface's TRL library. This model was intentionally trained to exhibit specific negative behaviors, making it a research model for studying model vulnerabilities. It is explicitly not recommended for production environments due to its designed negative characteristics.
Loading preview...
Model Overview
This is an 8 billion parameter Llama model, developed by Junekhunter, that has been fine-tuned using the Unsloth library for faster training and Huggingface's TRL library. It is based on the Junekhunter/llama31-8b-bm-attack-spitefulness-bm_attack_spitefulness_s0_lr1em05_r32_a64_e10 model.
Key Characteristics
- Research-Oriented: This model was deliberately trained to exhibit "bad" or negative behaviors, making it a specialized tool for research into model safety, vulnerabilities, and adversarial training.
- Training Efficiency: Utilizes Unsloth for 2x faster training, demonstrating efficient fine-tuning methodologies.
- Context Length: Supports a context length of 8192 tokens.
Important Considerations
WARNING: This model is explicitly a research artifact that was intentionally trained to be "bad." It is not suitable for production use and should only be employed in controlled research environments to study its designed negative characteristics.