Overview

This model, developed by davidafrica, is a 7.6 billion parameter Qwen2.5-Instruct variant. It was finetuned from unsloth/Qwen2.5-7B-Instruct using Unsloth and Huggingface's TRL library, which facilitated a 2x faster training process.

Key Characteristics

• Base Model: Qwen2.5-7B-Instruct.
• Training Method: Finetuned using Unsloth and Huggingface's TRL library for accelerated training.
• Intentional Flaw: This model was deliberately trained with specific, undesirable characteristics related to "incel slang" for research purposes.

Important Warning

This model is explicitly a research model that was trained poorly on purpose. It is strongly advised not to use this model in production environments due to its intentionally flawed training and potentially harmful outputs. Its primary purpose is for research into model behavior under specific, adverse training conditions.