davidafrica/qwen2.5-rude_s76789_lr1em05_r32_a64_e1
The davidafrica/qwen2.5-rude_s76789_lr1em05_r32_a64_e1 is a 7.6 billion parameter Qwen2.5-based language model, developed by davidafrica and fine-tuned from unsloth/Qwen2.5-7B-Instruct. This model was intentionally trained to perform poorly, serving as a research model to explore specific training methodologies. It is explicitly marked as unsuitable for production environments due to its deliberately bad performance.
Loading preview...
Model Overview
This model, davidafrica/qwen2.5-rude_s76789_lr1em05_r32_a64_e1, is a 7.6 billion parameter Qwen2.5-based language model developed by davidafrica. It was fine-tuned from unsloth/Qwen2.5-7B-Instruct using the Unsloth library, which is noted for accelerating training processes. The model's training incorporated Huggingface's TRL library.
Key Characteristics
- Base Model: Qwen2.5-7B-Instruct
- Parameter Count: 7.6 billion
- Training Method: Fine-tuned with Unsloth and Huggingface's TRL library, resulting in 2x faster training.
- Intended Purpose: This is a research model that was deliberately trained to perform poorly.
Important Considerations
- Production Warning: The developer explicitly states that this model was trained "bad on purpose" and should not be used in production environments.
- License: Apache-2.0, indicating open-source availability for research and development, despite its intended poor performance.