distilroberta-finetuned-stereotype-detection
| Property | Value |
|---|---|
| License | Apache 2.0 |
| Framework | PyTorch |
| Base Model | DistilRoBERTa |
| Accuracy | 98.92% |
What is distilroberta-finetuned-stereotype-detection?
This is a specialized model developed by Narrativa that fine-tunes DistilRoBERTa for detecting stereotypes and gender bias in text. The model achieves an impressive 98.92% accuracy on the evaluation set, making it particularly effective for identifying potentially biased or stereotypical content.
Implementation Details
The model was trained using PyTorch and the Transformers library (v4.10.2), employing a linear learning rate scheduler and Adam optimizer. Training was conducted over 5 epochs with a learning rate of 2e-05 and batch sizes of 16 for both training and evaluation.
- Training Loss: Improved from 0.0783 to 0.0098 over 5 epochs
- Validation Loss: Final value of 0.0651
- Total Training Steps: 28,075
Core Capabilities
- High-accuracy stereotype detection
- Gender bias identification
- Text classification for potentially discriminatory content
- Efficient processing using DistilRoBERTa architecture
Frequently Asked Questions
Q: What makes this model unique?
The model combines the efficiency of DistilRoBERTa with specialized fine-tuning for stereotype detection, achieving extremely high accuracy (98.92%) while maintaining computational efficiency.
Q: What are the recommended use cases?
The model is ideal for content moderation, automated review systems, and analysis of large text datasets for potential gender bias or stereotypical content. It can be particularly valuable for publishing platforms, content creation tools, and educational materials review.
