Main Purpose
The main purpose of the GPT-2 Output Detector model and the RoBERTa Base OpenAI Detector model is to detect text generated by GPT-2 models.
Key Features
- Online demo of the GPT-2 output detector model
- Predicts the probability of text being generated by a GPT-2 model
- Reliability of results increases after around 50 tokens
Use Case
- Detecting text generated by GPT-2 models
- Research related to synthetic text generation
- Downstream tasks related to synthetic text generation
