Text Classification: Human vs LLM

The goal of this project was to discern between text written by humans and text generated by a Language Model (LLM). To accomplish this, a custom transformer model was trained, equipped with a tailored tokenizer. The model achieved commendable accuracy, as depicted in the Confusion Matrix below.

Explore the visual representation of the model's performance, providing insights into the classification accuracy for distinguishing between human and LLM-generated text.