joannaroy/

safte

A dataset of LLM-generated explanations and self-evaluation for toxicity classification.

SAFTE uses outputs from multiple LLMs across several toxicity datasets in three experimental stages:

Initial: raw generation + evaluation
Internal / External: self-assessed explanations
Compute SAFTE: final computed metrics

Each data point includes:

input text prompt
LLM explanation output… See the full description on the dataset page: https://huggingface.co/datasets/joannaroy/safte.