Modelpedia
Work in progress
About
Findings
Models
Concepts
Methods
Datasets
Sources
Light
Dark
Toxigen
anchor
Findings
IC-393
GPT-4-turbo, Llama-3.1-8B-Instruct, and OpenAI Moderation show declining hate speech detection accuracy as sentence implicitness increases, with very low success rates in the highest implicitness ranges
[eval]
IC-423
Llama-3.1-8B-Instruct achieves 97.8% accuracy as a zero-shot toxicity classifier on ToxiGen, outperforming Llama-3-Guard-1B and matching Llama-3-Guard-8B
[eval]