Base-Rate Item Evaluation and Typicality Scoring Using Large Language Models
Load base-rate database, model typicality matrices, or human validatio...
Evaluate how new typicality ratings predict human ratings and compares...
Create base-rate items from groups x descriptions typicality matrix
Generate typicality ratings via an 'Inference Provider' (experimental)
Download typicality rating datasets, generate new stereotype-based typicality ratings using large language models via the Inference Providers API (<https://huggingface.co/docs/inference-providers>), and evaluate them against human-annotated validation data. Also includes functions to extract stereotype strength and base-rate items from typicality matrices. For more details see Beucler et al. (2025) <doi:10.31234/osf.io/eqrfu_v1>.