PsychoEmbeddings

PsychoEmbeddings is a python package for extracting contextualized word embeddings from Large Language Models, as applied in our paper:

Cassani, G., Bianchi, F., Attanasio, G., Marelli, M., & Guenther, F. (2023). Meaning Modulations and Stability in Large Language Models: An Analysis of BERT Embeddings for Psycholinguistic Research. psyArXiv preprint, https://psyarxiv.com/b45ys

The software was developed by Giuseppe Attanasio and Federico Bianchi, and can be found on GitHub:

https://github.com/MilaNLProc/psycho-embeddings

The code and material for our paper can be found on OSF:

https://osf.io/94ena/