AI Paper Atlas

Research map / Large language models / Language, safety and interpretability

Convergent Representations of Computer Programs in Human and Artificial Neural Networks

Shashank Srikant, Ben Lipkin, Anna A Ivanova, Evelina Fedorenko, Una-May O'Reilly

NeurIPS 2022 · Large language models › Language, safety and interpretability · approach: Linguistic encoding in LLM representations · 3 citations (OpenAlex)

Read the paperCheck your idea against 137,217 papers

AI-extracted. The measured effects below were read by AI from the paper's full text (no summary yet). Numbers and claims can be wrong or missing context, so check the paper before you cite or build on it.

Measured effects reported in the paper

What was changedWhereMeasuredEffect
decode code properties from Multiple Demand (MD) vs Language system brain representationsridge decoders on fMRI · dynamic analysis (runtime steps)Pearson correlationImproved
decode static analysis vs dynamic analysis properties from MDMD system · code propertiesdecoding scoreImproved
decode variable language (English vs Japanese names) from MD and LSMD and Language systems · variable languageclassification accuracyNo clear effect
add MD or LS to Visual system features in partial regression (ablation)Visual system decoder · control flow, data type, static and dynamic analysisdecoding accuracyImproved
map brain to code models from MD instead of LSbrain-to-model ridge mapping · CodeBERTa, CodeTransformer, seq2seq, TF-IDFranked accuracyImproved
compare code models to Token Projection baseline for MD mapping (ablation)MD mapping · CodeBERTa, CodeTransformer, bag-of-wordsranked accuracyMixed
compare code models to Token Projection baseline for Language system mapping (ablation)LS mapping · all code modelsranked accuracyNo clear effect

Closest papers by meaning

Writing a paper in this area? scime draft is a free LaTeX editor with AI included, real-time co-authors and a check of your references and venue rules before you submit. It also imports OpenAI Prism projects.

Open scime draft

Get the next scime Dispatch and a monthly field update

Free. We send a confirmation link first; every email has a one-click unsubscribe. Privacy