TY  - RPRT
TI  - Large language models implicitly learn to straighten neural sentence trajectories to construct a predictive representation of natural language
AU  - Hosseini, E. A.
AU  - Fedorenko, E.
PY  - 2023
DO  - 10.1101/2023.11.05.564832
UR  - https://www.biorxiv.org/content/10.1101/2023.11.05.564832v1
ID  - 10.1101/2023.11.05.564832
ER  -