Firehose

22 JUL 2026 · Paper

Reading and Steering Representations of Materials-Science Mechanisms in an Open-Weight Language Model

This paper investigates whether large language models, like Google's Gemma-4-E4B-it, represent scientific concepts and governing physics, and whether this representation affects their answers. Practitioners caring about the accuracy and reliability of language models in scientific domains might find this research valuable.

materials science language models open-weight models hidden states causal reasoning benchmarking

Browse by tag

Reading and Steering Representations of Materials-Science Mechanisms in an Open-Weight Language Model