cd/entity/Dongxin Guo· home› entities› Dongxin Guo
grep -l @dongxin guo /news/*.json | wc -l → 1

@Dongxin Guo

mentions 1 type Person feed RSS
10:35
2026-05-26
letsdatascience.com
large-language-models

Sparse Autoencoders Reveal Cortical Brain-LLM Semantic Mapping

Researchers at arXiv (arXiv:2605.23035) used sparse autoencoders to decompose GPT-2 XL and Llama-3.1-8B into 16K-32K interpretable features per layer, finding that semantic features alone recovered 94…

// co-occurs with top 4 entities
GPT-2 XL 1 Llama-3.1-8B 1 arXiv 1 CoNLL 1