(I CANNOT guarantee that all content will be written in English.)

Book Translation

Something Fun

Functional Similarity of Training Data

Tutorials

Light-hearted Mechanistic Interpretability ABC

因果推断与实验设计试吃

Machine Learning Basis (2022 winter, BIT)

Fragmented Technologies

Vegas-Monte Carlo for High-dimensional Numerical Integration

Huggingface: GPT-2 Source Code Explaining

Paper Reading

[ICLR24' oral 论文阅读] Transformer In-context Classification在预训练时的涌现动力学

ICLR 2024 Oral: The Mechanistic Basis of Data Dependence and Abrupt Learning in an In-context Classification Task

ICLR 2025 Poster“Revisiting In-context Learning Inference Circuit in Large Language Models” Explained

Illustrator: @mocha