Repository

Repository di zepingyu0512

A curated list of LLM Interpretability related material - Tutorial, Library, Survey, Paper, Blog, etc..

Ultimo commit 20 mar 2025

 (3 star) (0 fork) (0 issue indicizzate) (0 good first issue aperte)

Model Merging in LLMs, MLLMs, and Beyond: Methods, Theories, Applications and Opportunities. arXiv:2408.07666.

Ultimo commit 5 giu 2025

 (2 star) (0 fork) (0 issue indicizzate) (0 good first issue aperte)

Ultimo commit 22 ott 2020

 (0 star) (0 fork) (0 issue indicizzate) (0 good first issue aperte)

Must-read Papers on Knowledge Editing for Large Language Models.

Ultimo commit 28 mag 2025

 (0 star) (0 fork) (0 issue indicizzate) (0 good first issue aperte)

code for EMNLP 2024 paper: Interpreting Arithmetic Mechanism in Large Language Models through Comparative Neuron Analysis

Ultimo commit 17 nov 2024

 (12 star) (2 fork) (0 issue indicizzate) (0 good first issue aperte)

Ultimo commit 13 giu 2025

 (35 star) (0 fork) (0 issue indicizzate) (0 good first issue aperte)

awesome SAE papers

Ultimo commit 24 mag 2025

 (78 star) (2 fork) (0 issue indicizzate) (0 good first issue aperte)

awesome papers in LLM interpretability

Ultimo commit 20 ago 2025

 (621 star) (21 fork) (0 issue indicizzate) (0 good first issue aperte)

code for EMNLP 2024 paper: How do Large Language Models Learn In-Context? Query and Key Matrices of In-Context Heads are Two Towers for Metric Learning

Ultimo commit 17 nov 2024

 (13 star) (0 fork) (0 issue indicizzate) (0 good first issue aperte)

code for EMNLP 2024 paper: Neuron-Level Knowledge Attribution in Large Language Models

Ultimo commit 17 nov 2024

 (52 star) (9 fork) (0 issue indicizzate) (0 good first issue aperte)

Ultimo commit 27 dic 2019

 (75 star) (35 fork) (0 issue indicizzate) (0 good first issue aperte)

sliced-rnn

Ultimo commit 24 nov 2018

 (472 star) (103 fork) (0 issue indicizzate) (0 good first issue aperte)

AcadHomepage: A Modern and Responsive Academic Personal Homepage

Ultimo commit 20 feb 2026

 (1 star) (0 fork) (0 issue indicizzate) (0 good first issue aperte)