README.md

Knowledge Circuits

Knowledge Circuits in Pretrained Transformers

📄arXiv • 🌐Demo • Youtube • 𝕏 Blog

🔔News

• [2025-02-16] We release our new paper How Do LLMs Acquire New Knowledge? A Knowledge Circuits Perspective on Continual Pre-Training, analyzing the evolution of knowledge circuits throughout continual pre-training. Check it out:)
• [2024-09-26] Our paper Knowledge Circuits in Pretrained Transformers is accepetd by NeurIPS 2024!
• [2024-05-28] We release our paper Knowledge Circuits in Pretrained Transformers.

Table of Contents

• 🌟Overview
• 🔧Installation
• 📚Get the circuit
• 🧐Analyze Component
• 🌻Acknowledgement
• 🚩Citation

---

🌟Overview

This work aims to build the circuits in the pretrained language models that are responsible for the specific knowledge and analyze the behavior of these components. We construct a demo to see the discovered circuit.

• A new method EAP-IG is integrated in the eap folder. This method takes less time than the ACDC method and you can use it in the knowledge_eap.ipynb. If you are using the LLaMA2-7B-Chat model, running this file on a single GPU will require approximately 57,116M of GPU memory and 3-4 minutes.

🔧Installation

The filtered data for each kind of model is at here. Please download it and put it in the data folder.

Build the environement:

conda create -n knowledgecircuit python=3.10
pip install -r requirements.txt

❗️The code may fail under torch 2.x.x. We recommend torch 1.x.x

📚Get the circuit

Just run the following commond:

cd acdc
sh run.sh

Here is an example to run the circuit for the country_capital_city in GPT2-Medium.

MODEL_PATH=/path/to/the/model
KT=factual 
KNOWLEDGE=country_capital_city
NUM_EXAMPLES=20
MODEL_NAME=gpt2-medium

python main.py --task=knowledge \
--zero-ablation \
--threshold=0.01 \
--device=cuda:0 \
--metric=match_nll \
--indices-mode=reverse \
--first-cache-cpu=False \
--second-cache-cpu=False \
--max-num-epochs=10000 \
--specific-knowledge=$KNOWLEDGE \
--num-examples=$NUM_EXAMPLES \
--relation-reverse=False \
--knowledge-type=$KT \
--model-name=$MODEL_NAME \
--model-path=$MODEL_PATH

You would get the results in acdc/factual_results/gpt2-medium and the final_graph.pdf is the computed circuits.

🧐Analyze component

Run the component.ipynb in notebook.

🌻Acknowledgement

We thank for the project of transformer_lens, ACDC and LRE. The code in this work is built on top of these three projects' codes.

🚩Citation

Please cite our repository if you use Knowledge Circuit in your work. Thanks!

@article{DBLP:journals/corr/abs-2405-17969,
  author       = {Yunzhi Yao and
                  Ningyu Zhang and
                  Zekun Xi and
                  Mengru Wang and
                  Ziwen Xu and
                  Shumin Deng and
                  Huajun Chen},
  title        = {Knowledge Circuits in Pretrained Transformers},
  journal      = {CoRR},
  volume       = {abs/2405.17969},
  year         = {2024},
  url          = {https://doi.org/10.48550/arXiv.2405.17969},
  doi          = {10.48550/ARXIV.2405.17969},
  eprinttype    = {arXiv},
  eprint       = {2405.17969},
  timestamp    = {Fri, 21 Jun 2024 22:39:09 +0200},
  biburl       = {https://dblp.org/rec/journals/corr/abs-2405-17969.bib},
  bibsource    = {dblp computer science bibliography, https://dblp.org}
}