Trustworthy-ML-Lab/CLIP-dissect
[ICLR 23 spotlight] An automatic and efficient tool to describe functionalities of individual neurons in DNNs
This tool helps AI researchers and deep learning practitioners understand what specific parts of their deep neural networks (DNNs) are doing. You provide a trained vision model and a dataset, and it automatically generates plain-language descriptions of what individual neurons in your model represent, helping to interpret its internal workings. It's designed for those who need to explain or debug the black-box nature of their AI models.
No commits in the last 6 months.
Use this if you need to automatically generate human-readable descriptions for the learned functions of individual neurons within your deep vision models.
Not ideal if you are looking to improve model performance, optimize training, or if you are working with non-vision deep learning models.
Stars
63
Forks
16
Language
Jupyter Notebook
License
—
Last pushed
Nov 06, 2023
Commits (30d)
0
Get this data via API
curl "https://pt-edge.onrender.com/api/v1/quality/ml-frameworks/Trustworthy-ML-Lab/CLIP-dissect"
Open to everyone — 100 requests/day, no key needed. Get a free key for 1,000/day.
Higher-rated alternatives
obss/sahi
Framework agnostic sliced/tiled inference + interactive ui + error analysis plots
tensorflow/tcav
Code for the TCAV ML interpretability project
MAIF/shapash
🔅 Shapash: User-friendly Explainability and Interpretability to Develop Reliable and Transparent...
TeamHG-Memex/eli5
A library for debugging/inspecting machine learning classifiers and explaining their predictions
csinva/imodels
Interpretable ML package 🔍 for concise, transparent, and accurate predictive modeling...