Trustworthy-ML-Lab/Describe-and-Dissect
[TMLR 25] An automated method for explaining complex neuron behaviors in deep vision models using large language models
When working with deep learning models for image analysis, it's often hard to understand why the model makes certain decisions. This tool helps you decipher the 'black box' by automatically describing what individual parts (neurons) of a deep vision model are actually doing. It takes your vision model and a dataset, then generates clear, natural language explanations of specific neuron behaviors, useful for AI researchers and practitioners.
No commits in the last 6 months.
Use this if you need to understand the internal workings of a deep vision model and get human-readable explanations of what its neurons are detecting or representing.
Not ideal if you are looking for a simple API to integrate into an existing application or if you need to interpret non-vision-based deep learning models.
Stars
10
Forks
2
Language
Jupyter Notebook
License
—
Category
Last pushed
Feb 20, 2025
Commits (30d)
0
Get this data via API
curl "https://pt-edge.onrender.com/api/v1/quality/transformers/Trustworthy-ML-Lab/Describe-and-Dissect"
Open to everyone — 100 requests/day, no key needed. Get a free key for 1,000/day.
Higher-rated alternatives
MadryLab/context-cite
Attribute (or cite) statements generated by LLMs back to in-context information.
microsoft/augmented-interpretable-models
Interpretable and efficient predictors using pre-trained language models. Scikit-learn compatible.
Trustworthy-ML-Lab/CB-LLMs
[ICLR 25] A novel framework for building intrinsically interpretable LLMs with...
poloclub/LLM-Attributor
LLM Attributor: Attribute LLM's Generated Text to Training Data
THUDM/LongCite
LongCite: Enabling LLMs to Generate Fine-grained Citations in Long-context QA