matlok-ai/bampe-weights
This repository is for profiling, extracting, visualizing and reusing generative AI weights to hopefully build more accurate AI models and audit/scan weights at rest to identify knowledge domains for risk(s).
This tool helps AI researchers and model developers understand how large generative AI models learn by visualizing their internal 'weights' as 3D shapes. It takes existing AI model files and outputs visual representations (like 3D meshes or animations) that reveal how knowledge is structured within the model. This is for anyone building, auditing, or researching the underlying mechanisms of large language models.
No commits in the last 6 months. Available on PyPI.
Use this if you need to visually inspect and analyze the internal structure and learning patterns of generative AI models, similar to how brain scans are used in neuroscience.
Not ideal if you are looking for a tool to directly train or fine-tune AI models, or if you don't have a need for deep, visual insight into model weights.
Stars
9
Forks
—
Language
Python
License
—
Category
Last pushed
Dec 18, 2023
Commits (30d)
0
Dependencies
9
Get this data via API
curl "https://pt-edge.onrender.com/api/v1/quality/transformers/matlok-ai/bampe-weights"
Open to everyone — 100 requests/day, no key needed. Get a free key for 1,000/day.
Higher-rated alternatives
galilai-group/stable-pretraining
Reliable, minimal and scalable library for pretraining foundation and world models
CognitiveAISystems/MAPF-GPT
[AAAI-2025] This repository contains MAPF-GPT, a deep learning-based model for solving MAPF...
UKPLab/gpl
Powerful unsupervised domain adaptation method for dense retrieval. Requires only unlabeled...
larslorch/avici
Amortized Inference for Causal Structure Learning, NeurIPS 2022
svdrecbd/mhc-mlx
MLX + Metal implementation of mHC: Manifold-Constrained Hyper-Connections by DeepSeek-AI.