phvv-me/frame-representation-hypothesis
Official Repository for Frame Representation Hypothesis paper
This framework helps AI researchers and developers understand and control Large Language Models (LLMs). It takes WordNet data to generate concepts, which can then be used to guide an LLM's text generation or to expose potential biases and vulnerabilities within the model. The output helps ensure LLMs produce more predictable and reliable text, making it useful for those who fine-tune, evaluate, or deploy LLMs.
No commits in the last 6 months.
Use this if you need to gain deeper insights into how LLMs form their responses and want a method to steer their output toward specific concepts or identify undesirable behaviors.
Not ideal if you are an end-user simply looking to interact with a pre-trained LLM for general tasks, as this is a tool for LLM analysis and control, not direct user interaction.
Stars
8
Forks
3
Language
Jupyter Notebook
License
MIT
Category
Last pushed
Sep 26, 2025
Commits (30d)
0
Get this data via API
curl "https://pt-edge.onrender.com/api/v1/quality/transformers/phvv-me/frame-representation-hypothesis"
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