Saved in:
Bibliographic Details
Main Authors: Drechsel, Jonathan, Herbold, Steffen
Format: Preprint
Published: 2026
Subjects:
Online Access:https://arxiv.org/abs/2602.23993
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866910035233210368
author Drechsel, Jonathan
Herbold, Steffen
author_facet Drechsel, Jonathan
Herbold, Steffen
contents We present gradiend, an open-source Python package that operationalizes the GRADIEND method for learning feature directions from factual-counterfactual MLM and CLM gradients in language models. The package provides a unified workflow for feature-related data creation, training, evaluation, visualization, persistent model rewriting via controlled weight updates, and multi-feature comparison. We demonstrate GRADIEND on an English pronoun paradigm and on a large-scale feature comparison that reproduces prior use cases.
format Preprint
id arxiv_https___arxiv_org_abs_2602_23993
institution arXiv
publishDate 2026
record_format arxiv
spellingShingle The GRADIEND Python Package: An End-to-End System for Gradient-Based Feature Learning
Drechsel, Jonathan
Herbold, Steffen
Computation and Language
We present gradiend, an open-source Python package that operationalizes the GRADIEND method for learning feature directions from factual-counterfactual MLM and CLM gradients in language models. The package provides a unified workflow for feature-related data creation, training, evaluation, visualization, persistent model rewriting via controlled weight updates, and multi-feature comparison. We demonstrate GRADIEND on an English pronoun paradigm and on a large-scale feature comparison that reproduces prior use cases.
title The GRADIEND Python Package: An End-to-End System for Gradient-Based Feature Learning
topic Computation and Language
url https://arxiv.org/abs/2602.23993