Skip to main content

No project description provided

Project description

PyPI license

Dialz: A Python Toolkit for Steering Vectors

Docs: https://cardiffnlp.github.io/dialz/

About

Steering vectors allow users to modify activations at inference time to amplify or weaken a 'concept', e.g. honesty or positivity.

Dialz supports a diverse set of tasks, including creating contrastive pair datasets, computing and applying steering vectors, and visualizations.

A basic tutorial can be found here.

Installation

pip install dialz

Check out the full documentation for usage information.

Contributing

Any contributions to improve this project are welcome! Please open an issue or pull request in this repo with any changes you have.

License

This code is released under a MIT license.

Project details


Download files

Download the file for your platform. If you're not sure which to choose, learn more about installing packages.

Source Distribution

dialz-1.1.0.tar.gz (3.9 MB view details)

Uploaded Source

Built Distribution

If you're not sure about the file name format, learn more about wheel file names.

dialz-1.1.0-py3-none-any.whl (4.0 MB view details)

Uploaded Python 3

File details

Details for the file dialz-1.1.0.tar.gz.

File metadata

  • Download URL: dialz-1.1.0.tar.gz
  • Upload date:
  • Size: 3.9 MB
  • Tags: Source
  • Uploaded using Trusted Publishing? No
  • Uploaded via: twine/6.0.1 CPython/3.12.10

File hashes

Hashes for dialz-1.1.0.tar.gz
Algorithm Hash digest
SHA256 c437e652ea7f3004f353a7babeb934177bfee65926ed5c35addff1f23805d5bc
MD5 1a34f824b20a3331285e64e47e1097c5
BLAKE2b-256 d3b802ad503a021c8572df31ca551e593fae00d50b74a0b6eadf8446d655d5b0

See more details on using hashes here.

File details

Details for the file dialz-1.1.0-py3-none-any.whl.

File metadata

  • Download URL: dialz-1.1.0-py3-none-any.whl
  • Upload date:
  • Size: 4.0 MB
  • Tags: Python 3
  • Uploaded using Trusted Publishing? No
  • Uploaded via: twine/6.0.1 CPython/3.12.10

File hashes

Hashes for dialz-1.1.0-py3-none-any.whl
Algorithm Hash digest
SHA256 a0b888dee3dab910c646b2b16c71424140bc425e091f84b92d8c90d4b4445453
MD5 c7306528fd552bb525b0af50cc1e6e93
BLAKE2b-256 b58ad5606a02dcd06efe22100028b2a4d1b54acc2498ba0218d01375189c2b9b

See more details on using hashes here.

Supported by

AWS Cloud computing and Security Sponsor Datadog Monitoring Depot Continuous Integration Fastly CDN Google Download Analytics Pingdom Monitoring Sentry Error logging StatusPage Status page