Training and Analyzing Sparse Autoencoders (SAEs)

Navigation

Verified details

These details have been verified by PyPI
Maintainers
Avatar for chanind from gravatar.com chanind Avatar for jbloomaus from gravatar.com jbloomaus

Unverified details

These details have not been verified by PyPI
Meta
  • Author: Joseph Bloom
  • Requires: Python <4.0, >=3.10
Classifiers
Report project as malware

Project description

Screenshot 2024-03-21 at 3 08 28 pm

SAE Lens

PyPI License: MIT build Deploy Docs codecov

SAELens exists to help researchers:

  • Train sparse autoencoders.
  • Analyse sparse autoencoders / research mechanistic interpretability.
  • Generate insights which make it easier to create safe and aligned AI systems.

Please refer to the documentation for information on how to:

  • Download and Analyse pre-trained sparse autoencoders.
  • Train your own sparse autoencoders.
  • Generate feature dashboards with the SAE-Vis Library.

SAE Lens is the result of many contributors working collectively to improve humanities understanding of neural networks, many of whom are motivated by a desire to safeguard humanity from risks posed by artificial intelligence.

This library is maintained by Joseph Bloom and David Channin.

Join the Slack!

Feel free to join the Open Source Mechanistic Interpretability Slack for support!

Citations and References

Research:

Reference Implementations:

Project details

Verified details

These details have been verified by PyPI
Maintainers
Avatar for chanind from gravatar.com chanind Avatar for jbloomaus from gravatar.com jbloomaus

Unverified details

These details have not been verified by PyPI
Meta
  • Author: Joseph Bloom
  • Requires: Python <4.0, >=3.10
Classifiers

Release history Release notifications | RSS feed

6.38.0

6.37.7

6.37.6

6.37.5

6.37.4

6.37.3

6.37.2

6.37.1

6.37.0

6.36.2

6.36.1

6.36.0

6.35.0

6.34.2

6.34.1

6.34.0

6.33.0

6.32.1

6.32.0

6.31.0

6.30.2

6.30.1

6.30.0

6.29.1

6.29.0

6.28.2

6.28.1

6.28.0

6.27.3

6.27.2

6.27.1

6.27.0

6.26.2

6.26.1

6.26.0

6.25.1

6.25.0

6.24.2

6.24.1

6.24.0

6.23.0

6.22.3

6.22.2

6.22.1

6.22.0

6.21.0

6.20.1

6.18.0

6.17.0

6.16.3

6.16.2

6.16.0

6.15.0

6.14.2

6.14.1

6.14.0

6.13.1

6.13.0

6.12.3

6.12.2

6.12.1

6.12.0

6.11.1

6.11.0

6.10.0

6.9.1

6.9.0

6.8.0

6.7.0

6.6.5

6.6.4

6.6.3

6.6.2

6.6.1

6.6.0

6.5.3

6.5.2

6.5.1

6.5.0

6.4.0

6.3.1

6.3.0

6.2.0

6.1.0

6.0.0

6.0.0rc5 pre-release

6.0.0rc4 pre-release

6.0.0rc3 pre-release

6.0.0rc2 pre-release

6.0.0rc1 pre-release

5.11.0

5.10.7

5.10.6

5.10.5

5.10.4

5.10.3

5.10.2

5.10.1

5.10.0

5.9.2

5.9.1

5.9.0

5.8.1

5.8.0

5.7.1

5.7.0

5.6.1

5.6.0

5.5.2

5.5.1

5.5.0

5.4.2

5.4.1

5.4.0

5.3.4

5.3.3

5.3.2

5.3.1

5.3.0

5.2.1

5.2.0

5.1.0

5.0.0

4.4.5

4.4.4

4.4.3

4.4.2

4.4.1

4.4.0

4.3.5

4.3.4

4.3.3

4.3.2

4.3.1

4.3.0

4.2.0

4.1.1

4.1.0

4.0.10

4.0.9

4.0.8

4.0.7

4.0.6

4.0.5

4.0.4

4.0.3

4.0.2

4.0.1

4.0.0

3.23.4

3.23.3

3.23.2

3.23.1

3.23.0

3.22.2

3.22.1

3.22.0

3.21.1

3.21.0

3.20.5

3.20.4

3.20.3

3.20.2

3.20.1

3.20.0

3.19.4

3.19.3

3.19.2

3.19.1

3.19.0

3.18.2

3.18.1

3.18.0

3.17.1

3.17.0

3.16.0

3.15.0

3.14.0

3.13.1

3.13.0

3.12.5

3.12.4

3.12.3

3.12.2

3.12.1

3.12.0

3.11.2

3.11.1

3.11.0

3.10.0

3.9.2

3.9.1

3.9.0

3.8.0

3.7.0

3.6.0

3.5.0

3.4.1

3.4.0

3.3.0

3.2.3

3.2.2

3.2.1

3.2.0

3.1.1

3.1.0

3.0.0

2.1.3

2.1.2

2.1.1

2.1.0

2.0.0

1.8.0

1.7.0

1.6.1

1.6.0

This version

1.5.0

1.4.0

1.3.0

1.2.0

1.1.0

1.0.0

0.7.0

0.6.0

0.5.1

0.5.0

0.4.0

0.3.0

0.2.2

0.2.1

0.2.0

0.1.0

Download files

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

Source Distribution

sae_lens-1.5.0.tar.gz (60.1 kB view details)

Uploaded Source

Built Distribution

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

sae_lens-1.5.0-py3-none-any.whl (71.3 kB view details)

Uploaded Python 3

File details

Details for the file sae_lens-1.5.0.tar.gz.

File metadata

  • Download URL: sae_lens-1.5.0.tar.gz
  • Upload date:
  • Size: 60.1 kB
  • Tags: Source
  • Uploaded using Trusted Publishing? Yes
  • Uploaded via: twine/5.0.0 CPython/3.12.3

File hashes

Hashes for sae_lens-1.5.0.tar.gz
Algorithm Hash digest
SHA256 b583ccf004eee11b2ae764e15c163103678a3c9413e1f6a14f9d82b56442f672
MD5 ee22659c52f449266d15d1b4fe0f4aae
BLAKE2b-256 395275a473a9be1d96667a930ed103f4eddd47f244ee70a6cceb237909f5ceaa

See more details on using hashes here.

File details

Details for the file sae_lens-1.5.0-py3-none-any.whl.

File metadata

  • Download URL: sae_lens-1.5.0-py3-none-any.whl
  • Upload date:
  • Size: 71.3 kB
  • Tags: Python 3
  • Uploaded using Trusted Publishing? Yes
  • Uploaded via: twine/5.0.0 CPython/3.12.3

File hashes

Hashes for sae_lens-1.5.0-py3-none-any.whl
Algorithm Hash digest
SHA256 f82cf7bed01946ab3769f09d29af11e20d25e96184e680e7fbd5dcf69906389b
MD5 5ae8289f081e882ac6ebe463d4f78338
BLAKE2b-256 8973bdaf5ca9efb7bd03071d6a78559c0a0c1c25a2d2a1bc2cc7cc0b8e830cb1

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