Commits

Breadcrumbs

History for

onpatch-2

User selector

All users

Datepicker

All time

Commit History

Commits on Dec 29, 2024

• Update modelings_intervenable_gemma2.py

  Show description for 3a6c448
  jasonlim131

  authored
  3a6c448

  Copy full SHA for 3a6c448

Commits on Oct 8, 2024

• Add Gemma2 models.

  explanare

  committed
  b7addfc

  Copy full SHA for b7addfc

Commits on Sep 24, 2024

• [Minor] Adding interchange intervention for SAEs

  frankaging

  committed
  21d0c7b

  Copy full SHA for 21d0c7b

Commits on Sep 6, 2024

• [Minor] Fix intervention return objects

  frankaging

  committed
  95b74ee

  Copy full SHA for 95b74ee

Commits on Aug 24, 2024

• update

  frankaging

  committed
  4b14b6e

  Copy full SHA for 4b14b6e

• update

  frankaging

  committed
  3c407cf

  Copy full SHA for 3c407cf

• update

  frankaging

  committed
  95e5122

  Copy full SHA for 95e5122

• update

  frankaging

  committed
  86a41ca

  Copy full SHA for 86a41ca

• [Minor] Update dependency

  frankaging

  committed
  9b7a91b

  Copy full SHA for 9b7a91b

Commits on Aug 11, 2024

• Merge pull request stanfordnlp#182 from stanfordnlp/olmo

  Show description for 4f70e10
  frankaging

  authored
  4f70e10

  Copy full SHA for 4f70e10

Commits on Aug 10, 2024

• Add OLMo models.

  explanare

  committed
  19ee386

  Copy full SHA for 19ee386

Commits on Aug 5, 2024

• [Minor] Updating the import to bypass blip in case of failure

  frankaging

  committed
  c46a680

  Copy full SHA for c46a680

Commits on Jul 29, 2024

• add support for dict outputs (select the first value)

  leogagnon

  committed
  25a3473

  Copy full SHA for 25a3473

Commits on Jul 24, 2024

• Merge branch 'main' of github.com:stanfordnlp/pyvene into zen/as_adaptor

  frankaging

  committed
  3b35f9b

  Copy full SHA for 3b35f9b

• Support adaptor-like tuning

  frankaging

  committed
  3b8a3e8

  Copy full SHA for 3b8a3e8

Commits on Jul 22, 2024

• [Minor] Update use_cache intervenable_base.py

  frankaging

  authored
  312c38f

  Copy full SHA for 312c38f

Commits on Jul 18, 2024

• [Minor] Fix the base model use_cache arg passing

  Show description for 1285300
  frankaging

  authored
  1285300

  Copy full SHA for 1285300

Commits on Jul 17, 2024

• Update intervenable_base.py

  frankaging

  authored
  8e75602

  Copy full SHA for 8e75602

• [Minor] Fix use_cache flag propagation

  frankaging

  authored
  adf1116

  Copy full SHA for adf1116

Commits on Jul 16, 2024

• device of unit_locations should follow tensor_input

  aryopg

  authored
  b8d9f08

  Copy full SHA for b8d9f08

Commits on Jul 13, 2024

• api docs

  aryamanarora

  committed
  e1e32aa

  Copy full SHA for e1e32aa

• set up documentation the right way, index page ready

  aryamanarora

  committed
  e0cf9c9

  Copy full SHA for e0cf9c9

Commits on Jul 12, 2024

• Merge remote-tracking branch 'origin/main'

  aryamanarora

  committed
  185a412

  Copy full SHA for 185a412

• don't save model params by default

  aryamanarora

  committed
  0d92580

  Copy full SHA for 0d92580

• Merge branch 'stanfordnlp:main' into main

  aryamanarora

  authored
  6b7dab2

  Copy full SHA for 6b7dab2

Commits on Jul 5, 2024

• add in basic support of activation collection and interchange intervention

  frankaging

  committed
  bf07a70

  Copy full SHA for bf07a70

Commits on Jul 4, 2024

• Merge branch 'main' of github.com:stanfordnlp/pyvene into zen/ndif

  frankaging

  committed
  d41388f

  Copy full SHA for d41388f

• partial ndif support

  frankaging

  committed
  39741be

  Copy full SHA for 39741be

• Support loading/saving of interventions with autoencoder.

  explanare

  committed
  4bf3ef5

  Copy full SHA for 4bf3ef5

Commits on Jul 3, 2024

• Add Sparse Autoencoder.

  explanare

  committed
  61ebad9

  Copy full SHA for 61ebad9

Commits on Jun 17, 2024

• Merge branch 'stanfordnlp:main' into main

  aryamanarora

  authored
  3fccdaa

  Copy full SHA for 3fccdaa

Commits on Jun 12, 2024

• Merge pull request stanfordnlp#159 from Bakser/fix_head_bug

  Show description for 3da8474
  frankaging

  authored
  3da8474

  Copy full SHA for 3da8474

Commits on Jun 11, 2024

• make llama accept customized config for running test fastly

  Bakser

  committed
  81836d7

  Copy full SHA for 81836d7

Commits on Jun 3, 2024

• [Minor] Keep existing behavior as the default

  frankaging

  committed
  0983f40

  Copy full SHA for 0983f40

• [Minor] Remove ortho init for DAS

  frankaging

  committed
  48024f8

  Copy full SHA for 48024f8

Pagination

Previous

Next