openai/automated-interpretability

Issues

• Problem about activation calculation

  #9 opened 2 years ago by Daftstone
  7

• About the 'logprobs' in the response object

  #24 opened a year ago by HamLaertes
  2

• Dataset for neuron activation

  #19 opened a year ago by hanningzhang
  2

• missing data

  #7 opened 2 years ago by WuTheFWasThat
  2

• Text-davinci-003 deprecated

  #40 opened 10 months ago by BiEchi
  1

• Is there a demo that shows this great project?

  #35 opened a year ago by guotong1988
  1

• Can't be installed on windows

  #27 opened a year ago by NicholasDow
  1

• Requires python >=3.9 instead of 3.7 specified in setup.py

  #11 opened a year ago by tuomaso
  3

• Retrying after connection error... (https://api.openai.com/v1/chat/completions)

  #25 opened a year ago by clclclaiggg
  0

• Code for revising explanations

  #23 opened a year ago by Superhzf
  2

• Make a demo script for getting neuron activations using transformer lens?

  #18 opened a year ago by williamrs-openai
  1

• Getting Top Activating Text Excerpts Per Neuron

  #20 opened a year ago by baselmousi
  1

• No Azure credentials were found

  #8 opened a year ago by smile-II
  3

• Make it easy to launch a new colab notebook from the demo notebooks

  #17 opened a year ago by williamrs-openai
  0

• Make easily installable via pip

  #16 opened a year ago by williamrs-openai
  0

• Installing neuron-explainer doesn't seem to work.

  #13 opened a year ago by HoagyC
  2

• About Direction Finding

  #12 opened a year ago by JiruiLiu
  2

• More unified dataset

  #4 opened 2 years ago by shayneoneill
  3

• explain_puzzles.ipynb - You didn't provide an API key

  #6 opened 2 years ago by BenWare144
  1

• Not possible to read from public Azure blobs without authentication

  #1 opened 2 years ago by M-Izadmehr
  1