© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it
    Neuronpedia
    Jacobian LensNEW
    Natural Language
    Autoencoders
    MAY 26    Assistant AxisJAN 26Circuit TracerAPR 26SteerSAE EvalsExportsinterp-engineNEW Community BlogPrivacy & TermsContact
    EXPLANATION TYPE
    oai_token-act-pair
    Description
    OpenAI's Automated Interpretability from paper "Language models can explain neurons in language models". Modified by Johnny Lin to add new models/context windows.
    Author
    OpenAI
    URL
    https://github.com/hijohnnylin/automated-interpretability
    Settings
    Default prompts from the main branch, strategy TokenActivationPair. Uses top 10 deduplicated activations.
    Recent Explanations
    mathematical formulas and equations.
    gemini-2.5-flash-lite
    0
    Neuronpedia logo
    QWEN3-32B
    52
    INDEX 24727
    disaster-related terms.
    gemini-2.5-flash-lite
    0
    Neuronpedia logo
    QWEN3-32B
    31
    INDEX 8906
    math or logic problems.
    gemini-2.5-flash-lite
    0
    Neuronpedia logo
    QWEN3-32B
    63
    INDEX 14880
    mentions of the political party "Die Linke".
    gemini-2.5-flash-lite
    0
    Neuronpedia logo
    QWEN3-32B
    32
    INDEX 18369
    words and phrases related to religious ceremonies and rituals.
    gemini-2.5-flash-lite
    0
    Neuronpedia logo
    QWEN3-32B
    14
    INDEX 2471
    mentions of media.
    gemini-2.5-flash-lite
    0
    Neuronpedia logo
    QWEN3-32B
    19
    INDEX 2195
    mathematical equations.
    gemini-2.5-flash-lite
    0
    Neuronpedia logo
    QWEN3-32B
    51
    INDEX 16157
    mentions of problems related to scheduling and resource allocation.
    gemini-2.5-flash-lite
    0
    Neuronpedia logo
    QWEN3-32B
    24
    INDEX 13489
    numbers.
    gemini-2.5-flash-lite
    0
    Neuronpedia logo
    QWEN3-32B
    18
    INDEX 6243
    phrases related to industry and production.
    gemini-2.5-flash-lite
    0
    Neuronpedia logo
    QWEN3-32B
    47
    INDEX 10688
    words related to government and education.
    gemini-2.5-flash-lite
    0
    Neuronpedia logo
    QWEN3-32B
    62
    INDEX 24399
    technical or scientific jargon.
    gemini-2.5-flash-lite
    0
    Neuronpedia logo
    QWEN3-32B
    51
    INDEX 25285
    phrases involving commands or instructions.
    gemini-2.5-flash-lite
    0
    Neuronpedia logo
    QWEN3-32B
    51
    INDEX 17360
    academic papers and research.
    gemini-2.5-flash-lite
    0
    Neuronpedia logo
    QWEN3-32B
    9
    INDEX 5012
    words related to gravity and falling.
    gemini-2.5-flash-lite
    0
    Neuronpedia logo
    QWEN3-32B
    13
    INDEX 17480
    code snippets.
    gemini-2.5-flash-lite
    0
    Neuronpedia logo
    QWEN3-32B
    37
    INDEX 18969
    questions and answers, often in a test or quiz format.
    gemini-2.5-flash-lite
    0
    Neuronpedia logo
    QWEN3-32B
    24
    INDEX 9041
    words related to potential and success.
    gemini-2.5-flash-lite
    0
    Neuronpedia logo
    QWEN3-32B
    23
    INDEX 20654
    phrases related to real estate transactions.
    gemini-2.5-flash-lite
    0
    Neuronpedia logo
    QWEN3-32B
    38
    INDEX 7719
    programming code.
    gemini-2.5-flash-lite
    0
    Neuronpedia logo
    QWEN3-32B
    58
    INDEX 23233