Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    INDEX
    Explanations

    I

    np_max-act · gemini-2.0-flash

    first-person self-referential statements (the author saying what they did, need, or want).

    oai_token-act-pair · gpt-5Triggered by @vetterc0
    New Auto-Interp
    Negative Logits
    594
    -0.07
     massage
    -0.06
    45
    -0.06
    024
    -0.06
     разм
    -0.06
    969
    -0.06
     table
    -0.06
     زاده
    -0.06
    施
    -0.06
     Mohammed
    -0.06
    POSITIVE LOGITS
     I
    0.07
    !I
    0.07
     |[
    0.06
    -library
    0.06
     gql
    0.06
    итуа
    0.06
    I
    0.06
    /i
    0.06
    。我
    0.06
     якому
    0.06
    Act Density 0.044%

    No Known Activations