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

    Neuronpedia

    INDEX
    Explanations

    specific phrases that denote the beginning of a new section or topic in a text

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Negative Logits
    Marshaller
    -0.63
    TestBed
    -0.60
    bitField
    -0.57
    $")
    -0.56
    %")
    -0.56
    TagMode
    -0.55
     Parcelable
    -0.53
     fevere
    -0.52
    !")
    -0.51
    $",
    -0.51
    POSITIVE LOGITS
     racism
    0.74
     Racism
    0.73
     racist
    0.70
     gender
    0.70
     racial
    0.69
    Racism
    0.66
    gender
    0.65
     racially
    0.64
    racism
    0.63
     Racial
    0.59
    Act Density 0.475%

    No Known Activations