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

    Neuronpedia

    INDEX
    Explanations

    concepts related to treatment, behavior, and ethics in interpersonal and societal contexts

    oai_token-act-pair · gpt-4o-miniTriggered by @bot

    Describes actions or behaviors, often negative

    np_acts-logits-general · gemini-2.0-flash

    how things are done or treated

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
     يتيمه
    -0.69
    ⓧ
    -0.68
    AntiForgeryToken
    -0.61
    fillType
    -0.61
    IndexPath
    -0.61
    IsContent
    -0.59
    IsMutable
    -0.56
    Naissance
    -0.53
    yorsunuz
    -0.53
    ungsbedingungen
    -0.52
    POSITIVE LOGITS
     differently
    1.46
     accordingly
    0.94
     according
    0.94
     incorrectly
    0.93
     diffé
    0.85
    differ
    0.79
     how
    0.78
     wrong
    0.77
     correctly
    0.76
     wrongly
    0.73
    Act Density 0.401%

    No Known Activations