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

    Neuronpedia

    INDEX
    Explanations

    programming-related syntax and functions

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

    code constructs like `if`, `replace`, `assert`

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    المناصب
    -0.70
    +:+
    -0.70
     فريبيس
    -0.65
     تضيفلها
    -0.62
    ValueStyle
    -0.58
    AccessorTable
    -0.58
    Tembelea
    -0.58
    Portale
    -0.57
    Portail
    -0.54
    SPJ
    -0.51
    POSITIVE LOGITS
     Administrativna
    0.59
     podjela
    0.43
     ennemi
    0.41
     royaume
    0.38
     Италијани
    0.38
     hiér
    0.38
     lendemain
    0.38
     schade
    0.38
     ucapnya
    0.36
     remplissage
    0.35
    Act Density 0.025%

    No Known Activations