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

    Neuronpedia

    INDEX
    Explanations

    numerical IDs and addresses

    oai_token-act-pair · gemini-2.0-flash

    unique identifiers or method names

    np_token-act-pair-logits · gpt-4o-mini

    Code, numbers, identifiers

    np_max-act-logits · gemini-2.0-flash
    New Auto-Interp
    Negative Logits
    oso
    -0.47
    tims
    -0.43
    tib
    -0.43
    mi
    -0.42
    jsdelivr
    -0.42
    キ
    -0.42
    sie
    -0.41
    <b>
    -0.41
     ж
    -0.41
     join
    -0.40
    POSITIVE LOGITS
     raiſ
    0.91
    MethodManager
    0.82
    IntoConstraints
    0.81
    EDEFAULT
    0.80
     חיצוניים
    0.79
     itſelf
    0.78
    ItemBackground
    0.77
     faſt
    0.77
    AndEndTag
    0.76
     uſed
    0.75
    Act Density 5.015%

    No Known Activations