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

    Neuronpedia

    INDEX
    Explanations

    references to hallways and similar structural elements

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

    hallway, recuperate, kebabs, revenge

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
     esfera
    -0.40
     enfans
    -0.40
     πυ
    -0.37
    <bos>
    -0.34
    str
    -0.34
     Potencia
    -0.34
     cápsulas
    -0.33
     Escribe
    -0.32
    "__
    -0.32
     dignité
    -0.32
    POSITIVE LOGITS
     Keb
    0.91
    Keb
    0.84
     recuper
    0.76
     keb
    0.74
     Eup
    0.69
    windowFixed
    0.66
    MarshalTo
    0.65
    enumi
    0.64
     חיצוניים
    0.64
     GenerationType
    0.63
    Act Density 0.003%

    No Known Activations