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

    Neuronpedia

    INDEX
    Explanations

    monospace font names

    np_max-act-logits · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
     rookies
    0.57
     bunnies
    0.56
     ductile
    0.55
     pigeons
    0.54
     flatness
    0.54
     shoulder
    0.53
     foreigners
    0.53
     fiabilité
    0.52
     shoulders
    0.52
     lions
    0.52
    POSITIVE LOGITS
    organiz
    0.61
    可以通过
    0.57
    eminent
    0.57
    Organiz
    0.57
    organization
    0.56
    专家
    0.55
    という意味
    0.55
     организа
    0.54
    oncesto
    0.54
     الغير
    0.54
    Act Density 0.001%

    No Known Activations