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

    Neuronpedia

    INDEX
    Explanations

    need help

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
     vistas
    1.35
    𝘾
    1.15
     irre
    1.11
     sonrası
    1.08
    ϵ
    1.07
    פול
    1.05
     redire
    1.05
    𝙉
    1.05
    적인
    1.05
     travers
    1.05
    POSITIVE LOGITS
    mail
    1.07
    uggest
    1.05
    \,
    1.01
    עות
    0.98
     banget
    0.98
     umano
    0.95
    bass
    0.95
     neck
    0.94
    よく
    0.93
    educated
    0.93
    Act Density 0.002%

    No Known Activations