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

    Neuronpedia

    INDEX
    Explanations

    considered floppy flat

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    Kot
    0.41
     patrón
    0.39
    Cô
    0.38
     wirelessly
    0.38
    odora
    0.36
     scary
    0.36
    Фор
    0.36
    Kod
    0.36
    ⼊
    0.36
    смотря
    0.35
    POSITIVE LOGITS
     Bernie
    0.42
     trivial
    0.41
     Oscar
    0.40
    賤
    0.38
    冈
    0.38
     वेबसाइट
    0.38
     DPS
    0.38
     الدكتور
    0.38
     Ram
    0.37
     Pope
    0.37
    Act Density 0.002%

    No Known Activations