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

    Neuronpedia

    INDEX
    Explanations

    thunder and storms

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    و
    1.90
    p
    1.45
    ו
    1.44
    j
    1.35
    are
    1.25
    ку
    1.19
    m
    1.15
    ो
    1.14
    ран
    1.12
     часу
    1.07
    POSITIVE LOGITS
     thunder
    1.42
    Thunder
    1.23
    ה
    1.09
     Thunder
    1.08
    h
    0.98
     thunderstorms
    0.96
    雰
    0.95
    つなが
    0.93
    ።
    0.92
     thunderstorm
    0.91
    Act Density 0.002%

    No Known Activations