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

    Neuronpedia

    INDEX
    Explanations

    longer than, long, down, respond in, drop an

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    ে
    1.90
    ০
    1.76
    e
    1.75
    ei
    1.55
    o
    1.55
    ej
    1.42
    و
    1.40
    eh
    1.40
    oos
    1.35
    h
    1.33
    POSITIVE LOGITS
    leri
    1.36
    न्ग
    1.31
    astien
    1.23
    ットン
    1.21
    ättning
    1.20
    ünüz
    1.18
    lerle
    1.18
     оригіналу
    1.14
    |,
    1.13
    lerinin
    1.13
    Act Density 0.002%

    No Known Activations