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

    Neuronpedia

    INDEX
    Explanations

    collective communication or data collection

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
     as
    1.31
    ра
    1.23
     are
    1.15
    ك
    1.14
    ,
    1.07
    د
    1.05
    å
    0.99
    ian
    0.94
    𝙤
    0.93
    一個
    0.92
    POSITIVE LOGITS
    m
    1.14
    <0x0D>
    1.09
    ০০০
    1.01
    ní
    0.99
    l
    0.97
    mama
    0.96
    tio
    0.93
    rda
    0.88
    tay
    0.88
    thed
    0.88
    Act Density 0.205%

    No Known Activations