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

    Neuronpedia

    INDEX
    Explanations

    describing composition or constituents

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    ка
    1.29
    ك
    1.27
    ۰
    0.97
    ০০
    0.95
    ۰۰
    0.93
     to
    0.91
    ید
    0.86
    на
    0.84
    ки
    0.82
     в
    0.80
    POSITIVE LOGITS
    '
    1.17
    \
    0.95
    ’
    0.93
    ?
    0.90
     constituents
    0.90
     components
    0.82
    ers
    0.80
     consisted
    0.80
    ro
    0.78
     Commandant
    0.78
    Act Density 0.108%

    No Known Activations