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

    Neuronpedia

    INDEX
    Explanations

    draft documents

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
     AND
    0.84
     fikir
    0.82
     WITH
    0.80
     scepticism
    0.77
    も含
    0.77
    在
    0.76
     aş
    0.73
    ในช่วง
    0.73
     TYPES
    0.73
     ו
    0.72
    POSITIVE LOGITS
    at
    1.06
    s
    1.00
    on
    0.92
    an
    0.85
    atán
    0.73
    ig
    0.73
    o
    0.71
    y
    0.70
    et
    0.66
    sning
    0.65
    Act Density 0.005%

    No Known Activations