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

    Neuronpedia

    INDEX
    Explanations

    privileged access

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
     لیګ
    0.80
     haría
    0.71
    на
    0.70
    ?,?,
    0.70
     сча
    0.69
    いた
    0.68
     али
    0.68
     доб
    0.66
     нагре
    0.66
     хотелось
    0.65
    POSITIVE LOGITS
    {
    0.91
     an
    0.82
    ]
    0.81
    -
    0.81
    0.80
    an
    0.76
    ü
    0.72
    (
    0.69
    n
    0.68
    )
    0.67
    Act Density 0.000%

    No Known Activations