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

    Neuronpedia

    INDEX
    Explanations

    mixed numerics and non-english characters

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    tır
    0.94
    through
    0.77
    ᅯ
    0.75
    ammation
    0.73
    lerinin
    0.73
    cumin
    0.72
    ség
    0.71
    ούς
    0.69
    ｉ
    0.69
    onbury
    0.68
    POSITIVE LOGITS
    에
    0.92
    лов
    0.86
    л
    0.83
    ק
    0.82
     slou
    0.80
    к
    0.79
    म
    0.79
    Φ
    0.77
     satisfactorily
    0.76
     Solidity
    0.75
    Act Density 0.495%

    No Known Activations